Skip to the Main Content

Note:These pages make extensive use of the latest XHTML and CSS Standards. They ought to look great in any standards-compliant modern browser. Unfortunately, they will probably look horrible in older browsers, like Netscape 4.x and IE 4.x. Moreover, many posts use MathML, which is, currently only supported in Mozilla. My best suggestion (and you will thank me when surfing an ever-increasing number of sites on the web which have been crafted to use the new standards) is to upgrade to the latest version of your browser. If that's not possible, consider moving to the Standards-compliant and open-source Mozilla browser.

February 5, 2018


Posted by John Baez

Since nothing get parodied until it’s sufficiently well-known to make it worth the effort, this proves the nnLab is a success:


Click on the links!

Posted at February 5, 2018 6:23 AM UTC

TrackBack URL for this Entry:

17 Comments & 0 Trackbacks

Re: mLab

Hahahaha… According to my non-cat colleagues, that’s exactly how category theory looks to an outsider :)

Posted by: Paolo on February 5, 2018 11:01 AM | Permalink | Reply to this

Re: mLab

Of course, there’s nothing to prevent people asking questions about entries at the nForum. Hopefully at least some entries are largely comprehensible, and many at least somewhat comprehensible.

Some entries are even designed to be pedagogic, such as motivation for sheaves, cohomology and higher stacks. We could perhaps do a better job of signalling such pages. I should think that if you’re not getting much from these, then the nLab is not for you.

Also let us not forget just how few people contribute to the nLab, and how much work has been done by non-tenured people for no personal gain. Just because it has grown to a considerable size now, and is noticed from time to time, e.g., on MathOverflow, doesn’t mean it’s the product of a coterie of influential mathematicians promulgating their view of the world.

Perhaps the mLab architect would care to join the team.

Posted by: David Corfield on February 5, 2018 11:26 AM | Permalink | Reply to this

Re: mLab

Well, I think most of us knows that the nLab is precisely not just a bag of jargon.

Personally, for example, I really love that basically all entries have an “Idea” section, which is usually very useful and pedagogical, and makes the rest of the article much easier to understand. Moreover, while most concepts on the nLab are abstract and formal, they make a lot of sense, it’s not literally “abstract nonsense”. All of that has a precise and deep meaning, and I’m sure nobody in the community doubts it.

Still, I find the parody quite funny. Maybe exactly because I know that the real thing is not like that, it just may look like that at a first, superficial look.

Posted by: Paolo on February 5, 2018 2:44 PM | Permalink | Reply to this

Re: mLab

It made me smile, although I expect it’s largely based on an engine like Mathgen and there’s nothing particularly witty going on as parody. I’m vaguely imagining the “architect” loading a bunch of nLab vocabulary items and style formats into the engine and letting it rip, although the formulae in display lines might have required a bit more cleverness.

Posted by: Todd Trimble on February 5, 2018 3:18 PM | Permalink | Reply to this

Re: mLab

Yes, definitely nice work coding the engine that creates the mLab pages! Just hit reload to see lots more in short order.

Posted by: stefan on February 5, 2018 4:52 PM | Permalink | Reply to this

Re: mLab

I expect it’s largely based on an engine like Mathgen

If you poke around on mLab’s website you find in

The mLab

A satire generator making fun of the nLab, a wiki for higher mathematics and category theory. Generates totally legitmate articles about totally legitimate category theory. It works using my package nearley-generator, which turns a Nearley grammar into an efficient fake-text generator.

Posted by: RodMcGuire on February 5, 2018 5:08 PM | Permalink | Reply to this

Re: mLab

I should have mentioned that the mLab is generated from just a few smallish hand coded grammar files. Here they all are:

Posted by: RodMcGuire on February 5, 2018 5:37 PM | Permalink | Reply to this

Re: mLab

David wrote:

Perhaps the mLab architect would care to join the team.

The mmLab was created by Chase Meadors. Since his homepage features a pushout diagram, I suspect he’s probably spent a fair amount of time perusing the nnLab, and likes category theory.

It’s hard to tell what he thinks about the nnLab. But from my brief forays into the public eye I’ve learned that the only way to avoid parodies and other potentially hurtful criticisms is to remain invisible. If one is going to do anything that matters, it helps to have a thick skin. Even better, it helps to treat most attacks as signs that one is succeeding.

But you already know this from your work on philosophy!

Posted by: John Baez on February 5, 2018 6:00 PM | Permalink | Reply to this

Re: mLab

Yes, in announcing the mLab on Reddit, the creator wrote:

If you’ve ever googled any topic in the ballpark of category theory, you’ve probably ended up at the nLab, a wiki for [higher] category theory. Though it’s a great resource, the most notable thing about nLab for me is how utterly impenetrable most of the articles are if you’re not versed in the field. So, I made this page that generates fake nLab articles to deliver the same authentic experience as reading an nLab article.

Posted by: Alexis Hazell on February 5, 2018 9:24 PM | Permalink | Reply to this

Re: mLab

Hey John, author of mLab here. Found this topic and just wanted to express how pleased I am that one of the founders themselves found this project!

Rest assured, I intended this project to be merely a good-natured jest. I’m truly a fan of the nLab, and I used it a great deal when I was learning some Category Theory. I think it’s a great resource and, importantly, it does what a wiki does best - connect information with other information to explore topics from every possible angle. If anything, this is poking more fun at the particularly high amount of jargon in Category Theory in general.

Thanks for the mention and all the valuable contributions from yourself and others to nLab!

Posted by: Chase Meadors on February 7, 2018 10:51 PM | Permalink | Reply to this

Re: mLab

I was wondering if anyone else looks at that nLab logo and things of Matisse’s late cutout phase. Is the logo supposed to be reminiscent of eternal inflation do you think?

Posted by: mmultiplier on February 5, 2018 5:11 PM | Permalink | Reply to this

Re: mLab

See here.

The logo was inspired by my [Jacques Distler’s] 8 year old son (and loyal Instiki user)’s Science Fair project.

A quasi-two-dimensional droplet of ferrofluid, when subjected to a perpendicular magnetic field, forms labyrinthine patterns like the one depicted. The effect is a competition between magnetic energy (which makes the fluid want to spread out) and surface tension (which wants to make the droplet circular).

The image was auto-traced, in Adobe Illustrator, from a photograph my son took of a droplet of ferrofluid squished between two glass plates.

Posted by: Todd Trimble on February 5, 2018 5:46 PM | Permalink | Reply to this

Re: mLab

But yes, your comment about Matisse might be relevant as well. Here is Matisse’s La Gerbe, which is a term you’ll certainly find on the nLab. :-)

Posted by: Todd Trimble on February 5, 2018 5:56 PM | Permalink | Reply to this

Re: mLab

I clicked through several pages, and none of them contained any “theorems”. Only definitions along with some properties and examples. I wonder if that’s part of the joke.

Posted by: chris r on February 5, 2018 11:45 PM | Permalink | Reply to this

Re: mLab

Maybe! Good point!

I think the mmLab could be significantly improved by two fairly easy changes:

1) When you click on a linked term, right now you go to a randomly created new page with a randomly generated title. I think it should be easy to make the title match the term you clicked on.

2) When you click on the subject headers at the top of the page, you don’t go anywhere. It would be nice if you went to sections with those titles.

A bit harder:

3) It would be nice if the term in the title appeared repeatedly in the page.

However, I barely have enough time to spend much on improving the nnLab, so I won’t be spending any on improving the mmLab.

Posted by: John Baez on February 6, 2018 12:34 AM | Permalink | Reply to this

Re: mLab

One notable predecessor of this is Carl Westerholm’s book Mathematics Made Difficult, which came out in 1971. That was, of course, written by hand (and is occasionally very funny, though I think that there isn’t enough humour in the concept to stretch to an entire book).

Posted by: Graham White on February 8, 2018 11:36 AM | Permalink | Reply to this

Re: mLab

Randomly generated nLab music festival t-shirts, courtesy of Tim Hosgood.

Posted by: Tom Leinster on May 2, 2020 10:21 PM | Permalink | Reply to this

Post a New Comment