[sillysaurusx]

See also Nat’s twitter announcement: https://twitter.com/natfriedman/status/1712470683207532906

$700k is a life changing amount of money. I admit, it’s tempting to drop everything and go devote myself like a monk to the pursuit of ancient enlightenment via modern ML. I wonder where we’d start…

It’s also funny that the scroll might just be a laundry list.

[terhechte]

This is one of more than 600 scrolls that could be read afterwards if the method becomes scalable. What's more: "excavations were never completed, and many historians believe that thousands more scrolls remain underground." [0]

[0]: https://scrollprize.org

[radarsat1]

> if the method becomes scalable

the machine learning stuff is cool, but it's important not to discount the apparently pretty manual labour still involved in the virtual unwrapping:

> Early in the summer, a small team of annotators (the “segmentation team”) joined our effort. They began mapping the 3D structure of the scroll using tools initially built by EduceLab and improved by our community. By July we had segmented and “virtually flattened” hundreds of cm2 of papyrus.

So, it sounds like it was about a month or two of work, for a single scroll. Although, it probably could be partially or fully automated too, with some effort. Already they developed some tools to help, and I guess it's the kind of task that gets easier after you do it the first time.

[versteegen]

Actually it's much worse than that. Only a very small fraction of the first of the two scanned scrolls has been segmented/unwrapped after 5 months, and it's the easiest parts that are done -- about 1000cm^2 across something like 100 layers of papyrus 10cm wide. Only 50cm^2 of scroll 2 is done. Where the sheets are right against each other is much harder.

[gojomo]

But at the same time: scanning tech & software automation just keep getting better, including via spillovers from other unrelated projects.

The ability of an ML system to learn to mimic what the manual "virtual unrolling" process is doing, from a small number of examples-to-follow, is growing.

Each bit of success, once confirmed by other experts or correlation with other texts, improves the training data.

Eventually a fully-software pushbutton pipeline of "raw imaging to likely texts" should be possible.

And if, say, some of the scrolls are sufficiently 'read' nondestructively to embolden teams to risk destructive techniques – such as incremental ablation while reading the exact chemicals at every coordinate – even higher-resolution data could become available.

[madacol]

But that's expected!

When you are discovering to do a new thing that nobody knows how to do, you first try a bunch of things manually to learn what works and what doesn't.

Only after you have a procedure that reliably works, you automate it.

[ssnistfajen]

Being able to virtually unroll the scrolls within reasonable time/effort/cost will hopefully encourage the Italian government to approve further excavations of the villa for more papyri. If Vesuvius erupts again we don't know how much of the present excavation will survive.

[chakintosh]

Or a customer complaint: https://www.thearchaeologist.org/blog/complaint-tablet-to-ea...

[Wojtkie]

What I love about the Ea-Nasir story is the tablet was found in a pile of other tablets, suggesting that Ea-Nasir saved them. Why? Who knows, maybe he found them funny.

[vimax]

I heard somewhere it was common practice to reuse tablets. It was easier to scrape the surface clean than to make a new tablet.

You'd save any tablets you have, and might wait until you need it to scrape it clean.

In Mesopotamia there was a period where it was fashionable to use a more rare softer red clay on top of the white clay. Your stylus would cut through the top layer leaving nice white letters on a red background. It made it easier to scrape clean and reuse, but much less durable over time.

[jdminhbg]

Yes, the clay tablets were used over and over. The ones that are preserved have what was written on them when they were fired, accidentally, by being in a building that was destroyed by fire.

[cheese_van]

Yeah, and a large number of clay tablets found were training tablets. At any given time, there may have been more students working on properly learning the mechanics of writing on tablets then tablets in circulation so a surprising number of what is found is, basically homework, or apprentice practice.

[daveevad]

That's fascinating to this lay clay tablet thinker. It had never occurred to me that clay tablets were reused. In fact, it nearly fundamentally changes my estimation of them as a long-term durable storage medium.

[PeterisP]

Hardening clay is relatively simple (just as any other pottery) but optional - you can choose to keep the clay unbaked if you want to reuse it (it would dry out but can be made wet again), but you can also make them permanent.

[user3939382]

> You'd save any tablets you have, and might wait until you need it to scrape it clean.

The 3.5” floppies of yore.

[spiritplumber]

The first meme dump.

[wut-wut]

Well done.

[dmix]

> What do you take me for that you treat me with such contempt? …

Apparently "What do you take me for" is an extremely old phrase. Funny how things stick around. I wonder if that's a result of translation though.

[omoikane]

If you like the Ea-Nasir story, there is a whole subreddit dedicated to it:

https://www.reddit.com/r/ReallyShittyCopper/

Also: https://xkcd.com/2758/

[jdminhbg]

> It’s also funny that the scroll might just be a laundry list.

Most likely not, I believe they're starting with scrolls that were readable on the outside, which we know are minor works of Greek stoic philosophy. Also a laundry list would be written on a reusable wax tablet, rather than costly papyrus.

[michael_nielsen]

It's likely somehow a reference to the Emperor. Purple cloth was extremely rare and expensive, and it was the colour worn by the Emperors. Indeed, it eventually became a capital crime for people outside the Emperor's family to wear it. I don't know if that was yet true at the time of Vesuvius, although Wikipedia claims Caligula may have had someone killed for wearing purple.

[Terr_]

> Indeed, it eventually became a capital crime for people outside the Emperor's family to wear it.

Similarly, various Emperors far away in China had a similar enforced color-monopoly, except it was on yellow.

https://en.wikipedia.org/wiki/Color_in_Chinese_culture#Yello...

[thaumasiotes]

Under traditional Irish law, there was no significance to any particular color, but the number of different colors you could wear simultaneously was determined by your status.

The general phenomenon of legal protections on status signifiers goes under the name "sumptuary laws". https://en.wikipedia.org/wiki/Sumptuary_law

[opportune]

Interesting. Does that mean the Yellow Turbans choose Yellow to demonstrate a lack of legitimacy for the Eastern Han emperor?

[Terr_]

AFAICT the imperial color requirement came about a few hundred years later.

[Tao3300]

Tyrian purple was so incredibly expensive that you would have had to do something illegal to get it unless you were already an incredibly wealthy noble with the right political connection.

Still interesting that they found that word. As far as I know the sea snail it comes from didn't inhabit the waters off Herculaneum.

[Arete314159]

The other word visible is "oino", wine. Wine can be described as purple.

[stevula]

I see “oiōn” (οιων), not “oino” (οινο or οινω). I don’t see how it could be read as “wine”. οιων could be several things off the top of my head:

1. the beginning of some form of of οἰωνός “omen”

2. genitive plural of οἶς, meaning “of sheep”

3. a genitive plural of some other word with a stem ending in -οι-, but with the beginning of the word missing. For example, the demonstrative τοίων “of such”, relative οἵων “of which”, or ποίων “of a certain kind”. Or, as speculated in the article, ὁμοίων “same”.

The 3rd option seems most likely to me without any further context. ὁμοίων seems especially plausible since the preceding characters do resemble "ΟΜ".

[thaumasiotes]

Only if you assume that whoever scribed the scroll wasn't too concerned about what order the letters in a word should be written in.

The image is annotated OIWN and the article tentatively identifies the word as OMOIWN, meaning "similar".

[OfSanguineFire]

While modern people make that connection, that is culturally dependent. The color terms available to speakers of a language, and what objects those terms can be associated with, change over time. In the case of the Greek word for "purple", it was connected to a dye and therefore used for clothing, but one shouldn't expect it to be used for wine.

[fsckboy]

https://en.wikipedia.org/wiki/Wine-dark_sea_(Homer)

"Wine-dark sea is a traditional English translation of oînops póntos (οἶνοψ πόντος, IPA: /ôi̯.nops pón.tos/), from oînos (οἶνος, "wine") + óps (ὄψ, "eye; face"), a Homeric epithet. A literal translation is "wine-face sea" (wine-faced, wine-eyed). It is attested five times in the Iliad and twelve times in the Odyssey[1] often to describe rough, stormy seas. The only other use of oînops in the works of Homer is for oxen, for which is it used once in the Iliad and once in the Odyssey, where it describes a reddish colour. The phrase has become a common example when talking about the use of colour in ancient Greek texts."

[Ylpertnodi]

Says you.

[OfSanguineFire]

Says anyone familiar with Ancient Greek, and also anyone who has followed linguistics (even in pop-sci form like Deutscher’s Through the Language Glass) recently. The comment by fsckboy, which was already there when you commented, gets it. For the Greeks, the connotation of wine was different than purple, purple was not a basic color term in that stage of the language, and πορφυρός was used only in the context of purple clothing and the dye used to make it.

[londons_explore]

wine stains are decidedly more purple than the wine they came from too.

[dataflow]

> $700k is a life changing amount of money

Probably ~half of that will go to taxes?

[thrway63245]

Not sure why this is downvoted. Yes, in California half will go to taxes and the rest is enough for a downpayment on a shack. Hardly life changing.

[lazyasciiart]

Pretty damn life changing, actually, for the vast majority of even gasp Californians.

[HeyLaughingBoy]

Must be nice to be a billionaire.

I'm a fairly well paid engineer and I would certainly consider a $700k gift to be life-changing.

[hotnfresh]

I make good, but not FAANG, money, and $350k after-tax, all at once, would still put me most of a decade ahead, financially, and give me some much better options for dealing with a bunch of stuff. Yeah, life changing.

[dataflow]

> would still put me most of a decade ahead, financially

I think that might be the kind of the crux of it. Saving this money to get a decade "ahead" might help your retirement life (assuming things don't collapse before then), but not your current life. Whether that's considered "life-changing" might be in the eye of the beholder.

[lazide]

No one can not consider that life changing in that context?

[hobofan]

It's not a gift, it's a prize, which you are only likely to get if you spend a fulltime effort on it (thus losing out on other income sources). There also is a deadline until end of the year (not sure what happens after that).

So it's more of a (skillful) gambling prize than a gift.

[tiagod]

>Hardly life changing. Maybe your life is pretty good already then.

[ksherlock]

Not sure why this is downvoted :p I'd guess HN income is multimodal with a peak for broke-ass college students and 3rd world programmers (where it is life changing) and another peak for the gainfully employed (where it isn't).

[jedberg]

> It’s also funny that the scroll might just be a laundry list.

Even if it were, a laundry list from 2000 years ago would be a fascinating read.

[jakderrida]

>I admit, it’s tempting to drop everything and go devote myself like a monk to the pursuit of ancient enlightenment via modern ML. I wonder where we’d start…

I think you'd be shocked how well LLMs translate cuneiform in the CDLI notation. What's hilarious is my first attempt included examples in-context and Claude prefaced the translation by stating that there's nothing in my example translations about "bulls", "horns" or "grabbing" and that it will ignore that translation. I looked it up word-by-word and realized Claude was right. Blew me away. Yet Assyriology subreddits were as excited about my findings as lawyer subreddits are about LLMs. Not sure why, either. Just a bunch of, "So what? Does that mean it's useful?".

[dclowd9901]

I don’t think people in those days wrote bullshit. You had to know how to write, and if you did, no one was entreating you to write grocery lists.

[gambiting]

They absolutely did, we have plenty of proof of ancient romans and other cultures writing down jokes, little squabbles, "John was here" on walls etc, one of the oldest known pieces of writing is literally one merchant complaining to another merchant about some marble that wasn't as ordered.

[1-more]

on walls and tablets, but on papyrus? Papyrus was expensive.

[empath-nirvana]

it might cost more than $700k in compute.

[latchkey]

It certainly did. https://news.ycombinator.com/item?id=36312385

[versteegen]

Unfortunately some people are taking you seriously. The whole scrolls are a huge amount of data, but for this First Letters prize using more than one GPU would be overkill. The amount of training data available is actually too small for deep learning to even work well.

[renewiltord]

That's the capital cost of the cluster, not of the compute cost of this operation. Significant and cool, but it over-represents the amount of compute required.

[cosmojg]

Where do they say that the winners used that cluster?

[latchkey]

It is an assumption based on the fact that the codebase uses cuda and the main backer of the project owns the cluster.

[nyssos]

> It is an assumption

Then don't say "certainly"

[latchkey]

Why not? If they had bought the compute themselves, it might cost more than $700k.

[mr_toad]

Many amateurs compete in Kaggle, most of them use whatever hardware they have to hand, and a lot of them will use CUDA directly or indirectly.

[0xf00ff00f]

A laundry list with something purple...

[seydor]

purple was the color of nobility and rather rare. It might be the description of a king or a room or roman fashion items.

[riffraff]

Or a complaint about bad writers

https://en.m.wikipedia.org/wiki/Purple_prose