Talk:philerast
Latest comment: 14 years ago by Ruakh in topic Request for verification
Request for verification
editThe following discussion has been moved from Wiktionary:Requests for verification.
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.
I doubt that this spelling is attestable. There is one citation c. 1920s, however, so the same publisher might have used this in other works. DCDuring TALK 23:45, 9 September 2009 (UTC)
- Plato’s Symposium is a well-known work, so this term is verified according to the second criterion for inclusion, as I noted when I created the entry. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 00:57, 10 September 2009 (UTC)
- But not necessarily any individual translation/type-setting. DCDuring TALK 01:14, 10 September 2009 (UTC)
- How else could it apply? Anglophones read The Republic, not Πολιτεία. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 01:21, 10 September 2009 (UTC)
- Doremítzwr is not without precedent in making this claim, as others have claimed "well known work" status from translations of well known works (see the rfv discussion of this entry). Quite frankly, until that particular unit of CFI is clarified, I think his argument stands. However, I also tend to think that any translation of a well known work should not automatically qualify as a well known work itself, especially regarding something like this. In any case, I do not think this issue is likely to be resolved here. A conclusive BP discussion (and likely a vote as well) on "well known work" vs. "translation of a well known work" is required to settle this. -Atelaes λάλει ἐμοί 01:40, 10 September 2009 (UTC)
- I think that this point of typography attestation is an extra remove from the well-known work. It is a particular typesetting of a particular translation. Would any content of the translator's foreword to some edition of Plato be thereby automatically included as from a well-known work. That seems simply preposterous. DCDuring TALK 02:12, 10 September 2009 (UTC)
- Yes, well, “Would any content of the translator's foreword to some edition of Plato be thereby automatically included as from a well-known work[?]” doesn’t apply, since the quotation in question is part of the Symposium proper. I used Hobbes’ Leviathan, and only the Leviathan, to attest adhære, and I would have to flatly deny that such a thing is preposterous. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 02:37, 10 September 2009 (UTC)
- It is a quite analogous situation. The decisions of a book designer or typographer are as far removed from from what made the work well-known as the words of the foreword or the text on the book jacket. This is making a mockery of the concept of a well-known work. Please provide adequate citation for this headword. DCDuring TALK 10:11, 10 September 2009 (UTC)
- Another issue which this raises is that of orthographical normalization. In Latin, entries do not have macrons in the entry title. No matter how many times you might find a work quoting a word with a macron in it, it doesn't count for inclusion. Similarly, for Ancient Greek, there are multiple characters for theta (θ and ϑ), among others. Words are only spelled with the first theta, simply because that is the only allowed theta for Ancient Greek words. If we decide that st is not meaningfully different than st, we could decide to simply exclude it in entry titles altogether. Certainly, this is tricky business, but it's something perhaps worth thinking about. -Atelaes λάλει ἐμοί 11:45, 10 September 2009 (UTC)
- That sounds like a good way to go, unless someone can show a difference between some ligatured word and the corresponding unligatured one. Hard redirect: we do it for apostrophes and other things.—msh210℠ 18:41, 10 September 2009 (UTC)
- Another issue which this raises is that of orthographical normalization. In Latin, entries do not have macrons in the entry title. No matter how many times you might find a work quoting a word with a macron in it, it doesn't count for inclusion. Similarly, for Ancient Greek, there are multiple characters for theta (θ and ϑ), among others. Words are only spelled with the first theta, simply because that is the only allowed theta for Ancient Greek words. If we decide that st is not meaningfully different than st, we could decide to simply exclude it in entry titles altogether. Certainly, this is tricky business, but it's something perhaps worth thinking about. -Atelaes λάλει ἐμοί 11:45, 10 September 2009 (UTC)
- This entry should if anything be a hard redirect. You can't verify a typesetting artifact, I mean any number of books use (or used to use) the character under discussion, but we don't consider it a separate English letter (unlike æ for instance). It's just a caprice of printers and in my opinion is not an ‘alternative’ spelling at all, just alternative Unicode codepoints used to render the same spelling. Ƿidsiþ 17:44, 10 September 2009 (UTC)
- I'd go with a redirect too - last time I checked, Plato didn't write in 20th Century English, did he? Mglovesfun (talk) 22:57, 10 September 2009 (UTC)
I too would be fine with a hard redirect, just so long as the ligated spelling remains listed in the Alternative spellings section of philerast. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 23:05, 10 September 2009 (UTC)
- I'd like to know if it is a real representation of the word. I'd like to know if it counts as a word. I'd like to know if typographical variants are worth spending screenspace on. I'd like to know whether these things are in any way attestable in a way compatible with our existing system. I had thought that this was an open-and-shut case of unverifiability. I hadn't realized that it is also fairly open-and-shut on WT:CFI (not a word, just a word representation).
- If this one is accepted without attestation or a decision that it is worth including, then it may well be argued that it is a precedent for unattested deviant typography at every headword that contains "ae", "ff", "st", "oe", "ue", "ss", or any of the other character pairs and triplets type designers have spent time on. DCDuring TALK 23:22, 10 September 2009 (UTC)
- It is no such precedent; this term is verified. Vertically speaking (which is what matters), retaining it takes no additional screen space. It seems reasonable to treat these like we do the conjugated forms of idioms — shown in the entry but hard-redirected from the actual page back to the lemma. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 23:32, 10 September 2009 (UTC)
- The term may be verified, but we don't have a policy of "all words in all font and orthographic styles". This is an orthographic variant, and not a different form or different spelling. This should be a hard redirect, or preferrably deleted. --EncycloPetey 04:26, 11 September 2009 (UTC)
- It's not even an orthographic variant, it's a typographic variant. We definitely shouldn't be using such ligatures to create separate entries; I certainly hope that fish and flat turn out to be red links when I hit "Show preview", and if not, I certainly hope that they're redirects and not separate entries. Angr 10:41, 11 September 2009 (UTC)
- Well, it seems like everyone except for Doremítzwr is down with not treating this as a separate word. If so, we really ought to write up a policy, and define which orthographic/typographic variants are considered illegal (I know, I know, this sounds so evil, as if such a list should go on the same shelf as a banned books list or something). We should figure out exactly which Unicode characters we don't accept, and if it applies to all languages, or just English. There are a lot of languages which use the Latin script, and some variants might be meaningless in some languages, but meaningless in others. -Atelaes λάλει ἐμοί 11:49, 11 September 2009 (UTC)
- No, as I said, I’m fine with hard-redirection, just not with deletion. There is at least one very important reason why we should not ban certain Unicode characters: On the Internet, efficient users are wont to copy-and-paste words whencesoever into search engines such as ours; such words may contain typographical ligatures such as ffi and st. If we don’t have the redirects in place, their searches will not return hits; the fact that, in many of the fonts that support them, these typographical ligatures look virtually identical with their unligated equivalent digraphs and trigraphs means that a user is quite likely not to think of going through the search term substituting them with their permitted character combinations. Of course, this is only an issue if the term exists in use with that ligated spelling, so to avoid clouds of redirects (which you’ve warned against in the past), even those hard-redirect entries would need to be attested. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 16:37, 11 September 2009 (UTC)
- Yes, it definitely matters which language we're talking about! I don't think we should be using "æ" for Latin or Modern English, but we should certainly use it for Norwegian, Danish, and Old English. Likewise, despite the Modern Irish penchant for dotless lowercase i's in signage (see for example File:Gaeltacht Donegal cropped.jpg), we shouldn't use "ı" for Irish, though of course we should use it for Turkish and Azeri. Angr 12:26, 11 September 2009 (UTC)
- Sorry; let me know if I understood you properly: Do you propose that we delete or hard-redirect English entries spelt with (deprecated template usage) æ or (deprecated template usage) œ? You mention Danish; note that there are a number of præ–initial terms in English that are homographic with Danish terms. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 16:37, 11 September 2009 (UTC)
- And so we come back to the reason why hard redirects are seldom feasible on Wiktionary... I'm inclined to think that we cannot hard-redirect any ligatures that are potentially significant in some language. On the other hand, are there any languages where the st, fi, ffi, &c. ligatures are distinct from the separate letters? If not, I would favor hard-redirecting those (especially given Adobe's penchant for adding the fi, ffi, etc. ligatures to PDF files).-- Visviva 02:46, 13 September 2009 (UTC)
- AFAIK, none of the ligatures I have listed in “Useful symbols yet to be added to MediaWiki:Edittools” > “Latin / Roman” signify any different semantic meaning from their digraphic or trigraphic equivalents (I may be wrong about the four Croatian ligatures). This is certainly not the case for the English ligatures Æ, Œ, æ, and œ: They are largely etymological in function, usually deriving from the Latin diphthongs (deprecated template usage) ae, (deprecated template usage) oe and/or the Ancient Greek diphthongs αι (ai), οι (oi); terms of a different derivation — such as those from Germanic roots and Latinate or Græcian ones such as (deprecated template usage) aetheogam and each of the aer–prefixed words — cannot properly be spelt with those ligatures. (Also note the occasional borrowing such as Æsir (from Old Norse) that cannot properly be spelt without ligation.)
- As for how we deal with the typographical ligatures, indeed “given Adobe’s penchant for adding the fi, ffi, etc. ligatures to PDF files”, I think the proposal I wrote in reply to Atelaes above (timestamped 16:37, 11 September 2009) is simple and perfectly workable. As long as our requirements of attestation apply to them just like any other term, there shouldn’t be any problem. † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 13:26, 14 September 2009 (UTC)
- I think a far better solution would be to teach the equivalencies to the MW software. It already auto-redirects A --> a, when appropriate. Why can't we teach it to auto-redirect st --> st? I would think that this would be a million times easier and more sustainable than creating a billion redirects for every typographic variant. Of course, when legal characters in one language conflict with this in another, we can use
{{see}}
(e.g. if "st" is allowed in some language, and that language contains the word "philerast", we could have a{{see|philerast}}
at the top). -Atelaes λάλει ἐμοί 13:48, 14 September 2009 (UTC)
- I think a far better solution would be to teach the equivalencies to the MW software. It already auto-redirects A --> a, when appropriate. Why can't we teach it to auto-redirect st --> st? I would think that this would be a million times easier and more sustainable than creating a billion redirects for every typographic variant. Of course, when legal characters in one language conflict with this in another, we can use
- I doubt that any of these ligatures (ff, fi, fl, ffi, ffl, ſt, and st) are “legal” characters in any language, given their inclusion in the “Alphabetic Presentation Forms” Unicode subset. Is teaching such equivalencies possible? † ﴾(u):Raifʻhār (t):Doremítzwr﴿ 19:24, 14 September 2009 (UTC)
Redirected to [[philerast]], which already has the quotation in question. —RuakhTALK 18:55, 5 December 2009 (UTC)