Profile cover photo
Profile photo
Alex Fink
Alex's posts

Post has attachment
In commemoration of the life of Raymond Smullyan (1919–2017), here's some video +Sai took at Gathering for Gardner 10 of some close-up card magic, a little of it performed by Raymond. The majority is by Jorge Luengo, also very much worth watching:

Post has shared content
Practical issue: I need to rewrite my script for the BOS TSA video (intro, voiceover, outro), and decide how to compress the full length CCTV (40m) into something viral length (~6m total* including intro, outro, etc).

I'd appreciate help from someone willing to read the documents, watch the video, and help compose a better script & video edit.

I find dealing with this extremely distressing. I can do the mechanics of editing myself — 'video edit' above means an outline of what to do, eg "speed up from x to y, then go back to 1x speed, burn in subtitles at this time saying z, say this in voiceover during it, intercut this image here".

It's all the deciding and writing and having to look at it over and over that makes me have severe anxiety reactions to all of this, which is the biggest part of why I haven't been able to get it done yet…

* I'll be publishing the uncut video, but I want to put it up simultaneously with a non-tl;dw version.

Post has attachment
I've just discovered a coauthor I never knew I had!  This is the great Harry Q. Bovik of Carnegie Mellon, whose research contributions in the many fields of computer science of his interest have been inestimable (one might even say nonesuch).  The fact that our two joint papers, "On the Economic Feasibility of Programming Turing Machines" and "Self Coercing Codes", appeared a year or two before I was born is a minor quibble -- even the middle initial matches mine, so who else could it be?

/pseudovia +Craig Kaplan, who pointed me to which was a submission to the annual conference in Bovik's honour, .  For those in the Pittsburgh area, SIGBOVIK 2015 takes place Wednesday of next week!

Post has attachment
Myths rooted in puns
(well, in folk etymologies)

+Ed S draws my attention to a myth about the beaver which flourished in classical and medieval times.  Here's Aelian:

Now [the beaver] understands the reason why hunters come after it with such eagerness and impetuosity, and it puts down its head and with its teeth cuts off its testicles and throws them in their path, as a prudent man who, falling into the hands of robbers, sacrifices all that he is carrying, to save his life, and forfeits his possessions by way of ransom. If however it has already saved its life by self-castration and is again pursued, then it stands up and reveals that it offers no ground for their eager pursuit, and releases the hunters from all further exertions, for they esteem its flesh less.

In fact what the beaver was hunted for were not its testicles (which are internal) but its scent glands, present on males and females both but of similar dangling appearance, and whose extract had uses in medicine, perfumery, and cuisine.  But that's a small quibble in light of the self-mutilation story, which AFAICT has no factual basis.  Ed's source[0] speculates about the origins of the myth but entirely misses the crucial fact, which my main link does pick up on — though etymologically unrelated, the Latin for "beaver" was castor, and for "castrate" castrō.  That's just too good a coincidence to pass up!  Some Romans were bound to have told a just-so story about why the beaver was called the castrator, which would've easily been astonishing enough and quite possibly early enough to spread to Aesop[1], and stuck around with embellishments for a millennium or two.

This makes a nice second example to set alongside the one that's been my mainstay so far: the Cyclops.  We all know the Cyclopes as the giants with a single big eye smack in the middle of the forehead.  Some have attempted to trace the story to birth defects or skulls with big central holes[2], but I reckon likeliest the explanation of Paul Thieme and others[3]: the Cyclopes will have originally been giants of quite common appearance whose distinguishing feature was that they were cattle thieves, *pḱu-klōps in Proto-Indo-European, and stock characters in Indo-European tales the continent over.  In Greek this name developed to Κύκλωψ and chanced to look like it broke down as Κύκλ-ωψ, "circle-eye", and the Cyclopes' faces were adjusted to match.

Anyone know of further examples?

[1] I'm assuming this bit about the Egyptians is noise: even identifying animals in hieroglyphs can be a tricky proposition...
[2] Here's one, from a dwarf elephant:

Post has shared content

Post has attachment
Metafilter user Joe in Australia points out a cute fact about our written digits that I should have noticed: the number of prime factors of a digit D is 1 + b_1(D).   (Unless it's zero or undefined.)  

Oh no. The shape of a digit is determined by the number of factors it has. Prime digits do not divide the plane (i.e.; they have no loops). Digits with two prime factors divide the plane into least two parts: they have one loop.(*) The only digit to have three prime factors (8) divides the plane into three parts.

Surely it's worth seeing whether and how this relationship continues as we proceed through multi-digit primes.

(*) And now you know how the digit "4" is correctly constructed.

The reason I should have noticed this is that Kopczynski's 2004 entry in the IOCCC has long been a favourite of mine: .
Give it ASCII art of a number from 8 to 11 as input, like .

Post has attachment
After having a large pi thrown in my face I figured out what +Erik Corry was up to, so now it's my turn to host Text Zendo.  Rules just as before:

I am thinking of an unambiguous*, objective rule that determines whether a string of text does or doesn't have the Buddha nature.  For simplicity, I'll only allow the English letters a–z in the strings.

I'll start the game by giving one string that has the Buddha nature and one that doesn't.

To play, either
(a) comment with a string you want to know about, and I'll tell you if it has the Buddha nature or not; or
(b) guess the rule, and either I'll give you a counterexample or you win.

Don't hog the play, please.  If you win, you get to host the next game and post a comment here with a link to it.

* Meaning and sound, e.g., are too ambiguous and variable by person to use. We're talking about strings of text (in a font like this one), not words.

Post has shared content
Language people!  Do you have free texts in Zulu, Xhosa, Quechua?  Or other widely-spoken but lesser-known languages?

I am developing a test suite in order to verify that the very useful black box "" ( does what it is supposed to do.

At there are four lengthy lists of languages.  I need short (500 words), freely redistributable texts in these languages.  Top priority is table 2 ("covered by the default model, but not yet the test suite"); second priority is table 4 ("other languages with more than 7.4 million native speakers"); third priority is languages which are easily confused with, but not mutually intelligible with, any of the languages already listed.

Contributors may win a place in the acknowledgments section of an upcoming paper.

Post has shared content
+Alex Fink & +Sai designing vocabulary for UNLWS (, the non-linear written language.

We're starting with Herodotus' story of the Egyptians' investigation of who had the most ancient language.

Might do more if there's interest. For this one, you can participate w/ comments or q&a; future ones might include others on video or the like.

If you're curious as to how we work on UNLWS together, this is it.

Post has shared content
Scarfolk are bang on again.
NEW! Remember when the state rationed democracy in case it ran out? More information here:
Wait while more posts are being loaded