Yeah, these are different for me as a southern Englishman. Languages generally try to avoid glottal stops. French has a lovely system called liaison where a consonant sound will appear to separate two consecutive vowels. In English we like diphthongs so I guess we use this similarly to liaison. Some accents just use glottal stops, though, nay bother.
I don't think it's really possible to learn a new language unless you have to. Sure, you can learn some words and grammar from books and games like Duolingo, but you'll never really learn the language if you don't use it, and you'll only use it if you have to. This is why it's hard for English natives to learn another language: too many people speak our language already. And now people won't be learning computer languages any more either.
I like the analogy, even if I find it a bit depressing. However, this caused me pause:
> Instead of the most important question being “how to build this”, or “can we build this”, now the most important question is “what should we build”.
Do people really think software is just a matter of paying for the tokens now? If that were true we should be able to ask for a web browser that is faster, leaner, more secure and more capable than all existing browsers, or otherwise be told why it can't be done. Given that the standards are all written down I shouldn't need to give any further direction. I would have thought AI companies would have tried this before trying something like Navier-Stokes.
Treating code as a low-entropy amorphous blob is a quick way to see bridges collapse.
It works for concrete because its tensile and compressive strengths are well known for a given composition, but for high-entropy products like software you pretty much need to also generate a formal proof to have any assurances.
Which, ignoring the production costs and assuming that the proof is in fact correct, is only ever as good as the specification.
If you were looking for a physical analogy for software development, it's closer to growing crystals than pouring concrete.
That's a good point, and I don't think the author would disagree with your take. "How to build" can refer to many things at once, some which still matter and some which don't. In the case of the analogy it means you're no longer optimizing the micro-structure of cut stones, in favor a homogenous slurry that is quick to produce. Of course the macro decisions still matter. A well-built roman bridge could last centuries (depending on the nature of erosion it was contending with), a concrete/rebar structure cannot. When it comes to computer systems the tradeoffs are just different, for example, how we model, validate and persist data is much much more important than the details of the code that processes it.
I have a slightly different angle on the quoted sentence though. In my mind "what should we build" was always the most important question. Code becoming cheaper to produce just highlights that more than ever. But even before agentic coding, the world was full of software that was not fit for purpose because the people calling the shots lacked some dimension of how to translate a problem statement into a workable system (computerized or otherwise), and built by programmers who were content (or at least complicit) in not probing any of those questions with decision makers.
That's why I don't think AI is inherently bad for software—it at least has the potential to empower those with better judgment since the sheer cost of production can no longer be an excuse for why something makes absolutely no sense from the end-user perspective. There are other, larger risks with AI at a societal level, but I don't think it's inherently bad for software quality.
It's become a bit clichéd to say it but I think it's different this time.
The kind of "piracy" you're talking about wasn't depriving the authors of anything because you could always make the argument you weren't going to pay for it anyway. If, on the other hand, you were making copies and charging people for them, you could definitely say you were depriving the legitimate authors of that revenue. AI companies are very much doing the latter, not the former.
The other part of it is it's not just copying. Previously, if I decided to make a copy of a work without paying, I'm only copying the work, not the author's whole writing style. Now the AI companies are depriving authors of revenue from works they haven't even made yet.
If they were able to create a model de novo then they could truly claim it hasn't just been lifted from existing culture.
Why are people not learning to write at school? In most developed countries don't we spend like a whole decade or more learning to do stuff like writing? Yet people still feel like they can't write?
Writing is something we really, really need to learn from other people, because it's with people we are trying to communicate. There should be no need for "writing with an LLM". None at all. If what you are writing is important, ask a friend to proof read it. If you can't write, learn.
To be 100 years or older you would have been born in 1926 or before. Meaning you turned 18 in 1944 or before. Maybe something happened in Japan that killed of a large percentage of young adult males around that time?
Well, exactly, but this lends credence to the numbers being somewhat accurate. If there were almost as many male centenarians then it could probably be attributed to bad record keeping.
I had to double check the date. I even saw 2001 in the title but I still think of every post-2000 date as "fairly recent" so didn't make the connection to 9/11.
reply