But shouldn't it be 8 < 1 because the eight is heavier and squeezes the bars of the = together?
- Posts
- 9
- Comments
- 743
- Joined
- 3 yr. ago
- Posts
- 9
- Comments
- 743
- Joined
- 3 yr. ago
It's mostly about throwing ACID at the problem, sqlite just happens to be battle-tested to a ludicrous degree, it's light enough to not be unconscionable overhead in simple situations (unless you're on embedded), and performant enough to also deal with nastier situations so I prefer it over some random K/V store with the same guarantees. It's also a widely-used and stable data format which might come in handy.
That said, if you want to go lightweight do consider good, ole, POSIX filesystem guarantees, in particular that mv is atomic (as long as you stay on the same filesystem but that's easy to ensure by mv'ing within a directory). That's not durable on its own, you'll need to fsync for that, and consistency and integrity is up to your code.
Easier to do than to get never-exercised edge-case code to work flawlessly. Are you sure you can't just throw sqlite at the problem? It's often overkill but, hey, it's there on the shelf, might as well use it and I've seen it out-perform hand-rolled data structures. Non-persistent ones, written by very confident C coders. And remember crashes are unavoidable, if nothing else then someone can trip over the power cord.
Crash-only software. To be resilient you need some kind of ACID anyway which means that you can let go of your shutdown procedure and just send yourself SIGKILL instead.
- JumpDeleted
Permanently Deleted
We essentially do have the death penalty for corporations, it's called being declared a criminal organisation.
- JumpDeleted
Permanently Deleted
Unlikely, I'd say, In EU jurisdictions copyright requires creative authorship, not "sweat of the brow" which is why by default databases aren't included, which is why they're have their own protection regime.
Quote, emphasis mine:
In the meaning of the European Union Directive 96/9/EC on the legal protection of databases,the term database refers to a collection of independent works, data or other materials, which have been arranged in a systematic or methodical way, and have been made individually accessible by electronic or other means. In the meaning of the Directive the data or materials:
- must not be linked, or must be capable of separation without losing their informative content;
- must be organised according to specific criteria, which means that only planned collections are covered;
- must be individually accessible – mere storage of data is not covered by the term database.
In AI models the organisation is inferred from the data, it's not planned into the database. The first bullet point is on less shaky, a summary an AI can make of a book can reasonably be regarded to be "informative content", nothing about db protections says that they have to store full works it could also be references, citations, etc.
- JumpDeleted
Permanently Deleted
AI is right-out unregulated in the EU unless and until you actually use it for something where it becomes relevant, then you've got at the lower end labelling requirements (If your customer service is an AI chat, say that it's an AI chat), up to heavy, heavy requirements when you use it for stuff like sifting through job applications. The burden of proof that the AI isn't e.g. racist is on you. Or, for that matter, using to reject health insurance claims, I think we saw some news lately out of the US what can happen when you do that.
OpenAI's copyright case isn't really good to make the legal situation any clearer: We already know that using pirated content to train stuff isn't legal because you're not looking at it legitimately. The case isn't about the "are computers allowed to learn from public sources just as humans are" question.
- JumpDeleted
Permanently Deleted
OpenAI hasn’t disclosed the datasets that ChatGPT is trained on, but in an older paper two databases are referenced; “Books1” and “Books2”. The first one contains roughly 63,000 titles and the latter around 294,000 titles.
These numbers are meaningless in isolation. However, the authors note that OpenAI must have used pirated resources, as legitimate databases with that many books don’t exist.
Should be easy to defend against, right-out trivial: OpenAI, just tell us what those Books1 and Books2 databases are. Where you got them from, the licensing contracts with publishers that you signed to give you access to such a gigantic library. No need to divulge details, just give us information that makes it believable that you licensed them.
...crickets. They pirated the lot of it otherwise they would already have gotten that case thrown out. It's US startup culture, plain and simple, "move fast and break laws", get lots of money, have lots of money enabling you to pay the best lawyers to abuse the shit out of the US court system.
You know there's a seek bar on youtube videos?
Notably, he's also the inventor of the magneto-turboencabulator.
- JumpDeleted
Permanently Deleted
Another one was GPS: They had prepared two sets of maths for the satellites, Newtonian and relativistic. They started operating them with the Newtonian model, and the satellites went out of sync, nothing really worked. Then they flipped the switch to relativistic, and everything worked flawlessly.
Even before that they took an atomic clock, put it on a plane, and flew it around the earth to later compare to one that stayed on earth. They differed by the expected fraction of a fraction of a millisecond.
Neither of those two could be done right when Einstein proposed relativity, but experiments like that could already be envisioned, "move a sufficiently precise clock sufficiently fast and compare it to a stationary one" is kind of a no-brainer. That's not the case with string theory, noone has any idea how to test any of it.
OTOH, physics shouldn't feel bad about that stuff. E.g. number theory is notorious for results which are considered useless even by the people formulating them, only for an application to appear a century or two later.
Servo should be vastly more advanced and it's nowhere near ready.
Sure it makes sense: Pretty much noone, but you, is going to buy them, and stocking shelves and warehouses with product costs money. All that unmoved stock would make them more expensive, making even more people not buy them. It's inefficient.
the M.2 form factor drives that everyone is hyperfixated on for some reason
The reason is transfer speeds. SATA is slow, M.2 is a direct PCIe link. And SSDs can saturate it, at least in bursts. Doubling the capacity of a 2.5" SSD is going to double its price as you need twice as many chips, there's not really a market for 500 buck SATA SSDs, you're looking for U.2 / U.3 ones. Yes, they're quite a bit more expensive per TB but look at the difference in TBW to consumer SSDs.
If you're a consumer and want a data grave, buy spinning platters. Or even a tape drive. You neither want, nor need, a high-capacity SSD.
Also you can always RAID them up.
Not sure whether we'll arrive there the tech is definitely entering the taper-out phase of the sigmoid. Capacity might very well still become cheaper, also 3x cheaper, but don't, in any way, expect them to simultaneously keep up with write performance that ship has long since sailed. The more bits they're trying to squeeze into a single cell the slower it's going to get and the price per cell isn't going to change much, any more, as silicon has hit a price wall, it's been a while since the newest, smallest node was also the cheapest.
OTOH how often do you write a terabyte in one go at full tilt.
I'm not even Christian what would I care about depictions or inspiration being "problematically Pagan".
What I can say is that it's unlikely that much of the parallels (like the Osiris thing) existed before Rome became Christian as over in staunchly monotheist Palestine people wouldn't have taken inspiration like that, while turning multiple gods into one sounds quite reasonable for a people going from polytheism to monotheism.
For modern Christian, I think, the question is "How much did the Romans change". That is, how different was early, pre-Roman, Christianity to what's now considered authoritative, like the Bible, which wasn't brought down from a mountain by Jesus.
Sol Invictus, in particular it's also where the halo in depictions comes from... which isn't really "other sun gods", it's in particular the Roman sun god. Misremembered the resurrection part, that's Osiris who isn't a sun god and the Horus parallels have been shown to be bunk, aside from getting nursed by Mary depictions being inspired by Horus getting nursed by Isis.
"Knowing where people come from" does not imply ID'ing individual people, which is why I specifically mentioned that Mozilla technology. The legitimate interest is in aggregate data, and yes "lots of people come here from the brothel" is legitimate data. "This particular person did" is not: If you wear a suit and happen to come with the office crowd doesn't mean you're an office worker, you could be a travelling salesman.
Shouldn't it be possible to only do the negotiation part and otherwise bridge everything? Not having to do anything high-bandwidth actively should keep the silicon costs down.