Wikipedia:Articles for deletion/Katarina Žutić

The following discussion is an archived debate of the proposed deletion of the article below. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.

The result was no consensus. Not sure whether there's a consensus to keep, but there's certainly no consensus to delete. The broader issues related to autotranslated articles are probably best addressed outside individual AfDs. Sandstein 10:44, 23 December 2019 (UTC) Sandstein 10:44, 23 December 2019 (UTC)[reply]

Katarina Žutić[edit]

[Hide this box] New to Articles for deletion (AfD)? Read these primers!

(Find sources: Google (books · news · scholar · free images · WP refs) · FENS · JSTOR · TWL)

The consensus on Wikipedia is that machine translations of foreign-language Wikipedia articles should be deleted. This is because foreign-language articles may be updated or improved in the source language and because the translation algorithms are constantly improving. Anyone can generate a machine translation with a few clicks, so it's best done in real time. Pasting a machine translation into en.wiki crystallises the version from the date of the translation. Sadly the WMF failed to see this and they made a special tool for machine translating articles. This tool was disabled on en.wiki as a result of the consensus at this discussion following which it was decided to allow a special CSD for the 3,603 articles that had been generated by the tool. That CSD was enshrined on WP:CSD as CSD X2. It applies most strictly to BLPs such as this one. However, when I tagged this biography for deletion in accordance with these discussions, my tag was removed with the comment "OKish article", which it's not. Please will the community authorise its deletion. No prejudice against a fresh article being generated on the basis of actual research by good faith editors, of course. —S Marshall T/C 02:24, 16 December 2019 (UTC)[reply]

Note: This discussion has been included in the list of Actors and filmmakers-related deletion discussions. —S Marshall T/C 02:24, 16 December 2019 (UTC)[reply]

Note: This discussion has been included in the list of Serbia-related deletion discussions. SharabSalam (talk) 02:27, 16 December 2019 (UTC)[reply]

Note: This discussion has been included in the list of Women-related deletion discussions. ~riley (talk) 04:48, 16 December 2019 (UTC)[reply]

Note: This discussion has been included in the list of Bands and musicians-related deletion discussions. CAPTAIN RAJU^(T) 05:35, 16 December 2019 (UTC)[reply]

Keep I think that this article should remain. She is a notable actress and more refs. should be added with some style tweaks as well. Sadkσ (talk is cheap) 10:17, 16 December 2019 (UTC)[reply]
This is one of those rare cases at AfD that isn't basically about notability. This may well be a notable person, and there's no reason why a good faith editor couldn't write an article about her. We just need to delete this text in the meantime, for all the reasons I gave above.—S Marshall T/C 10:33, 16 December 2019 (UTC)[reply]
Keep. I sympathize with your opinion, and I advocate for WP:TNT myself, however, that is not the current policy and practice. The subject is notable, the article is now reasonably sourced and in a minimum viable stub state.
By the way, The consensus on Wikipedia is that machine translations of foreign-language Wikipedia articles should be deleted is not exactly true. I participated in the WP:AN/CXT cleanup myself, and in the end, over half of the translated articles were kept (WP:CXT/PTR), and the effort ended up with a silent consensus that it was a fiasco. Heck, it was actually me who recommended TNT on WP:CXT/PTR about this very article. No such user (talk) 09:41, 18 December 2019 (UTC)[reply]
- The AN/CXT business hasn't ended. There was a pause because I seem to be the only person still doing it and I had a Wikibreak, but a pause isn't a silent consensus that it's a fiasco! CXT/PTR has long since been superseded but the cleanup's still ongoing --- as this discussion demonstrates. :)—S Marshall T/C 14:33, 18 December 2019 (UTC)[reply]
  - To quote Elinruby from WP:AN/CXT#The interim period ends today: I'm afraid we're going to find out that we've all done a huge amount of work to delete 30 articles that need to be deleted and 350 whose authors will will not contribute again. and The rest are... sloppy english but accurate, unclear but wikilinked, or some other intermediate or mixed level. This has not, in my opinion, been a good use of my time and I have stopped doing any translations, personally, until we get some sanity here. The whole process, it seems to me, simultaneously assumes that translation is easy and also that it is of no value. There has been quite a moral panic about automated translation, but on average that large batch was just an average collection of poor-to-mediocre-quality articles, where faithfulness of translation was very low on the list of issues. But let's leave that discussion for somewhere else.
    In my opinion, the article in its current state is poor but just above the TNT bar, and while foreign-language articles may be updated or improved in the source language and because the translation algorithms are constantly improving you propose is an interesting prospect, it needs to be seen if this will become a policy in the future. Until then, every edition of Wikipedia will need to maintain their articles separately. No such user (talk) 12:54, 19 December 2019 (UTC)[reply]
    - Wait, what? Wikipedia:Translation has said this since Jmabel added it on 24 December 2006. The consensus on raw machine translations is old and strong. Fixups of raw machine translations are OK if the editor fixing them up has dual fluency and can check that the translation is actually right... but this is a BLP. Also, I realise that Elinruby has thrown their hands up in the air, but I haven't.—S Marshall T/C 17:33, 19 December 2019 (UTC)[reply]

- Keep. The consensus on the undesirability of unedited machine translations is strong. The consensus about what to do with them is another matter entirely. One thing is clear, there is no longer a consensus to remove them by speedy deletion. There is no consensus that the translation must be edited by experts. Double fluency is a ridiculous requirement, considering the quality of most articles in Wikipedia . WP is not an reliable source, and is not intended to be. Even articles without sources altogether cannot in general be speedy deleted--proposals to do this have been rejected many times. Even unsourced BLPs must go through BLPProd, a semi-sticky process, where anyone can add a source. If a pertinent reliable source is added, the article stays. We just verify the existence of a source that confirms the existence and principal notability. We do not verify accuracy unless the article is individually challenged and has an individual discussion. The CXT process was a terrible idea from the first, spearheaded by some very good translators who were under the impression that translation is not for amateurs. But everything in WP is appropriate for amateurs, and neither expert subject knowledge nor fluency even in English is necessary.

Most of the problems with the machine translated articles is the same as for any articles--dubious notability and variable quality of the original. Some of it is from the practices of the two most reliable WPs, the French and German, which use a more informal way of citing sources even for BLPs, and though their standards--especially those of the deWP are in general higher than ours, it can be difficult to convert their sourcing to our practices.

Most errors in machine translation are obvious and easily fixed--for languages like French and Spanish and Italian, the problems with converting tense usage, especially for discussing the past, the problems of translating from languages where all nouns have gender. Sometimes the machine gives a word so absurd that any reader can know it needs checking. Sometimes, of course, there can be an important and more subtle error. But this can occur in any article, especially when it uses nonEglish sources, or even English sources that may not be carefully understood by the contributor.

The proper way to deal with these articles is to check them like any other article. (There is one special problem--it is considerably trickier to check possible copyvio from a source in a different language because of computer asisted checks are worthless here)

S Marshall, of the thousands of these articles, the best way to proceed is the same as in any area of possibly questionable material--start with the ones that look look they would not be acceptable here., and the ones that simply do not make sense as translated. There are enough of both. It takes judgment, like anythign else here. DGG ( talk ) 07:53, 22 December 2019 (UTC)[reply]

De.wiki has a higher bar for notability than us, but I don't agree that de.wiki has higher standards than us. I think its culture is really different. De.wiki isn't as source-focused as us. Unlike us, it does have trusted editors (which is why flagged revisions works for them) and an obsessive focus on good German style. All of these articles would have been deleted from de.wiki for poor writing.
I don't agree that fr.wiki's reliability is in any way comparable with ours. It's got no intelligible focus on trustworthy sources or editors.
I don't agree that the AN/CXT process was led or directed by good translators. There was input from some active translators, but Black Kite and Xaosflux were in the driving seat, notably supported by Iridescent. I feel that in fact the discussion was led and directed by people who were motivated by frustration with the WMF.
I note your comments there and I see that you made very similar comments at the time. I'm afraid it does very much appear to me that yours is not the consensus view on this.
We could save ourselves from discussing all these articles individually, of course, by actually draftifying all the articles in the draftification list. Or even just the BLPs. At this stage I'd settle for just the BLPs.—S Marshall T/C 11:51, 22 December 2019 (UTC)[reply]

I think our dialog is making some progress. You and I understand the deWP in the same way. For the frWP I know mainly the academic bios--they are written differently than ours, in a much more impressionistic way, and needs editing to remove material that we do not usually include, but I have yet to find one in this field that isn't clearly notable, except 1 or 2 that are being simultaneously challenged at the frWP--other fields might be different.

I also agree that the articles on the list need review. Some have already been reviewed, and are acceptable by our usual standards. The others could well go to Draft, or be reentered in some manner in NPP. It makes sense to start with the BLPs. If they are going to get decent review without overloading an already overloaded process, we'll need to go in small batches, like 10 a day. I looked for those in my primary field, but almost all the living scientists have already been dealt with; the 2 or 3 that remain need editing, but no more so than any correctly translated article from their WPs. Most of the others are politicians, which are easy to verify, and where there is usually no doubt about notability because of their positions, popular performers where it is difficult to deal with notability from different cultures, & athletes--some of which seem clearly notable and easy to evaluate, but this is a field I avoid.

We also need to look at the ones that were deleted without being sent to AfC for review, for many of them seem quite verifiable and notable, tho some don't seem worth the effort of fixing.

For that matter, we need to look at all articles that have not been substantially revised or edited since their introduction in earlier years that were written by new editors. In some fields, I think perhaps 25% or more would not meet our current standards. I think that might be half a million articles.

But if we agree on a way of proceeding with the current problem, let's do it. DGG ( talk ) 20:05, 22 December 2019 (UTC)[reply]

My suggestion would be to draftify the whole 2017 draftification list until they can be reviewed. I expect we'll be needing a bot.—S Marshall T/C 22:58, 22 December 2019 (UTC)[reply]

The above discussion is preserved as an archive of the debate. Please do not modify it. Subsequent comments should be made on the appropriate discussion page (such as the article's talk page or in a deletion review). No further edits should be made to this page.