Wiktionary talk:Votes/pl-2018-05/Proficiency as a prerequisite for contribution

Rationale

Estimate: Requiring proficiency can lead to a great reduction of inaccuracies made by contributors who do not know the language and merely rely on sources. --Dan Polansky (talk) 06:56, 17 May 2018 (UTC)

Is "contribute entries" supposed to be "contribute to entries", or is this only about starting entirely new entries? --Yair rand (talk) 07:18, 17 May 2018 (UTC)

Thank you. I changed it to "new entries". Is it ok now? --Dan Polansky (talk) 07:21, 17 May 2018 (UTC)

Yes, this is clearer, thanks. --Yair rand (talk) 07:41, 17 May 2018 (UTC)

I have some queries:

• How is an editor's knowledge of a language or the fact that he or she has studied the language to be ascertained? Are we going to ask for documents?
• What is the consequence if an editor is deemed not to be sufficiently proficient – an administrator block? If so, how do we determine if an editor has achieved enough proficiency that he or she should be unblocked?
• Is our current blocking policy allowing for editors to be blocked if they "directly or indirectly, hinder or harm the progress of the English Wiktionary" insufficient?

— SGconlaw (talk) 07:25, 17 May 2018 (UTC)

1) I don't know, but asking for documents is not being proposed (is not in the text), and I don't see why anyone would tolerate such an unwiki requirement on wiki. 2) Why "ban"? A series of short blocks to give plentiful opportunity to change behavior should be enough. In any case, that is not specified in the proposal. 3) As for "directly or indirectly, hinder or harm the progress of the English Wiktionary", that is quite different from what is proposed; the proposal of the vote does not follow from it. In fact, someone might argue that the proposal itself would "directly or indirectly, hinder or harm the progress of the English Wiktionary", and that therefore, the proposal, rather than following from the quoted blocking policy, is antithetical to it. --Dan Polansky (talk) 07:32, 17 May 2018 (UTC)

What I was trying to convey by the third question is that since we already have a policy that allows for editors who hinder or harm to be blocked, that may be the only test that is required, rather than having an additional proficiency requirement which would be quite subjective to determine. — SGconlaw (talk) 07:36, 17 May 2018 (UTC)

And my response is, "... and that therefore, the proposal, rather than following from the quoted blocking policy, is antithetical to it". Therefore, given we know that WT:BLOCK is supported, we still do not know whether the proposal is supported. And surely WT:BLOCK, in its extreme vagueness, is much more "subjective to determine" than the proposal made; so much for "subjective to determine". --Dan Polansky (talk) 07:45, 17 May 2018 (UTC)

Note that my original response was written against this revision; hence my talk about "ban". --Dan Polansky (talk) 08:03, 17 May 2018 (UTC)

My sense is that it is probably easier to determine if edits are hindering or harming the Wiktionary in some way (for example, someone making multiple edits which editors familiar with the language in question find problematic) than to determine if a person has enough "proficiency". Alternatively, at least it could be said that the level of determination is roughly the same, in which case the blocking policy is sufficient and we may not need another policy. — SGconlaw (talk) 08:48, 17 May 2018 (UTC)

That's not my sense at all; the blocking policy is extremely nebulous. The minimum test of proficiency is self-report; for instance, I report to be unable to read and understand Danish; I need sources to determine what Danish words mean. Danish reminds me of German, which I know, but I still cannot understand written Danish. The proposal is indisputably much more specific, to say the least. --Dan Polansky (talk) 08:54, 17 May 2018 (UTC)

Are all of -sche's "water" entries bad now? Lack of proficiency should not be equated with incompetence. —Suzukaze-c◇◇ 08:44, 17 May 2018 (UTC)

There is a long-term recurring problem with people who are (or claim to be) experts in their specialized fields but don't know how to compose dictionary entries, often getting their parts of speech mixed up or making grammatical errors. I'm not sure that the problem needs a specific vote though. SemperBlotto (talk) 08:57, 17 May 2018 (UTC)

Reaction

Wyang (talk) 08:36, 17 May 2018 (UTC)

A whole new level of rudeness not seen in the English Wiktionary: a large derogatory image, instead of words, coming from someone who professes to honor the strength of argument. --Dan Polansky (talk) 08:39, 17 May 2018 (UTC)

This is your proposal:

"Voting on: In order for an editor to be allowed to contribute new entries in a language to the English Wiktionary, they must know or have studied the language. Careful work with reliable sources, knowledge of the English Wiktionary lemmatization practice for the language (e.g. Latin stare vs. sto), and willingness and ability to learn from one's mistakes are not enough."

A proposal so poorly formulated and discussed that the proposed change is not implementable in any way... no discussion of the topic on BP, and no desire whatsoever by vote creator to discuss this topic there. The whole vote essentially stemmed as a ragevote from the discussion at User talk:Dan Polansky#κλινικός. What an utter waste of everyone's time - you really think the whole vote system is an experiment land for your fits of pique do you? Wyang (talk) 10:37, 17 May 2018 (UTC)

You can propose a better wording and place it to the vote page. Whatever practice you think should be stablished can be put to the vote. I believe the formulation is good, but I am open to specific amends, as usual.

It is true that User talk:Dan Polansky#κλινικός was the last straw. However, I vaguely remember I have seen various admins make representations similar to what is currently proposed in the vote. My objective is to find how many people actually support that, and what reasoning they have. --Dan Polansky (talk) 11:30, 17 May 2018 (UTC)

• 
  Magna carta
• 
  Thomas Jefferson
• 
  Liberty, equality, fraternity
• 
  Truth prevails

Let's play the image game, shan't we. I posted some images above. --Dan Polansky (talk) 11:40, 17 May 2018 (UTC)

Some thoughts for discussion

(Numbered, for ease of reference in any replies.)

1. A conscientious editor (versed in Wiktionary practice) with sufficient general understanding of linguistics and without specific education in a language may still be able to improve the content of entries for that language. A specific example: I don't feel my grasp of Polish is sufficient (from a month of working in Warsaw and a Lonely Planet phrasebook) to add a "pl-1" Babel box to my user profile; and yet I believe it was right of me to add these entries. Or when I added Marmorkrebs, I wrote an entry for German as well as English, in order to keep the etymological chain (and then added the declined forms in German). (Essentially, I believe that your second sentence, "Careful work...", is enough - particularly for inflections.) My reading of your vote proposal is that you would rather I hadn't made those entries; is that right?
   1. While this suggested policy (if effective) would certainly reduce the amount of clean up work required from contributions of insufficient quality, I am concerned that it would also reduce the amount of contributions from users that currently add good quality entries.
   2. I suppose a counterargument would be that rather than create an entry oneself (if lacking specific education), one should instead "rfdef" currently-unlisted terms. We're currently missing a vast corpus of inflected terms in many languages; what's the best way to handle these for an editor that is not allowed to create a new entry for that language? (As an aside, I'd like us to be bot-creating some inflection entries, but that's not a conversation for here.)
2. What problem is this vote trying to solve that is not already handled by existing practices? Or is it more a pre-emptive policy which, if introduced, is intended to reduce the burden of administrators?
   1. I think the admin burden will not necessarily be reduced by this policy (if no automated barriers are in place) since admins will still need to police new entries and compare against a user's self-declared proficiency. And a user may either not use Babel boxes at all, or simply lie (either of which would also stymie an automated system).
3. If we mandate that a user cannot add new entries for a language that they do not know or haven't studied, do we also say that a user cannot add new entries for a dialect that they do not know or haven't studied? Should a British English speaker not add American English terms? Should a British English speaker from the South not add any Northern England English terms? Should someone from Yorkshire not add any Lancashire English terms (neighbouring county)? Should an American English speaker not add any Lancashire English terms? Should a Scottish English speaker not add Scots terms?
4. What and how much study counts as "studied the language"? I formally studied Arabic for a year at school and have a certificate to prove it. I worked in Madrid for three years with no formal study other than a phrasebook to get by. As a result of immersion in a Spanish-speaking environment, I can communicate better in Spanish than in Arabic, but have only "studied" the latter.

-Stelio (talk) 10:20, 17 May 2018 (UTC)

Let me explain that I do not necessarily support the proposal. I may abstain, but I will make the final decision after reading reasoned rationale of supporters of the proposal and similar proposals. --Dan Polansky (talk) 11:47, 17 May 2018 (UTC)

In response to you first point, I narrowed the proposal down to only concern lemmas. Thus, if the proposal passes as amended, you should still be able to create Polish inflected forms (or "declined forms"?). As for your German Marmorkrebs, some people really hate entries without pronunciation and inflection, and this hatred may be a driver for the proposal or similar proposals. --Dan Polansky (talk) 12:33, 17 May 2018 (UTC)

As for item 4., the relevant policy phrase is "they must know or have studied the language". I don't think the vote should contain a specific operationalization. Without a further specification, since you can communicate in Spanish, you know the language, Spanish. As for Arabic, you may not know it, but you have studied the language, so that is also a pass. This is just a common-sense interpretation, I belive. --Dan Polansky (talk) 13:12, 17 May 2018 (UTC)

This is the wrong way to write a rule.

My thoughts:

1. This type of rule is more likely to prevent a good entry than a bad one. The people who are likely to know and follow the rules are already the same people who solely contribute information they are confident is correct.
2. This rule is unenforceable, it seems like the only use that it is likely to see is as a weapon between long-term contributors.
3. This rule is completely subjective, both in the affirmative section (what constitutes "knowing" or "having studied" a language) and in the negative section (is not "careful work with reliable sources" the way to learn a language?), exacerbating points one and two.

If there is going to be an expertise rule (and I don't think there should be) it should be clear what the expectations and measures of expertise actually are. Since this project is mostly maintained by amateurs, an expertise rule seems extremely counter-productive. - TheDaveRoss 12:10, 17 May 2018 (UTC)

As for "The people who are likely to know and follow the rules are already the same people who solely contribute information they are confident is correct", I don't understand. If a newbie contributes in a language they do not know, someone can point the rule out to them. The newbie does not need to know the rule beforehand but only after an apparent problem occurs. A newbie could even be silently ignored in their contributing to various languages, and only after problems pop up could the rule be brought to their attention. --Dan Polansky (talk) 12:20, 17 May 2018 (UTC)

How is anyone to know what a newbie's language experience is? We might guess that they do not know the language if they are contributing incorrect information, but we already handle such cases without the need for additional rules. - TheDaveRoss 12:37, 17 May 2018 (UTC)

Self-report, for a start? If you see a mistake, you can ask: do you actually know the language or have you studied the language? If they say, no, I don't know the language but I am using sources so-and-so, you can say, well, that is not enough as per our policies. --Dan Polansky (talk) 12:40, 17 May 2018 (UTC)

And again, I vaguely remember I have seen various admins make representations similar to what is currently proposed in the vote. My objective is to find how many people actually support that, and what reasoning they have. I think the proposal is not that bad. It could be a good policy for a certain kind of wiki. However, I am far from sure it is a good proposal for a Wikimedia wiki. --Dan Polansky (talk) 13:32, 17 May 2018 (UTC)

Summary Deletion

@Wyang Hey Wyang, based on my interpretation of our policies this voting page should go through the RFD/O process, since it is not vandalism. You deleted the page, I restored it, and you have deleted it again. While I think it is clear that the consensus is not to pass this vote; I don't think it is a good-faith use of admin tools to delete the page without discussing it, and it certainly is not a good-faith use to re-delete after someone else (myself) has restored it. Please restore the page pending consensus to delete. - TheDaveRoss 20:30, 25 May 2018 (UTC)

This fails the first criterion at Wiktionary:Voting policy#Vote creation already. If the creator is actually serious about discussing and voting on this topic, he should bring it to BP before writing this absurdly unimplementable proposal. It is a waste of time for everyone to have to read and discuss it―including now that I am still taking the time to reply to this, when the vote is essentially concern trolling and an abuse of the vote system. All this time should have been spent on other more worthwhile things. Wyang (talk) 22:42, 25 May 2018 (UTC)

Whether or not a vote is a "waste of time" is not a valid reason for deletion. I get that you oppose the vote and that you don't think Dan created it in good faith. You can choose whether or not to engage with the discussion on the vote, you can choose how much time to invest (or not). It is up to other editors to choose whether they want to participate, it should not be a unilateral decision. - TheDaveRoss 01:00, 26 May 2018 (UTC)

Creating it should also not be a unilateral decision. Is there anyone who intends to support it? Even the vote creator had not decided to support it. —Μετάknowledge^{discuss/deeds} 10:55, 26 May 2018 (UTC)

I completely agree, and I think that point has been made abundantly clear on the talk page of the vote. I still don't think the deletion without discussion was appropriate. - TheDaveRoss 11:20, 26 May 2018 (UTC)

I agree as well. Vote creation without discussion is inappropriate; vote deletion without discussion is inappropriate. Now that we have finally had a discussion, we see that multiple people support deletion, and nobody supports the vote itself. This is an improper order for the usual process, but we now have reason to let this stay deleted. —Μετάknowledge^{discuss/deeds} 11:22, 26 May 2018 (UTC)

As for "Vote creation without discussion is inappropriate": I disagree; the discussion can be on the vote talk page, and we've had multiple productive votes created like that, including Wiktionary:Votes/pl-2017-05/Simplifying CFI about constructed languages, where Metaknowledge was very helpful in providing feedback on the vote page, based on which the voted text was made better. I still don't understand why Wiktionary editors cannot decide whether they want to have the vote via the "I oppose having this vote" section, which is on the deleted vote page. --Dan Polansky (talk) 09:18, 27 May 2018 (UTC)

As for "nobody supports the vote itself": I do support having the vote, and my plan is to place the text that I posted below ("Editors can contribute new entries even for languages that they do not know and have not studied...") to the vote page as soon as possible, which I cannot do when the vote page is deleted; I support the text. --Dan Polansky (talk) 09:33, 27 May 2018 (UTC)

(outdent) I believe the original proposal is implementable, but will probably not be supported. In Beer parlour, I put a proposal that can be put to vote, in response to the vehement opposition:

Editors can contribute new entries even for languages that they do not know and have not studied. However, in such case, they are strongly encouraged to work very carefully with sources, and get acquainted with the lemmatization practice of the English Wiktionary for the language. For instance, for Latin, some dictionaries use e.g. stare as the lemma while Wiktionary uses sto as the lemma.

Again, I find it inappropriate for a vocal minority to prevent votes from running. The above editor who deleted the vote repeatedly made representations to the effect that votes are fundamentally a bad thing, and they even erroneously invoked Arrow's Theorem in support of that thesis. The deleting editor is likely not to represent the English Wiktionary editorship overall. --Dan Polansky (talk) 23:13, 25 May 2018 (UTC)

Let's consider Wiktionary:Votes/pl-2018-03/Including translation hubs. In that vote, the concerned editor wrote "People even support this ridiculous proposal?" and "And no comments or harassments, thank you. This is a vote, not a discussion". The editor showed no appreciation for us finally having an approved policy for something for which there were only discussions over multiple years. The editor's anti-vote stances are unhelpful and are detrimental to the English Wiktionary progress and culture. --Dan Polansky (talk) 23:19, 25 May 2018 (UTC)