i generally dislike code reviews. they used to be something we'd opt to have only when we felt like our changes needed it. then one day everybody decided that mandatory code review should be a thing.
i suspect this comes from the OSS community and leaked into professional development. it makes sense in OSS where mandatory code reviews are essential: you're working with a larger team of mostly randos. fallout from bad changes falls on the maintainers and not the contributors, and therefore tighter control is essential.
but when you're in a team of 5-7 developers working on a product you're all responsible for the entire concept of code review is different.
mix in the usual personality types in the latter scenario and you generally end up with the following types of people:
---
- the "i will ignore your PR until i'm free and then 'lgtm'. if this takes a week well that just shows you how important my own work is so fuck you".
- the "i see your PR but i have changes in the same area so i'm going to merge my own changes before i event touch your PR. then you can deal with all the merge conflicts. and yes i did privately coordinate approvals of my PR ahead of yours."
- the "i pulled your branch and opened it in my IDE and tagged every single warning and suggestion the IDE made, pretended it to be my own. yes, i installed extra tools to increase the feedback. i think i'm a good code reviewer".
- the "honestly bro i don't care" type who will let others review and maybe make one comment just so they've been seen to review the code.
- the psycho: spends their entire time being toxic in the PR and starting arguments about things that don't matter.
---
"ah yes but" you might say "we have standards and patterns that must be adhered to. code reviews are essential for enforcing them".
yeah well... the truth is, when you're working in a team you have responsibility for that team's personal development and performance. if you are retroactively pointing out a failure to ensure the team understand the standards your team aheres to then you failed as a leader.
most code reviews are performative nonsense to pander to the egos of people who should be improving themselves and their colleagues, not reactively compensating for poor team development.
this is a hill and i am prepared to die here.
At a certain point we have to accept the ZIRP era of employment was incredibly abnormal. We can’t feel robbed when the job market isn’t as good as the best SWE job market in human history.
I hate AI when it's like this:
Notion used to have a one-click way to turn text into a quote. They removed this and now I have... AI
So instead of sub-500ms, it now takes 10+ seconds to turn selected text into a quote (type out the prompt, then wait ~4 seconds)
So backwards...
@GergelyOrosz An under explored counterargument to the AI maximalists is that large companies ALREADY shirk relatively proven developer productivity practices (CICD, blameless post-mortem, etc), so why would they be first in line to shake things up? Everything about their DNA forbids it.
Companies that won’t hire remote because of the perceived benefits of “in-person collaboration” are going to layoff their entire workforce for an army of digital robots. Sure.
@jasonmcgowan@kalanifsitake For younger fans, we respect LaVell, but we didn’t KNOW him.
Kalani is our LaVell. He has led this program to unbelievable heights. He is everything a man, a coach, and a disciple of Christ should be.
“If my son played football, I’d want him to play for Kalani.” -Matt Campbell
“OpenAI, as a single company, just needs to collect as much revenue as 1.5x total current spend on all software.”
Somehow, this is an argument IN FAVOR of OpenAI’s path to profitably!
Annual US spend on…
Software: $400B
Professional services: $2.8T
Logistics labor: $1T+
Healthcare admin: ~$1T
Education labor: ~$1T
That’s $6T+. Global is probably 3x.
Every market is up for grabs. ALL of them.
If you don’t get that, you aren’t ready for what’s coming.
$650B is a joke.
When you cultivate a reputation as a charlatan and fraudster (as @im_roy_lee has done) why would anyone believe you when you say you’ve “fixed” reliability issues in your product?
People desperately want “English is the new programming language” to be true because it will be the ultimate vindication of the “ideas guy.” But it will never be true, and good engineering will always require hard work.
For those saying "English is the new programming language."
Option A: "Make it dark green
... no, lighter
... no, a bit less blueish
... no, a bit more pinkish
... no, a little more greyish
... no, a hint darker"
Option B: use #7DB097
Which option is likely to stick?
The National Republican Senatorial Committee says Apple’s iOS 26, which is slated to be released in September, could cost them tens of millions of dollars in lost fundraising and also impact get-out-the-vote organizing opportunities.
The reason: iOS 26 includes significant message filtering features, so texts from political campaigns and committees will get filtered out into an unknown folder.
It’s fair to say Democratic committees and candidates will be impacted by this too.
The NRSC says 70% of small-dollar donations come from text messages now, so this could end up costing the GOP over $500 million in lost fundraising.
Via @PunchbowlNews
Overheard in America:
"Tyler, stop teasing Cooper. Go play with Hunter, Chandler, and Fletcher instead. And tell Carter, Sawyer, Harper, and Porter to come back and finish their lunch. Piper! Don't let little Dogcatcher run into the street!"