Updated Jun 23 2020 :: by Alex Yumashev

1) We still receive all errors via email

2) We still write jQuery code sometimes

3) We do not have an SPA and we barely use front-end frameworks. A couple of our most complex screens are written in Vue, but that’s it.

4) We... (gosh this one's tough)... we're on Microsoft stack! There, I said it. ASP.NET MVC + Windows + SQL Server. A couple of microservices run on linux (Python and Node), but the main app is Windows. Sorry.

5) We seem to be pretty happy (!) with this stack. No, really. Servers stay up for years.

6) ^^^ Planned outage does not count

7) We barely use Docker and I don’t know shit about Kubernetes. Every time I'm dealing with containers I spend 20 mins googling for a command and another 20 minutes figuring it out.

8) We don’t use TypeScript (my colleagues nudge me in the ribs) - OK, we do use TypeScript, the mobile application is written in it.

9) We don't use Jira, Trello, Notion, Confluence etc... In fact, we barely use anything but Slack (another nudge in the ribs) - yeah, OK, we also use Github

(that's a bit of a stretch since we're actively dogfooding our own product, but it sounds so cool I couldn't resist)

10) I don’t know anything about Agile, Scrum, Kanban, Shmanban, etc. Well, I know something. But we're still not using it.

11) We do not use JS compilers. Babel? Webpack? - nah, not using it. What's in the text editor is what users have in their browsers. Some stuff is being repacked and compressed over at the CDN though, but barely.

11.1) We also do not bundle scripts into one big lump, but that's intentional.

12) We write CSS in LESS. I guess LESS went out of fashion. About 8-10 years ago.

13) It's been only a year since we installed cloud-based CI. And I'm still not exactly sure what for - b/c we keep running all our builds and unittests locally out of habit.

14) I do not know Go, Rust, Kotlin or Dart. In fact, I barely know any fancy/trendy language other than C# or JS. Even my Python is rusty. I'm fluent in C++ but it's 2020 who cares.

15) We do not have a staging environment. Our CD rolls out to production directly.

16) My gaming laptop is way more powerful than any of our AWS servers

17) I don’t remember how flexbox works and have to google it every time. There are places in our markup where we have “float: right / left”. And sometimes even - (oh no) - tables.

18) We have no product managers

19) There were times when the whole team was religiosly using Macs AND MACS ONLY!!!!11111 Today there's only one Mac left in the company - the one we use to publish the mobile app. And Max (who does the publishing) gets really annoyed every time he has to take it out of his drawer.

20) We don't use anything "serverless" neither AWS-Lambda, nor CloudFlare Workers, nor analogues (does S3 count as "serverless"?)

21) One of our very, very important microservices is written in... PHP

22) I keep saying "microservices" but frankly 95% of our codebase is one giant ugly monolith, who am I fooling.

23) We have a Single Point of Failure - the database. Therefore, we fanatically back it up in 10 different ways.

24) In 12 years, we never ever had to restore from any of them

 

P.S. Stay tuned for "things we do right" I guess


'Things we do wrong at Jitbit' was written by Alex Yumashev
Alex Yumashev
Alex has founded Jitbit in 2005 and is a software engineer passionate about customer support.


Subscribe comments