HN user

suddenexample

381 karma
Posts2
Comments64
View on HN

Yeah I'm not so sure this CTO is on the mark here, but to be fair, I do think some of this IRL long tail/edge case data is important for Waymo. The simulation software is super interesting to me - the real world can be so chaotic, and even if they could generate every possible real life case, there needs to be validation on whether the Waymo driver is responding in the optimal way. They certainly haven't solved this problem, you can see some of their growing pains in all of these articles - floods in Austin, more and more interactions with emergency vehicles that first responders seem to believe are getting worse, etc.

Tesla on the other hand has billions of miles of data, yet because there is a limit to camera-only techniques, that data isn't that useful is it? They have no ground truth data to evaluate their camera system on, which is why sometimes you see those Teslas driving around with lidar rigs mounted on them. Going camera-only is just asking for trouble.

I don't particularly agree with Epic's victory in the Google/Epic case, but the one thing I hope it accomplishes is to convince those in charge of the Play Store that it's finally time to have developer-friendly policies (otherwise someone else will). Play Store policies constantly virtue signal about security and privacy while continually making it harder for developers to release high quality apps.

That's only enabled by the difference in culture though, right? Japanese culture has a much higher emphasis on order and following the rules - I don't know that this "open-by-default" system would work in, for instance, the US.

Oof. I actually really like the majority (with some notable exceptions) of what Material 3 Expressive does from a pure design perspective, but this article is the worst reflection of that.

"Big button easier to find" (let's think about whether "easy to find send button" is the top priority for an email composition screen, because these folks apparently didn't) and "We can make an existing UI less functional by taking up the entire screen" seem to be the writer's favorite parts of M3E.

It's ironic that they got rid of the tall bottom navigation bar and brought back the short one with less padding (likely after all of Google's own 1P properties decided it wasted too much space), because now it feels like that took that failed philosophy and applied it everywhere else.

A question for Googlers who may be responsible/adjacent - what is the intended function of this warning? It seems to be attempting to filter out low quality apps, but instead seems to be killing any attempt to change the status quo. If the app has fewer users than competing apps, the message Google is sending is "we don't need any new apps that do similar things to existing apps" and "if you're a small app, don't even think about unseating the dominant players."

Google's Play Store policies have been harebrained for quite some time - previously with the 15 reviewer approach they decided to make it even harder for developers with fewer resources to distribute their apps. It's ironic that even though the iOS App Store is arguably more of a walled garden, it's so much friendlier to human beings who are trying to build a product. But at this point it seems ingrained in Google to release self-defeating features (remember the finder network that prioritized "first of its kind privacy" over being able to find things?)

Weird - it's hard to beat widespread online narratives, but as someone who worked at Google there's no company I'd trust more with the "handling" part of my data. There's no doubt that on device is always a more private option, but if you've decided to keep data in the cloud, then Google is probably one of the most secure options you could choose.

I think the "LLM is intelligence" crowd has a very simplistic view of people. If you feel that natural language and the systems responsible it are pretty much the only things that human intelligence produces, then I can see the argument.

But I don't believe that. That a machine that can produce convincing human-language chains of thought says nothing about its "intelligence". Back when basic RNNs/LSTMs were at the forefront of ML research, no one had any delusions about this fact. And just because you can train a token prediction model on all of human knowledge (which the internet is not) doesn't mean the model understands anything.

It's surprising to me that the people most knowledgeable about the models often appear to be the biggest believers - perhaps they're self-interestedly pumping a valuation or are simply obsessed with the idea of building something straight from the science fiction stories they grew up with.

In the end though, the burden of proof is on the believers, not the deniers.

It's actually hilarious that whoever was in charge of Google's finder network decided to cripple the product's one and only function by prioritizing privacy.

In this tradeoff, Google gained a handful of articles mentioning the "innovative" privacy improvements (before the writers had a chance to test how terribly the network actually performs). For that, they sacrificed the chance to compete with Apple in this category, which outside of device revenue also weakens Android/Pixel ecosystem and market share. You really can't make up this level of incompetence.

The ones debating whether this is nefarious or not are the ones ruining the tech industry. This is absolutely nefarious. Whether or not it's a clever path to promotion due to corporate incentives is irrelevant.

I'm curious what part of Microsoft's culture enables these satirically slimy product decisions. In theory, other megacorps should be no better, but somehow they seem to maintain a bar that Microsoft always manages to stoop below

Not sure about the "less funds" bit, but Google started funding self driving cars in 2009. The continuous investment in tech is just now starting to make the dream feasible - as far as I know self driving cars were never more than a side project for Uber and Lyft.

With Cruise gone now there's basically just Waymo, Zoox, and Tesla (bit controversial) that are the names thrown around when talking about self driving market share, and out of those only Waymo has a functional service.

Google Pixel 9 Pro 2 years ago

Research: consumers are turned off by AI in marketing because it's a vague term that makes no promises about features or usefulness

Google: takes deep breath AI AI AI AI AI

Were royalties in the picture? Don't think it's crazy to think that a company obsessed with replacing artists wouldn't value one enough to pay royalties.

And in terms of collaboration potential... OpenAI is a big draw for businesses and a subset of tech enthusiasts, but I don't think artists in any industry are dying to collaborate with them.

The US White House press secretary issued a weak statement condemning Israel's action, but it was on April 1st and the costumed Easter Bunny overshadowed that statement.[7]

Wow you can't make this stuff up

News organizations are doing important work and I completely sympathize with them in a world where companies are claiming that they can replace your entire company with an ML model.

But I totally agree with the other comments that linking should not be something that you can demand payment for. And I find it equally absurd that they're demanding payment from platforms that funnel users over to them in the first place.

Some Google services like Google Meet and Google Earth were implemented for Chrome first.

Chuckled at this one - "yes now they support all platforms but they were on Chrome first! Malice!"

Transforming PR is different than transforming. I don't think Microsoft is any less scummy than it was in the past, it's just gotten better at recognizing that its audience includes both "normal users we can trick with dark patterns" and "gaming/dev enthusiasts who will lose it if we try to pull anything shady".

It's sad to me that so many in this community side with OpenAI/etc against the artists. It's understandable, because these companies have created a very cool and useful technology. But I think the arguments - from "it's not technically copying, it's just viewing copyrighted material and then producing its own version like a human is allowed to do" to "intellectual property/copyright is inherently flawed" - come from a lack of understanding.

Let's make an analogy that more people here will understand. If you built an app for your startup, and OpenAI trained on its source code such that any prompter could produce an app that's virtually identical with no effort - would that be OK with you? Would you claim that copyright is in fact irrelevant and that this doesn't constitute copying?