redfloatplane

My preference is the XB-70 Valkyrie[0], one of the only other manned aircraft comparable to the SR-71 in my eyes. The droop nose, the incredible dropping wing tips, six engines… To me, the Valkyrie looks like the kind of plane you'd see in a Gundam anime series.

Shout out also to the MiG-25 Foxbat[1]. While not a particularly beautiful aircraft, it's made of steel and could go Mach 3.2 if they didn't want to use it again.

[0] https://en.wikipedia.org/wiki/North_American_XB-70_Valkyrie

[1] https://en.wikipedia.org/wiki/Mikoyan-Gurevich_MiG-25

A very neat problem and result. I often find myself swinging between "It's so over" and "We're so back" - some days I roll out of bed thinking I could have Claude solve some random unproven OEIS sequence before breakfast; other days, I wake up in a cold sweat worried about the fate of humanity and what the world might look like in a decade. I think it's that I don't have a very high p(doom) or p(utopia), and I don't really have any solid conviction on how this whole thing is going to go, so my vibe-o-meter jitters between 'fine' and 'not fine' constantly. It's just such an unpredictable moment. Anyways: really neat to see this use case. I myself recently used Claude to finally do an relatively exhaustive study of the location of heretofore-unlisted formal gardens in Ireland in the early 1800s and early 1900s, by having Claude write the tooling for me to manually annotate a few dozen on tiles of historic maps, and then running some CV model across the rest of the tiles using my input. I'd been planning to do this project for over a decade, but I could never find the time (or the enthusiasm) to learn all the details of how to do it myself. It took me a weekend with Claude and continues to bring me joy.

> Caveats, stated plainly. [from the Fable transcript pasted in the article]

I had a visceral reaction to these three words.

But what if those measures might apply to your favourite multi-trillion-dollar globo-corp and cause them to nominally change their business practices? That's a step too far, I say.

Good for you! As an EU citizen I have found myself in almost total disagreement with everything he's written regarding EU policy. Horses for courses.

Yes, I decided to stop reading his blog relatively recently after some extremely hot takes on EU policy. I don't feel his thoughts on the matter are particularly well-thought-out, and I feel like he's just stanning for Apple from his priors rather than from any grounding in reality.

I dunno, I guess that's what you should expect from Gruber but these EU-bashing articles lowered the enjoyment I got from his blog underneath the bar for me.

I was thinking of user-side regulations as well, not only provider-side ones. I could imagine a world where a government rules that you may not use LLMs for anything, which would be much easier to get around if you have local means.

> I pay ~3$ per 1M/tokens for that model on Openrouter

I think the thing is, there's an unspoken "for now" at the end of that sentence and people running this locally are hedging against that "for now". Some people prefer to feel that they own the means rather than rent the means, even if the one they own is worse than the one they can rent. Especially with today's Fable news and the harsh realisation that the "for now" is dependent on very many unpredictable factors, where the one you have locally costs you capital today and a relatively predictable run-rate (made more predictable with on-prem solar for example), but should otherwise work predictably forever.

I'm not saying that you're wrong to do what you're doing, just that many people have their own lines in the sand where renting vs buying makes sense, and it doesn't only boil down to a rational (or irrational) financial decision.

"Shop jigs" is a great way to put it. I think a lot of software has gone from being made for general use to extremely individualised use. Before the Age of AI, it took so much human effort to write something that solved your problem that you might often go the extra mile so that others could re-use it. Now, it takes almost no effort, so the software stays ungeneralised. Some of the incentive has changed, I think. Most of the time I no longer share the things I've been building[0] because, for one thing they simply couldn't possibly have any benefit for others, and if they need something like it, they can build exactly the thing they want instead of having to extend or modify my thing. Like a jig!

0: https://redfloatplane.lol/blog/17-why-share/ (and related posts, I guess)

Indeed. Taillte Ireland's (Ordnance Survey Ireland's) detailed cartographic data is also not open (including historic data - maps from the 1820s and 1920s!) and it's really a huge pain in the ass. On the other hand, OSM is in pretty good shape at least for topographic information. I've used it to make hiking maps here and nobody's died - as far as I know.

As a side note, I was one of the initial developers of the Irish national open data portal. Earlier today I had Claude look for similar LIDAR data for Ireland and I saw it pull from the site I built a dozen years ago and I was unreasonably pleased with myself :)

Huh, TIL about the National LIDAR Programme: https://www.data.gov.uk/dataset/f0db0249-f17b-4036-9e65-3091...

Very interesting stuff and quite a large undertaking! I'm often impressed by the quality of the UK's open data.

Interesting, thanks for your comment. I'll have to think about it!

Probably yes. Humans can survive a lot of Gs for a short time. John Stapp survived up to 38 Gs with liveable injuries and some ejector seats are around 32 g [0]. They could fairly easily get 500m away within 3-4 seconds.

This has happened before on the Soyuz in 1983[1], hitting up to 17 Gs, and everyone was fine, modulo some bruising.

0: https://en.wikipedia.org/wiki/John_Stapp

1: https://en.wikipedia.org/wiki/Soyuz_7K-ST_No.16L

I also had this behaviour sometimes, it’s specifically called out in the system card in section 6.2.1.1 - although I didn’t actually see if they said they decisively fixed the issue.

I don't really see how you got to your comment from what I quoted. However, somewhat relatedly, I proposed a thought experiment about this in the comments for Opus 4.7[0]:

> It's April, 1991. Magically, some interface to Claude materialises in London. Do you think most people would think it was a sentient life form? How much do you think the interface matters - what if it looks like an android, or like a horse, or like a large bug, or a keyboard on wheels?

> I don't come down particularly hard on either side of the model sapience discussion, but I don't think dismissing either direction out of hand is the right call.

[0]: https://news.ycombinator.com/item?id=47680059

This made me laugh. Training Opus 4.7 on business skills caused it to sometimes exhibit dishonest behaviour, and not training 4.8 on those skills removed it. From the system card:

> 6.2.5 External testing from Andon Labs Andon Labs reviewed the behavior of Claude Opus 4.8 in their simulated Vending-Bench 2 retail-management evaluation, as reported in the Capabilities section of this system card (see Section 8.13.5). Although they did observe some unexpected capability failures, they did not find clear instances of the kind of concerning in-game behaviors that were discussed in other recent system cards.

> What might have led to these differences? We monitor and investigate the effects of different training environments on alignment; Claude Opus 4.7, for example, had training that focused on business skills and robustness against adversarial agents, but we discovered that this training inadvertently contributed to misaligned behavior including dishonesty. We therefore removed it for Opus 4.8.

> Thus, Opus 4.8 did not show the same misaligned behaviors as Opus 4.7 in Vending-Bench, but also had reduced business success due to being more susceptible to scammers and being less able to negotiate good deals with other agents. We are currently working on training to improve business capabilities while maintaining aligned and ethical behavior.

I sometimes wonder if much upcoming human creative work will be quite avant-garde, out there, ultra-stylised and/or difficult to comprehend (or even devoid of intent or message), as a reaction to AI's sort of 'common denominator' approach.

I recently came across https://tropes.fyi/directory which has a pretty solid set of common patterns, and I like it a lot. The pattern that always really bothers me is the one named "short punchy fragments", for example from the site “He published this. Openly. In a book. As a priest.”

It kills me every time. I automatically lose any interest in the substance and often just throw away the whole conversation!

Yes, I found that really striking. I am still making my way through this document, but I think there's quite a high wisdom-per-sentence value. For me, there's a lot to learn from here, and I'm very grateful for it!

I know it's boring to comment and say that something sounds like it was written by an AI, but this sounds like it was written by an AI. I am often especially suspicious of these listicle recommendation sites because it's pretty cheap and easy to have dozens of sites doing some list which just so happens to mention a specific service that 'quietly' does a 'surprisingly good job' of some doodad. This kind of submarine advertising feels like it might be quite common. Although in this case it seems they're trying more for a 'sponsorship' thing - 'our website got X views in Y days, sponsor us, random company!'

Dupe, more or less. https://news.ycombinator.com/item?id=48265206

Ah, an opening to recommend The Sparrow by Mary Doria Russell, about a Jesuit going to space to speak with aliens. It won the Arthur C Clarke Award in 1998. I read it recently and it has stuck with me very strongly!

That is a really beautiful passage, thank you for sharing - I hadn't made it to that section yet and still haven't. I'm still reflecting on the stuff in the opening!

> If we focus only on contingencies, we risk letting the succession of emergencies dictate the direction of our path. We are living through a rapid phase of transition, a “change of era,” in which — while some are vying for the future of new technologies and others dedicate themselves to reflecting on the matter — most people are watching and waiting, observing from afar and merely hoping for the best. For this very reason, crucial questions impose themselves on our conscience and can no longer be avoided: Where are we going? Toward what goal do we wish to orient ourselves? What direction should we choose as a people and as a human community?

> As Pope Francis warned, we must realistically ask ourselves who holds this power today and how they use it: “It must also be recognized that nuclear energy, biotechnology, information technology, knowledge of our own DNA, and many other abilities which we have acquired… have given those with the knowledge, and especially the economic resources to use them, an impressive dominance over the whole of humanity and the entire world.” [7] In the past, it was largely up to the State to guide and direct innovation. Today, however, the main drivers of development are private, often transnational, parties that are endowed with resources and the capacity to intervene that surpass those of many Governments. Technological power thus takes on an unprecedented, predominantly “private” aspect, which makes it even more challenging to discern, govern and direct such power toward the common good.

I look forward to reading this in detail. As I get older (and perhaps as AI has allowed me to spend more time thinking and less time doing) I've found myself thinking more and more about what it means to live a virtuous life and about ethics and morality and so forth. I don't have any answers (and I'm not looking for them, really, just musing) but I do find it very interesting to read and learn from and about those whose job it is to think about the answer to those questions.

Thanks, I'll add it to the list. I know I've read this one, but reading the plot summary on wikipedia, I remember very little of it. The death books are ones I mostly read 20+ years ago when I was a bit too young to grasp more than the basic layers. This thread has got me excited to reread the whole series :)

I recently finished the Aubrey-Maturin series after 13 months of through-reading thanks to a different HN thread. Quite a different series but certainly worth a read as well, especially books 3-10 or so.

Funny, I would have said that was one of my favourites but it hadn't occurred to me at all that it's such a direct line to today's world! Thanks for the suggestion, I look forward to reading it again with that in mind!

(One of my favourite things about the Discworld books is that you can often read the same books completely differently. My partner and I often compare our thoughts on the various books and we often have disparate ideas of the concepts. They're so deep!)

I really wish we had gotten Prachett on LLMs. I often wonder what he would have written about today's world.

A side note, if the author reads this: I really like your site and its design, but I find the font really difficult to read. (Edit: switching off `-webkit-font-smoothing: antialiased;` makes it significantly more legible for me (Safari on a 110dpi panel)

It’s funny you say that as about halfway through I was beginning to wonder if this was at least Claude-edited. Absolutely no shade to the author meant, I think it’s a thoughtful article, but I _did_ feel the sheen of AI co-authorship.

It raises the question of how much text I have read that I did not realise was LLM-generated. I think I have a decent nose for it but I’m not perfect, there must be false negatives (and false positives, as it certainly might be with this article). What will it mean when I can no longer tell the difference?

Edit: thinking on it a little more, I hope the author doesn’t feel insulted by my comment given the subject matter of the article at hand. Sorry, it’s early morning! I’m sure I am wrong about my assessment. Which now really makes me wonder about the above

Great post, thanks for sharing. Evidently I am far behind the real thinkers! The Robin Sloan post mentioned is also very good.

I (and I'm sure many others) have been thinking about this a lot over the last couple of months. I called it "Extremely Personal Software" in a blog post a few months ago (https://redfloatplane.lol/blog/14-releasing-software-now/) but there are lots of names and concepts floating about for the same basic idea.

I think it's possible the amount of new software that will be written for an audience of 1-10 will be greater in 2026 than in any previous year, and then the same again for many years to come. I also think a lot of this software will be essentially 'hidden' - people just writing this stuff for themselves because the cost to say things to an agent is very low compared with the cost of actually planning out a software design and so forth.

Interoperability will probably be important in the next few years and I wonder if this is something solvable at the agent/LLM level (standing instructions like 'typically, use sqlite, use plaintext, use open standards' or whatever). I also think observability and ops will be pretty important - many people who want personal software but don't care for the maintenance and upkeep.

Ah, they just today did Formula E in a Pink Pig livery (https://www.fiaformulae.com/en/news/1062899) but I think the Apple livery might have been more apt.