jomohke

I'm curious how much this is the cause or effect, though?

The publishers have been saying that their ability to promote books has drastically reduced with the internet, along with changes in reading and information habits.

It seems like a book needs a far bigger push today to rise above the noise of the internet (and people's over-abundance of content to consume), and this unfortunately meant that small publishers struggled unless they "joined together" to make a bigger push.

There's extremely small (self published) books and extremely large hits, but the middle is increasingly less viable, it seems. Similar to films.

You're validly critiquing where it is now.

The hype people are excited because they're guessing where it's going.

This is notable because it's a milestone that was not previously possible: a driver that works, from someone who spent ~zero effort learning the hardware or driver programming themselves.

It's not production ready, but neither is the first working version of anything. Do you see any reason that progress will stop abruptly here?

I've thought for a while now that we'll end up moving to stricter languages that have safer concurrency, etc, partly for this reason. The most prominent resistance against such languages was the learning curve, but humans like OP aren't looking at the code now.

This is standard practice. They need to use current lossless formats to display examples to people who don't have the format yet. They are still showing accurate examples of compression artifacts. I'm not sure what else you'd expect them to do.

Strange, as Cloudinary's test had the opposite conclusion -- jpegxl was significantly faster to decode than avif. Did the decoders change rapidly in a year, or was it a switch to new ones (the rust reimplementation)?

https://cloudinary.com/blog/jpeg-xl-and-the-pareto-front

If decode speed is an issue, it's notable that avif varied a lot depending on encode settings in their test:

> Interestingly, the decode speed of AVIF depends on how the image was encoded: it is faster when using the faster-but-slightly-worse multi-tile encoding, slower when using the default single-tile encoding.

It looks like folding@home is still going https://foldingathome.org/

I'm quite surprised these are still around as I hadn't seen them mentioned in so long.

I always assumed the phase out of screensavers (and introduction of CPU low power modes) were terminal for them.

In theory yes, but in practice they usually have the speaker up far higher than they are speaking themselves so we do only hear one side clearly.

I think the high distractability is a trifecta of volume, non-naturallness of the sound (compression etc: feeling out of place in the space) and this point.

Which models did you try?

Interesting. Even when nothing bad happens? It has always worked for me.

They likely have other things to do.

In this quote I don't think he means it from the business side. He's claiming more data allows a better product:

> ... the answers are a statistical synthesis of all of the knowledge the model makers can get their hands on, and are completely unique to every individual; at the same time, every individual user’s usage should, at least in theory, make the model better over time.

> It follows, then, that ChatGPT should obviously have an advertising model. This isn’t just a function of needing to make money: advertising would make ChatGPT a better product. It would have more users using it more, providing more feedback; capturing purchase signals — not from affiliate links, but from personalized ads — would create a richer understanding of individual users, enabling better responses.

But there is a more trivial way that it could be "better" with ads: they could give free users more quota (and/or better models), since there's some income from them.

The idea of ChatGPT's own output being modified to sell products sounds awful to me, but placing ads alongside that are not relevant to the current chat sounds like an Ok compromise to me for free users. That's what Gmail does and most people here on HN seem to use it.

Is this why everyone only seems to know the first half of Dario's quote? The guy in that video is commenting on a 40 second clip from twitter, not the original interview.

I posted a link and transcription of the rest of his "three to six months" quote here: https://news.ycombinator.com/item?id=46126784

Why do people always stop this quote at the breath? The rest of it says that he still thinks they need tech employees.

> .... and in 12 months, we might be in a world where the ai is writing essentially all of the code. But the programmer still needs to specify what are the conditions of what you're doing. What is the overall design decision. How we collaborate with other code that has been written. How do we have some common sense with whether this is a secure design or an insecure design. So as long as there are these small pieces that a programmer has to do, then I think human productivity will actually be enhanced

(He then said it would continue improving, but this was not in the 12 month prediction.)

Source interview: https://www.youtube.com/live/esCSpbDPJik?si=kYt9oSD5bZxNE-Mn

Some places resist this because it causes a "rich get richer" effect in popularity. But it's admittedly convenient.

My initial assumption is that this is more about Photomator than Pixelmator (ie. their Lightroom alternative rather than their Photoshop alternative).

Photomator has shown that you can add a lot of professional-level editing control to an Apple-Photos-like interface without making it difficult to use.

Their ML team also seems quite good — for instance, their spot/object removal tool was often more reliable for me than the one in Lightroom, despite being from a far smaller team than Adobe.

(I also feel that Photoshop has reduced in cultural significance in recent years, and that Lightroom is the more significant tool going forward, but that could reflect my own bubble)

(replying to myself)

I ran a quick, non-scientific Python script on my Macbook M2 using local Postgres and SQLite installs with no tuning.

It does 200 simple selects, after inserting random data, to match the article's "200 queries to build one web page".

(We're mostly worrying about real-world latency here, so I don't think the exact details of the simple queries matter. I ran it a few times to check that the values don't change drastically between runs.)

    SQLite:
    Query Times:
      Min: 0.007 ms
      Max: 0.031 ms
      Mean: 0.007 ms
      Median: 0.007 ms
    Total for 200 queries: 1.126 ms

    PostgreSQL:
    Query Times:
      Min: 0.023 ms
      Max: 0.170 ms
      Mean: 0.028 ms
      Median: 0.026 ms
    Total for 200 queries: 4.361 ms

(Again, this is very quick and non-optimised, so I wouldn't take the measured differences too seriously)

I've seen typical individual query latency of 4ms in the past when running DBs on separate machines, hence 200 queries would add almost an entire second to the page load.

But this is 4.3ms total, which sounds reasonable enough for most systems I've built.

A single non-trivial query required for page load could add more time than this entire latency overhead. So I'd probably pick DBs based on factors other than latency in most cases.

I think that's why the GP said "on the same machine" — I read the article as comparing to the typical DB that is located on a separate machine in the data center, accessed over a network, as is currently the common setup.

But if we're considering running SQLite, the apt comparison would be against other DBs running on the same machine, because we've already decided local data storage is acceptable.

I assume Postgres/MySQL/etc would have higher latency than SQLite due to IPC overhead — but how significant is it?

Deepmind's recent model is trained with Lean. It scored a silver olympiad medal (and only one point away from gold).

> AlphaProof is a system that trains itself to prove mathematical statements in the formal language Lean. It couples a pre-trained language model with the AlphaZero reinforcement learning algorithm, which previously taught itself how to master the games of chess, shogi and Go

https://deepmind.google/discover/blog/ai-solves-imo-problems...

And yet the data is ephemeral, so in many cases could be done faster without the "real" block storage guarantees.

Github Actions is pay-per-minute-used, unfortunately, so they may have a negative incentive to speed things up. Unless people become frustrated enough to switch to a non-bundled provider.

Yes, it's not arbitrary at all — they're only offering it on devices with at least 8GB of memory.

The iPhone 15 Pros were the first iPhones with 8GB. All M1+ Macs/iPads have at least 8GB of ram.

LLMs are very memory hungry, so frankly I'm a little surprised they support such low memory requirements (especially knowing that the system is running other tasks, not just ML). Microsoft's Copilot+ program has a 16GB minimum.

Note that buybacks must be announced beforehand by the company, so everyone selling their shares has the same knowledge.

Camping lanterns are still sold with this idea. For instance, I have one of these: https://hardkorr.com/au/product/u-lite-dual-colour-led-lante...

> The wavelength of our orange LEDs is 610nm, putting them outside of the spectrum visible to must bugs, but still well inside the boundaries of human vision. As a result, most bugs will not congregate around our orange light because they simply don’t know it is there.

That colour also feels reminiscent of a camp fire, which I've often thought could be related — a bug wouldn't survive long if it flew into fires.

There's definitely a lot that can be critiqued about that period.

Famously they divested their ARM-based mobile processor division just before smartphones took off.

The new CEO, as the article mentions, seems to have a lot more of a clue. We just hope he hasn't arrived too late.

The most upvoted answer does track it down to a Chinese source.

Presumably it generates revenue and gives them more data. Both can be useful for that goal.

Good point. Though that affects OpenAI too for new data.

I had assumed most of their web content was from Common Crawl, and the older pre-ChatGPT Common Crawl datasets used would still be available. But it looks like Twitter, for one, was not in Common Crawl.

To be fair, this isn't a ladder OpenAI used themselves, right? (The seemingly-extreme shortcut of training from an external LLM model)

Their ladder (using public data, and hiring humans to classify to taste) is still available I believe.

(slightly) more platform neutral link: https://overcast.fm/+_ztydHHNo

Re-using standards is a great idea, and should remain people's default, but I don't see the benefit here.

What's the advantage of re-using a format?

Ecosystem? That won't help SQLite here, who don't have dependencies.

Keep in mind that anyone trying to write a parser for this is also writing a parser for the entire SQLite file format (the only way to access bytes). And it's a spec simple enough to fit on one monitor.

Design?

BSON (and others?) seem to have different goals.

SQLite's format seems to minimise conversion/parsing (eg. it has multiple TEXT types depending on how much escaping is needed; BSON has one fully-parsed UTF-8 string type). BSON is more complex: includes many types not supported by json (dates, regexs, uuids...) and has a bunch of deprecated features already.

SQLite's on disk-format is something they intend to support "forever", and as with the rest of SQLite, they enjoy pragmatic simplicity.

As someone who's not really a gamer, hearing "it gets tiring after 85 hours" as a negative blows my mind. 85 hours!