amusingimpala75

Going to be ironic if it's still chromium-based

Note: this can be replaced with a bash script and https://github.com/kcrawford/dockutil for free

Meta needs to stop open-washing their product. It simply is not open-source. The license for their precompiled binary blob (ie model) should not be considered open-source, and the source code (ie training process / data) isn’t available.

Intel has a fab nearby as well which is actually net positive for potable water because they take in waste water and purify it enough that it goes back into the supply after they’re done using it.

> DeepSeek-R1 has been making waves recently by rivaling OpenAI's O1 reasoning model while being fully open-source.

Do we finally have a model with access to the training architecture and training data set, or are we still calling non-reproducible binary blobs without source form open-source?

Monopolies are not inherently bad. They are only bad when they abuse that position to retain a monopoly or allow a decline in product quality.

Historically (last two elections) the polls have been about 2% further left than the actual result. Thus a 50/50 could/should b be interpreted in Trump’s favor

And by fast we mean 2+ minutes to go to a link and fill in four fields

That's false. See AdGuard for iOS

https://apps.apple.com/us/app/adguard-adblock-privacy/id1047...

I really liked Brave Search but unfortunately it started giving me captchas on every search after 11 pm Pacific or so for some (unknown) reason. They’re not even image captchas, you just click the button and wait 15-20 seconds which is pretty inconvenient.

Sure but under what license? Because slapping “open source” on the model doesn’t make it open source if it’s not actually license that way. The 3.1 license still contains their non-commercial clause (over 700m users) and requires derivatives, whether fine tunings or trained on generated data, to use the llama name.

I may be missing it, but this article mainly talks about the issues of pyrolysis and advanced plastics recycling. But it provides data supporting traditional recycling, citing a 55-85% success rate. Are there any other issues with traditional recycling, or do we just need to make sure that the plastic being recycled is using the better/older method?

Except that before today you didn’t even need admin for access to the database, any process that is allowed to read things could access the Recall database.

How much is the AI necessary for this? At least for the targeting of sounds in the line of sight, that should be fairly easy to do without AI, but I don’t know about the human voice identification.

iOS 17.5.1 has addressed the issue, attributing it to a database corruption issue not properly deleting it the first time

Just to clarify, no it isn’t theft. For most games and software that cost money, you never really buy it, but rather you’re buying a license to use it, with strings attached for rules on the possibility of being banned, of which there is usually some vague, overarching “if the company feels like it” or “if the government asks us to” clause.

Not saying that this is the way it should be, but at least read the ToS before agreeing to it.

Recently saw an article about parents spending hundreds of thousands of dollars to have their kids work with professional resume builders just to get an advantage with college applications. Now that many colleges are trying to phase out SATs because “it discriminates against poorer households”, it’s being replaced with essay and resume based things that seem to provide an advantage to the richer households who can afford the absurd amounts to very carefully cultivate the documents.

People complain about the green/blue bubble thing that Apple does, but I like to be able to see if the other user is using iOS, because SMS has no encryption whatsoever, and this iMessage shim that Beeper is using effectively opens the possibly of MITM the iMessage traffic without one’s knowledge. From this security perspective, I don’t have a problem with Apple cracking down on this misuse of iMessage.

Orion + FF or Chrome extensions already exists for iOS, implementing the extension interfaces on top of WebKit. Which part is the one you want, the extensions or Firefox? ‘Cause if you just want uBO on iOS you already can.

I think that monopolies are generally bad for the user, but I don’t think that government regulation is the proper solution.

Yeah this is what I was intending. I don’t have a problem with individuals using Chrome (I think it’s a bad choice overall but it’s their choice). However “YouTube works best in Chrome” and “GMail works best in Chrome” because they depend on some tracking API, it’s not really the choice of the user anymore.

Good thing WebDRM got delayed ‘cause I don’t want to install Chrome just to use my banking app.

I’m just as much a Firefox fan as anyone, and I use Firefox exclusively on desktop, even going around explaining the benefits, etc. to others.

That said, as much as I have complaints about Safari, if this spreads beyond the EU, that’s practically handing Chrome a monopoly on a silver platter. What’s to stop them?

I would just like to retroactively reply to the objection that the government can just antitrust regulate Google. Ignoring my general distrust of large-scale government regulation on a free market, I would like to point out how long it took the EU to implement these changes since Apple started these behaviors. If Chrome does obtain a monopoly, we’re looking at years before the government even does anything.

And yet after disabling uBlock Origin and uMatrix entirely, it puts me in the endless loop of reloading the page again and again each time after clicking the verification button. I'm writing this from Safari just because it wouldn't work with my Firefox setup with the aforementioned plugins

Just wait for M3 Ultra which in past generations is literally just 2 Maxs. Imagine 256 of GB unified RAM (oh and I guess the 32-core CPU and 80-core GPU, though those aren't as impressive). Really would love to know how much that is going to cost

Wikipedia may not be a citable source itself, but it can still be a good place to find links to other sources more reputable.

What are some good alternatives for actually FOSS LLMs? The leading LLMs would seem to be ChatGPT-4 and LLama 2, but OpenAI isn't open, nor is a license that restricts usage even just larger limits free, which Llama is licensed under (and people are buying Facebook's lies that it is actually open-source).

If it ain’t broke don’t fix it. No one is going to be happy if you’re upgrading your system to allow diacritics in the name and suddenly all their money disappears. There’s a reason they use COBOL so much and that is because they haven’t needed to replace it. Replacing it just because it’s old just seems like chasing the new ideas for the sake of it.

I might be misremembering, but isn’t Arc also a Chromium-based browser?

Maybe maybe it’s meaningless without Apple agreeing as things stand, because of iOS. But if the EU proposal about side loading forces Apple to allow Blink-Chrome instead of just WebKit-Chrome on iOS, then there’s not much holding Google back.