This was the pre-Internet precursor of wikis. This program was likely venerable, shame it predates Ciro Santilli's era.
But the thing was much more bloated it seems, and also included visual programming elements, and WYSISYG UI creation.
Video 1.
Hypercard by The Computer Chronicles (1987)
. Source.
No open signup it seems. TODO CV of owner.
Originally by Springer, but later moved to the European mathematical society.
Written mostly by Eric W. Weisstein.
Ciro once saw a printed version of the CRC "concise" encyclopedia of mathematics. It is about 12 cm thick. Imagine if it wasn't concise!!!
Infinite Napkin is the one-person open source replacemente we needed for it! And will be the final multi-person replacement.
Ahh, this dude is just like Ciro Santilli, trying to create the ultimate natural sciences encyclopedia!
____ In 1995, Weisstein converted a Microsoft Word document of over 200 pages to hypertext format and uploaded it to his webspace at Caltech under the title Eric's Treasure Trove of Sciences. ____
Decent encyclopedia of mathematics. Not much motivation, mostly statements though.
Created by:
Unlike Wikipedia, they have a more sane forum commenting system, e.g. a page/forum pair:
Based on GitHub pull requests:
Joe Corneli, of of the contributors, mentions this in a cool-sounding "Peeragogy" context at
I earned my doctorate at The Open University in Milton Keynes, with a thesis focused on peer produced support for peer learning in the mathematics domain. The main case study was; the ideas also informed the development of “Peeragogy”.
A wiki that gathers mathematical proofs.
This appears to be the creator: "Joe George".
Cosmopedia is a dataset of synthetic textbooks, blogposts, stories, posts and WikiHow articles generated by Mixtral-8x7B-Instruct-v0.1.The dataset contains over 30 million files and 25 billion tokens, making it the largest open synthetic dataset to date.
App-only as of 2023, i.e. for children.
Humans make the table of contents, and then AI fills it. Ciro was thinking about doint the exact same thing at some point, maybe starting from Wikipedia categories.
This section is about wikis that are hosted on a blockchain of some sort.
Appears to be a Wikipedia clone but with much lower/no notability requirements guidelines, which overcomes one of Wikipedia's main issues: deletionism.
They do have the interesting idea of importing deleted Wikipedia pages as a source of content, which leads to some epic "most viewed pages" such as which currently reads:
Stop Being Pervs, Go Watch Lichfaop/Faoplich Instead and you can also visit MR Info 24 for more details.
We can for example see Ciro Santilli's deleted entry PsiQuantum at:, Wikipedia deletion page: Their attribution is atrocious however, e.g. it does not seem possible to find any mention of "Ciro Santilli" on the edit history, which just points to the delete article which is not visible anymore. They could really get into trouble for this one day.
Their main use case, as suggested by the website itself, if for people/brands to create pages about themselves.
This combined with the lack of "one version of each page per person" seems like an explosive invitation for unsolvable edit wars.
The website is backed by a French startup:
April 2024: merged with some fraud protection thing, is it sill a Wiki? Unclear, seem sto have lost that aspect:
Social media:
To state the obvious: Wikipedia is an incredibly useful website, but Gomila pointed out that notable companies and technologies like SV Angel, Benchling, Lisk and Urbit don’t currently have entries. Part of the problem is what he called Wikipedia’s “arbitrary notability threshold,” where pages are deleted for not being notable enough. (This is also what happened years ago to the Wikipedia page about yours truly — which I swear I didn’t write myself.)
Exactly! Deletionism on Wikipedia is so sad, and especially for companies. In particular e.g. Ciro Santilli tried to create a page for PsiQuantum, and it got reverted... and now golden has one of the largest Google hits for it:
TODO how do they do moderation?
As of April 2024
Login is currently disabled.
Asked at: Their last tweets were from August 2023, so maybe they just silently shutdown? Their name is too generic and hard to search for efficietnly...
They do have knowledge graph built-in which is cool.
WikiAlpha is an alternative to Wikipedia, where the main difference is that our deletion policy is far more lenient with regard to notability requirements. Basically, WikiAlpha is a near-indiscriminate collection of information in the form of articles on any topic: you can create an article about the band you just started, your pet dog, yourself, your house - as long as your content does not fall under our speedy deletion policy, it will likely remain on the site forever!
Video 1.
10k GitHub Stars by BookStack (2022)
. Source. Answering to an AMA unfortunately :-) But some OK small bits of information trickled through.
The dominating meme database as of 2020.
As of 2022 visible at:
Apparently they had a separate URL as just, so they were somewhat serious about it before shutting it down.
As of 2022 marked:
This page has been archived and is no longer updated
RIP. has last entry 2015, so presumably that's the shutdown year.
Self description:
Using our platform, you can customize your own eBooks for your students. Create an online classroom. Contribute and share content and connect with networks of colleagues.
so quite related to
Tree based organization at last.
Amazing WYSIWYG, including maths and tables, plus insane plugins like canvas mode, and specific file formats like code/mermaid diagrams/drawing mode.
Version history.
No multiuser features. Except for that, could have been a good starting point of an online multiuser thing such as!
Only possible to see one page at a time on output? Output chunking is a major feature of OurBigBook, I'm so proud.
Their tree based approach does have a problem however for the use case of sharing topics across users: every level forces is a scope. Which makes it basically impossible to reliably match topics across users.
HTML export keeps all data as HTMl is their native format. The files are mostly visible, but there is some CSS missing, it is not 100% like editor, notaby math is broken. There is also a hosted way of exposing:
Markdown export warns:
this preserves most of the formatting.
Architecture: runs on local SQLite database via better-sqlite3. Data apparently stored in SQLite database at ~/.local/share/trilium-data, no raw files.
Markup is stored as HTML as seen from: sqlite3 document.db 'SELECT * from note_contents'. HTML is their native storage format, quite interesting.
WYSIWYG based on which is a dependency. It is kind of cool that the view in which you view the output is exactly the same as the one you edit in, and there is no intermediate format, just the HTML.
Math is KaTeX based.
Why Wikipedia sucks: Section "Wikipedia".
The most important page of Wikipedia is undoubtedly: which lists the accepted and non accepted sources. Basically, the decision of what is true in this world.
Wikipedia is incredibly picky about copyright. E.g.: because "such portrait could be created". Yes, with a time machine, no problem! This does more harm than good... excessive!
Citing in Wikipedia is painful. Partly because of they have a billion different templates that you have to navigate. They should really have a system where you can easily reuse existing sources across articles! Section "How to use a single source multiple times in a Wikipedia article?"
Video 1.
What Happened To Wikipedia's Founders?
. Source.
Video 2.
Inside the Wikimedia Foundation offices by Wikimedia Foundation (2008)
. Source.
Video 1.
What Mental Breakdown Of a Wikipedia Moderator looks like by Vince Vintage
. Source.
Some exmaples by Ciro Santilli follow.
Of the tutorial-subjectivity type:
Notability constraints, which are are way too strict:
  • even information about important companies can be disputed. E.g. once Ciro Santilli tried to create a page for PsiQuantum, a startup with $650m in funding, and there was a deletion proposal because it did not contain verifiable sources not linked directly to information provided by the company itself: Although this argument is correct, it is also true about 90% of everything that is on Wikipedia about any company. Where else can you get any information about a B2B company? Their clients are not going to say anything. Lawsuits and scandals are kind of the only possible source... In that case, the page was deleted with 2 votes against vs 3 votes for deletion.
    should we delete this extremely likely useful/correct content or not according to this extremely complex system of guidelines"
    is very similar to Stack Exchange's own Stack Overflow content deletion issues. Ain't Nobody Got Time For That. "Ain't Nobody Got Time for That" actually has a Wiki page: That's notable. Unlike a $600M+ company of course.
There are even a Wikis that were created to remove notability constraints: Wiki without notability requirements.
For these reasons reason why Ciro basically only contributes images to Wikipedia: because they are either all in or all out, and you can determine which one of them it is. And this allows images to be more attributable, so people can actually see that it was Ciro that created a given amazing image, thus overcoming Wikipedia's lack of reputation system a little bit as well.
Wikipedia is perfect for things like biographies, geography, or history, which have a much more defined and subjective expository order. But when it comes to "tutorials of how to actually do stuff", which is what mathematics and physics are basically about, Wikipedia has a very hard time to go beyond dry definitions which are only useful for people who already half know the stuff. But to learn from zero, newbies need tutorials with intuition and examples.
Per-table dumps created with mysqldump and listed at: Most notably, for the English Wikipedia:
A few of the files are not actual tables but derived data, notably from Download titles of all Wikipedia articles
The tables are "documented" under:, e.g. the central "page" table: But in many cases it is impossible to deduce what fields are from those docs. contains a list of categories. It only contains the categories and some counts, but it doesn't contain the subcategories and pages under each category, so it is a bit pointless.
The SQL first defines the table:
CREATE TABLE `category` (
  `cat_id` int(10) unsigned NOT NULL AUTO_INCREMENT,
  `cat_title` varbinary(255) NOT NULL DEFAULT '',
  `cat_pages` int(11) NOT NULL DEFAULT 0,
  `cat_subcats` int(11) NOT NULL DEFAULT 0,
  `cat_files` int(11) NOT NULL DEFAULT 0,
  PRIMARY KEY (`cat_id`),
  UNIQUE KEY `cat_title` (`cat_title`),
  KEY `cat_pages` (`cat_pages`)
followed by a few humongous inserts:
INSERT INTO `category` VALUES (2,'Unprintworthy_redirects',1597224,20,0),(3,'Computer_storage_devices',88,11,0)
which we can see at:
Se see that
so it contains only categories.
We can check this with:
sed -s 's/),/\n/g' enwiki-latest-category.sql | grep Computer_storage_devices
and it shows:
There doesn't seem to be any interlink between the categories, only page and subcategory counts therefore.
Let's observe them in MySQL:
mysql enwiki -e "select page_id, page_namespace, page_title, page_is_redirect from page where page_namespace in (0, 14) and page_title in ('Computer_storage_devices', 'Computer_data_storage')"
| page_id  | page_namespace | page_title               | page_is_redirect |
|     5300 |              0 | Computer_data_storage    |                0 |
| 42371130 |              0 | Computer_storage_devices |                1 |
|   711721 |             14 | Computer_data_storage    |                0 |
|   895945 |             14 | Computer_storage_devices |                0 |
mysql enwiki -e "select cl_from, cl_to from categorylinks where cl_from in (5300, 711721, 895945, 42371130)"
| cl_from  | cl_to                                                                 |
|     5300 | All_articles_containing_potentially_dated_statements                  |
|     5300 | Articles_containing_potentially_dated_statements_from_2009            |
|     5300 | Articles_containing_potentially_dated_statements_from_2011            |
|     5300 | Articles_with_GND_identifiers                                         |
|     5300 | Articles_with_NKC_identifiers                                         |
|     5300 | Articles_with_short_description                                       |
|     5300 | Computer_architecture                                                 |
|     5300 | Computer_data_storage                                                 |
|     5300 | Short_description_matches_Wikidata                                    |
|     5300 | Use_dmy_dates_from_June_2020                                          |
|     5300 | Wikipedia_articles_incorporating_text_from_the_Federal_Standard_1037C |
|   711721 | Computer_architecture                                                 |
|   711721 | Computer_data                                                         |
|   711721 | Computer_hardware_by_type                                             |
|   711721 | Data_storage                                                          |
|   895945 | Computer_data_storage                                                 |
|   895945 | Computer_peripherals                                                  |
|   895945 | Recording_devices                                                     |
| 42371130 | Redirects_from_alternative_names                                      |
So we see that cl_from encodes the parent categories:
  • parent categories of categories:
    •, which has ID 711721, has parent categories: "Computer hardware by type", "Computer data", "Data storage", "Computer architecture". This matches exactly on the database. These are all encoded on the source code of the page:
      [[Category:Computer hardware by type]]
      [[Category:Computer data|Storage]]
      [[Category:Data storage|Computer]]
      [[Category:Computer architecture]]
    • has parent categories: "Computer data storage", "Recording devices", "Computer peripherals". This matches exactly on the database.
  • parent categories of pages:
    • whish is a redirect gets the magic category "Redirects_from_alternative_names", a humongous placeholder with many thousands of pages:
    • shows only two categories onthe web UI: "Computer data storage" and "Computer architecture". Both of these are present on the database and at the end of the source code:
      {{DEFAULTSORT:Computer Data Storage}}
      [[Category:Computer data storage| ]]
      [[Category:Computer architecture]]
      The others appear to be more magic. Two of them we can guess from the templates:
      {{short description|Storage of digital data readable by computers}}
      {{Use dmy dates|date=June 2020}}
      are likely Use_dmy_dates_from_June_2020 and Articles_with_short_description but the rest is more magic and not necessarily present in-source.
So to find all articls and categories under a given category title, say we can run:
mariadb enwiki -e "select cl_from, cl_to, page_namespace, page_title from categorylinks inner join page on page_namespace in (0, 14) and cl_from = page_id and cl_to = 'Mathematics'"
Definition, anywhere on article, likely ideally as the first usage:
<ref name="myname">{{cite web ...}}</ref>
And then you can use it later on as:
<ref name="myname" />
which automatically expands the exact same thing, or using the shortcut:
To cite multiple pages of a book:, the best method is to define and use the reference without adding the p or location in cite as:
<ref name="googleStory">{{cite book |title=The Google Story}}</ref>{{rp|p=123}}
Do not set the page in cite, otherwise it shows up on the references. Instead we use the {{rp}} template. And then use the reference with the {{r}} template as:
or for multiple pages:
{{r|googleStory|pp=123, 156-158}}
A good big sample definition:
<ref name="googleStory">{{cite book |last1=Vise |first1=David |author-link1=David A. Vise |last2=Malseed |first2=Mark |author-link2=Mark Malseed |title=The Google Story |date=2008 |publisher=Delacorte Press |url=}}</ref>
There is also title-link to link to a wiki page. But it is incompatible with url= for Internet Archive Open Library links which is a shame.
So, it turns out that Wikipedia does have a (ultra obscure as usual) mechanism for pull requests. You learn a new one every day.
OMG they have that. Slightly slightly overlap with
A 2022 clone of gives first commits from 2003 by:
  • Lee Daniel Crocker:
    He is best known for rewriting the software upon which Wikipedia runs, to address scalability problems.
    so that gives a good notion of the last major rewrite.
  • Brion Vibber
TODO when was wikipedia open sourced from Nupedia? The ealry days of Wikipedia are quite obscure due to its transition from Nupedia.
Cool tool that allows you to graphically visualize page viewc counts of specific pages. It offers somewhat similar insights to Google Trends.
The homepage shows views of selected pages, e.g. when Google had their 25th birthday:|Dog|Larry_Page Larry Page briefly beat "Cat" and "Dog".
/topviews shows the most viewed pages for a given month: It is extremelly epic that XXX: Return of Xander Cage, a 2017 film, is on the top ten of the August 2023 month. The page was around 8th place on a Google search for "xxx": at the time. XXXX (beer) was also on the top 20, followed by Sex on 21.
Because of edit wars and encyclopedic tone requirements. See also: Wikipedia.
One thing to note is that Jimmy was a finance worker before starting wikipdia, e.g. he had capital to hire Larry Sanger.
Maybe that's the way to go about it, make money first, and later on change the world.
Starting just after the beginning of the Internet can't hurt either. Though tooling must have been insane back then.
Video 1.
Meet the man behind a third of what's on Wikipedia
. Source.
Open source software engine created for and used by Wikipedia.
Their reference markup is incredibly overengineered, convoluted, and underdocumented, it is unbelivable!
Use the reference:
This is a fact.{{sfn|Schweber|1994|p=487}}
Define the reference:
*{{Cite book|author-link=Silvan S. Schweber |title=QED and the Men Who Made It: Dyson, Feynman, Schwinger, and Tomonaga|last=Schweber|first=Silvan S.|location=Princeton|publisher=University Press|year=1994 |isbn=978-0-691-03327-3 |url= |url-access=registration}}
sfn is magic and matches the the author last name and date from the Cite, it is documented at:
Unforutunately, if there are multiple duplicate Cites inline in the article, it will complain that there are multiple definitions, and you have to first factor out the article by replacing all those existing Cite with sfn, and keeping just one Cite at the bottom. What a pain...
You can also link to a specific page of the book, e.g. if it is a book is on Internet Archive Open Library with:
{{sfn|Murray|1997|p=[ 86]}}
For multiple pages should use pp= instead of p=. Does not seem to make much difference on the rendered output besides showing p. vs pp., but so be it:
{{sfn|Murray|1997|pp=[ 86-87]}}
A really good option to store educational media such as images and video!
Shame that like the rest of Wikimedia, their interface is so clunky and lacking obvious features.
This is basically what Jimmy Wales had originally set out to make Wikipedia, a peer reviewed thing.
But then he noticed the entry barrier was too high while inviding an economist to review an article he wrote, and just made the more open thing instead.
The venerable first wiki.
The pre-Eternal September feeling is palpable.
People could freely comment their thoughts and sign below, making it much closer to what Ciro Santilli wants to be. But with upvotes ;-)
Nothing can better encapsulate the nostalgia of early day Internet. Genius at times, banal at others, you will be forever in our hearts!