POPULARITY
Sponsors CacheFly Panel Charles Max Wood Shawn Clabough Caleb Wells Joined by Special Guest: Mads Kristensen Summary Mads Kristensen, from Microsoft, joins the panel to discuss building extensions in visual studio. Mads shares his story of getting into building extensions. Mads gives recommendations when getting started building an extension. The panel discusses what are important things to understand when you are new to building extensions. Charles Max Wood asks about the capabilities of extensions. Mads gives best practices for building extensions. Why writing extensions is getting easier and the place of extension in Microsoft is considered by the panel. The panel discusses the versions of visual studios and how they affect extensions. Mads shares what he is working on right now. Links 236 JSJ Interview with Mads Kristensen from Microsoft Ignite Getting started writing Visual Studio extensions Visual Studio 2015 Extensibility Checklist for writing great Visual Studio extensions https://www.facebook.com/Adventures-in-NET-373059030062837/ https://twitter.com/adventures_net Picks Charles Max Wood: Azure Functions Stranger Things Season 3 Shawn Clabough: https://slcnet.tech/ https://www.dotvvm.com/ Caleb Wells: Xbox Live for PC Mads Kristensen: https://hubitat.com/
Sponsors CacheFly Panel Charles Max Wood Shawn Clabough Caleb Wells Joined by Special Guest: Mads Kristensen Summary Mads Kristensen, from Microsoft, joins the panel to discuss building extensions in visual studio. Mads shares his story of getting into building extensions. Mads gives recommendations when getting started building an extension. The panel discusses what are important things to understand when you are new to building extensions. Charles Max Wood asks about the capabilities of extensions. Mads gives best practices for building extensions. Why writing extensions is getting easier and the place of extension in Microsoft is considered by the panel. The panel discusses the versions of visual studios and how they affect extensions. Mads shares what he is working on right now. Links 236 JSJ Interview with Mads Kristensen from Microsoft Ignite Getting started writing Visual Studio extensions Visual Studio 2015 Extensibility Checklist for writing great Visual Studio extensions https://www.facebook.com/Adventures-in-NET-373059030062837/ https://twitter.com/adventures_net Picks Charles Max Wood: Azure Functions Stranger Things Season 3 Shawn Clabough: https://slcnet.tech/ https://www.dotvvm.com/ Caleb Wells: Xbox Live for PC Mads Kristensen: https://hubitat.com/
פודקאסט מספר 372 של רברס עם פלטפורמה - אורי ורן מארחים בכרכור ערב החמסין את יאיר הרשקו מחברת Zadara לשיחה על אחסון, עננים ואנומליות בלוח השנה העברי.הפרק בחסות Dropbox, שהם גם ספונסרים של Reversim Summit 2019 (כבר יש וידאו, ותמונות וסיכום…)יאיר, נשוי + 3 מאלוני אבא, בתחום האחסון כבר ~20 שנה - תואר במדעי המחשב בטכניון ומשם התחלה בעולם האחסון, תחילה בחברה בשם StoreAge (שעסקה ב - Storage virtualization). ב-2006 החברה נרכשה ע”י LSI (שהתחילה בתוך הטכניון ומשם עברה לעיירת הקייט הציורית נשר). ויאיר המשיך כמפתח-ראש צוות-מנהל פיתוח.ב-2010 החטיבה נמכרה ל- NetApp וכולם הלכו הביתה (הלינק היום בכלל הולך לאתר של Broadcom). יאיר, יחד עם המנכ”ל וה-CTO הלכו להקים משהו חדש. אז Zadara קמה בתאריך העברי ז’ באדר א’ , ומשם הגיע השם Z Adar A . . . זהו תאריך חשוב במסורת היהודית - תאריך הולדתו ופטירתו של משה רבנו. לא מזיק - לאיש היה ניסיון עם Tablets (ובאופן כללי רעיון למגוון רחב של שמות לחברות).חוץ מזה שזה יכול לקרות רק בשנה מעוברת, וגם אז רק בשני, רביעי, חמישי או שבת . . .השנה היא כבר 2011 - יש עננים בחוץ, Cluodera כבר תפוס, CloudAge פחות תופס . . . Zadara it is.אז מה חסר עכשיו בעולם? נראה שיש פער בין יכולות האחסון שקיימות ב - Data Centers של הארגון הטיפוסי (מבוססי IBM ו-EMC וכו’), לבין מה שקיים בענן הציבורי (Public Cloud), שב-2011 זה כמעט רק AWS - ובעיקר S3 ו EBS. אפילו S4 עוד לא היה.גם EBS היה עדיין מאוד בסיסי - Block בלבד, אם אפשרות להגדיר Capacity ודי זהו.נראה היה שה-Storage מהווה את אחד החסמים העיקריים של ארגונים לעבוד לענן.מערכות האחסון בענן לא היו ברמה הנדרשת, וארגונים שהיו רגילים לביצועים, קיבולת ויכולת התאמות (Customization) ברמה מאוד גבוהה (Lift & Shift, שמצריך שכבת חומרה עם אותן יכולות).מי שכתב את האפליקציה שלו מחדש באותו תקופה (מי אמר Netflix?) כבר כתב אותן לתוך עולם ה-S3 ו - Object Storage, כבר עם Interface אחר - או מבוסס מכונות “commodity” עם אחסון כללי כלשהו - הנחה בעייתית, שצריכה להכיר בזה שמכונות נופלות (ואיתן הדאטה) כנתון. הגיוני אולי ל-Netflix כ - Cloud Native, פחות לארגון הממוצע.הרבה חברות לא רצו לכתוב מחדש - “פשוט” לקחת את ה - Workload הקיים ולהעביר לענן (משיקולי תקציב בדר”כ, נכונים יותר או פחות). זה נכון עד היום - Lift & Shift של Legacy systems ש”עובדות יפה” ב-Data Center on premise, וצריכות משיקולים כלשהם לעבור לענן.חשוב לשים לב שיש גם את הצד השני - רצון לקחת את ה - Cloud Economics (כל שמגיע כל כך יפה מ - AWS/GCP/Azure) “ולהרוויח” אותם גם On-premise - להפסיק לדאוג ל - Hardware refresh למשל (פחת, מערכת מתיישנת וכו’). הרעיון של IaaS הוא שהכל הופך לשירות - לא צריך לתכנן מראש, אין חוזים לטווח ארוך אלא רק עלות לפי שימוש בפועל. מי שנותן את השירות נותן הבטחה ל - SLA וזהו.איפה זה פוגש את מי שנמצא בענן ורוצה לעבור ל - On-premise? מערכות ה-Legacy שאנחנו מכירים היום לא מאפשרות את זה באמת.ב- Zadara מאמינים שעולם הענן קיים גם כ”ענן פרטי”, ורוצים לאפשר ענן כזה - למשל מאפשרים Storage as as service גם ב - Private cloud: מתקינים את המערכת אצל הלקוח, בלי שהוא קונה שום דבר אלא משלם רק עבור שימוש בפועל. Stop buying Storage . . .איך זה עובד ברמה הפרקטית? הרי מישהו צריך ללכת לחבר את הכבלים או לשדרג מדי פעם . . .המערכת בנויה כך שניתן לתפעל הכל מרחוק - שולחים ללקוח את ה - Hardware (שהוא Commodity - שרתים סטנדרטיים של Intel, Cisco, Dell וכו’), והמערכת היא Software Defined Storage. מבקשים Remote Access Permissions ומבצעים את ההתקנה מרחוק.מעיין Data Center מנוהל בתוך ה - Data Center של הלקוח.איך זה עובד בענן? נניח שאנחנו רצים על AWS - איך זה עובד? מכונות של EC2? משהו בחוץ?לא רצים בתוך ה - EC2 אלא בחוץ - מעבר לעלויות (AWS ישמחו כמובן…), Zadara מריצים את “הענן שלהם” (ה - clusters) ב - Data Centers שונים (עם Co-location, למשל Equnix) שנמצאים מאוד קרוב ל - Data Centers של “הגדולים” (Hyper-scale Clouds).מחזיקים כיום מעל 15 Data Centers כאלה בעולם (חוף מערבי ומזרחי של ארה”ב, אירלנד, גרמניה, יפן, סינגפור, קוריאה, אוסטרליה, קנדה . . .), כשבדר”כ החיבור ל-AWS הוא ב - Direct Connect (חיבור פרטי ומאובטח, גם מבחינת Bandwidth, ישירות ל-AWS), עם מודל עסקי שונה (משלם בעיקר “על הקו”, ופחות לפי Traffic in - Traffic out, למעט הגבלות שונות, בעיקר כשיש הרבה Write מה-EC2).אחד היתרונות של היכולת הזו “לרוץ מחוץ לענן” הוא היכולת ל - Multi-cloud solution - על אותה מערכת אחסון / Block File / Object, אפשר לגשת בו זמנית גם מ - AWS, גם מ - GCP גם מ Azure וגם מה - On-Premise.יש פתרון שנקרא Zadara Cloud on-premise, שמאפשר להגיע מה - On-Premise ל - Cloud (ב-Co-location), ומשם לגשת לאחד מספקי הענן לבחירתך: גם Multi-cloud וגם On-Premise עם חיבוריות ביניהם. מאוד מעניין.חזרה לרגע ל - Basics: הזכרנו מוקדם יותר שני סוגי אחסון ש-AWS הציעו - Block Storage ו - Object Storage. מה הם סוגי ה - Storage הנפוצים באופן כללי, או במילים אחרות - מה Zadara מאפשרת (Features) שלפחות בזמנו AWS לא אפשרו?לגבי Basics - מערכות אחסון מאופיינות בהרבה דברים . . . יש כמה אלמנטים עיקריים - סוג המדיה - המדיה המוכרת היא Hard Disk Drives (מאיזור שנות ה-50): פלטות מגנטיות מסתובבות עם ראש קורא, capacity גבוה (14Tb נשמע סביר לגמרי נכון להקלטת הפרק), וביצועים סבירים (כמה ms); סוג נוסף של מדיה הם ה-Solid State Drives (משנות ה-90 פחות או יותר): טכנולוגית Flash, כתיבה וקריאה הרבה יותר מהירות, אין חלקים נעים וביצועים ב Sub-ms.מעבר לזה יש מאפיינים נוספים - Data Protection Layer (ששומר גם על רמה מסויימת של יתירות Redundancy) יש כל מיני סוגי פרוטוקולים לגישה, וגם כאן העולם התקדם וזה משפיע על Hardware Refresh: אם למשל עברת ממערכת מבוססת SaaS או SATA שהייתה יותר איטית, ועכשיו יש NVMe ו NVMe over-Fabrics שהם פרוטוקולים הרבה יותר מהירים לגישה למדיה שהיא מבוססת Flash (כשיצא ה-Flash, צוואר הבקבוק של האחסון עבר מהמדיה (סיבוב הדיסק) לפרוטוקול, ואז הגיעו ה- PCI ומעל זה את ה - NVMf על מנת להתמודד עם זה…) - לקוחות רוצים לעבור ל-NVMe בלי לשלם יותר, ו - Zadara עושים את ה - Hardware refresh עבורם.רגע, רגע רגע . . .מה זה NVMe? למה זה חשוב?אוקיי - Non-Volatile Memory Extension הוא פרוטוקול שמאפשר לגשת ב-RDMA ישירות אל המדיה. בהפשטה - אפשר “לשים פוינטר” ישירות אל הדיסק, ולהתייחס אל הדיסק כאל זיכרון לא נדיף. זמן גישה? סדרי גודל של עשרות ns, ועבור NVMe Drives מדובר ב-ms בודדות (סדר גודל של כ - x1000 יותר מהר).המדיה הטובה ביותר כיום מגיע משילוב של Micron ו - Intel, ונקראת 3D Xpoint, וזמן הגישה הוא סביב ה -10ms (בתנאי שמדובר ב - Attached Storage - מחובר למכונה עם PCI).למשתמש בסוף חשוף איזשהו API -
Sponsors: Sentry use the code “devchat” for 2 months free on Sentry small plan Triplebyte offers a $1000 signing bonus CacheFly Host: Charles Max Wood Guest: Donovan Brown Episode Summary In this episode, Charles speaks with Donovan Brown, a principal DevOps Manager with Microsoft with a background in application development. Donovan talks about Azure DevOps and the transition from Visual Studio Team Services (VSTS) to Azure DevOps. They discuss the absolute need for continuous integration (CI) in developer teams and the ease of creating CI and continuous deployment (CD) pipeline to Azure. Links: https://devchat.tv/js-jabber/jsj-345-azure-devops-with-donovan-brown-live-at-microsoft-ignite/ Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops https://azuredevopsdemogenerator.azurewebsites.net/ https://azure.microsoft.com/en-us/services/devops/pipelines/ GitHub Azure DevOps’ Twitter https://twitter.com/TheLoECDA Picks: Donovan: YoTeam VSTeam Powershell Module Charles: Jet Blue Beta Tester
Sponsors: Netlify Sentry use the code “devchat” for $100 credit TripleByte Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Charles speaks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Sponsors: Netlify Sentry use the code “devchat” for $100 credit TripleByte Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Charles speaks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Sponsors: KendoUI Sentry use the code "devchat" for $100 credit Clubhouse Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Picks: Ed Podcast - All Things Git
Sponsors: KendoUI Sentry use the code "devchat" for $100 credit Clubhouse Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Picks: Ed Podcast - All Things Git
Sponsors: KendoUI Sentry use the code "devchat" for $100 credit Clubhouse Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Charles speaks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Charles speaks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Charles speaks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the React Round Up Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the React Round Up Charles speaks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Adventures in Angular panel talks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Adventures in Angular panel talks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Wood Special Guests: Ed Thomson In this episode, the Adventures in Angular panel talks with Ed Thomson who is a Program Manager at Azure through Microsoft, Developer, and Open Source Maintainer. Ed and Chuck discuss in full detail about Azure DevOps! Check out today’s episode to hear its new features and other exciting news! Show Topics: 0:59 – Live at Microsoft Ignite 1:03 – Ed: Hi! I am a Program Manager at Azure. 1:28 – Rewind 2 episodes to hear more about Azure DevOps! 1:51 – Ed: One of the moves from Pipelines to DevOps – they could still adopt Pipelines. Now that they are separate services – it’s great. 2:38 – Chuck talks about features he does and doesn’t use. 2:54 – Ed. 3:00 – Chuck: Repos and Pipelines. I am going to dive right in. Let’s talk about Repos. Microsoft just acquired GitHub. 3:18 – Ed: Technically we have not officially acquired GitHub. 3:34 – Chuck: It’s not done. It’s the end of September now. 3:55 – Ed: They will remain the same thing for a while. GitHub is the home for open source. Repos – we use it in Microsoft. Repositories are huge. There are 4,000 engineers working in these repositories. Everyone works in his or her own little area, and you have to work together. You have to do all this engineering to get there. We bit a tool and it basically if you run clone... Ed continues to talk about this topic. He is talking about One Drive and these repositories. 6:28 – Ed: We aren’t going to be mixing and matching. I used to work through GitHub. It’s exciting to see those people work close to me. 6:54 – Chuck. 6:59 – Ed: It has come a long way. 7:07 – Chuck: Beyond the FSF are we talking about other features or? 7:21 – Ed: We have unique features. We have branch policies. You can require that people do pole request. You have to use pole request and your CI has to pass and things like that. I think there is a lot of richness in our auditing. We have enterprise focus. At its core it still is Git. We can all interoperate. 8:17 – Chuck. 8:37 – Ed: You just can’t set it up with Apache. You have to figure it out. 8:51 – Chuck: The method of pushing and pulling. 9:06 – Chuck: You can try DevOps for free up to 5 users and unlimited private repos. People are interested in this because GitHub makes you pay for that. 9:38 – Ed and Chuck continue to talk. 9:50 – Ed: Pipelines is the most interesting thing we are working on. We have revamped the entire experience. Build and release. It’s easy to get started. We have a visual designer. Super helpful – super straightforward. Releases once your code is built – get it out to production say for example Azure. It’s the important thing to get your code out there. 10:55 – Chuck: How can someone start with this? 11:00 – Ed: Depends on where your repository is. It will look at your code. “Oh, I know what that is, I know how to build that!” Maybe everyone isn’t doing everything with JavaScript. If you are using DotNet then it will know. 12:05 – Chuck: What if I am using both a backend and a frontend? 12:11 – Ed: One repository? That’s when you will have to do a little hand packing on the... There are different opportunities there. If you have a bash script that does it for you. If not, then you can orchestrate it. Reduce the time it takes. If it’s an open source project; there’s 2 – what are you going to do with the other 8? You’d be surprised – people try to sneak that in there. 13:30 – Chuck: It seems like continuous integration isn’t a whole lot complicated. 13:39 – Ed: I am a simple guy that’s how I do it. You can do advanced stuff, though. The Cake Build system – they are doing some crazy things. We have got Windows, Lennox, and others. Are you building for Raspberries Pies, then okay, do this... It’s not just running a script. 15:00 – Chuck: People do get pretty complicated if they want. It can get complicated. Who knows? 15:26 – Chuck: How much work do you have to do to set-up a Pipeline like that? 15:37 – Ed answers the question in detail. 16:03 – Chuck asks a question. 16:12 – Ed: Now this is where it gets contentious. If one fails... Our default task out of the box... 16:56 – Chuck: If you want 2 steps you can (like me who is crazy). 17:05 – Ed: Yes, I want to see if it failed. 17:17 – Chuck: Dude, writing code is hard. Once you have it built and tested – continuous deployment. 17:33 – Ed: It’s very easy. It’s super straightforward, it doesn’t have to be Azure (although I hope it is!). Ed continues this conversation. 18:43 – Chuck: And it just pulls it? 18:49 – Ed: Don’t poke holes into your firewall. We do give you a lot of flexibility 19:04 – Chuck: VPN credentials? 19:10 – Ed: Just run the... 19:25 – Chuck comments. 19:36 – Ed: ...Take that Zip... 20:02 – Ed: Once the planets are finely aligned then...it will just pull from it. 20:25 – Chuck: I host my stuff on Digital Ocean. 20:46 – Ed: It’s been awhile since I played with... 20:55 – Chuck. 20:59 – Ed and Chuck go back and forth with different situations and hypothetical situations. 21:10 – Ed: What is Phoenix? 21:20 – Chuck explains it. 21:25 – Ed: Here is what we probably don’t have is a lot of ERLANG support. 22:41 – Advertisement. 23:31 – Chuck: Let’s just say it’s a possibility. We took the strip down node and... 23:49 – Ed: I think it’s going to happen. 23:55 – Ed: Exactly. 24:02 – Chuck: Testing against Azure services. So, it’s one thing to run on my machine but it’s another thing when other things connect nicely with an Azure set-up. Does it connect natively once it’s in the Azure cloud? 24:35 – Ed: It should, but there are so many services, so I don’t want to say that everything is identical. We will say yes with an asterisk. 25:07 – Chuck: With continuous deployment... 25:41 – Ed: As an example: I have a CD Pipeline for my website. Every time I merge into master... Ed continues this hypothetical situation with full details. Check it out! 27:03 – Chuck: You probably can do just about anything – deploy by Tweet! 27:15 – Ed: You can stop the deployment if people on Twitter start complaining. 27:40 – Chuck: That is awesome! IF it is something you care about – and if it’s worth the time – then why not? If you don’t have to think about it then great. I have mentioned this before: Am I solving interesting problems? What projects do I want to work on? What kinds of contributions do I really want to contribute to open source? That’s the thing – if you have all these tools that are set-up then your process, how do you work on what, and remove the pain points then you can just write code so people can use! That’s the power of this – because it catches the bug before I have to catch it – then that saves me time. 30:08 – Ed: That’s the dream of computers is that the computers are supposed to make OUR lives easier. IF we can do that and catch those bugs before you catch it then you are saving time. Finding bugs as quickly as possible it avoids downtime and messy deployments. 31:03 – Chuck: Then you can use time for coding style and other things. I can take mental shortcuts. 31:37 – Ed: The other thing you can do is avoiding security problems. If a static code analysis tool catches an integer overflow then... 32:30 – Chuck adds his comments. Chuck: You can set your policy to block it or ignore it. Then you are running these tools to run security. There are third-party tools that do security analysis on your code. Do you integrate with those? 33:00 – Ed: Yep. My favorite is WhiteSource. It knows all of the open source and third-party tools. It can scan your code and... 34:05 – Chuck: It works with a lot of languages. 34:14 – Ed. 34:25 – Chuck: A lot of JavaScript developers are getting into mobile development, like Ionic, and others. You have all these systems out there for different stages for writing for mobile. Android, windows Phone, Blackberry... 35:04 – Ed: Let’s throw out Blackberry builds. We will ignore it. Mac OS dies a fine job. That’s why we have all of those. 35:29 – Chuck: But I want to run my tests, too! 35:36 – Ed: I really like to use App Center. It is ultimately incredible to see all the tests you can run. 36:29 – Chuck: The deployment is different, though, right? 36:40 – Ed: I have a friend who clicks a button in... Azure DevOps. 37:00 – Chuck: I like to remind people that this isn’t a new product. 37:15 – Ed: Yes, Azure DevOps. 37:24 – Chuck: Any new features that are coming out? 37:27 – Ed: We took a little break, but... 37:47 – Ed: We will pick back up once Ignite is over. We have a timeline on our website when we expect to launch some new features, and some are secret, so keep checking out the website. 39:07 – Chuck: What is the interplay between Azure DevOps and Visual Studio Code? Because they have plugins for freaking everything. I am sure there is something there that... 39:30 – Ed: I am a VI guy and I’m like 90% sure there is something there. You are an eMac’s guy? The way I think about it is through Git right out of the box. Yes, I think there are better things out there for integration. I know we have a lot of great things in Visual Code, because I worked with it. 40:45 – Chuck: Yes, people can look for extensions and see what the capabilities are. Chuck talks about code editor and tools. 41:28 – Ed: ... we have been pulling that out as quickly as possible. We do have IE extensions, I am sure there is something for VS Code – but it’s not where I want to spend my time. 42:02 – Chuck: Yes, sure. 42:07 – Ed: But everyone is different – they won’t work the way that I work. So there’s that. 42:30 – Ed: That Chuck. 42:36 – Chuck: Where do people get news? 42:42 – Ed: Go to here! 42:54 – Chuck: Where do people find you? 43:00 – Ed: Twitter! 43:07 – Chuck: Let’s do Picks! 43:20 – Advertisement – Fresh Books! Links: GitHub Microsoft’s Azure Microsoft’s Pipeline Azure DevOps Erlang WhiteSource Chuck’s Twitter Ed Thomson’s Twitter Ed Thomson’s GitHub Ed Thomson’s Website Ed Thomson’s LinkedIn Sponsors: Angular Boot Camp Fresh Books Get a Coder Job Course Picks: Ed Podcast - All Things Git
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Adventures in Angular panel talks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Adventures in Angular panel talks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module
Panel: Charles Max Woods Special Guests: Donovan Brown In this episode, the Adventures in Angular panel talks with Donovan Brown. He is a principal DevOps Manager with Microsoft with a background in application development. He also runs one of the nation’s fastest growing online registration sites for motorsports events DLBRACING.com. When he is not writing software, he races cars for fun. Listen to today’s episode where Chuck and Donovan talk about DevOps, Azure, Python, Angular, React, Vue, and much, much more! Show Topics: 1:41 – Chuck: The philosophies around DevOps. Just to give you an idea, I have been thinking about what I want to do with the podcasts. Freedom to work on what we want or freedom to work where we want, etc. Then that goes into things we don’t want to do, like fix bugs, etc. How does Microsoft DevOps to choose what they want to do? 2:37 – Guest: We want to automate as much as we can so the developer has less work. As a developer I want to commit code, do another task, rinse and repeating. Minutes and not even hours later then people are tweeting about the next best thing. Do what you want, where you want. Code any language you want. 4:15 – Chuck: What has changed? 4:19 – Guest: The branding changed. The name wasn’t the most favorite among the people. The word “visual” was a concerned. What we have noticed that Azure will let me run my code no matter where I am. If you want to run Python or others it can run in Azure. People didn’t need all of it. It comes with depositories, project management, and so much more! People could feel clumsy because there is so much stuff. We can streamline that now, and you can turn off that feature so you don’t have a heart attack. Maybe you are using us for some features not all of them – cool. 7:40 – Chuck: With deployments and other things – we don’t talk about the process for development a lot. 8:00 – Guest talks about the things that can help out with that. Guest: Our process is going to help guide you. We have that all built into the Azure tab feature. They feel and act differently. I tell all the people all the time that it’s brilliant stuff. There are 3 different templates. The templates actually change over the language. You don’t have to do mental math. 9:57 – Chuck: Just talking about the process. Which of these things we work on next when I’ve got a bug, or a ... 10:20 – Guest: The board system works like for example you have a bug. The steps to reproduce that bug, so that there is no question what go into this specific field. Let the anatomy of the feature do it itself! 11:54 – Chuck comments. 12:26 – Chuck: Back to the feature. Creating the user stories is a different process than X. 12:44 – Guest – You have a hierarchy then, right? Also what is really cool is we have case state management. I can click on this and I expect this to happen... These are actual tasks that I can run. 13:52 – Chuck: Once you have those tests written can you pull those into your CI? 14:00 – Guest: “Manual tests x0.” Guest dives into the question. 14:47 – I expect my team to write those test cases. The answer to your question is yes and no. We got so good at it that we found something that didn’t even exist, yet. 16:19 – Guest: As a developer it might be mind 16:29 – Chuck: I fixed this bug 4x, I wished I had CI to help me. 16:46 – Guest: You get a bug, then you fix a code, etc., etc. You don’t know that this original bug just came back. Fix it again. Am I in Groundhog Day? They are related to each other. You don’t have a unit test to tell you. When you get that very first bug – write a unit test. It will make you quicker at fixing it. A unit test you can write really fast over, and over, again. The test is passing. What do you do? Test it. Write the code to fix that unit test. You can see that how these relate to each other. That’s the beauty in it. 18:33 – Chuck: 90% of the unit tests I write – even 95% of the time they pass. It’s the 5% you would have no idea that it’s related. I can remember broad strokes of the code that I wrote, but 3 months down the road I can’t remember. 19:14 – Guest: If you are in a time crunch – I don’t have time for this unit test. Guest gives us a hypothetical situation to show how unit tests really can help. 20:25 – Make it muscle memory to unit test. I am a faster developer with the unit tests. 20:45 – Chuck: In the beginning it took forever. Now it’s just how I write software now. It guides my thought process. 21:06 – Guest: Yes! I agree. 22:00 – Guest: Don’t do the unit tests 22:10 – Chuck: Other place is when you write a new feature,...go through the process. Write unit tests for the things that you’ve touched. Expand your level of comfort. DevOps – we are talking about processes. Sounds like your DevOps is a flexible tool. Some people are looking for A METHOD. Like a business coach. Does Azure DevOps do that? 23:13 – Guest: Azure DevOps Projects. YoTeam. Note.js, Java and others are mentioned by the Guest. 25:00 – Code Badges’ Advertisement 25:48 – Chuck: I am curious – 2 test sweets for Angular or React or Vue. How does that work? 26:05 – Guest: So that is Jasmine or Mocha? So it really doesn’t matter. I’m a big fan of Mocha. It tests itself. I install local to my project alone – I can do it on any CI system in the world. YoTeam is not used in your pipeline. Install 2 parts – Yo and Generator – Team. Answer the questions and it’s awesome. I’ve done conferences in New Zealand. 28:37 – Chuck: Why would I go anywhere else? 28:44 – Guest: YoTeam was the idea of... 28:57 – Check out Guest 29:02 – Guest: I want Donovan in a box. If I weren’t there then the show wouldn’t exist today. 29:40 – Chuck: Asks a question. 29:46 – Guest: 5 different verticals. Check out this timestamp to see what Donovan says the 5 different verticals are. Pipelines is 1 of the 5. 30:55 – Chuck: Yep – it works on my Mac. 31:04 – Guest: We also have Test Plant and Artifacts. 31:42 – Chuck: Can you resolve that on your developer machine? 31:46 – Guest: Yes, absolutely! There is my private repository and... 33:14 – Guest: *People not included in box.* 33:33 – Guest: It’s people driven. We guide you through the process. The value is the most important part and people is the hardest part, but once on 33:59 – Chuck: I am listening to this show and I want to try this out. I want a demo setup so I can show my boss. How do I show him that it works? 34:27 – Azure.com/devops – that is a great landing page. How can I get a demo going? You can say here is my account – and they can put a demo into your account. I would not do a demo that this is cool. We start you for free. Create an account. Let the CI be the proof. It’s your job to do this, because it will make you more efficient. You need me to be using these tools. 36:11 – Chuck comments. 36:17 – Guest: Say you are on a team of developers and love GitHub and things that integration is stupid, but how many people would disagree about... 38:02 – The reports prove it for themselves. 38:20 – Chuck: You can get started for free – so when do you have to start paying for it? 38:31 – Guest: Get 4 of your buddies and then need more people it’s $6 a month. 39:33 – Chuck adds in comments. If this is free? 39:43 – Guest goes into the details about plans and such for this tool. 40:17 – Chuck: How easy it is to migrate away from it? 40:22 – Guest: It’s GITHub. 40:30 – Chuck: People are looing data on their CI. 40:40 – Guest: You can comb that information there over the past 4 years but I don’t know if any system would let you export that history. 41:08 – Chuck: Yeah, you are right. 41:16 – Guest adds more into this topic. 41:25 – Chuck: Yeah it’s all into the machine. 41:38 – Chuck: Good deal. 41:43 – Guest: It’s like a drug. I would never leave it. I was using TFS before Microsoft. 42:08 – Chuck: Other question: continuous deployment. 42:56 – When I say every platform, I mean every platform: mobile devices, AWS, Azure, etc. Anything you can do from a command line you can do from our build and release system. PowerShell you don’t have to abandon it. 45:20 – Guest: I can’t remember what that tool is called! 45:33 – Guest: Anything you can do from a command line. Before firewall. Anything you want. 45:52 – Guest: I love my job because I get to help developers. 46:03 – Chuck: What do you think the biggest mistake people are doing? 46:12 – Guest: They are trying to do it all at once. Fix that one little thing. It’s instant value with no risks whatsoever. Go setup and it takes 15 minutes total. Now that we have this continuous build, now let’s go and deploy it. Don’t dream up what you think your pipeline should look like. Do one thing at a time. What hurts the most that it’s “buggy.” Let’s add that to the pipeline. It’s in your pipeline today, what hurts the most, and don’t do it all at once. 49:14 – Chuck: I thought you’d say: I don’t have the time. 49:25 – Guest: Say you work on it 15 minutes a day. 3 days in – 45 minutes in you have a CSI system that works forever. Yes I agree because people think they don’t “have the time.” 50:18 – Guest continues this conversation. How do you not have CI? Just install it – don’t ask. Just do the right thing. 50:40 – Chuck: I free-lanced and setup CI for my team. After a month, getting warned, we had a monitor up on the screen and it was either RED or GREEN. It was basically – hey this hurts and now we know. Either we are going to have pain or not have pain. 51:41 – Guest continues this conversation. Have pain – we should only have pain once or twice a year. Rollback. If you only have it every 6 months, that’s not too bad. The pain will motivate you. 52:40 – Azure.com/devops. Azure DevOps’ Twitter 53:22 – Picks! 53:30 – Advertisement – Get a Coder Job Links: Donovan Brown’s GitHub Donovan Brown’s Twitter Donovan Brown Donovan Brown – Channel 9 Donovan Brown – Microsoft Azure YoTeam Azure.com/devops GitHub Azure DevOps’ Twitter Sponsors: Angular Boot Camp Digital Ocean Get a Coder Job course Picks: Charles Jet Blue Beta Testers Donovan YoTeam VSTeam Powershell Module