could you expand on that?
I exist or something probably
could you expand on that?
programmer linguistigs is certainly something to behold.
read where?
Just going to point out: russian culture doesnt use nazi in the same way we do; largely they dont view nazis as ontologically bad because of the bigotry or the antisemitism or the genocide of many peoples, but because they were an existential threat to russia or russian ethnicities. a lot of the associations westerners have about nazis are just not widespread in russia. this is why there are a lot of seemingly idiosyncratic phrasings in this kind of stuff.
nothing you’ve said is worth responding to in a novel way, see earlier comment.
“when they’re drunk at 8am”, he didnt say “we”
also note the other quotes in the article that similarly single out Crow.
see certain optical illusions in a way that other vision models cannot.
eh… but not in a way that is really like what humans see. which is the articles claim, but it makes a clasically cs approach to nuerology: zero effort to prove the quite substantial claim.
Word soup
that is most certainly not word soup. it’s also an accurate statement, though uncharitable to the authors claims.
Also, the detail in description of their “quantum” inspiration (an effect not unique to quantum mechanics in fact, at that level of description) reads like they skimmed wikipedia’s intro to xyz topic, whether or not the author understands the topics more deeply.
removing unnecessary abusive misogyny != puritanism. hilariously different things in fact
why care that they removed it at all?
sorry are you saying people should pronounce their own names in ways they don’t prefer to be “correct”? Also etc etc language guides are descriptive not prescriptive.
I don’t know snough about the crowdstrike stuff in particular to have much of an opinion on it in particular, but I will say that software devs/engineers have long skirted py without any of the accountability present n other engineering fields. If software engineers want to be called engineers, and they should, then this may be an excellnt opportunity to introduce acccountability associations and ethics requirements which prevent or reduce company systemic issues and empower se to enforce good practices.
If they can target the underlying architecture of the models like nightshade does, it will actually be quite hard to deal with for the surveillance companies.
Trilium is good
Datas is correct if you are referring to multiple distinct populations of data. Which in this case works.
afaik no known cases
Did you actually look?
Sibling in existence I know asbestos must be airborne. You aren’t refuting anything by repeatedly saying that. Respond to the words I am saying or I can only assume you are copy pasting talking points.
It’s extremely easy to disturb asbestos, it does not take a large chronic exposure to get health consequences, it takes a very small amount of acute exposure or even less chronic exposure. Generally you will be fine from incidental one-off exposures, but if you live in a home with say, asbestos tiles in your kitchen, or asbestos in the paint or drywall, it can be very easy to build exposure from reno or damage from normal home wear. Not to mention it’s extremely expensive to modify because of the required controls, meaning it disproportionately effects low income households, who both struggle to afford preventative maintenance, and struggle to afford the reno.
There’s a reason asbestos ppe is decon controls roughly equivalent to mercury, lead, and beryllium.
Removed by mod
The internet as the internet companies percieved it would look like and sold it as absolutely and completely vanished, yeah.
also the answer to that question, shitloads of data for a better ai, is yes… with logarithmic returns. massively underpriced (by cost to generate) returns that have questionable value statement at best.