Hacker News Reader

Sometimes writing it down means very little. Like Perelman's Poincaré conjecture solution. It took some of the best mathematicians years to even prove it being correct.

Didn’t the Egyptians write it down? But we still have no idea how to understand their hieroglyphics. Or how they even figured out their math.

Some light info on it: https://www.storyofmathematics.com/egyptian.html

I don't think they proved the correctness, just read it very thoroughly and were convinced by Perelman's reasoning.

enter Space X: hold my beer

I think you're correct about the importance of the context in which the expertise exists in.

But even so, I think it's clear that with more broad documentation, we are in a better position to pick up the expertise of the past.

So in the future, all the wonderful, nutty youtubers that keep alive old arts and give demos would be especially invaluable. Not just in clarifying specific techniques and knowledge, but by demonstrating a general approach to problems that existed at the current time.

I've been watching all sorts of things the past couple of months like this: People making vacuum tubes, repairing vacuum tube equipment, older mainframe technologies like ancient versions of MVS, machining parts on equipment from the '30's.

None of it is how-things-are-done in a cutting edge factory or step by step directions on how to reproduce an important cultural artifact, but from the perspective of future people, it will be a valuable aid to gathering that context.

One I'm particularly reminded of is Clickspring on youtube who is building an antikythera mechanism. Even though we have the partial artifact, so little is known concerning metal working technologies of the time that it seems to be an anachronism completely alien to its time of origin.

His approach is to explore what technologies would make sense to a working craftsman of any era (which also could be surmised to have existed in the era). It's interesting to watch and it has apparently born fruit in terms discovering academically important information about the device.

Totally agree with you, with the one caveat that Youtube is not operated with long term accessibility in mind. Videos routinely are removed by their creator, or get taken down for silly reasons. And one day Youtube will disappear as well.

Archive the channels you care about - they might otherwise be impossible to find in a few decades, or a few years. Youtube-dl is one of my favorite pieces of software.

Digital artefacts could be, paradoxically, less durable going in the future - if you write and publish a physical book, it might get forgotten in a library somewhere and rediscovered 200 years after you die.

If you publish an ebook on your website and die and no one cares, the probability of anyone discovering it by happenstance 200 years later goes down to 0.

This is a very unsolved problem. It doesn't take much foresight to anticipate a Library of Alexandria v2.0 around the corner.

Certainly, we can and should individually archive what's important to us, but more generally (archiving information for the next 1000 years say), I dunno. I hope someone will figure it out.

"there was a library in the distant past that held the entirety of human knowledge. but all was lost when the ancient people who built it lost their 3rd-party cookie trade route to the European invaders"

> If you publish an ebook on your website and die and no one cares, the probability of anyone discovering it by happenstance 200 years later goes down to 0.

This seems overly pessimistic given the existence of archive.org (they accept donations!).

If archival institutions keep existing and improving their search and accessibility, it seems very possible that random websites will be accessible for a long time to come; 200 years is definitely within the realm of feasibility.

A relevant talk by Jonathan Blow: https://youtu.be/ZSRHeXYDLko

With more information going to a cloud, and storage formats becoming obsolete faster, I wonder if we're actually better now. Certainly transitioning from hand-written and typewriter paper documents to digital has allowed for data to be copied easier than ever; however, I can still go to the library and read a book from 50 years ago or more. I can't do the same for digital data, at least not without specialized equipment and systems.

Lots of documents are going 'paperless' these days to save the trees. Given what I've witnessed working in tech I'm worried that many of those systems will be backed by a single server with no backups. We are lucky that many of our documents are stored in email which is a little more robust but still it's no stone tablet.

Hell, pulp paperbacks, probably the worst books for longevity, from 70 years ago are usually fine as long as they've only been gently read a low-single-digits number of times and haven't gotten wet/mildewy. They're often past yellowing and into browning at the edges, but they're fine for reading.

Hardcovers from the 19th century are pretty common. Tons of low-value ones (textbooks, fiction by at-the-time very popular authors who've been forgotten) floating around at a couple dollars a piece in flea markets here in the US, and I bet even older books are somewhat common in Europe. Lots by famous authors but of course those are usually more expensive. Most of those weren't finely-printed on modern archival paper or anything like that, and haven't been kept in archival conditions, but often have a few years of hard-reading life left in them (and much more well-shelved), so 150-200 years is a fairly easy age to achieve for gently-read hardbacks using cheap (but not the very cheapest) materials. Maybe more if you re-bind them once or twice, or they're very rarely read.

That's without even getting into really nice material and binding, or very expensive non-paper for the pages if you're very serious about keeping a book in good shape for centuries. Not even tape comes close—we don't have a really good archival format for digital stuff yet, let alone one that normal people may have a bunch of in their houses (like books). Maybe one of the several long-lived storage formats under development will take off one of these days and our digital storage will get a bit less risky. Right now it's pretty bad, only saved by being extremely cheap compared to other options.

there's also john titor, the early internet small celebrity that traveled back in time because the future society run on some old obscure 8bit computer and they run out of parts

I don't know if NASA themselves could stick Saturn engines onto new rockets, but I reckon the Air Force probably could. Air force technicians are pretty good at fabricating and fitting parts from what I've seen.

Skills like hand fabrication and fitting aren't something that you can store. You can record videos or interview skilled tradesmen as much as you want, but the real skill is something that is learned through experience, it's muscle memory and intuition, not theoretical knowledge.

I worked in aerospace during the transition from Ariane 4 to Ariane 5. The Ariane 4 was a versatile modular launch system and widely considered a successful project. The Ariane 5, on the other hand, was struggling at the time. A large part was because it originally was designed to be a launcher for the Hermes space plane. Hermes was never built, but the design decisions that were made because of it had many negative consequences for the Ariane 5. Most importantly it was not very economic for launching smaller satellites.

It was quite natural to think of enhancing the successful Ariane 4 design instead of building Ariane 5. From the conversations I remember from that time this was not considered a viable path - mostly because the supply chain of the Ariane 4 parts had already gone dry and it was not considered realistic to rebuild them.

For me as a young engineer this was very surprising, but a few years later I understood two things much better:

1. The knowledge to build projects of this scale is very distributed. There were subcontractors over subcontractors (spread over several countries) and every one had some tacit knowledge viable for the project.

2. Setup cost for parts production is relevant, even for the small series productions common in aerospace.

While this is I’m sure true, the main point is about knowledge transfer. New people come onto the job, and they need to learn how to do it, and their skill level will always be basically zero. The question is, how do they learn how to do it, reading books or having a mentor watching over them? Certainly having a mentor will make that far easier, but if all the people who knew how to do it are now retired or especially dead, this isn’t possible, and so it becomes considerably more difficult for a new person to learn.

This is an example of institutional knowledge. It's part of the reason why you want to have some military production even if you have enough gear already. You might need people with this kind of expertise in the future.

Even with math, there is much more "unwritten shared knowledge" than usually assumed. The corpus of current mathematics lives much more in the shared understanding of mathematicians than in the pages of articles. Math fields may indeed be lost if they are not actively practiced. There was a beautiful essay on this by Bill Thurston, that I don't have handy at the moment, but should be easy to find.

[EDIT:] maybe this: https://mathoverflow.net/questions/43690/whats-a-mathematici...

Which episode is that?