Wiktionary:Votes/2019-08/Rescinding the "Coalmine" policy

Rescinding the "Coalmine" policyEdit

Voting on: Rescinding the "Coalmine" policy, as originally approved at WT:COALMINE and incorporated into the CFI at Wiktionary:Criteria for inclusion#Idiomaticity. That is, removing the following paragraph from WT:CFI:

Unidiomatic terms made up of multiple words are included if they are significantly more common than single-word spellings that meet criteria for inclusion; for example, coalmine meets criteria for inclusion, so its more common form coal mine is also included.

The effect of this vote, if passed, would therefore be that a multi-word term that means no more than the sum of its parts, such as coal mine, would become eligible for deletion under standard rules (assuming that no other exemptions apply), irrespective of the fact that an entry for the corresponding single word exists.

For the proposer's rationale for rescinding, as well as other statements in favour of or against the poposal, please see the Disussion page.

Be sure to vote the right way. A support vote is a vote to rescind the "Coalmine" policy; an oppose vote is a vote to retain the policy.


  • Vote starts: 00:00, 18 August 2019 (UTC)
  • Vote ends: 23:59, 16 September 2019 (UTC)
  • Vote created: Mihia (talk) 17:25, 11 August 2019 (UTC)

Other Discussion:


  1.   Support as proposer. What a so-far thoroughly disappointing response to the opportunity to abolish this ridiculous rule. Mihia (talk) 22:46, 20 August 2019 (UTC)
  2.   Support. Canonicalization (talk) 18:06, 21 August 2019 (UTC)
  3.   Support.​—msh210 (talk) 21:50, 31 August 2019 (UTC)


  1.   Oppose 1) Use the right lemma: Let us have compound term, compound-term and compoundterm, all attested. I consider the three items to be three different forms of what is essentially one word. The question is: where should we lemmatize it? If compound term is much more common than compoundterm, compound term is the proper lemma, in my view, even if sum of parts; alternatively, compound-term could be the lemma, depending on the frequency. 2) COALMINE is usually easy to administer, and from my experience often has us keep terms that I would instinctively like to keep for other reasons, which are harder to investigate and articulate. That is a bonus of COALMINE: it speeds up RFD for terms worthwhile for other less well articulable reasons. 3) Vanishingly rare solid forms compoundterm that we have good reasons to believe are misspellings should be deleted using WT:CFI#Spellings rare misspelling exclusion policy, thereby making COALMINE toothless: himand was deleted, so him and is not supported by COALMINE. (himand*10000), him and at Google Ngram Viewer provides frequency evidence of rare misspelling, and these are deleted. --Dan Polansky (talk) 19:20, 18 August 2019 (UTC)
    Other dictionaries: per coal mine at OneLook Dictionary Search, "coal mine" is in lexico.com[1], Collins[2], Macmillan[3], and dictionary.cambridge.org[4]; perhaps they found other rationale to keep than WT:COALMINE. For "iron mine", I find lexico.com[5]. iron mine, (ironmine*500) at Google Ngram Viewer shows an unfavorable ratio that could lead to the conclusion that "ironmine" is a misspelling or misformation; ironmine at OneLook Dictionary Search find no dictionary at all. --Dan Polansky (talk) 06:37, 23 August 2019 (UTC)
  2.   Oppose I am in favor of simple rules that are easy to apply. I think we should be less afraid to have "arbitrary" inclusion criteria that helps us avoid long philosophical discussions in RFD on what even really is a word, man (we just don't know). DTLHS (talk) 20:37, 18 August 2019 (UTC)
  3.   Oppose: If WT:COALMINE is abolished, then we will be left in the unenviable position of having entries for uncommon one-word forms of terms, but lacking entries for multiple-word forms of them (e.g. having a entry for coalmine but not coal mine). This will put us in the position of appearing to endorse rare one-word spellings of several common terms, which would confuse users and new editors alike while offering no real advantage. I don't believe that a notice placed at (for example,) coalmine reading "coal mine is the more frequent spelling" is enough to dispel the weirdness that this policy creates. As a aside, I also think that the fact that speakers choose to spell terms such as coal mine as one word could be used to adduce that they believe these terms to be non-SOP on some cognitive level. Hazarasp (parlement · werkis) 03:34, 19 August 2019 (UTC)
    "coalmine" is just an arbitrary spelling variation that makes no difference to the fact that it is "coal" + "mine". I would support deleting "coalmine" itself on the basis that it is SoP and/or written by people who don't really know how to spell, except that it is probably impossible to devise criteria to distinguish single-word SoP terms, and also to enable users who don't know how to break up the word to discover the meaning. But the fact that we have one unlegislatable SoP term is no reason to add another that is legislatable. Mihia (talk) 23:18, 22 August 2019 (UTC)
  4.   Oppose: I have reservations about WT:COALMINE, as I feel it can be manipulated in creating rare single-word forms of two-word terms, but on the whole I support keeping it. It can also be applied in creating compound adjectives, e.g. angellike vs. angel-like. The fact that many two-word terms probably aren't "coalmineable" shouldn't be grounds for deletion as SoP though; e.g. maple leaf and snap election. DonnanZ (talk) 11:26, 19 August 2019 (UTC)
    Getting slightly off-topic, but what we need is a "set phrase" exemption from SoP rules. Mihia (talk) 23:14, 22 August 2019 (UTC)
  5.   Oppose --{{victar|talk}} 18:24, 19 August 2019 (UTC)
  6.   Oppose Prosfilaes (talk) 07:59, 20 August 2019 (UTC)
  7.   Oppose Until we have better inclusion criteria. Matthias Buchmeier (talk) 21:30, 20 August 2019 (UTC)
  8.   Oppose (though not strongly) on the grounds that the existence of single-word forms often indicates at least a perception that a compound word/phrase is considered NSOP, and per Hazarasp above. Andrew Sheedy (talk) 20:53, 21 August 2019 (UTC)
    I'm sorry, but I think this argument is just nonsense. Something like "coalmine" is sum-of-parts whether written as one word or two. See my comment above. Mihia (talk) 23:21, 22 August 2019 (UTC)
  9.   Oppose /mof.va.nes/ (talk) 10:24, 22 August 2019 (UTC)
  10.   Oppose SemperBlotto (talk)
  11.   Oppose [ זכריה קהת ] Zack. — 22:39, 22 August 2019 (UTC)
  12.   Oppose I don't think this vote addresses my specific problem with COALMINE, which isn't the fact that it allows a (fairly acceptable) subset of SoP terms to be added, but rather that it sometimes gets abused to create blatantly SoP terms, by starting with an incredibly rare and non-standard single-word forms of a term that properly requires a space or hyphen (I seem to recall "oldman" and "stupidfuck"). DonnanZ's remark that "I feel it can be manipulated in creating rare single-word forms of two-word terms" is the same issue. Equinox 09:59, 23 August 2019 (UTC)
    I acknowledge that all too often people create vanishingly rare compoundterm forms to support compound term. I now sent stupidfuck to RFD since it appears to be a rare misspelling per stupidfuck, stupid fuck at Google Ngram Viewer. If the form gets deleted (I hope so), we may delete stupid fuck as well. --Dan Polansky (talk) 10:44, 23 August 2019 (UTC)
  13.   OpposeMahāgaja · talk 18:46, 31 August 2019 (UTC)
  14.   Oppose 𐌷𐌻𐌿𐌳𐌰𐍅𐌹𐌲𐍃 𐌰𐌻𐌰𐍂𐌴𐌹𐌺𐌹𐌲𐌲𐍃 (talk) 19:04, 31 August 2019 (UTC)
  15.   Oppose, as this is an excellent rule which prevents us from being restricted to less common forms of collocations that are frequent enough to have yielded single-word variations. bd2412 T 00:49, 1 September 2019 (UTC)
  16.   OpposeAryamanA (मुझसे बात करेंयोगदान) 02:56, 2 September 2019 (UTC)
  17.   OpposeNumberguy6 (talk) 22:18, 2 September 2019 (UTC)
  18.   Oppose — justin(r)leung (t...) | c=› } 02:05, 3 September 2019 (UTC)
  19.   Oppose Ƿidsiþ 06:29, 3 September 2019 (UTC)
  20.   Oppose. I think it's healthy that we continue to relitigate this as a community every few years (no, really). Despite the fact that this vote is going to be heavily defeated, I'd like to commend @Mihia for bringing up an important policy that should rightfully be somewhat controversial. —Μετάknowledgediscuss/deeds 00:22, 8 September 2019 (UTC)
  21.   Oppose, why rock the boat? Doesn't seem worth it to me, but hey it's good to have these types of discussions every now and again. --Robbie SWE (talk) 10:28, 13 September 2019 (UTC)


  1.   Abstain – Either delete both coalmine and coal mine as SOP compounds, getting a better understanding of material criteria, or neither. I would support a motion for excluding terms that are SOP though written together. It has been wrongly assumed by the community that terms can be included on the sole ground that they are written together – the pertinacious adherence to which has lost Wiktionary very valuable editors. On the other hand, the fact that a reference to this misconception is included in the CFI text voted upon here is a ground for deleting it. But the CFI are contradictory and widely misunderstood and misapplied in many cases. They are full of epistemological criteria, or external rules of thumb, in the guise of actual rules, foster anglocentrism, and fail to recognize the difference between rules of inclusion and rules of deletion. The fact that this document considered by the majority basic is called “criteria” and not “rules” alone disqualifies the document called WT:CFI as a yardstick. Criteria are not rules but building blocks for methods whereby rules can be obtained, but people are incapable to derive to rules from that document – it could be a serious dictionary otherwise. Either they try to buttress their common sense about what the rules are or should be by references to it, or they try to underpin their perverse sense with arguments in reference to it, both leveraging personal convictions because nobody is convinced of the CFI, or they try to lever their lack of any sense by applying everything in the WT:CFI literally and out of context or telos, which gives the worst results. I should write my own inclusion/deletion document according to all insight at some point and let it be voted upon for a Neubekanntmachung. Fay Freak (talk) 23:54, 22 August 2019 (UTC)
    To write to be understood or to bloviate? That's the question. --Dan Polansky (talk) 07:24, 23 August 2019 (UTC)
    "It has been wrongly assumed by the community that terms can be included on the sole ground that they are written together" is mindbogglingly confusing. The community correctly knows that it has nearly unbounded power to set the rules wherein terms can be included or excluded. Whether that is wise or not is a completely separate question.
    "But the CFI are contradictory and widely misunderstood and misapplied in many cases" is a little less mindboggling, but still, if you believe a set of criteria are contradictory and ill-advised, you are hardly in the best place to know how they are correctly understood and applied. Pragmatically, I'd say a set of rules constructed by discussion and voting mean what they are applied to mean, more or less. Certainly, rules created recently by the same group that's enforcing them are unlikely to be "misunderstood and misapplied", no matter how much you may not like how they're understood and applied.--Prosfilaes (talk) 07:31, 23 August 2019 (UTC)
  2.   Abstain - I don't like COALMINE, but it does allow us to keep some terms which I think should be kept when I have no other CFI rationale for doing so. It also allows us to keep even more terms which I think should be deleted, but I can't vote to rescind without a better version to propose, and I haven't figured that out yet. I would rather keep a useless term AND a good term than delete a useless term AND a good term. - TheDaveRoss 15:30, 23 August 2019 (UTC)
  3.   Abstain - We need some sensible way of reducing the number of silly entries that COALMINE permits. This is not it. I don't know of a good, non-arbitrary way. DCDuring (talk) 01:11, 6 September 2019 (UTC)
      Abstain - Why don't we just have a single page explaining how words that are sum of parts should be looked up as separate entries in Wiktionary ? I can see a Lua module that would try to break up a reference into separate words and automatically create links to the individual pages. We would still have REDIRECT links to this sum-of-parts-module page, so when someone tried to look up the pages for a word phrase or a hyphenated-word-phrase or a combinedwordphrase. This would allow our wiki to handled the words, but not have pages that are effectively duplications. Jawitkien (talk) 16:03, 17 September 2019 (UTC)
    Indented, as cast after vote ended. — surjection?〉 17:30, 17 September 2019 (UTC)


3-21-3, fails. — surjection?〉 17:29, 17 September 2019 (UTC)