juliendorra

Isn’t that a function of density? The denser the place, the more random chance someone wrong would pass by?

I loved my DAT recorder back in the days, recorded incredibly good digital audio on small tape cassettes with a recorder the size of a Walkman

Because coding agents are not idea compilers, they are idea upscalers.

https://www.linkedin.com/pulse/coding-agents-idea-compilers-...

I’m not sure it answers the original question. Understanding the plan is not understanding how the code exactly works.

Maybe not more professional widely distributed software, but personal software of limited scope and use has boomed around me: designers creating a tool for each new project, developers creating art software, super users asking for plugins or custom workflows… It is not unprecedented: HyperCard for example created a similar boom, of course on a smaller scale.

It seems more a regulate the nuclear missile watching agency problem than a regulate generic LLM tools problem. If my LLM tells me wrongly there’s missiles somewhere the impact is null

Indeed, in France the first to connect a home or a building of appartements has to let the other isp access

Several of the issues you point have been solved in France via regulations (on the ISP and on the tenants of condos). It tools several years, even decades, but we ended up with fiber regulations that are good (could have been better and faster).

If writing code is hard, how is it possible that so many kids learn it in a few months? I learned to program in basic at 9 with literally 2 books (no web back then), at 12 I was coding little programs in turbo Pascal, learnt with just the manual (it was a very good manual). I was not a genius. Granted it gets a certain type of mind to want to learn programming, but it’s far easier to learn than the pedestals it sits on (writing, reading, some maths and planning logic). Also code itself demands very few cultural background to start, hence why we have teenage dev genius but no teenage historian genius

Writing is not easy, and most children need years of training to write at sufficiently good levels. Writing is the technology we invest the most for training on as society

Magazines’ style is generally good but affected, with some distinct quirks mixed in. These quirks have been also picked up in the training and their automated repetition at scale made it weird and "AI-style"

You are right that at training the main goal is balancing between the sections to avoid certain paths becoming the only path. In the end the inference will be routed token by token to a mixture of say 3 or 4 sections of the model. The combination can change at each turn. It’s really a statistical optimization. As for many things in neural networks, the original intuition coming from anthropomorphism once implemented becomes something very non human!

For context in Paris (and probably other French cities) you would try that and the result would heavily depends on the driver and their mood of the day ! Because they too are not allowed to do it, but of course France is all about flexing the rules (a bit, not so much, and after some pleading)

Teen pregnancy (or unwanted pregnancy at all) has been extremely low in several countries, and the factors are not discipline but was of access, focus on health instead of morality, and public service campaign. If it is high in an otherwise rich country, it’s likely not a lack of discipline from teenagers but a lack of free and easy access to contraception and abortive means, social pressure and morality campaigns, lack of sex education (often by design sometimes by negligence after some period of activity)

> They don't keep every book though. In France they do! And other countries too I think. The French national library is also scanning everything (or trying at least) including copyrighted material (but you can’t access it except if you are a researcher and from their own physical place last I checked)

Suing is very costly, takes year and is mentally draining. And in the end they wouldn’t probably not get much if any money, and may even loose or be dropped. So in this case I understand why they wouldn’t want to do it.

Zelle seems more complex in terms of access, slower, and less universal (inside the covered zone) than IBAN based instant bank transfers. Users need to enroll using their phone number and / or an email, when IBAN instant transfer is available by default. They say that transfers "often" happen in minutes, when instant transfer via IBAN happen in a few seconds, always. The FAQ also covers issues enrolling, which means that users are not guaranteed to be enrolled. It’s an out of band process with it seems email or phone number as the ID. It seems the recipient has to enroll beforehand which is another complication (IBAN being the bank number you have the guarantee that your recipient exists and can receive the funds instantly) Zelle feels more like a system added on top, IBAN instant transfer is an extension of the existing bank transfer system, used and found at the same place UI-wise.

To give a little historical perspective (as you are young at 23) dads have been ignoring their kids for some times in our society: before it was magazines or maybe books. Or TV at home. The cliché "don’t disturb your dad while he is reading his newspaper" is sadly anchored in a real asymmetry in terms of child care (fathers tend to care for kids on their own terms, mothers for all the rest) Said otherwise: me time for dads is usually sanctuarized

It’s very bad PR for all US AI companies, not just Anthropic.

> Agents should have the same permissions as the user prompting them, nothing else.

In user support work, it won’t make them very useful. User support is the fallback when self-serve tools and public documentation, the one you have permission to read and use directly, are not allowing a solution.

By definition useful user support allows operations that are beyond the user’s permissions

"Our entire society has become black and white, overly tribal" It might be a recency bias, because the 19th and 20th were extremely polarized, politically… from entire nations split up on one issue, up to political assassination, civil wars… slavery, women’s vote, antisemitism, prohibition, civil rights, asylum, universal marriage, and much more.

Yes, it was obvious maybe even in the 60s for a few, and it has been fantasized by many, but you wrote it as a cohesive, nearly deterministic, and fluid story. Your deep understanding of some fundamental issues (like latency) that you turned into consequences instead of brushing them off is what made it so perfect as a very tangible and possible future. One read and it never left me

This is a really close equivalent to keep learning sketch and clay modeling in design school

ADHD and other mental issues are under-diagnosed in dysfunctional or toxic families, and of course exist in very stable caring families, so I would be very curious in which data link the very different symptoms you cite directly to trauma. It feels like going back to the era of shaming mothers for autism.

They are the basis of the prototype that Bas Ording used to design all the interactions we know today on touch: inertial scrolling with rubber band effect, row of icons for apps, pinch to zoom, etc. It was a fingerwork trackpad with his Director (in 2004!) interactions projected on! It was designed for a Mac tablet, but then the focus shifted to a phone.

I use an Air M2 8GB, and memory becomes an issues when I have Chrome with hundreds of tabs, a docker runner, an IDE, and several other apps like keynote all open. Add Claude and the machine is suffering! My guess is that 8GB should be fine for most users, but design students for example may stumble upon issues if they use a lot of big apps concurrently.

Nice write up! Even if I think that turn taking is a very simplified model of conversation! There’s collaborative overlapping, while the other continue, there is all the confirmations that the other agree, there’s the phatic messages maintaining the "listening channel open", and there’s even completion (filling a word or a name) that are not turn taking and should not be taken as such, yet that the model should be able to produce and accept. They are probably not modeled well or at all by a turn taking process

In the first years, I remember no other search engine was close to Google quality. We all ditched AltaVista because Google was incredibly better. It would have been awful to switch back to any other options. We can already switch between the 3 big proprietary models without feeling too much differences, so it’s quite a different landscape.

Consoles are not a big market enough to move a company like Intel. Consoles are a niche in computing. Gaming computing is niche compared to business computing (both in value and in number of users)

Wider range in bigger cities, I think it’s quite common everywhere? Small town restaurants need pauses