anonymousab

> the user base that relies on simplified solutions is largely already addressed

There is a wide spectrum of users for which a more white-labelled locally-runnable solution might be exactly what they're looking for. There's much more than just the two camps of "doesn't know what they're doing" and "technically inclined and knows exactly what to do" with LLMs.

The license agreements and ToS for all of this has various "we can terminate this in any way we want at any time we want for any reason we want" clauses, which is effectively a magical bulletproof solution in many countries.

> You want to make sure users aren’t stressing about making some final un-revokable decision before they even use the app

It has never come across that way as an end user. As it has always been on some terrible change or option where the answer should be "no, now bugger off forever", I have viewed "not now" as the same as "ask me again later":

an explicit thumb in the eye and statement by the software developers that they think their mandate takes priority over my preferences and that I am not qualified to make decisions about the software I run.

It is a very easy way to maximize my disdain for a developer/team/org.

The author is explicitly talking about non-racers/people who aren't racing at the level where it (explicitly does) matters.

The issue is Lewis going out of his way, in so many pubic forums, to present SBF as a good guy genius with only the best of intentions. At worst, as a good kid who made a couple of sad mistakes that you should forgive him for. Before and throughout SBF's trial, Lewis's biases were displayed more and more in his denials and ...fanciful interpretations and justifications of the things SBF did.

It is not that his book should have painted a bad guy as an unequivocal bad guy. It is that you cannot trust Lewis' accounting of events and people because of his known extreme bias in the matter; his descriptions of events and emotions at play during them are filtered through a very biased lens.

It can still be an entertaining read and story, it just can't be a trusted one.

They may have been paying for subscriptions, bit rewards/redemptions, prime, etc. Ads are only one source of Twitch's revenue.

Twitch started showing a purple "disable your ad blocker" screen when a block of ads failed to run. I think they probably would have made a lot more money and reduced a lot of churn by simply advertising Turbo instead, which many users don't even know is an option.

> It is nearly impossible to bring innovation to market because of panic and fear.

No, it's because of greed and managerial and executive rot. They could innovate as much as they want, nearly as fast as they want, and still be safe - as they and other airliners have done in the past! - but then they wouldn't save the "max" amount of money doing it. How dreadful.

Boeing's failures are all Boeing's fault alone. They rightfully deserve the pillorying.

> If a car has an issue we don't blink and we loose nearly 40k people a year in the US alone

That is more of an argument for stricter requirements around automobiles and that entire industry than anything else.

I haven't used pcbway so maybe they're much better, but I can say that messups and throwaway board issues are common enough with JLCPCB assemblies that their "sorry, here's a $5 coupon" response is a meme in hobbyist communities. They're very hard to trust for anything truly complex or critical.

For what it's worth, there are several municipalities that have different and conflicting rules about signalling in or with roundabouts than the provincial rules. So there is legitimate confusion about which rules to follow.

At least, in Ontario. The provincial police will give you a different answer from municipal police, and I've gotten a different answer still from MoT folks.

In practice, another issue I have seen is that some folks see any use of a signal near a roundabout as an indication that you are trying to "change lanes", which leads to a honk fest and more confusion immediately. So doing the "right thing" approaching or exiting a roundabout can lead to more confusion and dangerous situations than just... not signaling, unfortunately.

Modern Boeing has a consistent record for a reckless disregard for critical aspects of safety in favor of profits. So assuming that this is another case of that and calling for heads to roll at first sight is a fair reaction. That is the result of the reputation their executives have willingly built and the choices they have made.

> How do you know it's related to Boeing's actions in any way?

With all of their actions in recent memory, they currently have negative benefit of the doubt; problems and issues should be assumed, and if/when they occur, public blame and shame will generally start on them until proven otherwise.

That's the consequence of throwing caution to the wind and repeatedly trashing their own reputation for a few more short-term dollars.

You can be an expert of engineering and perform engineering work without being acknowledged or licensed by some authority. A country can have engineers before they even have a PE org or guild or system, after all.

Sure. And neither are mp3s of the same songs that were blowing up on Napster.

The existence of widespread illegal means to procure something doesn't mean that we don't and shouldn't require legitimate businesses to abide by the law or require them to make amends for their current transgressions.

Here's a breakdown of one of them: https://gizmodo.com/nytimes-bizarre-softball-article-ftx-sam...

It would be more accurate to say that the NYT has been far too eager to positively misrepresent and run interference for crypto in general, so of course the face of crypto got his fair share of that for a couple of years.

The "reasonable" singular removal is more about coming up with ways to block prompts that can produce infringing content, and having filters on the other end to catch infringements before they are published to the user. It's an endless whack-a-mole that never actually addresses the problem but might look good enough to the legal system or to keen supporters.

Barring some major breakthrough, the actual answer is to train a new model without the infringing data.

I think some of the people saying "remove it from your model" are aware of this and are simply being glib and needling; "you've created this infringement monstrosity, so surely you made sure to include a way to deal with this problem without throwing away all of your work, right?"

Or they can be forced to destroy or retrain their models without any copyright materials for which they don't have or do not now attain licenses for. These are multi-billion/trillion dollar companies. They can afford to be responsible members of society here, however much their shareholders and C-suite might hate it.

The mullvad browser defaults would get it chucked in the bin very quickly by the average web user.

The extreme bent towards privacy and not retaining any identifiable or fingerprint-able behaviors is exactly what its target audience wants, of course. But it leads to a markedly more inconvenient experience than Chrome or default Firefox or Safari.

> their world consists solely of software they have been allowed to install by their platform overlords

The same will be true of most cars within a generation, and is effectively true for most car owners now; they do not really know how to do much with their car beyond drive it, use the infotainment as-is and bring it in for repair when anything seems off.

> If Google didn't pay for being the default search engine, or tried lowballing, someone else would make a deal instead

This already happened with Yahoo. Mozilla got a ton more money, too.

It was utterly disastrous for them. It turned out that Firefox needs Google as the default more than Google needs them - well, for the stated purposes of the deal, of course.

Is the code actually there on non-nightly builds or is it just a simple text response for the experimental flag?

For the former then sure, it should just let you do the thing and simply fail loudly if it fails. For the latter, there's not much they can really do without switching you over to nightly rust.

They likely view the idea off competing with Chrome - a key chess piece of the company keeping Mozilla around merely as a pet to ward of regulators - as impossible if not counterproductive to their mid-term survival at this point.

Which is to say, at some point they might become annoying enough for Google to not want to pay them anymore, and then Mozilla would cease to exist shortly thereafter.

So with that in mind, Firefox as a dedicated separate browser with its own engine is an albatross around their neck. A source of significant ongoing investment and cost for comparatively little return. For the Mozilla "mission" and the CEO's personal goals and purpose, a simple rebadged chromium fork skinned up with open-web platitudes is a far better idea in the long term.

Because the megacorps should have to pay the people creating the works they are training their multibillion/eventual multitrillion dollar systems on, and should get a nice rake to the face when they try to do an end run around it.

> Trying to prohibit this usage of information

It's not trying to prohibit. If they want to use copyrighted material, they should have to pay for it like anyone else would.

> prevent centralization of profit to the players who are already the largest?

Having to destroy the infringing models altogether on top of retroactively compensating all infringed rightsholders would probably take the incumbents down a few pegs and level the playing field somewhat, albeit temporarily.

They'd have to learn how to run their business legally alongside everyone else, while saddled with dealing with an appropriately existential monetary debt.

They still load you up with many more adtech trackers on YouTube premium than you get when using adblock.

But yeah, if you use youtube a lot, at least premium does exist as a more morally righteous choice if you don't really care about the tracking side of things and are mostly just annoyed at ads themselves. Which I'd imagine is most people.

Substack's claim and argument is about free speech, though. Not "free speech, but".

From what I have seen from some tests at my company, you can almost always get hallucinations that are potentially very destructive.

But for front line "get this person to shut up and go away in frustration" customer service - not something for solving actual customer issues - it's a no-brainer.

I would guess that looping, no-human phone systems are just cheaper to keep for now.

> police can force you to open your door, your safe, or virtually any other container of secrets.

Is it different from compelling someone to enter a text password to unlock a vault? What if it's self-destructive otherwise?

What happens if the password itself - or act of unlocking - is something self-incriminating (in form, in contents, or otherwise)?

A not terribly uncommon belief is that continuous growth (pick your metric) is the raison d'être of all entities operating in a capitalist society, and so achieving and maintaining some high level of profitability and then staying there means that you and your company are failing to do their job.

But "growth" is a moving and indulgent target. For some, increasing profits isn't the growth that matters. Rather, increasing the rate at which profits are increasing is the true metric. So even vastly increased profits can still be a form of failure that requires more actions be taken to wring more money out of the platform and its users.

Those cheating devices usually use a first party controller for authentication and the major ones were not affected by the recent changes at all.

The major corps and tech companies loved to claim otherwise for years as it suited their interests.