Wiktionary talk:Votes/pl-2014-03/CFI: Removing usage in a well-known work 2

Rationale
The well-known work item allows inclusion of terms that no one uses, such as Joyce's "bababadalgharagh...". In Wiktionary talk:Votes/pl-2011-04/CFI: Removing usage in a well-known work, there was the objection that we need the item to include terms of poorly attested languages; this has been meanwhile solved by adding a dedicated regulation for them to WT:CFI: CFI. --Dan Polansky (talk) 14:01, 2 March 2014 (UTC)
 * I think there is room at Wiktionary for nonce words encountered in well-known works, including those by James Joyce and WWII-era Finnish songs with pseudo-Russian words in them. —Aɴɢʀ (talk) 14:12, 2 March 2014 (UTC)
 * Do you think "there is room" at Wiktionary for words used to refer to objects from a fictional universe, such as Mordor or Ungoliant?
 * What do you mean by "there is room"? Assuming that you are not referring to digital storage running out, is this some sort of idiom? --Dan Polansky (talk) 14:15, 2 March 2014 (UTC)
 * Room is not the question. Wiktionary should only be housing terms that are part of the language, which at least requires reuse and shared meaning, on both of which scores terms fail that are only found in a single author's well-known works and commentaries thereon.
 * For one thing for many of the nonce terms we favor because of their literary source, the meaning is not at all obvious to a would-be definer of the term, eg palp, sense 2. Joyce probably liked the ambiguity. possibly because he thought it better represented mentalese or however we would recharacterize stream of consciousness in terms used in modern cognitive psychology.
 * For a term to be first attestably used in what turns out to be a well-known work and yet fail to be reused is a clear indication that it is not part of the language, much more than would be the case where the same result follows for a term that did not benefit from such a happy accident of birth. The first case seems to indicate rejection; the second might merely indicate neglect. I would favor having such terms in concordance and appendix space where textual and cultural context, allusion, etymology of such terms can get all the chapter-length speculative treatment they deserve and more. DCDuring TALK 14:44, 2 March 2014 (UTC)
 * WT:CFI doesn't say "Wiktionary should only be housing terms that are part of the language", it says "A term should be included if it's likely that someone would run across it and want to know what it means." A person is likely to run across nonce words in well-known works and want to know what they mean. Now, if I was reading Finnegans Wake and came to the part with the thunderclap, I wouldn't actually turn to a dictionary to find out what it meant, because I'd be able to tell instantly that it was a word he'd just made up on the spot. But if I was reading Ulysses and came across palp:, I might very well wonder what it meant and turn to Wiktionary to find out, and would consider it an omission on Wiktionary's part if the term wasn't there. As for Mordor, judging from uses like this and this it does seem to have entered the language as a common noun. In cases like Ungoliant, though, which doesn't seem to have entered the language that way (but see this), I'd probably be less inclusive of fictional proper nouns, because it would normally be obvious to the reader that these words have no meaning beyond the person or place they refer to within the text. —Aɴɢʀ (talk) 20:25, 2 March 2014 (UTC)
 * There are lots of terms whose meaning I'd like to know, but which actually have no meaning in a linguistic sense. We do not and cannot offer a supportable definition. If we form a definition based only the single usage, then we have no opportunity to validate the meaning in other uses. If we cannot offer a definition that meets our normal standards, I think it is misleading to present any definition. Concordances and appendices are the only homes for this in Wiktionary that do not do a disservice to our users. Redirects to such places are the most that we can offer our readers while maintaining some level of consistency of quality in our entries. If you think that someone should do what Joycean commentators do, there might be a case for an article at WikiJoyce or, more generally, WikiLitCrit. DCDuring TALK 21:42, 2 March 2014 (UTC)

What's new?
We've had exactly this vote before, and it failed by a wide margin. What has changed since then that having the same vote again makes sense? &#x200b;—msh210℠ (talk) 16:59, 16 March 2014 (UTC)
 * In that vote (on the talk page), people argued that we need the well-known work provision to save poorly attested languages. That is no longer valid, since poorly attested languages have their dedicated regulation now. So people opposing the vote now will need to have a different reason. They will need to explain why including a would-be word invented for a single well-known work and never used again is a good thing. Alternatively, the voters can vote "oppose" with no reasoning whatsoever, as they did at Wiktionary:Votes/pl-2011-04/CFI: Removing usage in a well-known work. --Dan Polansky (talk) 17:30, 16 March 2014 (UTC)
 * In the three years since that vote, some new users have joined the project, and some of the users who participated in the previous vote may have changed their minds. I voted "lean oppose" in that vote, but plan to vote "support" now. Dan himself ended up voting "oppose" last time but created this vote. Had the two of us voted "support" rather than "oppose" in the previous vote, the numbers would have been 5S, 4O, a weak majority for "support", rather than the 2-to-1 majority for "oppose". - -sche (discuss) 18:40, 16 March 2014 (UTC)

Using Template:only in to direct users to an appendix of famous nonces
Expanding on DCDuring's comment that "concordances and appendices are the only homes for [nonce words] in Wiktionary that do not do a disservice to our users", I would just like to point out that in the same way we use Template:only in to direct people who look them up to our Appendix of English dictionary-only terms, we could use the template to direct users to an appendix of famous nonces. - -sche (discuss) 18:45, 16 March 2014 (UTC)
 * Using to direct readers to an appendix of nonce forms used in well-known works would be acceptable to me. If we did that, I could support this vote. —Aɴɢʀ (talk) 11:44, 17 March 2014 (UTC)
 * I hope we can incorporate this into the vote without starting a new vote. Obviously we coulda/shoulda had most of this discussion before the vote. DCDuring TALK 14:15, 17 March 2014 (UTC)
 * The vote was scheduled to start March 9, but the not-started notice has not been removed. DCDuring TALK 14:18, 17 March 2014 (UTC)
 * I think we are having this discussion before the vote, in that the vote has not started yet. The vote creator did say "Let us postpone the vote as much as the discussion will make necessary." Accordingly, I've pushed the start date back to March 22nd. - -sche (discuss) 17:18, 17 March 2014 (UTC)
 * What do you think of this wording: Removing the item "use in a well-known work, or" from WT:CFI, placing ", or" at the end of the item "clearly widespread use". Thus, no longer having full entries for words which are only used in one or two well-known works. Instead, entries for such nonces can use Template:only in or an equivalent to redirect users to an appendix of nonces words found in well-known works. Cases where a string is a well-known nonce in one language and an attested word in another can be handled like this. Please critique and improve it as much as necessary. - -sche (discuss) 17:33, 17 March 2014 (UTC)
 * @User:-sche: Thank you for postponing the vote.
 * As for : from what I can see, the use of the template is not regulated by CFI in anyway. CFI does not mention the template and CFI does not make any allowance for e.g. dictionary-only terms to host template . Continuing in this tradition, I would not want to incorporate anything about into a modification of CFI proposed by the vote. --Dan Polansky (talk) 11:21, 22 March 2014 (UTC)
 * Giving it a thought, I would like the simple vote as it is run its course, and if it fails, we can create one that explicitly allows the use of . The current practice is to place in a variety of entries for terms that do not meet CFI, without this being regulated by policy AFAIK. --Dan Polansky (talk) 11:45, 22 March 2014 (UTC)
 * Ok. I agree with your interpretation that it isn't necessary to explicitly allow . - -sche (discuss) 19:07, 23 March 2014 (UTC)