alexchamberlain

I think you should highlight your exemplary pre-AI writing too.

https://www.kcom.com/about-us/history/ they were operated by a separate company, so the crown wasn't appropriate.

It seems like inlining futures that aren't holding variables over the await point might solve a large part of these issues?

Government support won't work for OSS at scale — it's too globally decentralized... We recently received US 501(c)(3) tax-exempt charity status.

If this is successful in the first iteration, I'd love to see a UK and EU based charities too. That would allow european donors to support on a gross pay basis, and may simplify grants to european nationals too. (I'm sure similar things apply in other jurisdictions too.)

I'm not going to comment on the security implications of either situation, but is there a companion piece by the facilities team complaining about the amount of paperwork required to install turnstiles only for a software engineer to come along and lock them out of Jira on a whim?

There are 3 mines on Manhattan; is that correct?

Apologies if this is obvious and I missed it. Does this define a way to store the strings in various languages?

You can have 240V outlets too... at least, a lot of woodworkers seem to install them in their garages. Are they legal in the kitchen?

Would this be effective at smaller volumes? Could it get down to say the size of a washing machine for use at home?

Thanks, that's really interesting. Do they correct for spelling mistakes or internationalised spellings? For example, does `colour` and `color` end up in the same token stream?

I'm probably one of the least educated software engineers on LLMs, so apologies if this is a very naive question. Has anyone done any research into just using words as the tokens rather than (if I understand it correctly) 2-3 characters? I understand there would be limitations with this approach, but maybe the models would be smaller overall?

My point was more about the original comment is fine from the perspective of an American, but for the rest of the world, it doesn’t really matter if it is USD or rubles - it’s still a foreign transaction. I appreciate that for a large percentage of the world, consumers can probably do an approximation of the USD conversion in their head, and not a rubles one, and therefore, USD may be more friendly. That being said, the sales page has already got the approximation in USD anyway, which would be enough for me.

charging in rubles is most probably confusing, and that a flat $10 usd/month would be easier

As a Brit, I'd rather GBP...

Isn't this comment a form of US defaultism?

But what if my battery runs out?

The developers of git will continue to be motivated to contribute to it. (This isn’t specific to Rust, but rather the technical choices of OSS probably aren’t generally putting the user at the top of the priority list.)

This is the problem though, right? It’s not one league table of environmental goodness - there are tradeoffs that as an educated consumer are impose to navigate.

I think banning plastic completely in packaging is a much harder ask, as whether it is needed is rather nuanced (if I understand it correctly). For example, it's perfectly possible to deliver cucumbers to an end customer without them being shrinkwrapped. However, to deliver enough cucumbers to enough customers for a supermarket scale, I understand from several documentaries that plastic is still required in that case. (For those outside the UK, plastic covered cucumber is the social barometer for plastic packaging.) Banning plastic bags was easy and simple, and our laws don't tend to deal with nuance very well...

The UK banned single use plastic bags at major supermarkets. We all moaned about it for a few minutes, forgot our reusable bags a couple of times and then got on with it. Even the small plastic bags you put fruit or pastries in are now gone in a few super markets - initially, they replaced them with transparent paper-based windowed bags, but then I think people realised you really don't need to see inside the bag, and brown paper bags are back.

Setting up a scaffolding for a new website? LLMs are amazing at it.

Weren't the code generators before this even better though? They generated consistent results and were dead quick at doing it.

This is a really unhelpful attitude. There are periods of life where buying doesn't make sense and it's financially impossible - landlords provide people with a home at these times. I'm not saying there aren't bad landlords - there are - but being a landlord isn't inherently bad; they are providing an essential service for society.

Why aren’t landlords a good thing? Is it unreasonable for people to provide a service to people seeking it?

I'm not sure how, and maybe some of the coding agents are doing this, but we need to teach the AI to use abstractions, rather than the whole code base for context. We as humans don't hold the whole codebase in our hear, and we shouldn't expect the AI to either.

There is a category called "Annie Maunder Open Category", which allows for creative use of generative AI, as long as it is declared, otherwise it is banned. You have to take that slightly with a pinch of salt though, as many of the images don't represent what could be seen with the naked eye. It could be as simple as a wide spectrum sensor (which I don't think anyone could claim is AI) through to alignment of many 100s of images (for which the algorithms in use may have been considered AI a decade or 2 ago).

Assume that the road has a single entry, no exits, and is infinitely long

I couldn't help but think that the author forgot to assume the road is inelastic and has no mass...

Those are part of the modern day commentary, rather than the historic document that starts later in the article. The historic document itself seems to use capitalised nouns fairly consistently, though I haven't tried to find exceptions.

The first thought that came to my mind reading the DaisyUI website was "Is this an April Fool's joke?". I wouldn't normally post something like that, as it's entirely unfair to the hard work and dedication that someone has put into this. However, I think it captures my surprise by how smack on the nose this is in terms of the spiral of tech abstractions - this is exactly what CSS was designed to solve, and things like tailwind appear to be leading to people forgetting that.

For a source package based on setup tools, setup.py is executed with a minimal environment and can run arbitrary code.

I think it depends on the project. I think most of us could eye ball a blog directory pretty quickly and get more or less the same idea. However, give it a gnarly bit of legacy code in a language you haven't used for a while, and indeed, 11c is pretty cheap.

That is a global static relationship rather than a local one dynamic one, which is the sense in which Go users use wrapping.

In practice, the error type will be defined quite close to where the conversion is applied, so the static nature of it doesn’t feel too big.

I think you've highlighted the real engineering challenge of using LLMs - overcoming the uncertainty. In some contexts, a reasonable level of uncertainty is fine, as the tool itself is being used by experts ( ie the coding tools), but in other cases, you need to engineer a lot more guardrails to ensure a result of sufficient quality.