Do all the things like ++ or -- rants, post your own rants, comment on others' rants and build your customized dev avatarSign Up
From the creators of devRant, Pipeless lets you power real-time personalized recommendations and activity feeds using a simple APILearn More
pdinklaedch156305dSounds to me you made some bad experience with some projects (inside your company?).
I for example am glad someone is reviewing my PR before I merge it. That way I can point at the reviewer and say "but he approved it" lol.
@pdinklaedch I am more anarchic. You can come in and commit on my dev branch. I trust you are good enough. If you make a mistake, I'll fix it. If I make a mistake, you may fix it too, otherwise I will fix it.
But hey, the PR thing was just an example. I also hate it when I see lots of tests for a stupid crud app that has no business logic whatsoever, because, you know, you *must* write tests... and so on.
pdinklaedch156305d@jiffier Seems that whoever you work for is very strict when it comes to merging / testing
I mean, I totally agree with you if it's a small /personal project. Wouldn't want to be bothered with this bureaucracy as well. But if it's a client project, it's kind of understandable corporate wants to make sure everything works
I'm definitely not in support of one size fits all development, I think each team should structure its own process and I don't work with firms that assert there's one platform to rule them all for an entire org.
That said, the platforms you've listed are the kiddie pool. They're the market segments that grew up from a desire for cheap labor at the expense of quality. So when they try and make them deliver quality, they naturally can't get it. So the people in charge Google "how to make my cheap team deliver quality" and get results like "pull requests," and "mindless boilerplate test coverage."
Wanderer37305dWell I am currently facing the opposite of your experience. I WISH I get some code/peer review at my work place. All the projects I do I finish by myself (which is great in some ways) however, if I need help or second opinions I have no one to turn to.
PR is a double edged sword. It can be a great, beneficial learning opportunity for both parties. Or it can be a power move opportunity for a more "senior" dev to bully/"teach" the junior dev. It really depends on the team/workplace.
@SortOfTested I'm 47 now, and to be honest I can no longer be bothered with this stuff anymore. 15 years ago I was the one overengineering things with a lot of XML in Java.
Kids nowadays overengeneer because they want to forge a CV. I am instead a freelancer, and I want to think I get paid to get things done, with a good balance between quality and cost.
Time to go into management then.
15 years ago I was grating at how irritating the rise of xml was, so I wrote something I preferred in aspectj as an extension for Apache ORB bean map registration, config and code gen. Thankfully that segment of the industry caught up.
Nowadays I'm doing full time FP and RFP with a competent team of heavy hitters and I've never been happier. If you want less enterprisey bureaucratic bullshit in your life, go FP.
@SortOfTested Mm.. I'm not sure what do you mean by FP and RFP?
Functional programming, reactive functional programming. People in that spectrum don't tend to fuck around with fads. Scala, scalaz, clojure, F#, Haskell, reactivex, rxjs, ramda
Same here, two kids in college, running an engineering firm in the US and the UK.
You've definitely covered why I don't freelance. The division of attention drives me nuts.
@SortOfTested Oh boy, my only experience with functional programming has been with emacs lisp, which I just cannot get my head around. It feels like speaking backwards, hehe.
The thing that's getting me excited nowadays is Rust. I'm still quite newbie with it, but I have to say I am enjoying it.
react-guy1456305d@jiffier I have the exact opposite experience myself, weirdly.
I started development early as a student job, I had no interest in testing or peer reviews. I believed it was a waste of time.
Today, when it comes to business logic, I test as much as I can, integration tests if possible.
It's time consuming to write, but it makes me more confident with the production code, helps to avoid regression bugs, and makes me write production code faster.
I'm not enforcing it to anyone, however. I found it helps me, but if my colleague does not need it, it's fine.
Angry-dev78304d@jiffier about tests - I was also trying to write lot of tests. Becaue I learned from Robert Martin book I think. So the problem is that those stars teach you wrong things. Now I am more picky about where to write tests and so spend less time on them and more time on features. I just wanted to be better developer, so I did what better developer said.
Angry-dev78304d@jiffier I also hate when those tools make me slower or take lot of time to learn them and still you are slower. And also if somebody takes the probgram without knowing the framework, like angular.js for example he has hard time understanding and again has to spend time learning. And there are lots of those frameworks. There are some things good in angular but overall when you compare how much time it takes to learn, its bad. But probably thats the reason it is now only used for legacy projects.
jiffier191304d@Angry-dev Well spotted on tests. I do the same. For example, last week I started writing a 2048 clone in Rust. I just couldn't do without writing tests first. They were so helpful. Those unit tests just came in naturally, they were screaming to be written. Instead, if I'm writing some crappy crud web app, writing tests there is just loosing time imho.