gpm

25 ms·

511.
by gpm 9mo ago
Using it as a reference is a high bar not a low bar. The AI videos aren't trying to be accurate. They're put out by propaganda groups as part of a "firehose of falsehood". Not trusting an AI told to lie to you is diffe
512.
by gpm 9mo ago
If GP's prediction that no one is buying ICE cars 5-10 years from now, and the average car is 12 years old, by 20 years from now we're down to less than half the current ICE fleet by natural replacement. But the replacement isn&#x
513.
by gpm 9mo ago
Neither - it's a triumph of our ability to do increasing complex things in both software and hardware. An earbud should be able to make good use of the extra computing capacity, whether it is to run more sophisticated compression savin
514.
by gpm 9mo ago
No, I don't think so, not unless there's some feature of Haskell type classes I'm completely unaware of. If anything it's closer to SFINAE in C++ where it tries to implement methods but then doesn't consider it an e
515.
by gpm 9mo ago
It isn't really type inference. Each closure gets a unique type. Rather it's an automatic decision of what traits (think roughly "superclasses" I guess if you aren't familiar with traits/typeclasses) to impleme
516.
by gpm 9mo ago
I've been using /rewind in claude code (the terminal, not using vscode at all) quite a bit recently without issue - if that's the feature you're asking about. Not discounting at all that you might "hold it" dif
517.
by gpm 9mo ago
I have encoded rules against this in CLAUDE.md. Claude routinely ignores those rules until I ask "how can this branch be reached?" and it responds "it can't. So according to <rule> I should crash instead" an
518.
by gpm 9mo ago
An expectation of professionalism, training and written material on software design, providing incentives (like promotions) to not produce crap, etc. It's not a world where everything produced is immediately verified. If a human consis
519.
by gpm 9mo ago
> with tedious manual checks for specific error conditions And specifically: Lots of checks for impossible error conditions - often then supplying an incorrect "default value" in the case of those error conditions which would r
520.
by gpm 9mo ago
PS. In the 5 minutes between starting and finishing writing the parent comment https://claude.ai/settings/usage just stopped displaying my quota usage... fun. Edit: And 3 minutes later it is back...
521.
by gpm 9mo ago
Not my experience at all when I occasionally try making something purely coded by AI for fun. It starts off fine but the pile of sub-optimal patterns slowly builds towards an unmaintainable mess with tons of duplication of code, and state t
522.
by gpm 9mo ago
I have an impression that the two I named (cocroachdb, influxdb) are actually commercially successful, but I could be wrong.
523.
by gpm 9mo ago
> close to the price of the underlying metal Never? Never for plastics either. It seems like there's always going to be a lot of cost in shaping these materials through carefully controlled very high temperature environments. On the
524.
by gpm 9mo ago
Is there any reason to believe there would be any? My understanding of PFAS is that they are used in the application of various coating like things (teflon originally, also since then waterproofing applications, paints, makeup, and firefigh
525.
by gpm 9mo ago
I feel like there's numerous database companies that rewrote an existing database faster/with slightly better features and turned it into a successful product. Just about all of the successful ones really. It's a market where
526.
by gpm 9mo ago
> the description as "the next evolution of sqlite" is offensive That marketing is really the one thing that keeps me from considering this as a serious option. To callback to an article a few days ago, it's a signal of di
527.
by gpm 9mo ago
MVCC is a non-locking algorithm for concurrent writers that the big databases like postgres use (with caveats like aborting some transactions if conflicts would exist). It's not a matter of pushing locks around but allowing multiple th
528.
by gpm 9mo ago
I assume old failures aren't kept in the context window at all, for the simple reason that the context window isn't that big.
529.
by gpm 9mo ago
As I understand the LGPL - not a lawyer - you have to somehow enable all your users to relink your application against a different version of Skip (4.d.0 since 4.d.1 isn't possible on iOS). This means that your application must do some
530.
by gpm 9mo ago
Huh, what does one have to do to comply with the LGPL on iOS anyways? I'm sort of surprised that only the largest plan ($5000/month) and not the ($10/$500/$2,500/month plans) includes a license that doesn't inv
531.
by gpm 9mo ago
It doesn't look to me like Valve delegated triaging bug reports though, rather triaging security reports. It seems fair to me that the security reporter vendor triaged this as not a security report. It feels like saying "the w
532.
by gpm 9mo ago
Do I misunderstand that to be HackerOne staff - not Valve staff - marking it as "not a security vulnerability" - not "won't fix"?
533.
by gpm 9mo ago
It doesn't matter what the original purpose of the terms was it matters what they do.
534.
by gpm 9mo ago
See term D.4., the relevant part of which is > You grant us and our legal successors the right to store, archive, parse, and display Your Content, and make incidental copies, as necessary to provide the Service, including improving the S
535.
by gpm 9mo ago
Not a lawyer but I'm pretty sure: 1. Microsoft does not gain the license, but will be able to argue that they aren't intentionally committing copyright infringement in the cases where that distinction matters. 2. If Microsoft do
536.
by gpm 9mo ago
Yes - we agree on how the pricing and odds work now :) The hitman shouldn't expect to capture the value of the entire open interest. The market here is serving to negotiate the bounty with speculators betting that too much was offered
537.
by gpm 9mo ago
> I think that means 4,500 shares, or $450,000, is on the "No" side and 500 shares, or $50,000 is on the "Yes" side. Do I have that right so far? No - there's always an equal number of contract outstanding on bot
538.
by gpm 9mo ago
> So unless they have big money to wager, they don't have big money to gain. It requires that they put down collateral (the purchase of the the yes bets) that they lose if they don't meet the contract, so they do have to have s
539.
by gpm 9mo ago
> Why go through the "prediction market" at all then It's there. It's not actually easy to find hitman for hire. This is a publicly advertised market for it. Plausible deniability. We weren't paying for the witne
540.
by gpm 9mo ago
It's stayed perfectly flat across Trump taking office? Consistent with the tariffs not changing the price at which China sells goods at all and the US consumer bearing the entire cost of the tariffs. Click to, say, 3 year view of http

More ›