Hacker News Reader

Comment by WikipediasBad | Hacker News Reader

Agreed. Also, if IPFS is not the preferred way to go about this kind of endeavor, there are the guys at https://lunyr.com/ who are building a Wikipedia in blockchain architecture. About time someone actually attempted this. Additionally, the guys at Everipedia also have the entirety of English Wikipedia imported and forked very nicely so there is at least another large scale effort with a robust dump of Wikipedia already in production.

Lots of different attempts at keeping the vision of "sum total of human knowledge" on the web. :)

View on HN

>Sum total of human knowledge

I like wikipedia, but I always felt like it was missing crucial pieces of information to make it the "sum total of human knowledge." For example, I can look at the articles for ISO_31-11, Spheres, and Sphere Eversions, but I don't think that information alone is enough for me to go from "society of post-apocalyptic 'we lost how to do math'" to "we can prove sphere eversion is possible once again." I chose this example because I'm shit at math and never learned much of it anyway so I can maybe be considered a mad-max post-apocalyptic uneducated peon? Anyway point is I think there's the crucial "teaching" step missing.

Perhaps with a collection of a great deal more articles together, the result "prove eversion is possible" can be taught, but missing for example the ISO_31-11 article alone would make it monumentally difficult, I feel.

https://en.wikipedia.org/wiki/ISO_31-11

https://en.wikipedia.org/wiki/Sphere

https://en.wikipedia.org/wiki/Sphere_eversion

s/ total/mary/

I don't understand this comment :(

They mean "Sum total" -> "Summary"

As in, that's a more accurate descriptor(?)

That's the geekspeak for '"sum total" should be rephrased into "summary"'.

Ah I see, thank you. Is it some sort of programming syntax or like, scientific publishing notation?

It's regular expression syntax for, roughly, "substitute 'mary' for ' total'", which transforms "sum total" - " total" + "mary" --> "summary.

relevant xkcd: https://www.xkcd.com/208/

Only in the sense that it's something to do with regular expressions.

I think sum total is probably overstating it, but Wikipedia provides more than enough of a scaffolding to massively accelerate rediscovery of math. It's less about teaching and more about giving inquiring minds a roadmap with some answers filled in. How far do you think Euclid would have gotten if he could've read Wikipedia for the entirety of his life?

> How far do you think Euclid would have gotten if he could've read Wikipedia for the entirety of his life?

Much less far than he actually went, given that Euclid is known solely for compiling existing work. There's not so much reason to do that when it's already been done and you're happily reading the compilation.

In Euclid's day, "compilation" meant taking disparate and sometimes halfheartedly proved theorems with wildly different terminology and framing and laying them out in one consistent chain of deductions. So I have trouble writing him off as a trumped up textbook author.

It's like saying the formalization of continuity by Weierstrass and Bolzano was compilation because prior mathematicians had already been working with the concepts.

I believe, at some point, Wikipedia was talking about partnering with Archive.org to turn all their citation links into Wayback Machine snapshots—including the ones for journal PDFs.

Presuming such a thing ever gets done, an archival copy of Wikipedia + the Wayback Machine pages Wikipedia references would then make a pretty good "sum total of all human knowledge."

There's a lot that's not written down. Didn't NASA try to coax Apollo engineers out of retirement for their new big space stuff? The engineers had knowledge that wasn't written down, wasn't accessible, or was simply lost.

Sure, with sufficient time you could recreate the knowledge with brand new engineers but you'd be rediscovering lost information.

Did you look into Lunyr much at all? They are just using IPFS on the backend (or plan to use, per their ICO communication to raise money), and using an Ethereum-based ERC20 token to 'incentivize' people to contribute and edit.

Feels like everyone in the ethereum community has sick frontend design skills

Because every such project starts with crowdfunding. Some of them may actually get some code written but not always. /s

Reply on news.ycombinator.com