Hacker News new | past | comments | ask | show | jobs | submit login
Ask HN: So what was Knuth's "earthshaking announcement"?
61 points by Kliment on July 1, 2010 | hide | past | web | favorite | 31 comments
Knuth's mysterious announcement at the TeX user group gathered some interest on here, but even several hours later no specifics about it seem to be anywhere online. Can someone who was there or found out about it share what it was about?

I did a quick twitter trawl an hour or so after, and it seems like it was actually a joke: a new successor to TeX, using XML, arbitrary precision, and auto-layout.

Also, my favourite was this one: http://twitter.com/elmindreda/statuses/17444964495 "Anyone looking to profile Twitter users can easily identify all programmers today simply by searching for "Knuth"."

Knuth's sense of humor shouldn't be underestimated-- there are a lot of great jokes hidden in the index of TAOCP, among other places.

And, of course, his first publication was in Mad Magazine.

I tweeted about it from the event but the corresponding news.yc submission got killed.



I'm not sure that HN generally looks too kindly on individual tweeds being submitted as news items.

Probably would have had more credibility if you had made a blog post about it then put the link here on HN.

I truly worry about the overzealous killing of yc news stories. I can't be the only one who's noticed a recent uptick in the number of stories killed for no apparent reason.

  1. It's a tweet submitted as 'news'.
  2. It seems that the 'earth shattering' announcement was boring.
I think there's a couple of good reasons why it got killed.

Unfortunately, when they killed it I assumed that it was because it was a hoax submission. It was news that this community was waiting days (months?) to hear about, so, although perhaps not a full news story, it was certainly of interest here. And probably of more interest than "X reasons I love/hate Y" for a lot of us.

Same thing happened to Reddit years ago. As if a battalion 4chan script kiddies had nothing better to do than sick a downvoting botnet on Reddit to decrease its signal:noise ratio. I hope they haven't targetted HN now too. :/

Downvotes don't kill stories, moderators kill stories.

I thought I read that enough flags if a story has less than 10 points will automatically kill it.

Am I the only one who's a little disappointed?

In retrospect, if Knuth had an actual "earthshaking announcement" to make, he probably would have pre-announced it a bit more prominently, or not at all.

Knuth played a joke and "announced" iTeX.


The problem is, now if he really proves P=NP, nobody will believe him.

If Knuth announces that he proved P=NP a lot of people will believe him.

That was kind of the joke...

Most would not believe him straight away, since it is highly unlikely that P=NP. If the proof holds up to scrutiny though, it is not an issue of belief any more (apart form the social process of proof of course).

Actually, in formal areas of academia (like math, where I studied), there can be considerable use of intuition even after a 'proof' goes out.

I would say this is a good thing, as there are also many cases where a 'proof' is later shown to have holes, areas where everyone made an assumption and didn't realize it.

A good example of this involves limits; for many years mathematicians proved a number of interesting results using infinitesimal limits. It was many decades before a mathematician (Riemann? Maybe?) noted that you can't assume that all the limits in an equation approach their number at the same rate.

This was a super smart thing to notice, and junked a huge number of 'proofs' that had all relied unobtrusively on this idea. All that to say, good theoreticians spend a lot of time wondering if an idea makes sense to them or not.

This is more or less what I wanted to allude to with the social process of proof. The number of things that are actually rigorously formally proven is surprisingly small IMHO.

I think the problem is that we can't write down all our assumptions, even if we try REALLY hard. This is some of what inspired the mathematical formalism movement; trying to make sure that we know what we think we know.

Fortunately mathematics is based on rigorous formal proofs rather than belief, so if he really achieved that he would just publish the proof and become even more famous.

I may have missed a joke here, but even mathematics is based on belief.


I heard that the iTeX rumor was created by my neighbor. Same with HiTeX and TeXML.

Knuth didn't prove that P=NP. He proved that P-NP=42.

You can see a video of the talk, uploaded with Knuth's permission, here:


TeX 3.15

Tex 6.28

Tex Tau

A "joke".

I wasn't aware adults actually used twitter. I thought it was for 12 year old girls

There are twits of all ages.

Applications are open for YC Summer 2020

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact