It does not. It states that Jagwar has made copyright violation by entering copyrighted material into Malagasy Wiktionary, which he has later deleted. It does not explain why he had to create so many definitionless entries. Thanks to Jagwar's bot, Malagasy Wiktionary is a huge repository of unattested inflected forms and copyvios of other Wiktionaries. Some of that can be seens from mg:Dinika amin'ny mpikambana:Jagwar. --Dan Polansky (talk) 18:25, 20 July 2012 (UTC)[reply]
The great number of definitionless entries were due to copyright violations (maybe I should learn to write properly in English). That's all. I also do not agree with the huge repository of unattested infected forms : are referring to Volapük verb forms or Malagasy noun forms? on 440,000 malagasy language pages created. I can assure you that only 30,000 (seemingly very needed) pages were recreated due to copyright. That's all ; bot-created forms in Malagasy are perfectly attestable, as well as my volapük verb forms. Our CFI ain't the same as yours here, because comparing to English, we don't have many resources in Malagasy language. You don't find anything about Volapük lemma forms on Google? Then open a Volapük grammar book! --Jagwar (talk) 18:51, 20 July 2012 (UTC)[reply]
Thanks for that link, Dan. I noticed that Jagwar did not answer my last question, but I thought that meant he wasn't involved; now I see otherwise. I have struck my vote until I get full clarification from you, Jagwar, about why this time around, in enwikt, it will end of differently, and a good reason why I should believe it. AFAICS, the mess is not cleaned up yet over at mgwikt, and it's been a while. --Μετάknowledgediscuss/deeds18:55, 20 July 2012 (UTC)[reply]
For those who need to understand why there is so much definitionless entries on the Malagasy wiktionary :
Their addition were primarily made following a misunderstanding :
In a mail to a Malagasy pnline dictionary site owner, I have asked to the owner if definitions in his dictionary can be used in Malagasy Wiktionary under a free licence. The site owner has replied (to simplify) "Yes, you can use them". So I did. Implicitly thinking that the dictionary from which the definitions were retrieved was a P.D dictionary.
But a few months later, I have received from the WMF staff a copyright infringement notice (from the real owner of the definitions, supposedly dead for 70 years!), following which I had to delete (and re-create) all entries related to the site in question. And also learnt by the WMF that the real copyright owner of the content was a publisher to which I sent a mail for authorisation (which has been refused) to copy and release it under CC-by-SA licence. Now, you know all the story.
I must also notice that since that story, my PC on which I usually run my Wiktionary bot has crashed and don't have the motivation to create the magic script anymore (which required ~500 lines of code, using web and online dictionary researches for attestability). So I decided to keep the noun forms not to waste hundreds of hours of work.
I had to rewrite all my scripts ex-nihilo. I have BTW made a translator using a bridge language which works quite well (the first 2,000 test translations were fail-less), making a quite reliable tool to make the mg.wikt more exhaustive, and increasing the number of languages translated into Malagasy from a few tens to more than 100 languages.
You must also consider that my programming skills are not static. Because in my life, I study industrial computer science at University. I've been working on wikis for almost four years now. And creating scripts working on them for two.
I understand your point of view, but here every entry has a definition, or at least a translation. So do you really think that I will be going to import the 290,000 entries on the English Wiktionary? Only word form of existing words in this Wiktionary will be created. No worries. --Jagwar (talk) 17:19, 20 July 2012 (UTC)[reply]
If you don't know what I am planning to do on this Wiktionary, why don't you read the bot purposes? Maybe you need further explanations? Why not asking for it instead of vetoing just because you don't like what I've done on another Wiki? I can do some good work here. I propose my help on this wiki, but if the local community don't want so. It is no more my problem. --Jagwar (talk) 17:39, 20 July 2012 (UTC)[reply]
A note on attestation: Above, I have listed several inflected forms that point to a definitionless entry. Several of these inflected form have one web hit, on the Malagasy Wiktionary: google:"agilahinao", google:"agilahinay", google:"agilahinareo", google:"agilahiko", google:"agilahiny", google:"agilahintsika", google:"agilahim-", google:"agilahin'", google:"agilahin-", and google:"agilahinjareo". Thus, they are not attested by any stretch. The owner of the bot seems to think that having a grammar book from which the inflected forms theoretically follow is enough for their entry into Wiktionary. By contrast, I still think that attestation of each form in use should reign supreme. Even if Malagasy is one of the poorly documented languages and thus, unfortunately, is subject to extremely weak attestation criteria that allow for a single quotation in use or mention, these forms do not seem to have even that. --Dan Polansky (talk) 23:29, 20 July 2012 (UTC)[reply]
I encounter many problem with the way you attest Malagasy words because according to that, the phrase Nikapahinjareo ny agilahiko (My Tragia cordifolia have been cut by them) would be totally unattestable because you can not find two most important words on Google : nokapahinjareo (they have cut) -- which can also be written as "nokapahin-jareo" --, and agilahiko (agilahy is Tragia cordifolia in Betsileo dialect). To avoid having such gaps in the Malagasy Wiktionary, we don't need Google attestation to authorise or not the creation of an Malagasy noun form. Because noun forms you are referring to are actually possessive and compounding forms: All nouns without any exception can be declined following this same scheme. So this is why we should accept the creation of such inflected forms.
The only thing for which we use attestation for Malagasy (and other highly agglutinative) language is when creating a compounded word. If that word is a noun, then it is theoretically and technically possible to have forms of it. If never it isn't the case, how would I translate "my appointment" in "my appointment to the dentist"? I would say "ny fotoako any amin'ny mpanao nify".
The French Wiktionary seems to share my opinion, as they also create very rare verb forms : cf. google:rererecommençât or google:rererecommençassions which have quite rare hits for a language spoken on the five continents.
This is way I (we) do create these entries even if they can not be attested by Google. I know I can not change your opinions on attestability, but this language is my native language, and I must tell you that I practice it everyday in real life and have learnt my grammar lessons enough to have some weight in my statements. As far a --Jagwar (talk) 10:41, 21 July 2012 (UTC)[reply]
Actually I don't understand your reticence. As an admin, you are basing your arguments not on the way how my bot works here. Am I right? --Jagwar (talk) 19:59, 20 July 2012 (UTC)[reply]
Oppose after reading all arguments on this page and the relevant talk pages. I will not respond to any discussion of my vote, especially not on this page. Gauss (talk) 20:25, 22 July 2012 (UTC)[reply]
Hi Ruakh, misformatted bot edits have been re-formatted. And I don't think that taking out the "/" are a good idea. --Jagwar (talk) 17:00, 24 July 2012 (UTC)[reply]
Re: first sentence: I'm aware, since I'm the one who re-formatted them! (Though I see that you've now run a new batch of test-edits that is not misformatted. Thank you.) Re: second sentence: I have no idea what you're talking about. Who said anything about taking out the /? —RuakhTALK17:50, 24 July 2012 (UTC)[reply]
You have not done all the work (the bot still found other pages to changes when searching IPA pronunciations). About taking out the /? No one actually, it was just a note. --Jagwar (talk) 18:10, 24 July 2012 (UTC)[reply]
Re: first sentence: Au contraire. I literally went through every single one of your bot's misformatted pronunciations and fixed them. You later made some more misformatted pronunciations and fixed those, but that doesn't win you any points from me. Re: second sentence: Why? What are you getting at? Who was the note directed at, and why did you feel it was necessary/helpful? —RuakhTALK19:06, 24 July 2012 (UTC)[reply]
Eh bien soit ; now they are all fixed, as well as the script I use. The note was directed to myself. It is necessary to put the slash to make it indicate phonological pronunciation. --Jagwar (talk) 21:56, 24 July 2012 (UTC)[reply]
User talk:Jagwar (local talk page) -- mg:Dinika amin'ny mpikambana:Jagwar (home wiki talk page)
Support --Μετάknowledgediscuss/deeds 18:11, 20 July 2012 (UTC) per talkpage, which seems to adequately explain away all of Dan Polansky's concerns. --Μετάknowledgediscuss/deeds 18:11, 20 July 2012 (UTC)
Oppose Dan Polansky (talk) 17:14, 20 July 2012 (UTC) I oppose adding inflected forms in the absence of base form entries, which is what the user has been doing in Malagasy Wiktionary. For reference, there are currently around 3000 entries in Category:Malagasy nouns in English Wiktionary. The corresponding category in Malagasy Wiktionary is mg:Sokajy:Anarana iombonana amin'ny teny malagasy (Category: Malagasy nouns) and has 31 418 entries, many of which are definitionless, which can be seen by their definition reading "Tokony hofenoina" and the wiki code of the definition reading {{...|mg}}. There is mg:Sokajy:Endrik'anarana_amin'ny_teny_malagasy (Name forms of Malagasy language; endrika-->form of; anarana-->name, term?; teny --> language) with 291 214 entries. Many of these inflected forms point to definitionless base entries. A case in point if mg:agilahy (definitionless), to which mg:agilahinao, mg:agilahinay, mg:agilahinareo, mg:agilahiko, mg:agilahiny, mg:agilahintsika, mg:agilahim-, mg:agilahin', mg:agilahin-, and mg:agilahinjareo point. --Dan Polansky (talk) 17:14, 20 July 2012 (UTC)
