Talk:⦰

Deletion discussion

The following discussion has been moved from Wiktionary:Requests for deletion.

This discussion is no longer live and is left here as an archive. Please do not modify this conversation, but feel free to discuss its conclusions.

---

⦰

When you reverse an empty set, you get an empty set: "reversed empty set" is not a definition that makes sense.

But this is just one entry of many. I have seen tons of "symbol" entries created (I think mostly by Equinox) simply by using the Unicode character name as the definition. This is wrong for several reasons. First, the Unicode consortium gets character names way wrong not that rarely, like with Ƣ ("LATIN CAPITAL LETTER OI" later corrected to "LATIN CAPITAL LETTER GHA") and ꀕ ("YI SYLLABLE WU" corrected to "YI SYLLABLE ITERATION MARK"). Second, Unicode character names are, well, names of the characters, not their meanings; there is nothing in "MATHEMATICAL SCRIPT CAPITAL P" to tell you that the symbol 𝒫 is used as the power set operator. It does not help you understand the symbol at all. Third (though admittedly weakest), Unicode often unifies characters with similar appearance, but distinct meanings; ∅ is used both in mathematics and linguistics, but "EMPTY SET" will not tell you that.

But back to the entry at hand: while I think it possible that this be attestable as an alternative form of ∅ (though it will be hard to do even if true), the definition we have here now is not worth keeping. — Keφr

I think they mean it’s reversed in the visual, not mathematical sense. — Ungoliant ^(falai) 18:30, 26 December 2014 (UTC)

The empty set reversed in the visual sense is still indistinguishable from itself before the reversion. — Keφr 18:45, 26 December 2014 (UTC)

I think you should close this discussion and take this to RfV. Purplebackpack89 19:51, 26 December 2014 (UTC)

I think you should actually find some attesting quotations instead of telling people what they should do. — Keφr 21:31, 26 December 2014 (UTC)

Indistinguishable from itself before the reversion? You can't see a difference between ⦰ and ∅? —Aɴɢʀ (talk) 20:31, 26 December 2014 (UTC)

I can see a difference between ⦰ and ∅, but I cannot see the difference between the empty set and a reversed empty set, whatever definition of "empty set" and "reversed" I make up. — Keφr 21:32, 26 December 2014 (UTC)

It's not the empty set itself itself that's reversed, it's the empty set symbol that's reversed. That's what Ungoliant meant by reversed in the visual sense: the symbol itself is reversed (written in mirror writing). —Aɴɢʀ (talk) 22:47, 26 December 2014 (UTC)

You're both right. The symbol is reversed, making this a (reversed) (empty set). But that's what the glyph is, not what it means. Does anyone besides the Unicode Consortium Archivist know of the use of this symbol in a paper? --Catsidhe ^{(verba, facta)} 22:52, 26 December 2014 (UTC)

RFV seems right. Is this used in any human languages? Renard Migrant (talk) 17:21, 27 December 2014 (UTC)

It's not something that Unicode probably stressed out about much. It's from the STIX project by the American Mathematical Society, and the whole batch of characters was pretty much handled as a bunch. It's shown here, where the STIX names is bemptyv and it's described as "reversed circle, slash".--Prosfilaes (talk) 12:43, 28 December 2014 (UTC)

Can we move this to RFV? I know w with a line through it was RFV'd because it seemed not to be used in any human languages, just in lists of characters. I would link to the debate, but alas, I haven't found a way to enter it on my keyboard. A little help, please? Renard Migrant (talk) 16:51, 29 December 2014 (UTC)

Well, if you insist… I doubt it would change the outcome, though. As for "w with a line through it", there seems to be no such thing in Unicode. Unless you mean U+20A9 WON SIGN, but I think that should be relatively easy to attest. — Keφr 17:01, 29 December 2014 (UTC)

No not that. Before your time perhaps? Back when Liliana-60 was Prince Kassad. Renard Migrant (talk) 17:06, 29 December 2014 (UTC)

Perhaps you meant Talk:ẘ? — Keφr 17:51, 29 December 2014 (UTC)

yes, thank you. Renard Migrant (talk) 18:11, 29 December 2014 (UTC)

Keep in RFD. If we take this to RFV, the question is how we can effectively search for a use. While google:⦰ finds nothing, google:∅ (for the non-reversed empty set) does not find anything either, although we do not doubt the symbol is used. Since I do not see anyone presenting a way how to search for use of unicode symbols, I would avoid RFV and just keep it, perhaps with a usage note linking to http://www.ams.org/STIX/private/stixprv-E4.html. As for whether the symbol has to mean something: we include e.g. Latin letters as letters, that is, units that do not have meaning but are rather used to compose larger units that carry meaning. We use {{non-gloss definition}} to enter a descrition of the letter as contrasted to meaning, e.g. in d#English. In ↑, one of the "senses" is "upwards arrow", which describes the glyph rather than a meaning, it seems. In ↔, Unicode says "left right arrow" while our current definitions consist of the sole one saying "material equivalence; if and only if"; the left right arrow is used to denote a multitude of things, and I would find it worthwhile to define it as "left right arrow", apart from "material equivalence; if and only if". As for which definition to use, we might either stay with "reversed empty set" with things being implied, or go for {{non-gloss|reversed symbol for empty set}}, to making clear it is the symbol that is reversed, not the empty set. --Dan Polansky (talk) 09:55, 30 December 2014 (UTC)

That page is out of date (it assigns the character U+E41A, which is in the Private Use Ares) and no more informative than a Unicode code point chart. We might as well link to those — and in fact we do already, through {{character info}}.

A definition of "letter" at least tells you that the character is used in constructing larger meaningful units (words). Letters are also often paired with their usual pronunciations; you can include this pronunciation in a definition, since the letter arguably stands for the sound. Merely describing a glyph does not establish its meaning even in this weak sense. "Reversed symbol for the empty set" is not a definition, it is an etymology. It tells you nothing about why would anyone choose to use this glyph for anything instead of some other glyph. An "arrow" definition is similarly meaningless; I would delete those too.

As for searchability — the STIX font package for LaTeX (and I think some others) provides \revemptyset for this symbol; searching arXiv for it turns up nothing (compare searching for \varnothing). This might be one of those times when a symbol was added to a font because "it might be useful someday", and later to Unicode for the sake of maintaining feature parity. — Keφr 13:03, 30 December 2014 (UTC)

As for the uselessness of definitions like "arrow", I disagree. (Dan Polansky, signed below.)

Articulate as always. — Keφr 15:35, 30 December 2014 (UTC)

As for searchability, as you can see, http://search.arxiv.org:8081/?query=\varnothing finds mere 33 hits, which suggests a very low coverage. And the search only covers uses via LaTeX markup, not direct uses. Thus, compared to our search facility for words, we do not have a decent search facility for symbols. --Dan Polansky (talk) 14:21, 30 December 2014 (UTC)

What do you mean by "not direct uses"? A PDF version is two clicks away. arXiv is also a much smaller corpus than Google Books, the modest list of results is kind of expected. You can also look for \emptyset (which Unicode apparently considers merely a glyph variant). Given you have a real definition for ∅, it also gives you a clue where to look for attestation: just open a random set theory or topology textbook. Many more common mathematical symbols can be attested this way. While it may not be as convenient, this is better than \varnothing. — Keφr 15:35, 30 December 2014 (UTC)

You seem to agree: we do not have a decent search facility for symbols. --Dan Polansky (talk) 16:11, 30 December 2014 (UTC)

Depending on what you mean by "symbol", we might never have. This does not mean we should give up attesting them. I think what we have at our disposal is good enough to at least make reasonable conjectures. — Keφr 16:44, 30 December 2014 (UTC)

We may make a conjecture about "⦰" not being used, but, absent good search facility, the conjecture is too secure against refutation. This is one of the reasons for which I prefer to keep Unicode codepoints per being Unicode codepoints, having attesting quotations available or not. --Dan Polansky (talk) 16:58, 30 December 2014 (UTC)

But it has no meaning. Delete DCDuring TALK 17:02, 30 December 2014 (UTC)

In the edit summary, DCDuring said: "no meaning ⇒ no entry". I refuted this principle by pointing to our letter entries. --Dan Polansky (talk) 17:10, 30 December 2014 (UTC)

Does this mean we should create U+1F574 MAN IN BUSINESS SUIT LEVITATING now? — Keφr 17:09, 30 December 2014 (UTC)

@Dan Polansky: Do we have any entries for things that neither have a conventionally accepted meaning, nor are components of terms that have conventionally accepted meaning, nor are proper names? DCDuring TALK 18:05, 30 December 2014 (UTC)

@DCDuring: Thank you; now you have stated a workable criterion for exclusion: the item has no attested meaning, and is not attested as being part of larger lexical items with attested meaning. Even then, I think it worthwhile to include symbols that are demonstrably used albeit not with any particular conventional meaning. Furthermore, as for "⦰" specifically, my reservations about our search facilities remain. --Dan Polansky (talk) 18:15, 30 December 2014 (UTC)

…any examples? — Keφr 19:01, 30 December 2014 (UTC)

@Dan Polansky: What is the possible rationale for having them? This is far outside WT:CFI and any accepted concept of what a dictionary is or could be. Few indeed would even know how to enter any characters outside those available on their native-script keyboard. Shouldn't we await the development and spread of a technology that made it possible to input a free-hand drawing and search for images, first among Unicode, them other similar standards, then line drawings available on-line? Is there any chance at all that someone would come to Wiktionary for such items? DCDuring TALK 20:24, 30 December 2014 (UTC)

Shapecatcher [1] actually does allow you to look up Unicode characters by drawing freehand. Equinox ◑ 21:49, 30 December 2014 (UTC)

Cool tool. Added it to my toolbar. Now lets add it to MediaWiki Search. DCDuring TALK 02:16, 31 December 2014 (UTC)

Any accepted concept of what a dictionary is or could be? The Matematika Vortaro (Mathematical Dictionary), page 271, defines ∅ as "malplena aro". The fact that it doesn't define ⦰ is based on frequency, not some "accepted concept of what a dictionary is". Webster's Condensed (1887) offers us a section straightforwardly titled "Arbitrary Signs used in Printing and Writing".--Prosfilaes (talk) 01:32, 31 December 2014 (UTC)

@Prosfilaes: What evidence do you have that it is a frequency-based omission rather than one based on the absence of any meaning? "malplena aro" is not a name or description of the symbol, it is the definition. The symbol in question has no meaning that anyone knows or can find. DCDuring TALK 02:16, 31 December 2014 (UTC)

That anyone here knows or can find. Our limitations are not limitations on reality, and I much assume if it was worth putting in fonts, someone used it, and therefore it has meaning, if only in a certain narrow context.--Prosfilaes (talk) 02:42, 31 December 2014 (UTC)

Never "assume". If someone used it, show it. — Keφr 00:40, 2 January 2015 (UTC)

• This entry has a description of the symbol not a definition, its meaning. Apparently it has no meaning assigned to it by any authority, let alone a generally accepted meaning, let further alone one that we know how to attest. How does this fit even with our slogan? At best this would be a use for {{no entry}} with either {{in appendix}} or {{in Wikipedia}}. DCDuring TALK 15:05, 30 December 2014 (UTC)

  WT:CFI slogan "all words in all languages" (italics mine) is unhelpful with morphemes (un-), provebs (curiosity killed the cat), symbols (∅), and letters in particular (d), which are not words; their inclusion is in part mandated by WT:CFI#Terms. Therefore, slogan "all words in all languages" does not help in this discussion. Furthermore, as I said above, letter entries are examples of entries that, in their Letter sections, do not state meaning on their definition lines. --Dan Polansky (talk) 16:11, 30 December 2014 (UTC)

  I'd actually noticed some limitations and qualifications that render our slogan operationally useless, except as a way of disabling brain function. DCDuring TALK 16:15, 30 December 2014 (UTC)

BTW, the reason I've created some of these more esoteric symbols is that people (not me) tend to add them at Wiktionary:Wanted_entries, and it's easier to clear the request pages by creating an entry than by deleting something as inappropriate and starting a tedious fight, as sometimes happens at WT:REE. Perhaps this is cynical and I shouldn't, but otherwise the request pages get so large and almost nobody is helping. Equinox ◑ 19:00, 30 December 2014 (UTC)

• Keep; we've already committed to including most of Unicode; I don't see the win to not including ⦰ and 🕴 (MAN IN BUSINESS SUIT LEVITATING).--Prosfilaes (talk) 01:32, 31 December 2014 (UTC)
  • But neither of these glyphs has a definition. Do you think that we also need an entry for U+1F47E 👾 ALIEN MONSTER or U+1F43F 🐿 CHIPMUNK, or U+1F31A 🌚 NEW MOON WITH FACE? For domino tiles? For mahjong? Blocks? How would you define them? — Keφr 15:38, 31 December 2014 (UTC)

Including all of Unicode is not all our job. All the ones used in human language, yes, of course! But not those which aren't. Renard Migrant (talk) 22:02, 31 December 2014 (UTC)

@Dan Polansky you think we should keep this whether it's attestable or not? When did you become such a big fan out flouting the rules? Renard Migrant (talk) 22:02, 31 December 2014 (UTC)

• Move meaningless Unicode forms to an appendix. Oh, wait, guess what? bd2412 T 02:37, 6 January 2015 (UTC)

{{look}} DP and Prosfilaes say "keep", DCD says "delete", Equinox seems not to care, RM's and BD's statements seem to be to the effect of "delete" (the former would prefer an attempt at attestation first, though). Anyone else? — Keφr 20:24, 8 January 2015 (UTC)

I'm also in the don't care category, by virtue of the fact that I believe this is the wrong forum. Purplebackpack89 21:19, 8 January 2015 (UTC)

On a pure "keep/delete" vote, I would say delete; however, we have this glyph in Appendix:Unicode/Miscellaneous Mathematical Symbols-B, with the exact same definition. By redirecting (or using an "only in" signal), we avoid having an entry on a meaningless symbol, but retain access to all of the information that already exists. I certainly agree that the entry should not exist as a freestanding entry. bd2412 T 17:41, 10 January 2015 (UTC)

Redirect per bd, or delete, if it's not attested. - -sche (discuss) 05:37, 19 January 2015 (UTC)

Note: See Wiktionary:Beer parlour/2015/January#Is documenting all Unicode characters within the scope of Wiktionary?, where redirection to the appendices has been proposed for all meaningless Unicode characters. Cheers! bd2412 T 13:53, 28 January 2015 (UTC)

Based on the direction of the above discussion, I have gone ahead with the proposed redirect. Cheers! bd2412 T 21:24, 16 February 2015 (UTC)

• RFD closed: redirected by bd2412 above per Beer parlour discussion. I think it should be a soft redirect, but anyway. --Dan Polansky (talk) 10:47, 22 February 2015 (UTC)