|
Ok for context: The company’s entire software engineering team is me, him, another guy, and a manager. This guy is effectively the lead frontend engineer. So I have some obligation to help him get better at this, if for no other reason that it won’t otherwise happen, and I work closely with him every day, so it affects me directly. “Approach it from the perspective of unit tests” is an interesting idea, but he’s the one writing the tests, so I don’t think it’s a silver bullet.
|
# ? Dec 13, 2022 05:22 |
|
|
# ? May 11, 2024 14:57 |
|
raminasi posted:Ok I'm having real trouble figuring out constructive communication in response to some stuff that a guy on my team does. I just had an quintessential exchange. Here was the existing code (please forgive any syntactic errors, they were just introduced by my obfuscation): The first thing is a bad pattern precisely because the 2nd thing always happens. Usually when the 2nd one shows up, I try to make them both disappear instead: can the code be structured so that neither is necessary, like can we just shove them in a `class ConfigApi`, for example. I find that this usually provokes a more useful line of thinking than "have you considered writing code that isn't bad", even if its very appropriate in this case.
|
# ? Dec 13, 2022 06:39 |
|
raminasi posted:“Approach it from the perspective of unit tests” is an interesting idea, but he’s the one writing the tests, so I don’t think it’s a silver bullet. Sounds like your org is very early in the software dev maturity. Writing your own tests sounds flimsy. "I can just hack it to make the test pass, what's the point?" I get it, I was there too early on - it's just more code for the same result! But as somebody said earlier, you end up reviewing the tests rather than the code. The test cases don't cover "thing 1 true, thing 2 false"? Make them add that test case. If the code is broken, they'll figure it out real fast, fix it, and ensure they don't break it again in the future. It's easier to say "your tests are wrong" than "your code is wrong". It might not sound like a silver bullet but it absolutely is, and it sounds like this is the next big step forward in your dev journey.
|
# ? Dec 13, 2022 13:45 |
|
raminasi posted:“Approach it from the perspective of unit tests” is an interesting idea, but he’s the one writing the tests, so I don’t think it’s a silver bullet. I think in this sort of situation it actually is a silver bullet. code:
In this case, you could even leave comments like "Needs test case for thingDoer(thingThatIsFalse, thingThatIsTrue)" or "What happens if they are both false?"
|
# ? Dec 13, 2022 15:46 |
|
Yeah, with N boolean inputs, it's immediately clear something is missing when there aren't 2^N test cases
|
# ? Dec 13, 2022 16:58 |
|
If you don't have tests, you don't have a feature.
|
# ? Dec 14, 2022 00:25 |
|
Bongo Bill posted:If you don't have tests, you don't have a feature. Asset.That(main(argc, argv), false); Done!
|
# ? Dec 14, 2022 00:59 |
|
raminasi posted:Ok I'm having real trouble figuring out constructive communication in response to some stuff that a guy on my team does. I just had an quintessential exchange. Here was the existing code (please forgive any syntactic errors, they were just introduced by my obfuscation): why not use dot notation? config,API.importantStuff = {} config,API.importantStuff.importantKey = importantValue config,API.importantStuff.otherImportantKey = otherImportantValue
|
# ? Dec 14, 2022 22:17 |
|
Doktor Avalanche posted:why not use dot notation? I simplified the code a bit for the post; I don’t have the original in front of me but I think that wouldn’t work. (Or maybe it would and we both just missed it!) Thanks for your thoughts, everyone.
|
# ? Dec 15, 2022 16:58 |
|
Bongo Bill posted:If you don't have tests, you don't have a feature. We all test in production one way or another, friend.
|
# ? Dec 15, 2022 20:13 |
|
Volmarias posted:We all test in production one way or another, friend. Especially with networking code. Try unit testing the bugs out of that!
|
# ? Dec 16, 2022 11:47 |
|
That's what a smoke test is for. You see where the smoke leaks out of the tubes and wires
|
# ? Dec 16, 2022 17:58 |
|
Well, sure. You can't test everything. But that's not a reason to not test anything.
|
# ? Dec 16, 2022 18:25 |
|
I tested that my code compiled, what else do you want from me!?
|
# ? Dec 16, 2022 18:33 |
|
Bongo Bill posted:Well, sure. You can't test everything. But that's not a reason to not test anything. Look, I have a robust alerting mechanism. If I hear screams after I deploy, that means I should take a look on Monday. Tests would just slow me down!
|
# ? Dec 16, 2022 18:46 |
|
One of the few things I miss about in-person work is blaring "Entrance of the Gladiators" when something failed a scream test
|
# ? Dec 16, 2022 18:58 |
|
CPColin posted:One of the few things I miss about in-person work is blaring "Entrance of the Gladiators" when something failed a scream test I still work in an office a few days per week, I'm stealing this (I'm a networking engineer, the scream test is one of my favourite tools )
|
# ? Dec 16, 2022 19:11 |
|
Volmarias posted:Look, I have a robust alerting mechanism. If I hear screams after I deploy, that means I should take a look on Monday. Tests would just slow me down! Tests are important, but I'm a senior engineer. Tests are for the juniors to write. (actual belief by developer of the code I inherited at my current job)
|
# ? Dec 17, 2022 06:37 |
|
Aka, “here’s a branch with a proof-of-concept I did, can you flesh it out?”
|
# ? Dec 17, 2022 14:46 |
|
Hey thanks for agreeing to help me with the coding work for this project. I just finished the HLD and LLD. Just let me know when it's done thanks!
|
# ? Dec 17, 2022 16:26 |
|
Bongo Bill posted:If you don't have tests, you don't have a feature.
|
# ? Dec 17, 2022 16:37 |
|
Users are just surprise tests.
|
# ? Dec 17, 2022 16:45 |
|
My place of work has an internal design library. To my mind, the benefits of our design library are:
I'm often tasked with building new pages based on design sketches provided by one of our designers who don't work directly on the design library, and he constantly deviates from the library. He'll design his own buttons, use colors that aren't in the library guidelines, and just repeatedly reinvent the wheel for no apparent reason. And, of course, he fails to account for accessibility. I don't think the design library has to be treated as word of God or anything, but if someone's gonna deviate from it there should be a clear reason for doing so. Whenever I question him about weird design choices he'll basically wave me off because he's a Professional Designer™ and knows what he's doing, while I'm just a developer with no formal design training. My product owner generally sides with him, arguing that the design library doesn't update fast enough to keep up with our needs and that it "looks good". I don't know what kind of response I'm hoping for on this, mostly I'm just venting because I'm real tired of this situation. And it's incredibly demoralizing to have to build stuff that I know is bad.
|
# ? Dec 19, 2022 14:59 |
|
Developers shouldn’t be the ones enforcing use of the design library, so that sucks.
|
# ? Dec 19, 2022 16:03 |
|
Have you explained to the person up the chain who cares about these things that deviating from the design system costs money? That's usually how I get buy-in on adhering to our existing systems.
|
# ? Dec 19, 2022 16:38 |
|
A design with no regard for accessibility is not a complete design. If he's really a Professional Designer he should account for these things.
|
# ? Dec 19, 2022 17:00 |
|
Welcome to every experience I've ever had at every company implementing a component library. This only works if the designers are the people maintaining the library, which they never are since the bar is generally "can lay something out in Photoshop" without any other tech skills required. YanniRotten fucked around with this message at 19:19 on Dec 19, 2022 |
# ? Dec 19, 2022 19:16 |
|
It’s also inevitable that designers get bored of working within the library and start pushing the boundaries. “Every page looks the same” is good and bad.
|
# ? Dec 20, 2022 01:10 |
|
smackfu posted:It’s also inevitable that designers get bored of working within the library and start pushing the boundaries. “Every page looks the same” is good and bad. In my experience there's a big difference between expanding the library and ignoring the library.
|
# ? Dec 20, 2022 02:56 |
|
prom candy posted:Have you explained to the person up the chain who cares about these things that deviating from the design system costs money? That's usually how I get buy-in on adhering to our existing systems. Clanpot Shake posted:A design with no regard for accessibility is not a complete design. If he's really a Professional Designer he should account for these things. smackfu posted:It’s also inevitable that designers get bored of working within the library and start pushing the boundaries. “Every page looks the same” is good and bad. prom candy posted:In my experience there's a big difference between expanding the library and ignoring the library.
|
# ? Dec 20, 2022 09:57 |
|
Woebin posted:Money is always the last thing I think about, so no, I don't think I've tried that angle. I will! Depending on how large the company is and how far your Manager's budget is divorced from the practical effects of your actions, this may not work.
|
# ? Dec 21, 2022 19:58 |
|
Volmarias posted:Depending on how large the company is and how far your Manager's budget is divorced from the practical effects of your actions, this may not work. My experience with larger companies is that anyone making a decision has no idea about the real consequences and that it's a miracle any money is made at all.
|
# ? Dec 22, 2022 09:02 |
|
Large enterprise companies are like those huge wild animals with a thousand gigantic fat ticks hanging off them. It owns if you are with one of the tick companies.
|
# ? Dec 22, 2022 14:57 |
|
Which one of you idiots did this https://www.pcgamer.com/software-engineer-busted-after-being-inspired-by-office-space-scam/
|
# ? Jan 3, 2023 16:01 |
Cup Runneth Over posted:Which one of you idiots did this https://www.pcgamer.com/software-engineer-busted-after-being-inspired-by-office-space-scam/ jesus christ how does someone actually think they can get away with that
|
|
# ? Jan 3, 2023 16:19 |
|
ChickenWing posted:jesus christ how does someone actually think they can get away with that quote:His involvement was apparently uncovered when a document was found on his computer that detailed a plan to alter logs of audits and alarms to cover up evidence of theft. It was called "OfficeSpace project". Not exactly the brightest guy here.
|
# ? Jan 3, 2023 16:25 |
|
MyCrimes.md
|
# ? Jan 3, 2023 17:11 |
|
Volmarias posted:Not exactly the brightest guy here.
|
# ? Jan 3, 2023 17:16 |
|
Cup Runneth Over posted:Which one of you idiots did this https://www.pcgamer.com/software-engineer-busted-after-being-inspired-by-office-space-scam/ further proof that you don't have to be smart or even not dumb to do this job
|
# ? Jan 3, 2023 18:57 |
|
|
# ? May 11, 2024 14:57 |
|
We've got an entry-level dev on our team who is turning out to be a bad hire. He's already pissed off 3 of the 4 people on his subteam. Since the senior dev on his subteam who had been reviewing his code was out on vacation for the holidays, my manager asked me if I could pick up where that other senior dev left off. Sure thing, no problem. Needless to say, there were a ton of issues. But the icing on the cake is the gall of this guy. In response to my review, he's now:
This fuckin guy
|
# ? Jan 6, 2023 06:19 |