|
Started the day with a candid conversation with two layers of management above me that I was having trouble. I took the Lead position on the team only after confirming it was 75% 'actually do my own engineering' and 25% 'mentoring, etc' Instead it's become like 95% mentoring and I'm burning out. On the upside, they basically said its because I'm too drat helpful and valuable as a resource and that I had carte blanche to just work from home any time I wanted to actually get work done. So that's a plus. Combined with some reflection that they're going to step in and try to start rebalancing things and strongly encouraging team members to be independent again. Which is necessary. Because my days often end up a bit like today. Where I watch Mid-Senior devs struggle to implement an IEnumerable<int> that counts up. Just as a 'let's learn how IEnumerable' works exercise. Mid-Senior. 
|
#
?
Mar 21, 2019 20:12
|
|
|
|
| # ? May 17, 2024 23:50 |
|
|
Protocol7 posted:I like to vent about my job a lot. I still love it, even if my coworkers are dinguses. But that's a problem you'll encounter literally anywhere. https://www.youtube.com/watch?v=co_DNpTMKXk But yeah,  I never want to give up remote, even if sometimes it's a pain that you don't get noticed as much for what you do (dealing with everyone lining up to jerk off the new guy on my team for this latest huge project...that I did 2/3 of the work for  ). It's so nice to not have to deal with people coming over to my desk and ignoring that I have headphones on, the universal sign for "gently caress off I'm concentrating", to just be able to take off for a while to run errands in the middle of the afternoon when poo poo isn't busy, etc.That said, I just wish I could work on something I cared about. This is not what I went to school for, it was a desperation move when I graduated into the worst job market imaginable, let alone with a degree in a field that had its funding slashed as a result. The upside is all this experience means I can probably make the transition back to at least doing development for scientific applications really easily once I get out of this god-forsaken state.
|
|
#
?
Mar 21, 2019 20:34
|
|
|
Cuntpunch posted:Started the day with a candid conversation with two layers of management above me that I was having trouble. I took the Lead position on the team only after confirming it was 75% 'actually do my own engineering' and 25% 'mentoring, etc' Being a mentor can be great, but is so reliant on the recipients. Talented and passionated individuals are fun to mentor. Dull and lazy individuals are not. Helping someone along who has a bit of a poo poo background or is dealing with something nasty outside of work, can also be great. But handholding those who should and could do better is horrible. Get out if it's burning yout out. Being burnt out on your profession is bad enough - burning out on working with other people is worse.
|
|
#
?
Mar 21, 2019 22:31
|
|
|
Cuntpunch posted:Started the day with a candid conversation with two layers of management above me that I was having trouble. I took the Lead position on the team only after confirming it was 75% 'actually do my own engineering' and 25% 'mentoring, etc' That's crazy for a mid-senior level dev. On the one hand stuff like this is perplexing, but on the other I feel like its given me a bit of leg up because there are so many devs kinda like that.
|
|
#
?
Mar 22, 2019 00:58
|
|
|
downout posted:That's crazy for a mid-senior level dev. On the one hand stuff like this is perplexing, but on the other I feel like its given me a bit of leg up because there are so many devs kinda like that. What's horrible to me is these are the kind of people interviewing me.
|
|
#
?
Mar 22, 2019 01:26
|
|
|
Rocko Bonaparte posted:What's horrible to me is these are the kind of people interviewing me. The kind of people failing you because you can't solve the Knight's Tour in 30 minutes are also the kind of people who don't understand memory constraints and know way less about development than their title represents. The poo poo I've seen. 
|
|
#
?
Mar 22, 2019 13:20
|
|
|
Pollyanna posted:The kind of people failing you because you can't solve the Knight's Tour in 30 minutes are also the kind of people who don't understand memory constraints and know way less about development than their title represents. The poo poo I've seen. That sounds more like one of those ugly 'haha gotcha!' trivia questions. It sucks because I think there's a lot of cross-contamination between people who use *those* 'code tests' or 'whiteboard problems' to surprise candidates - and that puts people off of the idea of checking algorithmic thinking at all.
|
|
#
?
Mar 22, 2019 14:21
|
|
|
The question was definitely cargo culted since the dude that asked it couldn’t really implement it either. Same dude that didn’t understand how things were loaded into memory, of course.
|
|
#
?
Mar 22, 2019 14:28
|
|
|
Software industry: requiring an MSc level understanding of algorithms to implement CRUD apps
|
|
#
?
Mar 22, 2019 14:34
|
|
|
But it’s what senior engineers do, and we need senior engineers to fix the problems we created!!
|
|
#
?
Mar 22, 2019 14:41
|
|
|
Pollyanna posted:But it’s what senior engineers do, and we need senior engineers to fix the problems we created!! We want name brand quality at great value prices!
|
|
#
?
Mar 22, 2019 15:13
|
|
|
Edit on second thought this is probably a bad idea.
Pollyanna fucked around with this message at 15:39 on Mar 22, 2019 |
|
#
?
Mar 22, 2019 15:35
|
|
|
What's gonna make these 50,000 lines of code maintainable is this bitchin' algorithm.
|
|
#
?
Mar 22, 2019 16:25
|
|
|
Oh good, we’re not gzipping the JSON we’re sending over HTTP either. No wonder we’re hitting timeouts. God, do I have to do everything around here? This dude took months to not even figure out what I figured out in less than a day.
|
|
#
?
Mar 22, 2019 20:31
|
|
|
Vulture Culture posted:Software industry: requiring an MSc level understanding of algorithms to implement CRUD apps The persistent joke at Google is that everyone else is working on amazing projects while you're just busy shuffling protobufs around. It's more true than we'd like.
|
|
#
?
Mar 22, 2019 21:18
|
|
|
Pollyanna posted:Oh good, we’re not gzipping the JSON we’re sending over HTTP either. No wonder we’re hitting timeouts. God, do I have to do everything around here? This dude took months to not even figure out what I figured out in less than a day. But you'll ruin the integrity of the package! How do we know the unzipped blob is the same as the blob before it was zipped??
|
|
#
?
Mar 22, 2019 21:20
|
|
|
Polio Vax Scene posted:But you'll ruin the integrity of the package! How do we know the unzipped blob is the same as the blob before it was zipped?? Triggered
|
|
#
?
Mar 22, 2019 21:27
|
|
|
Volmarias posted:The persistent joke at Google is that everyone else is working on amazing projects while you're just busy shuffling protobufs around. Is protobuf the thing at Google? Need to send an int? Protobuf. Need to send a string? Protobuf. Need to send a snail mail? Protobuf. How many protobuf messages do you guys send before your first coffee?
|
|
#
?
Mar 23, 2019 02:45
|
|
|
Well I feel sufficiently stupid Figured I would leave early since it was Friday, and after getting home, one of my automation pieces started puking in production. Had to drive back to the office to fix it because I couldn't get into the VPN. Of course, it was the one part of that app that doesn't (yet) have unit tests because it's hard to test. It's been a good long while since I've goofed up that bad.
|
|
#
?
Mar 23, 2019 05:15
|
|
|
Volguus posted:Is protobuf the thing at Google? Need to send an int? Protobuf. Need to send a string? Protobuf. Need to send a snail mail? Protobuf. How many protobuf messages do you guys send before your first coffee? It is the efficient wire format that all our tooling and libraries support (and we have great libraries and tooling). Are protos perfect for everything? No, but they do a good enough job at serialization for RPC/storage and are a super reliable and mature tech.
|
|
#
?
Mar 23, 2019 08:15
|
|
|
Paolomania posted:It is the efficient wire format that all our tooling and libraries support (and we have great libraries and tooling). Are protos perfect for everything? No, but they do a good enough job at serialization for RPC/storage and are a super reliable and mature tech. If your huge corporation is like my huge corporation, then the next frontier will be protobuf metaframeworks so everybody can act like they are the hot poo poo. A decade from now, it will be impossible to switch to anything newer.
|
|
#
?
Mar 23, 2019 10:57
|
|
|
Volguus posted:Is protobuf the thing at Google? Need to send an int? Protobuf. Need to send a string? Protobuf. Need to send a snail mail? Protobuf. How many protobuf messages do you guys send before your first coffee? "Well obviously you have a  "
|
|
#
?
Mar 25, 2019 18:13
|
|
|
Me, to guy installing the software: "Ok they need these files to output to D:\ImportWatchFolder on NET-SRV-01 so make sure you set up a network folder or something that the program can put them in then put that in the config file" Guy installing the software: "got it"  e: bonus, he put the .config in the task as the thing to run instead of the .exe 
Polio Vax Scene fucked around with this message at 20:28 on Mar 28, 2019 |
|
#
?
Mar 28, 2019 20:15
|
|
|
I love it.
|
|
#
?
Mar 28, 2019 20:21
|
|
|
 I once asked for a copy of a backup disk - they put it into the copier and faxed me the copy. Gallatin fucked around with this message at 22:53 on Mar 28, 2019 |
|
#
?
Mar 28, 2019 22:31
|
|
|
Some online form: enter the exact text you want to appear on your cake: I admire whoever filled this order, clearly stars and sprinkles are premium options and they were trying to circumvent the system.
|
|
#
?
Mar 28, 2019 22:52
|
|
|
I wonder if those things can handle emojis.
|
|
#
?
Mar 28, 2019 23:03
|
|
|
*actually poops on the cake*
|
|
#
?
Mar 28, 2019 23:14
|
|
|
Gallatin posted:
I'm going to admit that I didn't get this right away.
|
|
#
?
Mar 28, 2019 23:20
|
|
|
One of my favorites from 2005 while supporting a rollout. Me: Type in all lowercase letters [password]. User: It's not working. Me: I'll be over in a bit. *looks at screen* Password: *********************************** Yep. She typed "in all lowercase letters [password]" into the password field.
|
|
#
?
Mar 29, 2019 00:05
|
|
|
During my support days, we used LogMeIn for remote assistance. I have had the following conversation more times than I can count: "Alright, I need you to type in 'log me in one two three dot com' in the address bar" "it's not working" "Ok, can you read back to me what you typed?" "Log on me three two one dot com" The terrifying part is when you realize there's nothing *special* about technology that makes things difficult, it's the same problem in every situation for the vast majority of humanity.
|
|
#
?
Mar 29, 2019 08:52
|
|
|
GI_Clutch posted:One of my favorites from 2005 while supporting a rollout. Reminds me a bit of how Apple’s two factor login for iCloud is supported in older devices that were never updated to have the new auth flow. You type in your (correct) password and the server side triggers two factor and returns an error saying “enter your password plus the authentication code.” You literally have to type in “password1234” in the password box. No one does this right.
|
|
#
?
Mar 29, 2019 13:11
|
|
|
Cuntpunch posted:The terrifying part is when you realize there's nothing *special* about technology that makes things difficult, it's the same problem in every situation for the vast majority of humanity. The problem is most people aren't far beyond basic literacy. People who read text on forums for fun, on the other hand...
|
|
#
?
Mar 29, 2019 14:13
|
|
|
smackfu posted:Reminds me a bit of how Apple’s two factor login for iCloud is supported in older devices that were never updated to have the new auth flow. You type in your (correct) password and the server side triggers two factor and returns an error saying “enter your password plus the authentication code.” You literally have to type in “password1234” in the password box. No one does this right. There's some financial poo poo I had for my last 401k that did that as well. I can't wait for passwords to die.
|
|
#
?
Mar 29, 2019 15:57
|
|
|
I'm working on a firmware integration test plan for my company's blootooth low energy device I was supposed to have the device communicate over ble to a server and have it act as a ble proxy for a client. The client should run scripts and tests on the data. So I made a server and client in C# and and used Nunit for the tests. I had gotten mixed messages about whether I should use TCP or UDP so I made it work with either protocol and you could select which one it used with a config file. Since I'm in a junior position I consult with a staff engineer. I showed the staff engineer what I made and he doesn't want the team to deal with C# so I'd have to write it in C or C++ instead. winsock.h looks pretty ugly, and although I could probably use cygwin or WSL to use sys/socket.h, I'm not sure how to debugging would work if I used cygwin or WSL (I don't want to deal with command line GDB, I want to use an IDE). So I decided to use asio (non-boost). It uses UDP and I pray that I don't have to rewrite it support TCP. He also doesn't like how any of the C++ testing frameworks work so I'm rolling my own. I thought google test would meet all our requirements but he doesn't like working with xml our json output. So I've spent this past week porting my code from C# to C++ just to get to where I was a week ago. Feels bad man. Is it normal for places to roll their own test framework instead of using something like google test, boost test, catch2, etc? And is it normal for a team that uses C to decide that code-reviewing something in C# is too much trouble?
|
|
#
?
Mar 30, 2019 20:23
|
|
|
Faith For Two posted:Is it normal for places to roll their own test framework instead of using something like google test, boost test, catch2, etc? There might be good reasons to do this but “I don’t want to deal with XML” doesn’t sound like one. quote:And is it normal for a team that uses C to decide that code-reviewing something in C# is too much trouble? Yes, absolutely. If you’re a C shop, you need a real good reason to create something new in C#.
|
|
#
?
Mar 30, 2019 20:29
|
|
|
My Rhythmic Crotch posted:Well I feel sufficiently stupid At least you cared enough and got notified? Our internal controlling ETL processes have been throwing errors for a few days, and the colleague who implemented the processes basically said "yeah, that external data source is having issues, whatever", which still doesn't change the fact that the only reason we are getting notifications at all is because he tries to iterate over the result, which, if the source is borked, is a boolean, and that causes an unhandled exception, which the scheduler happens to report via email (because I set it up to do so). It's not just that none of his code has any tests (unit or integration), but he doesn't even bother to handle Exceptions. At least our boss was happy he finally got his controlling stuff, I guess? Just half a year longer, then I'll hopefully be able to leave this place... 
|
|
#
?
Mar 30, 2019 21:27
|
|
|
It depends what "decide it's too much trouble" means. If it means "we reject C# code out of hand because we use exclusively C", that's fine. If it means "we accept C# code without looking at it because we only know C", that's bad.
|
|
#
?
Mar 30, 2019 21:27
|
|
|
Faith For Two posted:He also doesn't like how any of the C++ testing frameworks work so I'm rolling my own. I thought google test would meet all our requirements but he doesn't like working with xml our json output. Sorry for your loss.
|
|
#
?
Mar 30, 2019 21:30
|
|
|
|
| # ? May 17, 2024 23:50 |
|
|
Faith For Two posted:I'm working on a firmware integration test plan for my company's blootooth low energy device So my dept builds tools and infrastructure for firmware testing. The firmware itself has no unit tests as far as I know, because none of the engineers liked the available unit test frameworks, and no one felt like making something. We're not a small mom 'n pop operation, mind you... this is a massive company. There's no excuse not to have unit testing as far as I'm concerned. So all firmware testing is done in a home-brewed interpreted scripting language that looks basically like C, but has kid gloves in a few areas (I don't think malloc and free are allowed, for example). A lot of the test engineers are ... stunted, I guess I would say, in their abilities. So the test scripts are utter poo poo. "I'll just copy-paste this ten times because I don't know what a for loop is" kind of bad. That's a long winded way of saying yes on both counts
|
|
#
?
Mar 31, 2019 05:25
|
|