# Talk:2606: Weird Unicode Math Symbols

Apparently, nobody knows what U+237C ⍼ means (https://ionathan.ch/2022/04/09/angzarr.html)

For me it looks very like as designation of where electrical cable is burrowed. It should come with numbers near angle hands designating depth and offset from sign. But it just a guess, of course 141.101.76.221 05:30, 14 April 2022 (UTC)
There are at least four people on (https://news.ycombinator.com/item?id=31012865) who claim to have seen the symbol in the wild: German/Dutch proof by contradiction, diode with a gate, Finnish proofreaders mark indicating when and how to split a word, and indicating which way EM waves are polarized --162.158.62.180 11:42, 14 April 2022 (UTC)
I think that is "larry potter" as the lightning symbol, and the L comes from the L shape the lightning is over imo. 172.70.34.191 14:47, 14 April 2022 (UTC)Bumpf

Can someone add a column where we try to crowdsource a description for the "mathematical use of symbol" ?

I'm curious what those symbols actually mean, and the unicode titles don't give that much information.

I do not doubt that have enough math geeks on here to find the answer to most of then :-D

Thanks! Flekkie (talk) 01:20, 14 April 2022 (UTC)

i agree w Flekkie's comment Blue in real life (talk) 02:27, 14 April 2022 (UTC)

If you make edits, please don't immediately delete all of mine because you think yours are better. It shows you conflicting edits for a reason. Some explanations are nonsensical, like defining a smash product as the "result of dividing two product spaces." The smash product is specifically the quotient of the underlying spaces of two pointed spaces where points in the product spaces are identified if they contain either labeled point as an element. Other claims are simply mistaken. For instance, the ≝ symbol is used to introduce a definition, not to declare that the definition has been achieved in a proof. The claim that "A union on smash product appears to be one where the sets are nit isomorphic" is totally meaningless. The symbol ⩩ is not merely decorative but is intended as a supplemental math symbol like all the others. I couldn't track down its purpose. The APL symbol description somehow never bothers to mention APL. This is all very cursory, which is fine, but just please don't delete my work while doing it. 172.70.130.5 04:15, 14 April 2022 (UTC)

I'm done, page is yours. I'd been working on it for an hour and every time I tried to save, a new edit came in. I did my best to reconcile them but by the fifth I just saved a copy of what was there and pasted mine over. I immediately got to work on recovering what I had pasted over as indicated in the edit comment and like I said, I'm done. I've merged to the best of my ability and have no more interest in this page.
--FrankHightower (talk) 04:37, 14 April 2022 (UTC)
I'm very sorry if I was one of those who stepped on your edits. I was originally trying to fix brokenness in the table.
^^^^^^ is the symbol for preferring many small edits over not saving your work often in a batch-mode collaborative editing environment. 172.70.207.8 04:55, 14 April 2022 (UTC)
since adding a column means adding a cell to every row, I thought it would be easier/better to put something in each cell. I wasn't expecting this --FrankHightower (talk) 18:12, 15 April 2022 (UTC)
Don't worry about it, idk why I was so upset. It's that feeling you get when you click "save" and then suddenly it's gone. If anything's still missing, I'll just restore it. 172.70.131.106 05:41, 14 April 2022 (UTC)

I wish we had as much research on ⧍ and ⩩ as we have for ⍼. 172.70.207.8 05:16, 14 April 2022 (UTC)

No clue what those are for. The first is a triangle with serifs? The triangle symbol (not capital delta) is often used in geometry to represent a triangle, but why in the world would you give a geometric figure serifs? I think I may have seen the triple cross-hatch somewhere, but I couldn't say where. Maybe it's the chess commentary symbol for being mated so badly it's embarrassing. 172.70.131.106 05:41, 14 April 2022 (UTC)
I'm guessing what must have happened is the code page in question was filled in by a typographer with more graphic design than mathematical experience, who likely added made-up "missing" symbols as the design-logical extensions of the symbols they were given, presumably to be on the safe side in case they had what appeared to be a possibly incomplete set. There are some awesome ones in there, like "⩐" TEST-TUBE PARTHENOGENISIS, and "⨻" THE ILLUMINATI IS DEAD. 162.158.255.171 06:07, 14 April 2022 (UTC)
The ⧍ character has the same origin as ⍼, namely that they were both part of ISO/IEC TR 9573-13 with no explanation given. You can find it in some old charts online, like here: [1], under `trisb`/codepoint E27E. ionchy (talk) 07:03, 14 April 2022 (UTC)
I am sure that I have seen maps with such a symbol used for campsites, and without the bold strokes of the top sides as the National Park Service draws it. I haven't found any yet, but I have found one without the center base -- like _/\_ -- on a 1960s era map. 172.69.134.131 07:07, 14 April 2022 (UTC)
As for ⩩, the character just before it in the Unicode code chart [2], ⩨, has the text "identical and parallel to", so it's possible the horizontal lines in this one also means "identical to". I don't know what three vertical lines mean though, and in Unicode there's three (!) different characters with similar glyphs: U+2980 ⦀, U+2AF4 ⫴, and U+2AFC ⫼. ionchy (talk) 07:28, 14 April 2022 (UTC)
Identical both vertically and horizontally? 172.69.134.131 07:32, 14 April 2022 (UTC)
I found one suggestion that the dodecathorp should be used to refer to very big numbers...172.69.79.223 08:33, 14 April 2022 (UTC)
Three vertical bars means "similar to", so that'd be identical to and similar... to. OK that doesn't work. 172.70.230.75 04:05, 19 April 2022 (UTC)

Apparently, Randall reads Hacker News like the rest of us... --172.68.110.141 09:38, 14 April 2022 (UTC)

Did you know that ⩩ was a logo of the Romanian fascist group Iron Guard?172.68.238.67

Wow. "Saint Michael's Cross, the (Iron Guard) movement's symbol designed by Codreanu." I prefer the 4x4 tic-tac-toe board.... 172.69.33.181 12:29, 14 April 2022 (UTC)

Not that I have special knowledge (above aspiring-polymath level), but for "Rightwards Arrow Above Reverse Almost Equal To" and its leftwards sibling (the current suggestion being that it's an assignment of an approximation) I'd posit that it's a directional approximation specific to chaotic systems. For a precise a=b in a system where small changes to one side can effect large changes in the other, uncertainty or deliberate approximation of either of them may still map well enough to the value across the desired ≈ relationship, but that is only acceptably so unidirectionally. Hence (and I wouldn't know which would be the better convention) the arrow depicts the direction of either the dependence or the more dominant option, whichever it is that almost-ties the values tovether. (I also really ought to check what the reverse almost-equal-to sign is used for, it could be radically different from the 'forwards' version in some field, for which the directionality might even be trivially understood by context.) 172.70.91.36 11:25, 14 April 2022 (UTC)

Any sources? I found several papers on the topic you describe naively searching on those keywords, and while they are packed with notation, I couldn't find any arrows or approximate equality signs, let alone any sort of composition of the two. 172.69.33.51 12:47, 14 April 2022 (UTC)
No sources except in my (above-IP's, just to confirm) head. It's just something I might use it for, if let loose on a whiteboard for some reason, not something for which I've seen it used. It's a situation I've encountered often enough in my own amateur numerical analysis, but under self-taught/self-experimenting circumstances where I was probably just (inefficiently) re-inventing the wheel. And if I've been commenting code (itself strings of conventional syntactical symbols), as a usually futile attempt to remind future-me what I was doing, I would have been using longhand to describe what I was either basing my functions on or actually asking them to test... So no surprise that a paper-search didn't match much of my description. ;) 172.70.162.155 14:39, 14 April 2022 (UTC)

No idea how Randall got that wrong, but U+2A50 (⩐) is totally catching a snowflake with your tongue. 172.68.110.141 15:02, 14 April 2022 (UTC)

While cute, I think it'd probably be better to remove the Unicode letters in the chart for the second hashtag (⩩). Not only would it mess up screenreaders, but it doesn't actually look like the image in question. Perhaps it would be better to upload an image and give it some alt text to describe it. --Trlkly (talk) 00:12, 15 April 2022 (UTC)

Could the strange-looking characters be made to look like the image in question using Unicode combining diacritical marks? BunsenH (talk) 15:31, 15 April 2022 (UTC)
A screen reader should read it as "Double struck H, Double struck a, double struck s..." etc. which seems a pretty accurate readout to me --18:12, 15 April 2022 (UTC)

## Calming down - Sigma long S

Calming down may refer to the pronounciation of the Letter Esh https://en.m.wikipedia.org /wiki/Esh_(letter)

"Other links: XKCD #2606 mentions ⍼ and its Explain XKCD entry cites this post" yooo people the blog post now mentions us! Someone put this in trivia or something Mushrooms (talk) 09:01, 15 April 2022 (UTC)

Ah, yes, so the article we reference that we think Randall references has now referenced both him and us and so now we ought to reference that? Seems perfectly Ok to me! :-p
(P.S. Hello to that author. You should just be glad that you're not Wikipedia... ;) ) 172.70.162.147 11:19, 15 April 2022 (UTC)
The blog author is User:Ionchy, who has contributed to the discussion above. 172.70.211.72 13:47, 15 April 2022 (UTC)

Any of these that are particularly obscure and someone doesn't find and post here what its actual existing use is will be impossible to research later, as people talking about this comic will end up dominating the results of any attempt to search for the symbol.--108.162.216.27 00:08, 19 April 2022 (UTC)

Regarding the angzarr symbol, ⍼ - I can't remember exactly where I got this idea from, but someone might be able to check it out. I think it might have been a symbol from WW2 to enable 3-D coordinates to be mapped onto 2-D paper. It might have been used in the study of 3-D trajectories. The right-angle in the image represents the X and Y axes. The zig-zag represents the z axis. The arrow pointing downwards or outwards towards the viewer represents the 3rd dimension. This symbol was placed in the z axis to indicate that this axis which continues in a straight line from the y axis below the x axis, is not the negative y axis, but the z axis. Regarding the triangle symbol, ⧍ - this is the delta math symbol, e.g. https://sciencing.com/delta-math-6678201.html - 172.70.162.147 19:08, 6 May 2022 (UTC)

Responses from GPT-3's davinci-002 model with https://beta.openai.com/playground default settings in bold:

What does the symbol ⍼ mean?
The symbol ⍼, known as a "diaeresis" or "umlaut", is used in some languages to indicate that the two vowels it is placed over are to be pronounced separately, rather than as a single vowel sound.
No, it isn't either of those.
It means "integral" in APL.
Well, that's a better guess, but also wrong.
It's actually a symbol used in chemistry to indicate that two atoms are bonded together covalently.
No, that's a colon, :. Try again?
It's actually a symbol used in physics to indicate that two particles are interacting.
That seems very plausible! What is your source?
The symbol ⍼ is actually used in mathematics to indicate that a function is continuous.
In math, might it indicate an indefinitely-valued relation over a finite domain.
I.e. for the domain {1,2,3,4} the relation "is even" would be symbolized
⍼ 2 4 2 4 2 ...
The symbol ⍼ is not typically used to indicate an indefinitely-valued relation over a finite domain.
More like, the relation "nearest other evens" for the domain {1, 2, 3, 4} would map to, in order, {{0,2}, {0, 4}, {2, 4}, {2, 6}}. But that's not an indefinitely-valued relation or a finite domain, as each integer maps to exactly two other integers.
The symbol ⍼ is not typically used to indicate an indefinitely-valued relation over a finite domain.
Well it doesn't indicate that a function is continuous, either. So what does it actually mean?
The symbol ⍼ is actually used in set theory to indicate that a function is one-to-one.
No. I liked your physics idea, meaning two particles interacting.
But that's not it either.
The symbol ⍼ is actually used in logic to indicate that a formula is valid.
How? Give an example?
The symbol ⍼ can be used to indicate that a formula is valid if the following holds
⍼(p ∨ q) ↔ (⍼p ∨ ⍼q)
In other words, the formula is valid if the statement "p or q" is true if and only if the statement "p is valid or q is valid" is true.
What are p and q there?
P and Q can be any formulas. So, for example, if P is the formula "x+1=2" and Q is the formula "x=1", then the formula "P or Q" is true if and only if either "x+1=2" is valid or "x=1" is valid.
What does it mean for a formula to be valid?
A formula is valid if it is true in all possible circumstances. So, for example, the formula "x+1=2" is valid if and only if, no matter what value x takes, the statement "x+1=2" is always true.
Is the formula "x+1=2" valid?
No, the formula "x+1=2" is not valid, because there are some values of x for which the statement "x+1=2" is not true. For example, if x=0, then "x+1=2" is not true.
So is ⍼x>x-1 true?
No, the formula "x+1=2" is not valid, because there are some values of x for which the statement "x+1=2" is not true. For example, if x=0, then "x+1=2" is not true.
Is the formula "x>x-1" valid?
Yes, the formula "x>x-1" is valid, because no matter what value x takes, the statement "x>x-1" is always true.
What if x is infinity?
The formula "x>x-1" is still valid if x is infinity, because no matter what value x takes, the statement "x>x-1" is always true.
Is infinity minus one less than infinity?
No, infinity minus one is not less than infinity, because infinity minus one is still infinity.
So how can the formula "x>x-1" be valid?
The formula "x>x-1" is valid because it is true in all possible circumstances. So, for example, if x=0, then "x>x-1" is true because 0 is greater than 0-1. If x=1, then "x>x-1" is true because 1 is greater than 1-1. If x=infinity, then "x>x-1" is still true because infinity is greater than infinity minus one.
Is infinity greater than infinity minus one?
No, infinity is not greater than infinity minus one, because infinity minus one is still infinity.