itsnotlupus

> you can't express emotions if you don't have any

That feels off. When I watch an actor on screen conveying emotions, there's no actual human being feeling those emotions as I watch their movie. Very dumb machines have already been rendering emotions convincingly for a while in that way, and their rendering impacts our own emotional state.

Emotions expressed through tone of voice are just one mean of nonverbal communication. We should expect more of those to develop and become more widely available next.

In a way, we're lucky all that gpt-4o seems to be hell bent on communicating how cheerful and happy it is so far, because it's certainly not the only option.

Humans can be manipulated through nonverbal communications, in a way that's harder to consciously spot than through words, and a model that's able to craft its "emotional output" would not be far from being able to use it to adjust its interlocutor or audience's frame of mind.

I for one look forward to the arrival of our increasingly charismatic and oddly convincing LLMs.

Fair points, within the context that IP rights are fundamentally made up and shifting rights crafted to balance the concerns and interests of various parties.

I can't tell where this is going to land. Right now, we're seeing a number of parties trying to put metaphorical barbed wire in the newfound prairie of ML models, each struggling to influence the prevalent wisdom of how IP rights should apply to this context.

We could easily end up in a universe where LLMs are a licensing minefield where every copyright owner of any part of their training data gets rights on the model, become essentially unmanageable without relying on helpful licensing middlemen that smooth out the right for LLMs to exist, at a cost.

We could just as well end up with LLMs being recognized as not being derivative works themselves, albeit able to generate derivative works, a much less advantageous situation for creatives and their middlemen who see their creative output as being pirated, to reuse a familiar term of IP rights propaganda.

It'd be a little surprising to me if we ended up in a situation where the work needed to produce good LLMs wasn't associated with any commensurate IP rights on the results, and I expect the megacorps investing billions into this will find it in their heart to throw a few millions toward lobbying efforts to ensure that this isn't the outcome.

Cracking is about removing anti-features, typically copy protections. It pairs well with pirating, but neither one strictly requires the other.

Notably that show features a gargoyle who seems very proud to be walking the streets with all their cool hardware.

With the Vision Pro, Apple is going to let us become the prettiest gargoyle at the ball.

New GPT-4-powered business model just dropped.

I think scraping websites that don't require authentication is legal in the US nowadays.

Monetizing data acquired through scraping could run afoul of copyrights, so the love letters would probably only arrive once OP attempts monetization.

By far the best feature of the minitel 1B for a kid like me with no budget was that it was free, as long as you could get your hands on one.

I have fond memories of endlessly annoying my parents and the local post office in my quest to get that model, the only free one with a plug i could connect to my PC. Good times.

Chrome sets it if you tell it you want to use its spellchecker with some languages.

I'm not sure what the standard says, but i suspect you'd surprise users considerably less if you used the first language listed in that header rather than the last (assuming equal or missing q values).

Right. Once upon a time, it used to be a decent resource for computer security-oriented news items.

Speaking of python things, there's an actively developed JavaScript implementation of tensorflow called tensorflow.js, which has its own set of backends to leverage GPUs in browsers or in node.js though webGL, webgpu or node bindings into c++ stuff to get cuda support, alongside wasm and pure js implementations.

https://github.com/tensorflow/tfjs

Hmm how long before we start to see companies extracting money from ransomware gangs by deploying honeypots and selling access to them?

Imagine getting blacklisted by ransomware gangs for being an untrustworthy company.

It's definitely there, and updated often:

NewPipe (Lightweight YouTube frontend) - https://f-droid.org/app/org.schabi.newpipe

Right. Criminal prosecutors, like many other criminals, are the product of their environment.

Unfortunately their jsdoc stuff only exposes a subset of their typing capabilities, so you're leaving some advanced type goodies on the table.

They do. I used that "feature" by accident recently. I think one of the accounts was a shopping account, the other started as an AWS account. Both accounts have the same name, same billing address, same credit card.

I think the logical next step is to give them the same password and see how bad my foot hurts afterwsrd.

There's a csrf token set in a php session that's hard to guess that needs to be provided in form data, yes.

Probably difficult to exploit that way without first finding another bug to retrieve that token from a random origin.

I expect we'll find out that Libra's default set of trusted validators will be perfectly willing to accommodate local regulations.

More than that though, AML/KYC and any other checks deemed necessary can occur at almost every point of contact with fiat currencies.

Sure, I knew exactly what I was doing by daring to use the popular parlance rather than the GNU/preferred nomenclature. Being labelled as "unfortunate" is certainly the least that I deserve for that transgression.

However, at no point did I blame anybody. Everybody here probably has very good reasons for doing what they do. The parent comment asked "what part of the above portions of copy fail to convey what Guix is?", and rather than equivocating around that profound mystery, I answered truthfully.

It's however perfectly acceptable to dismiss me and my comment and assert that there is nothing wrong with that website copy. For example, it may be doing a great job of self-selecting and separating the enlightened from the casuals.

Maybe if you know what the "GNU operating system" is, it's a great description.

As for me, I had to keep reading until I saw "Linux" to go "oh ok, it's a Linux distro and they're being a little weird about it because it's GNU/Linux, nevermind it's GNU. It's just GNU."

Got a new phone, tried to activate it through my cell phone provider, was only given worse plan choices than what I already had, used the chat widget, waited a while, was asked a bunch of seemingly silly and redundant questions, then the new phone was activated within my existing plan.

What did they exaggerate?

They mentioned the attack can work with roughly a week worth of unprivileged runtime, as long as the ECC mode of the ram chips in the targeted system has been previously sufficiently reverse engineered.

Is that too alarmist? To me, it sounds like something perhaps too cumbersome for casual drive by attacks, but it seems right down the alley of so called "persistent threats", or whatever it is we call those guys nowadays.

I really want a way to get a generated voice recognition transcript for all that content people really think needs to be in video form. 2x playback is still 14:45 too long in this instance.

I think the ability to mine is incidental here. This is more likely to be targeted at dapps developers by offering them an easy setup for both test and production environment.

Probably mobility, hence increased ability to seize opportunities as they present.

They used to do a lot of fancy things, back in the days, including running bits of suspicious executables in heavily sandboxed interpreters to spot behavioral patterns.

I'm guessing those kind of approaches have largely gone away, being replaced with signatures that are hopefully fuzzier than a wholesale cryptographic hash, but still essentially only catching things after the fact, which works well with subscription business models.

Common startup wisdom is that having to deal with too many users is a "good problem" to have, along with notions of launching fast first and solving scale later.

But of course sustained rapid user growth without a robust preexisting scalability approach means you're caught with your pants down, and it's a struggle to put them back up, at least until new user traffic starts to die down.

Solidity needs formal verification much more than syntactic sugar. There's a chance solidity is already too sweet as it is.

Words of praise for fish food.

I remember hearing stories of homes in Houston that had been elevated after previous floods to make them flood proof. They were elevated high enough to survive the highest flood on record. But this one was higher..

Annoyingly enough, that's not sufficient to make those ICOs scam.

The easiest counterexample would be Ethereum itself, who had a teenager as its public face, and raised large amounts of funds on top of mostly just a whitepaper, yet turned out to actually deliver all the crazy stuff they promised.