Wiktionary:Votes/2014-11/Entries which do not meet CFI to be deleted even if there is a consensus to keep

Entries which do not meet CFI to be deleted even if there is a consensus to keep

 * Voting on: making it official policy to delete entries which do not meet WT:CFI to be deleted even if there is a consensus to keep.

Rationale: deletion debates have turned into a matter of personal opinion, whether certain editors like a certain entry, made it themselves, etc. This vote would meet that admins have an obligation to delete entries that do not satisfy the criteria laid out in WT:CFI even if there as a consensus (by which I mean a numerical consensus, by counting the number of votes) to keep the entries. Instead of simply counting up the number of votes, the closing admin should look at the arguments made to see if the entry meets CFI (rather than keep because I made this entry or keep because my mother used this word 20 years ago).

As a result, debates would become about whether an entry meets CFI rather than just about the number of votes an entry gets.


 * Vote starts: 00:01, 24 November 2014 (UTC)
 * Vote ends: 23:59, 23 December 2014 (UTC)
 * Vote extended to 23:59, 30 January 2015 (UTC) --Dan Polansky (talk) 09:50, 20 December 2014 (UTC)


 * Vote created: Renard Migrant (talk) 18:47, 17 November 2014 (UTC)
 * Discussion:
 * [[Image:Wikt rei-artur3.svg|20px]] Beer parlour/2014/November
 * [[Image:Wikt rei-artur3.svg|20px]] Wiktionary talk:Votes/2014-11/Entries which do not meet CFI to be deleted even if there is a consensus to keep

Support

 * 1)  Consensus should reflect CFI. If not, then CFI itself is the problem and it should be modified, not just circumvented whenever people feel like it. —CodeCat 13:28, 24 November 2014 (UTC)
 * Then why did User:CodeCat create on 6 March 2014 to label and keep certain words not meeting current CFI, without first changing CFI? --Dan Polansky (talk) 11:23, 14 December 2014 (UTC)
 * Why do you ask? —CodeCat 14:38, 14 December 2014 (UTC)
 * 1)  Renard Migrant (talk) 18:04, 24 November 2014 (UTC)
 * 2) . — Ungoliant (falai) 21:37, 24 November 2014 (UTC)
 * 3)  Agree with CodeCat. I also more or less agree with the opposing comment that it's impossible to implement, but still support in theory. Equinox ◑ 21:55, 24 November 2014 (UTC)
 * 4)  It would be nice if arguments in favor of keeping were couched in terms of CFI. CFI can be changed. DCDuring TALK  18:19, 7 December 2014 (UTC)
 * 5)  Consensus should reflect CFI. Leasnam (talk) 10:49, 11 December 2014 (UTC)
 * It is the other way around: CFI should be updated to reflect consensus. Since we have not managed to make CFI reflect consensus so far, rigid application of CFI is undesirable. We have plentiful evidence about the history and origin of CFI, and we know that it did not originate by consensus. --Dan Polansky (talk) 09:31, 14 December 2014 (UTC)
 * Right, CFI should reflect consensus. —Stephen (Talk) 20:37, 14 December 2014 (UTC)
 * 1)  Votes are great and all, but if we base our entries on whether the beliefs of more individuals say we should keep, this dictionary will end up having lots of unwelcome content. If you know what I mean by that. I believe that entries should only be voted on if there is some way that CFI isn't working too well for the process. That being said, by the way, I still think cult film should be deleted. When I made this entry, I was not used to the site, so I didn't know what an SOP was.  Rædi Stædi Yæti  {- skriv til mig -} 20:42, 26 December 2014 (UTC)
 * I think some of your recent entries come under the classification of "unwelcome content" - e.g. catfucker. And then you have the nerve to say that SoP entries should be removed. The mind boggles. Donnanz (talk) 12:42, 29 December 2014 (UTC)
 * Single words cannot be SOPs. And if you think this entry should be removed, challenge it in RFV. I never said you couldn't do that, and as this is a wiki, my contributions here are treated the same as everyone else's. Rædi Stædi Yæti  {- skriv til mig -} 21:22, 29 December 2014 (UTC)

Oppose

 * 1)  Impossible to implement.  Requires little or no ambiguity of CFI; CFI will likely forever remain ambiguous and will probably never satisfactorily address every possible case.  I continue to believe that RfD discussions should be decided primarily on consensus of the participants, and am disheartened that this proposal could potentially lead to the deletion of a greater number of articles. Pur ple back pack 89   05:15, 24 November 2014 (UTC)
 * I do agree. However, we should keep the CFI rule, but if CFI doesn't have a solution to the entry's specific problem, we should rely on consensus. That is my personal opinion. Rædi Stædi Yæti  {- skriv til mig -} 05:26, 27 December 2014 (UTC)
 * 1) * Query:
 * How often do we get situations where there is consensus to keep a term, but that consensus is in opposition to WT:CFI?
 * If there is consensus to keep a roterm, there should presumably be a rationale, something based on Wiktionary norms and policies that backs up this consensus. If CFI does not describe these norms and policies sufficiently, then the error is in the CFI.
 * Note: We do not need CFI to be explicit in all details and to cover all cases. That would be ideal, but that just ain't gonna happen in the real world: process documentation never exactly matches reality, that's just a given.  That said, CFI can be updated to cover the broader majority of cases, and to leave wiggle room for those cases that it does not cover.  The fact that CFI cannot be revised to describe everything is not a valid argument for ignoring it entirely.  &#8209;&#8209; Eiríkr Útlendi │ Tala við mig 06:41, 24 November 2014 (UTC)
 * How many? This is really a question for you, not I.  I know you believe it to be a significant number, and a too-high one, but I don't know what that number. Pur ple back pack 89   21:24, 3 December 2014 (UTC)
 * Of course it comes down to interpretation of CFI. The question I would ask myself is how many entries are kept where there's no interpretation of CFI to justify them. It would be very difficult to talk about numbers. Maybe a couple a week in recent weeks? Renard Migrant (talk) 21:45, 3 December 2014 (UTC)
 * 1)  We should add various other rules to CFI, namely - translation target when an English term is a SoP but it is a single term in a few foreign languages, Lemming test - when a term is used in certain dictionaries (the list can be agreed on). It still won't cover all situation, like gas station, mobile phone, foreign language, nominative case, lung cancer, etc. where a vote would still be required. Translation target and Lemmings would heavily reduce valuable time spent on RFD's. --Anatoli T. (обсудить/вклад) 06:54, 24 November 2014 (UTC)
 * 2) * It sounds like you're saying that CFI should be updated, in order to make CFI more useful in evaluating RFDs.  This sounds like you don't oppose the underlying idea -- that CFI should be used for arbitration, rather than just group agreement without regard for CFI.
 * Am I understanding you correctly here? If so, your vote would be support or abstain, no?  &#8209;&#8209; Eiríkr Útlendi │ Tala við mig 07:59, 24 November 2014 (UTC)
 * Yes, until CFI is updated to include the above two, I am opposing. I would change or reconsider, if there are changes to CFI. --Anatoli T. (обсудить/вклад) 08:02, 24 November 2014 (UTC)
 * I've always supported the proposal of such a policy (allow words from other major dictionaries). I would oppose it but, I'm not convinced it would be massively unpopular. I'm surprised nobody's proposed it at all in vote form. Renard Migrant (talk) 18:09, 24 November 2014 (UTC)
 * Yeah, I've been meaning to get around to it. But if I propose it, that'll automatically be 2-3 oppose votes right there.  BTW, what's the answer to Eirikr's question about how often there is a consensus to ignore CFI? Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   19:00, 24 November 2014 (UTC)
 * 1)  If there is a consensus to keep a word and CFI states that it should be deleted, then CFI must be changed accordingly, because CFI should reflect this consensus. Discussions are useful when intelligent arguments can be exchanged: if reflection is forbidden because CFI must be applied blindly, then discussions are useless. Lmaltier (talk) 21:01, 24 November 2014 (UTC) In other terms, the sequence should not be: 1. delete. 2. improve CFI. 3. restore (this is illogical). but rather: 1. take the keep decision. 2. discuss on how CFI should be improved to reflect this consensus. 3. improve CFI. Lmaltier (talk) 21:18, 24 November 2014 (UTC)
 * CFI has a much stronger consensus than a single RFD discussion. So it's not illogical; it would be illogical if the consensus of the few people who participated in the RFD would trump the consensus of the dozens of people who were involved in shaping CFI over the years. —CodeCat 21:34, 24 November 2014 (UTC)
 * 1) * Similar to CodeCat's argument, I would caution strongly against assuming that consensus should be taken at face value just because it's consensus. Iff the consensus is based on reasoned argument, and that reasoned argument calls for an outcome in contravention of CFI, then yes, CFI probably needs changing.  If, however, the consensus is not backed up by reasoned argument, then this consensus must be ignored in favor of CFI, which at least has been through an extensive vetting process that (presumably) involves reasoned argument and consensus.  &#8209;&#8209; Eiríkr Útlendi │ <small style="position: relative; top: -3px;">Tala við mig 21:59, 24 November 2014 (UTC)
 * 2) ** Of course, I fully agree with you. Lmaltier (talk) 17:50, 25 November 2014 (UTC)
 * 3) *** I think that's more of an argument for deleting CFI all together and just going purely with consensus. I mean, how could you possibly update CFI after every deletion debate. And more importantly, why bother? Renard Migrant (talk) 12:02, 28 November 2014 (UTC)
 * 4) ****There's a happy medium between the two poles you posit. It's called making CFI a guideline.  Something that's generally accepted, but need not be followed to the letter. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   22:29, 3 December 2014 (UTC)
 * 5) *** Anyway, nothing is perfect, and certainly not CFI. CFI must be improved with time. Lmaltier (talk) 06:50, 8 December 2014 (UTC)
 * 6)  We can create policies. We can't compel everyone to interpret said policies in precisely the same manner. There will always be cases in which it isn't clear-cut whether or not a term meets CFI because not everyone has the same interpretation of things like "sum of parts." Such cases require a subjective judgment about the term's fitness for inclusion to be made. The RfD process exists so that all Wiktionarians can partake in such determinations, sharing their individual thoughts on the inclusion-worthiness of nominated words, and thus help reach a consensus. Allowing for the outcome of RfD discussions to be overridden at the personal discretion of administrators would be completely contrary to the community-driven nature of this project. -Cloudcuckoolander (talk) 08:55, 26 November 2014 (UTC)
 * “Allowing for the outcome of RfD discussions to be overridden at the personal discretion of administrators” -- I don't follow you. Where does administrator whim come into the discussion?  &#8209;&#8209; Eiríkr Útlendi │ <small style="position: relative; top: -3px;">Tala við mig 10:03, 26 November 2014 (UTC)
 * When it's closed. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89  17:09, 26 November 2014 (UTC)
 * "Instead of simply counting up the number of votes, the closing admin should look at the arguments made to see if the entry meets CFI."
 * That means that, under this proposal, the closing admin would make a subjective judgment about which arguments hold the most merit, and thus decide whether the entry warrants keeping. But there is no single correct interpretation of CFI (particularly SOP). Two people can look at the same entry and reach differing conclusions about its inclusion-worthiness. And that's why the RfD process allows us to weigh different viewpoints and reach a consensus. Leaving it up to admin whim, in short, would be a disaster. It would be giving the closing admin license to favour arguments with which they agree over arguments with which they disagree, and thus give them the power to impose their own personal personal preferences over community consensus. -Cloudcuckoolander (talk) 22:47, 27 November 2014 (UTC)
 * But if the result of RFD is consensus, then how do we know what that consensus is unless someone decides what it is? —CodeCat 23:03, 27 November 2014 (UTC)
 * No one needs to decide that consensus exists. It will be objectively apparent. -Cloudcuckoolander (talk) 23:24, 27 November 2014 (UTC)
 * Nothing will ever be apparent to Purple unless it is what he wanted. Equinox ◑ 02:51, 28 November 2014 (UTC)
 * What's consensus is apparent to me, thank you very much. I see no reason why you needed to say it wasn't.  This is about CFI, not about me. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   14:19, 28 November 2014 (UTC)
 * (This is Zeggazo btw). I believe the existing process is fine. There are some words which are very common in blogosphere but have unluckily not been picked up in permanently recorded media. If we allow some grey areas wiktionary will get more of these blogosphere type words. 2.96.221.56 10:23, 28 November 2014 (UTC)
 * I don't think unregistered users are allowed to vote. If I knew the name of the vote I'd look it up. Renard Migrant (talk) 11:44, 28 November 2014 (UTC)
 * Indeed they are not (the instructions cannot be understood any other way). Struck. — Keφr 09:51, 29 November 2014 (UTC)
 * See Votes/2010-04/Voting policy. --Dan Polansky (talk) 11:47, 29 November 2014 (UTC)
 * 1)  --Dan Polansky (talk) 22:43, 28 November 2014 (UTC). It must be possible to override CFI on a case-by-case basis, as e.g. for olinguito. The long-standing tradition in RFD of keeping terms for which there is no consensus for deletion should be continued. Translation target should continue to be a consideration for those who feel this is a valid extra-CFI concern, as well as lemmings (keep a possibly sum-of-parts term when certain monolingual dictionaries, not WordNet, include the term; see also the Beer parlour discussion), and set phrase. The CFI as a whole as it is without modification is not supported by consensus, and the principle of consensus should prevail over any statutory document like CFI. That said, CFI should be used as a recommendation (rather than an inviolable rigid rule), with exceptions being justified, and thus should not be entirely ignored when deciding whether to keep an entry. Each argument for keeping should be based on CFI as far as possible. The non-rigid approach I have outlined seems to be implemented in WT:ELE as per WT:ELE, a section that is lacking from CFI: "While the information below may represent some kind of “standard” form, it is not a set of rigid rules." --Dan Polansky (talk) 22:43, 28 November 2014 (UTC)
 * If this vote fails, what we oughta do is demote CFI from policy to guideline. I think you're correct that most of the major editors here are dissatisfied with CFI as written.  Having the "what's in, what's out" page as policy and not guideline is unusual for a project anyway.  If CFI was demoted to guideline, that would allow the overriding you want. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   02:14, 29 November 2014 (UTC)
 * What is a "guideline"? — Keφr 19:13, 29 November 2014 (UTC)
 * A guideline is one step down from policy. The main difference is that policy has to be followed 100% of the time, but guidelines don't. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   19:38, 29 November 2014 (UTC)
 * I don't think we'll need to 'demote' it. It'll become a lame duck. In fact I don't think any policies will have any weight any more. People will just say if there's no vote forcing me to apply this policy, why bother? Renard Migrant (talk) 22:16, 29 November 2014 (UTC)
 * You seem to be taking the de facto, and I the de jure. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89  00:20, 30 November 2014 (UTC)
 * Re: "I don't think any policies will have any weight any more.": I don't agree with this all-or-nothing view. CFI is a useful policy and is largely followed, with exceptions being argued on a case-by-case basis. The entries kept despite current CFI are fairly few and are mostly exceptions to WT:CFI rather than WT:CFI; have a look e.g. at and  to see how many they are. "Sum of parts" is still a valid reason for nomination for RFD, just that multiple editors are in the habit of examining redeeming qualities of terms so nominated. CFI still does a useful service as a policy applied with a certain, fairly low, flexibility. --Dan Polansky (talk) 09:43, 30 November 2014 (UTC)
 * There's a term for a policy with flexibility. It's called a guideline.  And CFI should be one, preferably one with SOP stripped from it.  The problem with CFI being policy is not only the lack of flexibility afforded, but the need for it to apply universally.  I think it's clear there are blind spots in CFI.  If CFI were a guideline, that wouldn't be as much of an issue, and we could address the blind spots as they come up with routine consensus. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   21:51, 3 December 2014 (UTC)
 * If there is truly "routine consensus" then you ought to have no problem getting it voted in as policy! Equinox ◑ 22:08, 3 December 2014 (UTC)
 * You seem to be arguing that I posit that WP:IDIOM WT:IDIOM be merged into WP:CFI WT:CFI. There's no reason for CFI to mention every little thing that survived an RfD.  Far better to have it be a guideline with gray areas that can just be addressed when they come up. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   22:21, 3 December 2014 (UTC)
 * What is "WP:IDIOM" and "WP:CFI"? — Keφr 22:33, 3 December 2014 (UTC)
 * 1)  The situation does not arise. If there's a consensus to keep, then there's a consensus that the entry does meet WT:CFI. CFI has more than one possible interpretation; it's not objective, nor should it be. —Aɴɢʀ (talk) 07:32, 29 November 2014 (UTC)
 * Re: "If there's a consensus to keep, then there's a consensus that the entry does meet WT:CFI": Not really. If I vote based on translation target consideration and I know that it is not part of WT:CFI, then my vote does not indicate that I want to keep the entry via CFI. It is fair to say that translation target consideration is not part of current CFI. --Dan Polansky (talk) 08:25, 29 November 2014 (UTC)
 * Yeah this is clearly false; I could vote to delete an entry based on a prejudice because of who created the entry; would you argue that that's part of CFI? If so where is it? Renard Migrant (talk) 13:17, 29 November 2014 (UTC)
 * If you wanna talk about stuff like that, look up how often "redundant" and "redundancy" appear in CFI. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89  19:40, 29 November 2014 (UTC)
 * Why? Renard Migrant (talk) 22:16, 29 November 2014 (UTC)
 * Because neither word is used in CFI. There are a number of legitimate arguments that you, I, or anyone would agree to that are not explicitly spelled out in CFI <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   00:20, 30 November 2014 (UTC)
 * Indeed. In fact CFI says nothing about content being correct. Renard Migrant (talk) 12:26, 30 November 2014 (UTC)
 * Maybe I missed some sarcasm, but I think "correctness" is covered by WT:CFI. — Keφr 12:49, 30 November 2014 (UTC)
 * I suppose it is. Renard Migrant (talk) 13:06, 30 November 2014 (UTC)
 * Redundancy and correctness are different issues <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89  16:01, 30 November 2014 (UTC)
 * I fully agree (in case that wasn't clear). Accord to CFI, there's no reason not to have two identical meanings, because if one meets CFI, so does the other. Renard Migrant (talk) 16:49, 30 November 2014 (UTC)
 * 1)  I can't think of any good reason to support this proposal. The main problem with CFI is that the rules are overinterpreted or misinterpreted by beady-eyed would-be deletionists. There is nothing in CFI which specifically says that SoP entries should be deleted. Yet birthday present was deleted and Christmas present survived for devious reasons. I know common sense should prevail, thus red dress doesn't qualify, but little black dress does, and little black number has been overlooked. One problem I find is with translation targets, especially from languages where compound words are commonplace: I was trying to find a home for barnesoldat (child soldier) on the English side, and had to settle for an entry under soldier and hope that it would be seen there. So I think provision should be made in CFI for translation targets. Donnanz (talk) 11:50, 8 December 2014 (UTC)
 * Actually, [[Christmas present]] was deleted in the nominated sense. That another sense was considered idiomatic is a separate matter. — Keφr 12:08, 8 December 2014 (UTC)
 * I think that using soldier for the translation barnesoldat was the right place. I fully agree with you, except for using translation target as a criterion in CFI, which would make the objective of the project very confusing (e.g. thousands of little ... pages would be created). The only criterion should be does this belong to the vocabulary of the language?. This would be a clear and simple basic principle. Lmaltier (talk) 22:26, 10 December 2014 (UTC)
 * I am happy for you agreeing with me, but why are you replying to me right here? — Keφr 18:10, 11 December 2014 (UTC)
 * I think Lmaltier's comment was directed at me. Donnanz (talk) 18:42, 11 December 2014 (UTC)
 * I agree. That's what paper dictionaries use, limited by the amount of paper they have. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   23:12, 10 December 2014 (UTC)
 * Paper dictionaries are limited by what they can squeeze into one volume. The more comprehensive dictionaries are published in more than one volume, and are rather expensive I imagine. The Oxford Dictionary of English has 2088 pages; my Duden Deutsches Universalwörterbuch has 2016 pages and is falling apart due to poor binding. An online dictionary such as Wiktionary doesn't have this restriction with modern technology, and can be super-massive and cater for translation targets. It may not be as bad as you fear, Lmaltier. Donnanz (talk) 09:45, 11 December 2014 (UTC)
 * Of course, it would be feasible, I agree. But it would be confusing: users of a dictionary don't look for translations targets, they look for a term of a language. For the same reason, I would group all phrasebook entries in topical appendices and remove them for the mainspace, but I would add attested words from fictional universes to the mainspace from appendices (with special rules for proper nouns, of course). Lmaltier (talk) 21:29, 11 December 2014 (UTC)
 * Kindersoldat appears in Duden online, but not my copy; this may prove my point about online dictionaries, they can be updated more quickly than paper dictionaries anyway - in fact paper dictionaries may be phased out in the future, who knows?. Donnanz (talk) 10:15, 11 December 2014 (UTC)
 * Paper dictionaries are likely to be used less and less, but they keep real advantages: 1. The pleasure of paper book pages. 2. They are available at any moment, no need to start a computer. 3. They must be bought once, but consulting them is 100% free. We are not bought, but users consulting us spend some money at each consultation (think to power they use). Lmaltier (talk) 21:36, 11 December 2014 (UTC)
 * Paper dictionaries have a lease on life until we adopt the lemming principle. As long as there are words print dictionaries have that we don't, people will have to buy them rather than get those words here for free. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   21:50, 11 December 2014 (UTC)
 * 1)  —Stephen (Talk) 03:06, 9 December 2014 (UTC)
 * Stephen: Any particular reasons? — Keφr 11:29, 14 December 2014 (UTC)
 * CFI has too many flaws. It was written and edited by inexperienced amateur lexicographers. If we start to attract editors with more experience in lexicography, I don’t think they should be hobbled by earlier editors with little knowledge of the field. CFI can be edited and rewritten, but it seems to be rather difficult to make any substantive changes to it. Perhaps if there is a consensus to keep some terms, it will be an incentive to improve CFI. —Stephen (Talk) 20:34, 14 December 2014 (UTC)
 * Though a few users argued that habitually ignoring CFI at the whim of circumstance will disincentivise improving CFI (as in "why bother conducting bureaucracy over updating something that is ignored on a daily basis anyway?"). Your comment on that? — Keφr 21:12, 14 December 2014 (UTC)
 * I think it’s nonsense. The difficulty in getting enough agreement to improve CFI, coupled with the problem of inexperienced lexicographers trying to come up with improvements, is what disincentivizes improving CFI. —Stephen (Talk) 23:06, 14 December 2014 (UTC)
 * How much incentive we have to fix CFI doesn't override my belief that CFI is broken beyond repair. There's no way it's ever going to be able to mention every reason we've used to keep an article.  This is why it is foolish for CFI to be policy. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   23:35, 14 December 2014 (UTC)
 * Nonsense. It's true that WT:CFI can never specify every case, but it doesn't have to, as long as it doesn't claim to. Some things can be explicitly included by policy, and some things can be explicitly excluded by policy, and some things can be (either explicitly or implicitly) left open by policy. In fact, there are already some cases where WT:CFI explicitly doesn't specify whether or not they can be included: see Criteria for inclusion. —Ruakh <i >TALK</i > 05:08, 15 December 2014 (UTC)
 * Word. Agree absolutely with Ruakh. --Dan Polansky (talk) 09:11, 20 December 2014 (UTC)
 * Stephen G. Brown, if you think that users on this project are not competent enough to enact sensible criteria for inclusion, then you probably have no reason to believe they can make sensible decisions by "consensus" in individual deletion discussions either. You might as well give up on this project entirely. It is not just CFI that is written by amateurs: everything is written by amateurs, a fact that is not going to change very soon. Those "editors with more experience in lexicography" you speak of are not going to fall from the sky — but when they do come, I expect they will provide some input on how CFI should be changed, which surely will not be ignored. For the time being, I can stick to rules written by amateurs.
 * And I even disagree on the "inexperienced" part: Category:Archived discussions contains over ten thousand pages of RFVs and RFDs. I assume at least some still-active regulars have witnessed or participated in a large part of these discussions. There is plenty of precedent on which to build something. If only someone cared enough. — Keφr 18:23, 26 December 2014 (UTC)
 * You make a lot of assumptions about what I think, and jump to a lot of bad conclusions about what I believe. As for what I should do, I think I am a better judge of what I should do than you, which may account for the fact that I have not sought your opinion in the matter. As for what you can do, I’m sure you know best and I wouldn’t have it any other way. And as to the "inexperienced" part, I was talking about a different kind of experience, not experience in RFVs and RFDs. —Stephen (Talk) 01:48, 27 December 2014 (UTC)
 * Re: "...habitually ignoring CFI...": We are not ignoring CFI, since, in RFD, we are taking it into consideration, albeit not as the ultimate arbiter allowing no exceptions; in RFV, exceptions to CFI are very rare. Multiple of us override CFI in RFD not on a whim but rather in a reasoned manner, applying common sense and lexicographical sense of what is good for a multilingual dictionary with a broad range of international readership with various backgrounds, needs, and language skills. Furthermore, despite e.g. "translation target" being mentioned by RFD voters for past multiple years, that did not stop us from improving CFI. One of the latest substantive improvements even made CFI more strict, including less: Wiktionary:Votes/pl-2014-03/CFI: Removing usage in a well-known work 3. --Dan Polansky (talk) 09:11, 20 December 2014 (UTC)
 * 1) per Stephen's arguments, which I agree with. --Neskaya sprecan? 16:47, 8 January 2015 (UTC)

Abstain

 * 1) I abstain for reasons I already stated in the linked BP discussion. — Keφr 21:40, 24 November 2014 (UTC)
 * 2)  --Romanophile (talk) 18:04, 26 November 2014 (UTC)
 * 3)  We've already had enough problems with admins enforcing their unilateral interpretations of the CFI by listing words at WT:RFV and making up rules that the citations supposedly have to meet. The proposal here seems to be that even admins who would prefer to make decisions consensually, will instead be expected to make them unilaterally. I definitely recognize the problem that this proposal is trying to address, and I agree with most of the "support" voters' comments — if there is a consensus that something should be kept, then we should fix WT:CFI rather than flout it — but I just can't bring myself to support the proposal as written. —Ruakh <i >TALK</i > 22:01, 14 December 2014 (UTC)
 * @Ruakh: any specific case comes to your mind? — Keφr 18:24, 26 December 2014 (UTC)
 * One class of examples is sub-word morphemes. WT:CFI says only that they are acceptable, with no details, but that has not stopped people from crediting it with surprisingly detailed rules. (See Talk:-os for one example.) —Ruakh <i >TALK</i > 04:29, 27 December 2014 (UTC)

Decision

 * Closed as not enacted: No consensus to enact. 7 in favor and 9 against. <font face="Verdana"><font color="#3A003A">Pur <font color="#800080">ple <font color="#991C99">back <font color="#CC33CC">pack <font color="FFBB00">89   00:08, 31 January 2015 (UTC)