binarybits

Defining who "really" invented something is often tricky. For example I mentioned in the article that there is some dispute about who discovered backpropagation. A

According to Wikipedia, Nvidia released its first product, the RV1, in November 1995, the same month 3dfx released its first Voodoo Graphics 3D chip. Is there reason to think the 3dfx card was more of a "true" GPU than the RV1? If not, I'd say Nvidia has as good a claim to inventing the GPU as 3dfx does.

Author here. What do you mean? The article this thread is about has hardly any opinion in it. Or consider this piece explaining how LLMs work: https://www.understandingai.org/p/large-language-models-expl...

What isn't really true? The passage is about a hypothetical LLM so obviously the exact steps depicted here don't correspond to any particular LLM. But LLMs undoubtedly modify hidden states to reflect context gleaned from other words, right? I don't understand what point you're making.

I actually don't have a job! I'm a self-employed writer who made low five figures last year.

I discussed this in the final section: "The level of employment across the economy is ultimately driven by macroeconomic factors: If consumers spend more money, then businesses will respond by hiring more workers. The last three years have illustrated how powerful this can be: In the wake of the pandemic, Congress and the Fed worked a little too hard to boost the economy, producing a super-tight labor market and rising inflation. If AI starts replacing workers in the coming years, that will put downward pressure on wages and prices while growing the economic pie. That will give the Fed more leeway to cut interest rates and give Congress more room to raise spending or cut taxes. As long as Congress and the Fed are doing their jobs, there’s no reason for the total number of jobs, economy-wide, to decrease."

My argument that it won't cause mass unemployment is that most jobs in the economy are not jobs AI can automate because (1) they involve physical work that current robots can't do (like plumbers or any of the building trades), or (2) they involve interaction with other human beings (child care workers, nurses, baristas, waiters, etc.). To the extent jobs are eliminated in AI-adjacent sectors, people will need to shift to these other jobs that AI can't do.

Yes exactly.

I'm old enough to remember them too! I'm not saying that system was better. I'm just saying it's weird that the new system is so much more expensive to run.

And then people frequently complain about these varieties—for example, that the BLS produces a "core" version that excludes food and energy even though there's also a version with that stuff included.

OP author here. I wrote a shadowstats debunker back in November:

https://fullstackeconomics.com/no-the-real-inflation-rate-is...

tldr: they don't actually reconstruct the old methodology. They add a fudge factor based on how much they think the new methodology increases the inflation rate. And that estimate is based on a basic math error.

There wasn’t a 2010 long form. The 2010 census had about 10 questions: sex, race, age, ethnicity, as well as your relationship to the primary homeowner and whether the home was owned or rented. Nothing about money or jobs.

There used to be a long form in 2000 and before, but this was spun off around 2005 to become a separate annual sample-based survey called the American community survey.

Right, that was my point: the nanny does more than just keep the kid safe.

No, this was before she became a successful author. About half of it was an inheritance on her side. The other half came from his side.

What do you use?

Sorry, but this is not an accurate summary of the state of the law. Whether Android's implementation of Java was fair use is one of the questions the Supreme Court is considering. Google was absolutely trying to achieve interoperability with Java, so if Google loses it would make it more difficult for others to claim fair use. Fair use is complicated so others might succeed where Google failed, but a loss for Google would be bad news for interoperability generally.

No, it's closer to the opposite. The statute says that the courts "may, only to the extent necessary" charge fees for access to PACER. I don't think there would be any legal barrier to the courts eliminating the paywall other than the courts have become dependent on the revenue stream. I suppose in that sense Congress has forced it to exist by failing to provide an alternative funding source.

Man, you're reading way too much into my headline. The headline describes the article accurately: I created a deepfake. It took me two weeks and I spent $552. The headline doesn't say anything about whether it was convincing or not.

Absolutely, but they have that influence and so far they've mostly been using it in ways that make journalism worse (creating powerful incentives for shallow articles with clickbait headlines). This seems like a step in the right direction.

The question is how you want to characterize an iPhone. If the phone is Apple's property that they're letting you use, then Apple's app store restrictions are analogous to McDonalds deciding what going on their menu. If the iPhone is a product that you bought from Apple, then Apple's app store policies are equivalent to GE selling me an oven that will only cook food purchased from GE's grocery store.

Luckily, oven technology doesn't really make that business model feasible so GE doesn't try to do it. But I think there are reasons for antitrust authorities to take a careful look at the behavior of companies that do have this kind of power over their customers.

Nope.

Mainstream news sites aren't going to give you the kind of super-deep insights you're going to find in academic research or trade press or whatever. 90 percent of any site's audience knows very little about any given topic, so laying out obvious-to-insiders ideas in a clear way is a significant value for the average reader. Ars readers are more technically savvy than most sites but most of them are not experts on corporate finance.

I do not know very much about the specifics of Ars's ad sales strategy. I would guess Google pays more to established sites. Also, if you're seeing a Google ad that might be because they didn't have a higher-paying direct-sale competitor for that particular impression.

Author here. What do you mean by "short?" The piece is 1,500 words, which is about double the length of a typical newspaper column.

I'm not going to get into the exact economics, but Ars writers write about a piece a day, and the average piece gets tens of thousands of visitors. Revenue per article (for most mainstream news site, not just Ars) is significantly better than $0.001 per visitor. High-end advertisers are willing to pay a premium to put their ads next to high-quality content and to know the demographics of the people seeing their ads. None of us is getting rich doing this though!

We're also fortunate to have a significant base of paying subscribers, which makes it easier for us to write in-depth pieces that might not generate enough revenue from ads alone.

Timothy B. Lee here. You are correct, except that I think it's a mistake to treat ADAS systems as "self-driving cars." I'm bullish on true self-driving car projects like Waymo and Cruise. I think ADAS systems have a lot of problems that have become more clear to me as I've covered them in more depth.

Tesla cut ties with Mobileye in 2016, and this was a 2017 Model X, so it probably wasn't running Mobileye's software.

(1) Yes. (2) Yes. (3) Yes.

The main point is that Waymo had a 4-6 year head start over virtually everyone else in this industry. So it's not surprising that their technology seems significantly better! Obviously we don't yet know how much better, or if it's good enough to improve on the average human driver. But they seem to be pretty close.

The Jaguar I-PACE deal envisioned designing a custom version of the I-PACE so that the sensors and computing hardware could be installed at the factory. I haven't seen specifics on the Pacifica deal but it seems likely they'd take the same approach.

What do you mean by "anymore?" The last decade has seen faster progress than almost anyone expected a decade ago.

Yes.

That's the point: I'm pretty sure that somebody will create automated driving technology that's better than humans within the next decade or so. But it won't necessarily be Tesla. If I had to guess I'd say Waymo is likely to get there first. And as I say in the piece, there's good reason to think that Tesla's incremental approach—start with ADAS, evolve into full self-driving—may be an evolutionary dead end because humans are really bad at supervising a driving technology that works 99.99 percent of the time but gets in a fatal crash the other 0.01 percent.

And yet: https://twitter.com/arstechnica/status/991636171309625344

https://twitter.com/elonmusk/status/991752018338304000