|
Pollyanna posted:People at my work place are making plans to jump ship. It's sinking pretty drat fast. Looks like it's time to crank up the , maybe I'll find a place with front-end engineers that actually understand that not everything works on all browsers. It's terrifying to think that I'm the most competent engineer in the room. If I remember where you work correctly, and given my wife's experience with said company, I'm totally not surprised. Good luck.
|
# ? Apr 18, 2017 18:28 |
|
|
# ? May 11, 2024 12:21 |
|
The Fool posted:If I remember where you work correctly, and given my wife's experience with said company, I'm totally not surprised. Good luck. She wouldn't happen to also be doing front-end there, is she? (And yeah, it's a mess.)
|
# ? Apr 18, 2017 18:30 |
|
Pollyanna posted:She wouldn't happen to also be doing front-end there, is she? (And yeah, it's a mess.) No, she worked as an agent for a while. Moved to a competitor.
|
# ? Apr 18, 2017 18:31 |
|
Pollyanna posted:People at my work place are making plans to jump ship. It's sinking pretty drat fast. Looks like it's time to crank up the , maybe I'll find a place with front-end engineers that actually understand that not everything works on all browsers. It's terrifying to think that I'm the most competent engineer in the room. I doubt such a place exists. Or that I would be hired by such a place. Only shambolic trash fires hire me.
|
# ? Apr 18, 2017 18:32 |
|
The Fool posted:No, she worked as an agent for a while. Moved to a competitor. Doesn't surprise me that it sucks all over. Oh well, hope she's doing better now! Hope I will, too.
|
# ? Apr 18, 2017 18:32 |
|
Another Q: is it considered kosher to decline a PR because it contains too many changes that are unrelated to the specific ticket in question? A basic parameterization ticket from my coworker is massive because it includes Jenkins changes, CSS tweaking, removal of a shitload of files, whitespace changes, React prop type changes, and some webpack insanity, and it's way too big for me to review. I want him to break the PR up, but in afraid of him getting mad at me if I do even though he declines our PRs for things like alphabetization. Is it okay to say "no, this poo poo is too big, please restrain yourself"?
|
# ? Apr 18, 2017 20:18 |
|
Pollyanna posted:Another Q: is it considered kosher to decline a PR because it contains too many changes that are unrelated to the specific ticket in question? A basic parameterization ticket from my coworker is massive because it includes Jenkins changes, CSS tweaking, removal of a shitload of files, whitespace changes, React prop type changes, and some webpack insanity, and it's way too big for me to review. I want him to break the PR up, but in afraid of him getting mad at me if I do even though he declines our PRs for things like alphabetization. Is it okay to say "no, this poo poo is too big, please restrain yourself"? That's a totally valid and correct reason.
|
# ? Apr 18, 2017 20:21 |
|
good jovi posted:That's a totally valid and correct reason. Yeah if you can't confidently review the PR (and there's almost certainly no way you can with that mess), politely declining it is the right move. That sounds like it has way too many ways to break something for comfort.
|
# ? Apr 18, 2017 20:27 |
|
revmoo posted:Wait, what project are you working on with 37 minute builds? Geez, in the future can you please preface all your small-team commit bullshit with "I can't imagine a build that takes more than 37 minutes"? Would really make these things go quicker.
|
# ? Apr 18, 2017 20:45 |
|
JawnV6 posted:Geez, in the future can you please preface all your small-team commit bullshit with "I can't imagine a build that takes more than 37 minutes"? Would really make these things go quicker. The Linux kernel needs like 8 minutes on decent hardware. 37 minutes is either a neural algorithm designed for HFT, modeling the human genome, or poo poo loving code.
|
# ? Apr 18, 2017 21:39 |
|
revmoo posted:The Linux kernel needs like 8 minutes on decent hardware. 37 minutes is either a neural algorithm designed for HFT, modeling the human genome, or poo poo loving code. Enterprise level business requires a robust amount of poo poo code to run poorly on our magnificent and amazing future technology.
|
# ? Apr 18, 2017 21:42 |
|
Now talk about Dunning-Kruger some more.
|
# ? Apr 18, 2017 21:42 |
|
Btw my last gig was writing Siem software backed by a 200TB hdfs cluster, several million LoC, commits going back to 2006, and I managed the entire CI and VCS infrastructure through several different revisions. So attack my ideas instead of my background.
|
# ? Apr 18, 2017 21:46 |
|
Ask me about baking multiple 20 minute AMIs in a row before I can see my changes in a production-like environment every time I push to a branch.
|
# ? Apr 18, 2017 21:45 |
|
revmoo posted:The Linux kernel needs like 8 minutes on decent hardware. 37 minutes is either a neural algorithm designed for HFT, modeling the human genome, or poo poo loving code. It's that and poo poo like this: revmoo posted:Look, I'm right, and several of you are dead wrong. And who spends three loving days in git bisect? It is literally divide-and-conquer. after 24 work hours you should be able to literally bisect tens of thousands of commits. It's totally fine that you've only worked in places that didn't have these kinds of constraints, but could you perhaps work on the utter disregard for tact and relentless condescension you're applying to folks whose work you're blissfully unaware of?
|
# ? Apr 18, 2017 21:49 |
|
JawnV6 posted:It's totally fine that you've only worked in places that didn't have these kinds of constraints, but could you perhaps work on the utter disregard for tact and relentless condescension you're applying to folks whose work you're blissfully unaware of?
|
# ? Apr 18, 2017 21:53 |
|
revmoo posted:You literally just did this exact thing to me. I implied that your workflow avoided a particular need for a flavor of testing. That's a peculiarity of that code base, not some global truth that must hold across Literally All Disciplines.
|
# ? Apr 18, 2017 21:55 |
|
JawnV6 posted:It's totally fine that you've only worked in places that didn't have these kinds of constraints, but could you perhaps work on the utter disregard for tact and relentless condescension you're applying to folks whose work you're blissfully unaware of?
|
# ? Apr 18, 2017 22:06 |
|
Maybe his tact is a WIP.
|
# ? Apr 18, 2017 22:14 |
|
CPColin posted:Maybe his tact is a WIP. Busy rebasing, brb
|
# ? Apr 18, 2017 22:19 |
|
Pollyanna posted:It's terrifying to think that I'm the most competent engineer in the room. When that happens and you're not running your own company, it's time to move on.
|
# ? Apr 18, 2017 22:47 |
|
revmoo posted:The Linux kernel needs like 8 minutes on decent hardware. 37 minutes is either a neural algorithm designed for HFT, modeling the human genome, or poo poo loving code. Just imagine if it were ported to C++, it'd take two hours to build.
|
# ? Apr 18, 2017 23:12 |
|
If it was ported to Haskell it'd take 15 hours to build after which the type system will have discovered 100+ security vulnerabilities and violations of contracts with userspace programs and correctness which are all absolutely true but nobody will ever actually encounter in the wild. After that, it will be shown through a TLA+ implementation of the kernel that the original presumptions of the Haskel program's type system were, indeed, incorrect and that most of those vulnerabilities are non-exploitable.
|
# ? Apr 19, 2017 00:56 |
|
My entire project takes 40 minutes to build from scratch. It's a complete embedded Linux build though.
|
# ? Apr 19, 2017 03:47 |
|
Last project I worked on had a 5 minute build type, most of this was setting up websphere liberty. The team is moving to tomcat now btw.
|
# ? Apr 19, 2017 10:10 |
|
Our underpointing has gotten a little out of hand. 4 points is a week, 2 points is more than one day, and 1 point is anything less than that. So we have 30 minute stories pointed the same as 4 hour stories, which is making our velocity erratic since the same dev is doing 1-5 points in a day. We need to reset our points lower but that's going to be an effort.
|
# ? Apr 19, 2017 12:23 |
|
smackfu posted:Our underpointing has gotten a little out of hand. 4 points is a week, 2 points is more than one day, and 1 point is anything less than that. So we have 30 minute stories pointed the same as 4 hour stories, which is making our velocity erratic since the same dev is doing 1-5 points in a day. Points are dumb. Sprints are dumb. I loving hate modern development cargo culting.
|
# ? Apr 19, 2017 12:37 |
|
On the plus side, no one is making me do hour estimates of an entire release before we write a line of code.
|
# ? Apr 19, 2017 12:40 |
|
smackfu posted:On the plus side, no one is making me do hour estimates of an entire release before we write a line of code. Points may get dumb sometimes but I promise it's better than the hell described above. "Looks like 400 hour project was finished in 463 hours. We're really aiming to keep projects within 10% of the estimate."
|
# ? Apr 19, 2017 16:11 |
|
"Ship something every day' is a really good goal to strive for, but stories like this always make me wonder whether the problem lies within the process of estimating, or whether handsy project managers keep dividing stories into units of work that are too small in the first place.
|
# ? Apr 19, 2017 16:15 |
|
Also ask me about babysitting Jenkins' buggy broken rear end through our overengineered build and deployment process and manually clocking APPROVE SPPROBE SBBPORRD as my entire loving day I hate you Jennings or whoever set you up,oh right it was the guy who went in vacation the week of release requiring us to scramble to fo the shut he has to do and is barely available abyway cause he goes drak like constantly please help me I do not like being a babysitter.
|
# ? Apr 19, 2017 17:11 |
|
Pollyanna posted:Also ask me about babysitting Jenkins' buggy broken rear end through our overengineered build and deployment process and manually clocking APPROVE SPPROBE SBBPORRD as my entire loving day I hate you Jennings or whoever set you up,oh right it was the guy who went in vacation the week of release requiring us to scramble to fo the shut he has to do and is barely available abyway cause he goes drak like constantly please help me I do not like being a I think someone broke the markov chain generator...
|
# ? Apr 19, 2017 17:29 |
|
Pollyanna posted:Also ask me about babysitting Jenkins' buggy broken rear end through our overengineered build and deployment process and manually clocking APPROVE SPPROBE SBBPORRD as my entire loving day I hate you Jennings or whoever set you up,oh right it was the guy who went in vacation the week of release requiring us to scramble to fo the shut he has to do and is barely available abyway cause he goes drak like constantly please help me I do not like being a And Jenkins claims another victim.
|
# ? Apr 19, 2017 17:43 |
|
I hated Jenkins the first time I used it for CI. The second time around I took a week and hacked the poo poo out of it during a big OSS push where we were ditching Bamboo/Atlassian. I got it running absolutely beautifully, and cranked out a theme that put Bamboo to shame to go with it. It's a great platform but it needs careful and methodical setup to start with. One thing I always do now, is put my entire CI process into shell scripts, depending on Jenkins for just Git, credentials, and artifact generation, and nothing else. A nice bonus is that you can switch CI platforms in an afternoon since all of the build logic lives outside of it.
revmoo fucked around with this message at 19:29 on Apr 19, 2017 |
# ? Apr 19, 2017 19:26 |
|
Also, protip; generate a 1mb file, commit it to Git, and then put the md5 of it in your deploy script as a canary. Abort deploys if the hash ever fails to match.
|
# ? Apr 19, 2017 19:28 |
|
Dare I ask why that ever came up as a concern?
|
# ? Apr 19, 2017 19:29 |
|
revmoo posted:Also, protip; generate a 1mb file, commit it to Git, and then put the md5 of it in your deploy script as a canary. Abort deploys if the hash ever fails to match. this is insane cargo cult nonsense
|
# ? Apr 19, 2017 19:35 |
|
One weird trick to ensure your build fails for inscrutable reasons years after you leave the company, employers and devs hate it!
|
# ? Apr 19, 2017 19:40 |
|
That would make for a good prank once I leave.
|
# ? Apr 19, 2017 20:10 |
|
|
# ? May 11, 2024 12:21 |
|
Clanpot Shake posted:One weird trick to ensure your build fails for inscrutable reasons years after you leave the company, employers and devs hate it! But is your build time under 37 minutes?
|
# ? Apr 19, 2017 20:10 |