Granted some human will likely have to review it. Or packages flagged by Al could require users to explicitly allowlist them.
4,987 karma · joined April 6, 2016
Granted some human will likely have to review it. Or packages flagged by Al could require users to explicitly allowlist them.
That's a really good description, I should update my linkedin profile :)
But often results are incremental and obvious.
But also if you jumped from A to A''''', nobody would understand why or what it relates to.
IMO the concurrency model in go is the biggest reason, I'd hesitate to use it.
Managed memory, single threaded with lots of lints and good tooling. Is IMO what can raise my confidence in code, before I even review it.
Granted golang has a really good stdlib. Which counts for a lot.
Example: a plugin architecture with a solid API that makes degenerate behavior limited to usages of said plugin.
If taste is maintainability, etc.. the no quite the opposite.
The startup that has no taste and vibecodes mindlessly might end up with a codebase they can't maintain or refactor. A product that can't advance.
But honestly, I don't feel like the toil is gone. Landing code in big projects remain hard.
But toil is not gone. We have still have code owners, lawyers, physical hardware, thousands of configurations that we can't test them all.
And notably: we will still be asked to write software we can maintain in the future.
Sure, there are short-lived non-critical software projects where LLMs will rule.
But for many big complex projects, you can't land code without reviews. You can argue its because taste is already being enforced in the name of maintainability.
The dashboard clearly says 89.15% uptime!
Who says nines need to be leading?
NP problems gets solved with heuristics every day.
And what is the failure mode?
Doubtful it'd hold in court; this admin would have to show that it's not corruption, because we'd all assume otherwise.
But yeah, writing specs is usually harder than reviewing the code 4 times :)
Sentences are not added up. So yes, trying to shot a police officer will definitely get you decent sentence -- it's not hefty by American standards.
Regardless, this is enormously dumb. If you want to search and arrest an activist who crosses the line, you make it as boring as possible.
Shipping is very important, sometimes more important than what you ship :)
Ideally also ability to drag faces like plasticity.
I'm not sure what kind of special torture you'd employ to make people display that kind of enthusiasm :)
Mythos being drummed up as a cyber weapon is not the same as it being a weapon -- and treating it as such is just dumb.
Yes, it may find vulnerabilities, but "find" is the keyword -- whatever it finds was already there.
My point is: Remember to enjoy your dreams. And 99% of the time let them be just that: "dreams".
CS 1.5, OpenTTD, AoE2, the likes..
Fancier graphics doesn't make games better. Gameplay, stories and jokes do.
Spying on employees is not free. If you want to spend serious resources doing it, there has to be an upside.
It's not anymore, right?
The idea that officials aren't personally liable for mistakes made in good faith isn't bad. But somehow the US tends to produce a lot of cases where good faith requires a lot of faith :)
But, yes, I would rather not have an outbreak of ebola.
If their license is void, then distribution is illegal, right?
If Josef really wants to go after BambuLab over AGPL violations he can do what the music/movie industries did: have them blocked at the ISP level.
Cease and desist for all servers outside China. Block traffic at ISP levels.
Unlike various pirate websites there won't be a hundred copy cats to also hunt down.
But when higher prices stick around industries close or never opened.
Stop there. The costs of today state-of-the-art hype-ware will be lower tomorrow.
In a year, there will be fewer easy to find bug, and models like this will be cheaper.
Doing math based on hardware deployed today is a bit naive.
Assuming any of this is intentional about reducing devs is just dumb.