misnome

I’d infinitely prefer a relevant wikipedia article to an AI “Answer” that is almost always wrong.

Google lens image search used to be amazing, I tried a repeat of a search I did before of a piece of art, it showed the same piece but confidently listed the artist and year wrong by about 300 years.

I’ve had relatives do “research” about things I mentioned I needed to do, and they’ve just sent screenshots of the incorrect AI answer.

It’s made google almost entirely useless, there is zero incentive for them to try to make search better (vs incentive to make it worse) and even if they did want to make it better the sheer volumes of slop have made that even harder.

We’ve completely sabotaged out ability to collate information at scale as a civilization, for the benefit of a few companies that were already the largest in the world to begin with. And it turns out, very few people notice or even care about this.

> used to be

I mean Andrew is an extreme case. If he weren’t as out-of-favour I imagine nothing would have happened, and this has been _entirely_ forced by external information.

I assume that otherwise they would have less of an issue. It’s not like he married someone slight off-white, that would be real grounds for excommunication.

> And I see it doing more planning and making fewer mistakes than before

Anecdotally, maybe this is the reason? It does seem to spend a lot more time “thinking” before giving what feels like equivalent results, most of the time.

Probably eats into the gambling-style adrenaline cycles.

> How many degrees of temperature increase is worth one .. whatever the unit of AI gain-of-function is?

Billionaire. And they are definitely willing to make the trade.

But “PhD level” reasoning a year ago.

> That rule is not optional, and it was not followed here.

It’s not optional, but wasn’t followed, with zero repercussions.

Sounds optional.

OTOH I find it pretty funny that the instant they manage to make a model that breaks general containment of popularity and usefulness (4.5), the toxicity of the business model kicks in and they instantly enshittify.

Yes, I’ve been evaluating since the start of the year and since 4.6 suddenly the most innocuous requests will sit there “thinking” for 5+ minutes and if I can get it to show me the thinking it’s just going round in circles.

Or, it decided it needs to get API documentation out and spends tens of thousands of tokens fetching every file in a repo with separate tool use instead of reading the documentation.

Profitable, if you are charging for token usage, I suspect.

But I’m reaching the point where I can’t recommend claude to people who are interesting in skeptically trying it out, because of the default model.

And their status history isn't much better. It's just that they are so much smaller it's not Big News.

When did Jack Dorsey unban personal friends of his that had gotten banned for posting CSAM?

10% revenue to google?

If only there was an extremely high profile example of how this isn’t at all true in the past several days!

Funnily enough, that’s also all the people who made the rules in the first place knew

We do rendered manifest pattern. The chart gets rendered into a single yaml, that get checked into it's own branch and PR. That way, any changes can be easily inspected before merging and can work with confidence that e.g. changing a setting or updating isn't going to change ALL of the objects. It's also extremely easy to (trustingly) roll back to previous states.

The only downside is that you can't really prune excess objects with this method. We're pushed to use Argo for deployment which I don't really gel with, but I trust it to apply the yaml, and at the very least it highlights when objects need to be removed.

Yes! Sorry Kitty for mixing you up!

I have various issues with it (e.g. wrapping on resize is just broken) and miss iTerm a little, but the built-in tabs aren’t too bad (unlike Kitty’s hardline stance) and it’s available cross-platform, so I can have the same config on many machines.

Read your own post?

> Off-Topic: Most stories about politics > If they'd cover it on TV news, it's probably off-topic.

I shouldn’t be surprised when people flag such things.

It’s not really anything to do with tech though, is it?

“This doesn’t belong here” is perfectly valid reaction to stuff covered on a million other sites.

Congratulations, you know the local weather

Everything up to (and including) Unreal Tournament had software rendering. It was one of the selling points when its competitor (Quake 3) was Hardware-accelerated-only.

> "Why Git Does Not Use SQLite"

For one, you probably don't want your git repo permanently corrupted if you happen to be running it over an NFS mount.

There is a lot of value in "scattered write-once files" as a storage mechanism.

I spent some time working with jujitsu and found it relatively pleasant... except it doesn't seem at all ready for collaborative use beyond a single working folder.

Sharing the repo state seems impossible without going through the (somewhat painful) multi-step process to tag and push commits into git, and you lose all of the value that jujitsu actually adds (e.g. no evolog, no saved conflicts, no anonymous commits). There is no sharing solution beyond this - no way for me to work on the same working state in two locations beyond rsync the entire folder tree, which is no solution.

A SCM in 2026 that only works locally without being a clunky way of talking to git doesn't seem like a solution ready for prime-time. Maybe the google-internal-only backend systems work better.

My main other pain points were a) zero integration with pre-commit hooks and zero ideas how to do so (yes, CI does this, but it's nice to get all the auto-formatting and sanity checks without pushing, waiting, pulling new commits), and b) Automatically picking up all changes is great except when it doesn't work in which case it is _horrible_ and takes manual unpicking at best. Remembered to add something to .gitignore before doing something in the working directory? Great! Just never ever try to checkout a commit from before you added the .gitignore because now it's permanently absorbed into the repo. Also, if you every accidentally have a secret in your repo directory then that's also permanently in there, as far as I can tell there is no way to manually verify that things are purged and all the commentary on the jujitsu discussion forums were "Just avoid doing this in the first place".

FS will literally sell you heavy-duty Armored (e.g. thicker/stronger sheath) cable and the packet it comes in will be labelled "military grade". That's literally your scenario.

Is one supposed to send it back for a refund and order the much thinner, less-durable cable? Or is perhaps the landscape not as black-and-white as your "this is automatically snake-oil"?

> strangely and disproportionately pushed on Hacker News

There is literally nothing strange or disproportionate. It's incredibly obvious that new languages, that were designed by people who found older languages lacking, are of interest to groups of people interested in new applications of technology and who want to use their new languages.

> then those from outside the project shouldn't really have any say on it. It's their project.

People outside the project are allowed to say whatever the hell they want, the project doesn't have to listen.

I think this is a different, new rename? It was renamed last year, but I don't recall seeing office in it's entirety being called "The Copilot App"?

Even so, the borrow checker repeatedly profiles as an insignificant part of compile times, so wouldn’t make a difference.

When does this cease to justify any possible retribution? How many murdered palestinian children, or emergency workers, or aid workers balances this out? How much torture of prisoners?

It’s honest. Time was when the number of improvements in every release was impressive and sometimes radical improvements.

For at least a couple of years it’s been nothing but AI, I am happy to ignore updates and should probably just turn them off now.

Oh christ, absolutely this. We spent some time evaluating FPGA for our purposes and ended up GPU instead (algorithms we running can be adapted to strength of either).

The concepts are easy enough but learning the toolsets are an exercise in frustration… the documentation/onboarding is either nonexistent or extremely unhelpful, and getting past the stage of “the entire thing doesn’t work because you misclicked a button in the gui several hours ago”. In theory everything can be scripted, usually in TCL, but this is also unstable and seems liable to break every different version of the toolsets.

Alongside Xilinx, we also looked at Altera/Intel OneAPI/dpcpp and this seemed promising until we realised we were encountering so many toolchain/actual compiler bugs that nobody else could have been actually using this, except the oneapi cloud platform that seemed it had been hotpatched to fix some of the issues. In the end, after selling us some compatible cards they dropped the OS and card from support. I guess this taught us not to trust Intel!

We decided teaching to Juniors would be an exercise in frustration unless hiring explicitly for, and decided to go the GPU route.