Wikipedia talk:Wikidata/2017 State of affairs/Archive 11

Archive 5

←

Archive 9

Archive 10

Archive 11

Archive 12

Archive 13

Archive 14

A possible solution to some issues

One common argument against using Wikidata is that Wikidata short descriptions are less visible to Wikipedia editors, with potentially lower quality and susceptibility to vandalism. To fix this, I propose putting the description in smaller, gray text to the right of the article title, in the desktop UI, with an edit button next to it. Clicking that edit button would provide a form for editing the short description, and perhaps also the title. Editing the short description would edit the Wikidata value, and editing the title would move the page. —{{u|Goldenshimmer}}|✝️|ze/zer|😹|T/C|☮️|John15:12|🍂 02:56, 30 October 2017 (UTC)

Or we could use a simpler solution and not use Wikidata short descriptions at all and be free of all the complications they cause.· · · Peter (Southwood) ^(talk): 16:31, 30 October 2017 (UTC)

Pbsouthwood: In my opinion, the potential advantages to be gained by using Wikidata makes overcoming the technical challenges worthwile. —{{u|Goldenshimmer}}|✝️|ze/zer|😹|T/C|☮️|John15:12|🍂 15:40, 31 October 2017 (UTC)

Then perhaps indicate what the potential advantages are, the things having descriptions on Wikidata achieves which are not (or in a worse way) achievable on enwiki? Fram (talk) 15:48, 31 October 2017 (UTC)

The main potential advantage is that Wikidata has already the (technical and social) system for creating such short descriptions, while enwiki doesn't have it yet. Jo-Jo Eumerus (talk, contributions) 16:01, 31 October 2017 (UTC)

Technical, yes, for now. Social, not so much (many items, even new creations on topics which have an enwiki article, don't get descriptions at Wikidata). Fram (talk) 16:06, 31 October 2017 (UTC)

Goldenshimmer, when I look at the advantages and disadvantages, the disadvantages of not having control over the content outweigh any potential advantages, particularly since they are not very apparent to start with. The technological challenges are another issue, secondary to the data reliability and curation issues. · · · Peter (Southwood) ^(talk): 17:42, 31 October 2017 (UTC)

I have been waiting years for the abilities to have ONLY the changes to WD items / properties used on EN WP appear in my watchlist. Until this becomes possible I strongly support pulling all usage of WD on EN WP. I do not have the ability to build this myself so I am simple unable to "fix it". Doc James (talk · contribs · email) 20:54, 1 November 2017 (UTC)

Does that include interwiki links? · · · Peter (Southwood) ^(talk): 05:42, 5 November 2017 (UTC)

@Doc James: But you can? Preferences -> Watchlist -> "Show Wikidata edits in your watchlist", which I think works the way you want? Albeit in a more limited fashion at the moment, given that there's currently a maximum number of pages that the Wikidata edit is connected to. Or am I missing a nuance? Thanks. Mike Peel (talk) 11:10, 5 November 2017 (UTC)

That shows any change to the WD item not just changes to the statements used within WP. Doc James (talk · contribs · email) 16:22, 5 November 2017 (UTC)

Might it help if the edit summary showed the diff, so that you can see each change on the Watchlist without having to check the diff? Jo-Jo Eumerus (talk, contributions) 16:51, 5 November 2017 (UTC)

@Jo-Jo Eumerus: From my experience in pt.wiki, it basically floods your watchlist with irrelevant stuff. Even with improved summaries, it's not something I would ever use to monitor vandalism. I'm already watching two projects, I don't need a third - even more one where I don't even participate.-- Darwin ^Ahoy! 18:09, 5 November 2017 (UTC)

Well, I know that and it's indeed useless in it's current form. I was thinking that displaying the actual text change in the edit summary may make it more useful. Jo-Jo Eumerus (talk, contributions) 19:33, 5 November 2017 (UTC)

@Jo-Jo Eumerus: I believe your suggestion is a good improvement, and it would help, yes. But the huge amount of noise would still be there.-- Darwin ^Ahoy! 19:53, 5 November 2017 (UTC)

Relevant discussion at Portuguese wikipedia

The portuguese Wikipedia is currently holding a relevant discussion[1] about the use of Wikidata infoboxes (similar to ours, as they use the same Lua module). Please don't go and join the discussion there, whatever your position on this; but read it and look at the arguments and whether they apply here as well or not. It is interesting to see a discussion without the same familiar names but in many cases with the same arguments. Fram (talk) 08:45, 2 November 2017 (UTC)

This is rather circular, as they are watching this discussion, and the arguments posted here are influencing the discussion there... Thanks. Mike Peel (talk) 09:10, 2 November 2017 (UTC)

I think the editors at Portuguese Wikipedia are mature and intelligent enough to decide which arguments are valid (completely or somewhat) and which have any importance or not. They are not influenced by biases and prejudices we may have about each other (I mean "if X says it, it is probably right or wrong" kind of thinking). Fram (talk) 09:37, 2 November 2017 (UTC)

I'm working with the (Brazilian) Portuguese community on Wikidata infoboxes as well, and my experience so far is that there is a lot of some thinking along the lines of "If English Wikipedia says it, it is probably right" (sadly). That seems to partially be a cultural thing (I think they even have a ptwp article about the concept, although I can't remember the name right now). So The two conversations are quite intricately combined, and I don't think you can consider them to be independent. Thanks. Mike Peel (talk) 11:03, 2 November 2017 (UTC)

Any actual evidence that this Portuguese discussion actually does this? Fram (talk) 11:10, 2 November 2017 (UTC)

Yes, but I can't find the links right now - it will take me a bit of time to dig them out. Thanks. Mike Peel (talk) 11:50, 2 November 2017 (UTC)

@Fram and Mike Peel: Greetings from the Brazilian community. It is definitely true that discussions on this community have influenced the development of the discussion on infoboxes that are integrated to Wikidata on Wikipedia in Portuguese. You can easily check the change of direction (and tone!) once an editor in our community brought up for the first time in our community a link to the long set of discussions that were going on on Wikipedia in English.

Let me bring some perspective from an editor in an emergent community. I won't go over all the positive aspects on Wikidata-integrated infoboxes (WiI), that were mentioned on the trend on English Wikipedia several times, and I will just focus on two aspects that I believe are extremely relevant from my perspective. I hope you don't mind.

1. WiI are a means of international catch-up. Wikipedia in English has over 5 million entries --OMG, good job!--, a lot of the work we are trying to do has already been done by you guys (we are not even at the one-million-entry level yet!). Just to give an example: when our community decided it was time to edit on the Art Institute of Chicago, a beautiful collection that had remained a stub for ages (it actually just became a good-enough article a month ago [2]), we were able to get an impressive infobox via one single command, because the work had been done from your side (Q239303). The impact of your work (on Wikipedia/Wikidata) on our work (on Wikipedia) is definitely very impressive --and I am thankful. Nevertheless, I believe a strong community like the English Wikipedia will also benefit from work we are doing (on Wikipedia/Wikidata). At some point in history, an editor on English Wikipedia will be really glad to find out the Brazilian community has put a lot of energy to improving the Wikidata entries on the institutional acts, such as the AI-5 (you guys have created only one of the 17 documents that have institutionalized the Brazilian military dictatorship!). This is the Wikidata entry you can benefit from (Q2642410 in the case of the AI-5, and we have done an equivalent work for all 17 institutional acts); you might also be interested in taking a look at the WiI we have designed for documents, that you can also easily and eventually adapt to your environment (AI-5). I have myself used as a reference the WiI on books that was developed in English to create the one in Portuguese (thanks to the developers!). Moreover, I truly believe that once an editor in English comes to the decision that it makes sense to create a specific entry on the Brazilian Constitution of 1891 --the first document of this kind in the Brazilian Republic era--, he/she will be very pleased to find out we have already done the work of checking features of this constitution on Wikidata (Q16143192), including the list of hundreds of subscribers. I hope what I just said makes sense.
2. WiI benefit from the integration of several Wikimedia projects. Since editors from different projects are ideally editing Wikidata, they are easily letting editors from other projects know what they can connect with. From the perspective of a Wikipedian, at some point the Commons will be so large --maybe it already is-- that it will take a real effort to find what he/she wants to find; the reciprocal perspective is also likely true. The increasing effort of finding what one wants to find could occur across projects, across languages. Wikidata provides a collaborative ground to helping each other finding what is out there, and this can be expressed on WiI. For instance, as we have developed a large GLAM initiative with the Brazilian National Archives, that has had an impact on three projects (Commons/Wikisource/Wikipedia), we have come to the decision that Wikidata remains the ideal project to let us know what we were doing on each front, and integrating automatically efforts on Wikisource and the Commons onto infoboxes on Wikipedia. How so? Well, once an editor fed Wikidata with info on the fact that users on Commons had compiled the pages of the AI-5 (that we had received as individual pages in the context of the GLAM) and users on Source had created a full transcription of the AI-5, links to this new content automatically popped up on the infobox on Wikipedia. Again, you can see the result here: AI-5 in Portuguese --the link to the Commons full version and the transcription on Source just popped up. (BTW, you can find more on this GLAM we are trying to integrate as much as possible with Wikidata and we are proud of here: piece on the GLAM newsletter).

Sorry for the long post. I hope I have contributed. Good work to all! --Joalpe (talk) 12:37, 2 November 2017 (UTC)

That the tone and direction of the Portuguese discussion changed once they read our discussion may well be true. That however doesn't mean that the participatnts there decided "oh, we like Wikidata, but if enwiki says it bad, we have to follow suit". It could equally mean (and this seems more likely to me) that they saw the actual complaints, examples, problems, ..., checked for themselves, and then decided "hmm, it's true, using Wikidata in general isn't as unproblematic as it looked at first". Fram (talk) 12:59, 2 November 2017 (UTC)

Well, this is the sort of statement that cannot really be falsified. So, from any reasonable logical standpoint this won't really move forward.

As I have tried to make the case for on my earlier post (possibly without enough clarity), there is a specific discussion on integrating Wikipedia-Wikidata from the perspective of emergent communities, and this specific discussion was overridden as the discussion on Wikipedia in Portuguese became a replicate of the discussion on English Wikipedia (whatever reason led to this).

Is there a specific point from the discussion in Portuguese you would like to bring up here, as for some reason --you might want to clarify-- you thought it would be for some reason relevant to discuss Wikipedia in Portuguese herein? --Joalpe (talk) 13:19, 2 November 2017 (UTC)

No, my reason for posting this here is that other groups of editors (in this case the Portuguese) are also discussing the use of Wikidata infoboxes, and many editors seem to agree that there are significant problems with these infoboxes. Which means (assuming that don't blindly believe what some us say here) that the problems with Wikidata are not constructs from the minds of some anti-Wikidata editors, but are more widely perceived. They are a form of control-group, an indication that the discussion here is not something aberrant but something which is perceived similarly by others as well. Fram (talk) 13:43, 2 November 2017 (UTC)

Not sure if you can have a "control group" without independent testing in an experiment; so, again, I don't think this approach is helpful, as the discussion on Wikipedia in Portuguese is undoubtedly influenced by what has happened on Wikipedia in English.

Moreover, I am not sure what your reading skills in Portuguese are but the consensus tends to be, so far: (1) to prefer infoboxes that are manually generated on Wikipedia over Wikidata-integrated infoboxes; and (2) to rely on Wikidata-integrated infoboxes whenever an infobox is missing (if it makes sense to do so, of course). The consensus on Wikipedia in Portuguese is not formalized yet. This consensus certainly differs from what the community on Wikipedia in English appears to have decided. There are specific benefits from the perspective of an emergent community to rely on Wikidata-integrated infoboxes.

The false dichotomy between anti/pro Wikidata editors that has emerged from the tense discussion on Wikipedia in English is unhelpful, I believe. I think the best rationale is to consider that Wikidata is already a resource out there (thanks fellows who contribute to this project and are reading this!) and that the real question is to devise an idea on how it could ideally contribute to the goals of our projects (and general mission of providing open, free knowledge of high quality to the world) and move collaboratively in the direction of this idea. This seems more appropriate to the general, underlying principles that have kept us moving forward.

From my perspective --from someone who is not an editor on Wikipedia in English and have normally been admirative of the good work that has been done in this project--, I wish the discussion here would have been of higher quality and we could have benefited more from the devising of a constructive agenda for our projects and community. This was certainly not the case. --Joalpe (talk) 14:14, 2 November 2017 (UTC)

Wikidata is a resource, but not a reliable one. Their approach to sourcing is fundamentally different from ours. To use this as the basis for data here is fundamentally wrong. This doesn't mean that there aren't subsets on Wikidata which are good (or that many Wikidata-editors aren't well-meaning). Asking how it could ideally contribute is ignoring that after 5 years, the communities have completely grown apart, with Wikidata attracting (at least as some of the core editors) a group of people whose ideas are fundamentally different from what the basic principles at e.g. enwiki are (to put it extremely simple, in theory they go for quantity, we for quality). Approving datasets with known errors or approving unreliable sources is common. See e.g. Wikidata:Requests for comment/Familypedia links removed for "described at URL", a current discussion where one lone voice of reason is being dismissed, and the use of crowd-sourced genalogy websites is being promoted. Similar discussions for IMDb and Findagrave can also be found. That last one especially is fascinating reading. Fram (talk) 14:45, 2 November 2017 (UTC)

I know your perspective, as we have been reading what your posts. I thought that when you indirectly summoned me up on this discussion --as a member of Wikipedia in Portuguese-- you were actually interested in reading my take and getting a sense of how the discussion has evolved in the community I volunteer in --as from what I get you are likely to be unable to read Portuguese (or have never put any skill on reading Portuguese into use in any project in Portuguese). I might have been wrong, as I don't see how what I said might have led you to produce a "they-we" kind of speel. I tend to think more in general, global terms: (1) "we all" want good quality content to flourish in our projects; (2) "we all" believe there is a transition cost in the process of integrating Wikidata and Wikipedias. Given this "we-all" approach --and hopefully foreclosing your vague depiction of Wikidata editors as a wiki-Fronde--, there is room for discussion; you don't give me the impression to be interested in going along these lines, or is there anything you *really* want to know about the discussion in Portuguese I have not brought up? Anyway, I am happy to provide more information on the discussion on Wikipedia in Portuguese to anyone else who is interested. --Joalpe (talk) 15:04, 2 November 2017 (UTC)

I was not aware that I summoned you here, directly or indirectly. Anyway, I don't know who you include in the "we-all", but no, looking at the discussions I linked (and others) from Wikidata, it doesn't look to me as if "we-all" want good quality content. Good quality is a core requirement of enwiki (even though we way too often don't have or produce it), and is a nice-to-have for Wikidata (in reality). The approach openly advocated by many at Wikidata is "first mass-import a source, and then piecemeal deprecate it if and when we can lay our hands on a better source". We (myself included) have done the same here on enwiki with outdated sources (e.g. the 1911 Britannica), but never (or certainly not since a very long time) with fundamentally unreliable sources. So you start from an incorrect position when you claim "I think the best rationale is to consider that Wikidata is already a resource out there (thanks fellows who contribute to this project and are reading this!) and that the real question is to devise an idea on how it could ideally contribute to the goals of our projects (and general mission of providing open, free knowledge of high quality to the world) and move collaboratively in the direction of this idea." Unless Wikidata or enwiki is willing to change fundamentally (and I have seen no indication that most editors there or here are willing to even consider this), there is no such collaborative move possible and Wikidata is not a resource we should be using. Fram (talk) 15:22, 2 November 2017 (UTC)

Joalpe you said Portuguese Wikipedia consensus tends to be, so far: [] to rely on Wikidata-integrated infoboxes whenever an infobox is missing. I had to use Google-translate to read the discussion, but it looks to me that there is majority opposition to wikidata in general. Perhaps I miscounted, or perhaps I had difficulty due to machine translation. Or perhaps confirmation bias influenced your estimate of consensus. Could you either (1) confirm that I am correct in counting a majority opposed on that issue, or (2) provide the list of names you view as !voting for/against to justify your estimate of consensus-so-far in favor on that issue? Alsee (talk) 23:34, 3 November 2017 (UTC)

@Alsee: Congrats on the effort! I know using Google Translate can be a pain, so good job! At this point, there are six editors who have clearly indicated they agree with the full use of Wikidata on Wikipedia in Portuguese; one editor who has identified himself as being neutral to the proposal; and six who have used the "against" sign in the discussion. A number of these editors who have used the "against" sign have done it for qualitatively different users: one editor has claimed to be against the use of the infoboxes if they are not adapted to providing versions in Portuguese from the several countries that are part of the project --there are important differences in how Portuguese is written in Portugal, Brazil, Angola, and the claim here is that Wikidata-integrated infoboxes should fully acknowledge these differences, which is actually a very good point I agree with--; one editor has claimed we should never use Wikidata-integrated infoboxes over content that is manually created by editors on Wikipedia, but as it is better than nothing we might want to use when there is no other infobox; and so on, until a couple of editors who have clearly and strongly identified themselves as being against the use of Wikidata-integrated infoboxes, relying on arguments that are similar to what you have seen on this community. I think it is also worth noticing that the link you have read is not a poll, but a discussion, so in order to get what is being collectively formulated there one definitely needs to get the nuances, I believe. It is from this qualitative reading that I said that though the consensus is not formalized yet it tends to be a sort of compromise and to indicate the use of Wikidata-integrated infoboxes under a specific circumstance. I am ping-ing an editor who has used the "against" sign to see how he is reading our discussion and perhaps he can provide some inputs without any risk of confirmation bias--I hope you don't mind, @DarwIn:. --Joalpe (talk) 00:45, 4 November 2017 (UTC)

Joalpe I agree a discussion isn't a poll, and I agree that nuances are important. However a "poll count" is a useful measure. It is unusual for a consensus to go against the majority, and there needs to be a very good reason for a consensus against the majority. Could you list the names you view on the two sides of "Wikidata-integrated infoboxes whenever an infobox is missing"? I would find that much more helpful than the description you gave. Even when I try to generously count in your favor, it looked to me like "no consensus with majority opposed". Alsee (talk) 12:45, 4 November 2017 (UTC)

First of all, I'm appalled by the impression Mike Peel has of our community, reducing it to "If English Wikipedia says it, it is probably right", but the fact that he basically does not participate there is surely not alien to that. I, for one, can certainly think by myself. I found this debate after jbribeiro1 mentioned it at the VP over there, and when I started reading the arguments, they actually answered/reinforced many questions I was asking for some time already to Wikidata involved folk, always having wishful thinking kind of replies such as "If we don't try, we don't know!" "Believe me, that's not a problem. And if it is, someone would eventually come out and fix it" - and so on. So I changed my initial openness about Wikidata experimenting at pt.wiki, to a much more wary approach, and started asking a lot of questions - a significant number of them, assumedly, inspired by this debate - and also doing some experiments on my own to test it. Some were OK, some failed terribly, canalizing a significant number of vandals from pt.wiki into Wikidata, where they could bypass the protection we had put in place, and thus vandalize the infobox at will for weeks. Shortly after I mentioned the case here, the Wikidata entry was permanently protected (something - I've read above - is problematic as well).

Anyway, I must say that I, jbribeiro1 and possibly a few others, we're indeed following this debate, and taking our own conclusions about it, and applying them to the pt.wiki reality. And if we are doing so, it's not because we want to blindly copycat whatever wiki.en says or decides. It's because THE debate is taking place here, and we clearly benefit from such an extended exchange of ideas and experiences. We at the Portuguese Wikipedia may be a community facing a number of severe problems and challenges, but we are not laboratory rats or gullible simpletons who would uncritically accept whatever the nice people from Wikidata are so kindly offering to us, to brighten up our dull little lives.

I must say, as well, that despite having now quite a dire image of the Wikidata project, and at least part of the community involved there, I still think it has a tremendous potential, if used correctly - if we are allowed to choose what to use and not to use of the stuff over there, as we generally do with our other sister projects - and I've actually debated that potential and possibilities in a very enthusiastic way with a number of people, as Joalpe knows very well. But I've not the least shred of doubt that the current approach of Wikidata, pesting sister projects and imposing their rotten or easily prone to rotting stuff, really is not the way to go, as it's impossible to miss the mammoth in the Wikidata room - its very well established incapacity to consistently provide reliable data, as well as to deal with vandalism. But I certainly hope all this we are discussing here now, and the result of the RfCs that will inevitably come out of it, can contribute to change that grim scenario.-- Darwin ^Ahoy! 06:05, 4 November 2017 (UTC)

The WMF has been pushing a promotional campaign. My own first impression of Wikidata was 'cautious but willing to give it a try'. Most people accept the initial promotional sales pitch that wikidata can/will do good things. Wikidata enthusiasts are invited to go ahead and experiment. However consensus increasingly turns against wikidata when people start dealing with wikidata in reality. The problems massively outweigh the benefits. People show up with torches to burn down the gross experiments. Half of the problem is the technology, foreign anti-wiki content invading wikipages. The other half of the problem is the grossly incompatible wikidata community itself. The wikidata community does not remotely share the interests, concerns, or priorities of the Wikipedia community. They are also grossly short of population to manage the content. Wikidata can exist as a separate wiki, but the more I deal with the technology and the more I see of the community, the more clear it is that wikidata needs to stay the hell out of articles. Alsee (talk) 11:54, 4 November 2017 (UTC)

What a load of nonsense Alsee. Every community event that I have visited over the past 3 years, has loads of people very enthousiastic about wikidata (and actually many of the other parts of our movement, wikisource, wikivoyage, glam and so much more). The Foundation is providing their service to that interest and that is fully within their merit and purpose. Just because sometimes they fail, or just because some people in this wiki seem to live inside a bubble where they think the current form of the English Wikipedia is the only form of an encyclopedia it should ever be and that stible any dissenting voices, is ill informed at best and malicious at worst. I know that I won't convince you and Fram of that, but I'm saying it anyways, for those 80% of 'silent' voices that I talked to today at WCNL 2017, who say they just turn around at any indication of an aggressive debate as we have on this page. —TheDJ (talk • contribs) 22:16, 4 November 2017 (UTC)

User:TheDJ Are you really not hearing the objections? That is a real question. Jytdog (talk) 01:02, 5 November 2017 (UTC)

I'm hearing the objections. I'm also hearing a lot of hyperbole however. And that irks me. I'm fully convinced that Wikidata on en.wp is totally tainted for the next 5 years (as so many things often are here), don't worry. It's that kind of overreaction that I vehemently object against however. And to say that the foundation is running a promotional campaign, while as far as I can tell, they only ever tried to use the Wikidata description (as a suitable and workable fix for a problem they encountered) seems a good example of such overreaction and is more closely tied to Alsee's personal feelings about the foundation than anything else from my point of view. Wikidata is a project that runs on a lot of community enthusiasm, and is facilitated by Wikimedia Germany. This enthusiasm is real and closely resembles the enthusiasm so many initially felt for Wikipedia. The foundation doesn't just cater to English Wikipedia and they never have. And for that reason I actually have my own HUGE objections against the foundation's actions where they triple the cost of every project simply to bend over backwards for this wiki. Instead, they should spend that effort on Commons, Wikidata, Wikisource, smaller wikipedia's etc, where i think the investment would create much higher returns. —TheDJ (talk • contribs) 10:52, 7 November 2017 (UTC)

@TheDJ: The huge advantages and potential of having a native and completely integrated database in the Wikimedia projects, as is the case of Wikidata, are self-evident. While I'm still not using it, I foresee a lot of great stuff I would be able to do by querying that database - Making lists of missing geographical subdivisions (and possibly populating the initial stubs) is one of them. This is one thing. Another entirely different thing, is seeing (what we later found to be) Wikidata-connected people coming into our project, presenting themselves as goodwill newbies, and starting doing their sales pitch, telling how they had found some fabulous thing that is going to change our lives, and they are so willing to help us that they are already implementing it on our project. Then when we start making questions, they suddenly turn aggressive and arrogant, treating us as the archetypal wild man that does not want to know fire because he fears it (I believe they may have even invented a new kind of logical fallacy, "appeal to luddism" - which you have just used yourself, BTW).

The Wikidata database is something huge and fabulous, yes - but it should be kept on its own space, and we get things from there at will. No effort should be put by Wikidata people into pesting sister projects forcing what have been proven to be very dangerous uses of that database. If we want it, we use it. And that's true for short descriptions, infoboxes, lists, whatever. Please, stop shoving Wikidata down our throats, that is only giving a terrible name to that project, and raising a lot of antibodies against it. -- Darwin ^Ahoy! 12:45, 5 November 2017 (UTC)

@DarwIn: Odd to talk to you in English, but here we go. You seem to be making two relevant points. Firstly, you indicate the "huge advantages and potential" of Wikidata, described by yourself as "something huge and fabulous". Secondly, you emphasize some circumstances that were not that good in the process of integrating Wikidata and Wikipedia in Portuguese, specifically: misunderstandings with the Wikidata community, perhaps suggesting discussions among communities should take place at higher standards, and limits to the integration, giving the leading role on how the integration Wikipedia-Wikidata would happen on Wikipedia to Wikipedians. The second point is strongly connected to your own experience with people in the Wikidata community --and I am sorry it happened the way it happened--, but let me stress despite this unfortunate bad experience I read what you are saying as a compromise standpoint. You are expressing this compromise in general terms; in my response above, that was focusing exclusively on Wikidata-integrated infoboxes I presented a more specific compromise that I believe is in accordance to what you just said (and let me know if you disagree). Anyway, what you just said --at least from my angle-- is what tends to be the direction in which the discussion on Wikidata is going on the community in Portuguese. --Joalpe (talk) 14:05, 5 November 2017 (UTC)

@Joalpe: It all boils down to two points: 1) Wikidata is unreliable; 2) Wikidata is actively pesting other projects. I can live with the first, as it still is very useful, even with the reliability problems. With the second, no.-- Darwin ^Ahoy! 18:16, 5 November 2017 (UTC)

"Wikidata-connected people coming into our project" "If we want it, we use it." And there's one of my biggest problems. Some pretend that these people are totally seperate groups of people. There not. There is huge overlap and interaction and even more so between their various 'non-editor' usergroups. But instead of recognizing this, we pretend like they are separate and even that that is a good thing. It's not that simple. —TheDJ (talk • contribs) 10:52, 7 November 2017 (UTC)

TheDJ, isn't it rather ironic to accuse us of living in a bubble, and to use as evidence people you meet at WCNL, which I presume is some Wikimedia event in the Netherlands? People who attend such events are not representative of the 80% of silent voices but are a specific subgroup of people who are much more likely to be enthusiastic about WMF initiatives and the like. I think I found the event you are discussing, and I note among the speakers multiple Wikidata enthusiasts like the reliable, productive and friendly art lover Jane023, and the (no description as no personal attacks are allowed) GerardM, and a discussion specifically about Wikidata with statements like "Viaf is switching from Wikipedia to Wikidata. YOU SHOULD DO THAT TOO!"[3], and whole discussions abotu data quality which are solely about internal consistency and not at all about sourcing, verifiability, that kind of stuff. I really don't think that the people you meet at that bubble are really representative either. Fram (talk) 11:41, 6 November 2017 (UTC)

You mean that being in multiple bubbles (en.wp, MediaWiki, event attendee, nl.wp, wikipedia weekly, strategy group) is less representative than being in one bubble ? Maybe we should wonder why conferences are talking more about wikidata than about wikipedia and include things like "move away from Wikipedia", instead of questioning the relevance of the people attending the event. —TheDJ (talk • contribs) 10:52, 7 November 2017 (UTC)

Yes, being in those bubbles of often non-enwiki editors is hardly representative of how some silent majority (if such a thing exists here) at enwiki may think. Nlwiki has a completely different culture in many ways, so it wouldn't surprise me if they feel differently about e.g. Wikidata as well. Why are they talking more about Wikidata? Because it is new, and many people love "new". Part of the appeal of editing Wikipedia is that you were building something new, and that aspect has waned for many people. Moving over to Wikidata may again give them that buzz. Which is a good reason for those editors to move over to Wikidata, but is not a good reason for content to switch over to there. Fram (talk) 11:05, 7 November 2017 (UTC)

Thank you for providing this insight and confirmation of your point of view. I think we have now clearly established that I'm of an opinion that is differing from yours and that "i'm not going to convince [you]". :) —TheDJ (talk • contribs) 12:39, 7 November 2017 (UTC)

Hint: if some people say "we don't want this" and provide a list of reasons (e.g. unreliability, different requirements, hard to watch, ...) and your reply is "I talked to a lot of people who loved it", your chances of convincing anyone are very, very slim. Argumentum ad populum is considered a fallacious argument for good reasons. If you can't or don't want to address the actual concerns, you are rarely going to convince anyone this way. That's fine, and your conclusion that you are not going to convince me is based on this technically correct, but I hope you or others are not then going to use it as fodder for the "Fram is too stubborn and can never be convinced" meme which seems to be spreading among some people. I can be convinced, but not by fallacious arguments (or at least those I recognise of course). Fram (talk) 12:55, 7 November 2017 (UTC)

If you want a different slice of the "silent 80%", take a look at Wikipedia:Administrators' noticeboard/Incidents#Wikidata spam campaign, where some other uninvolved people experienced editors all of them) note surprise, dismay, and a generally negative view of the use of Wikidata on enwiki. Now, this slice isn't any more conclusive or definitive than another slice, but it is a good indication that the pushback against the use of Wikidata here isn't limited to a few vocal people on these pages. Fram (talk) 08:16, 10 November 2017 (UTC)

I'm late to the party and most participants here already know about this I guess, but in case the link gets archived from ANI: see also Wikipedia:Arbitration/Requests/Case#Crosswiki issues (but don't contribute there unless you are already or want to become involved). —David Eppstein (talk) 01:49, 11 November 2017 (UTC)

More damage to en-WP on mobile

We just had this report at ANI, noting vandalism persisting in en-WP on mobile related to Gastroenteritis since Nov 10 on a medical topic. The vandal replaced the short description in Wikidata with fucken butt mation of the gastrointestinal tract that involves the stomach and/or the small intestine. It was fixed 10 minutes later, in this diff. I don't know why it persisted somewhere this long.

WMF reading team that is your responsibility. Yours. Jytdog (talk) 15:17, 18 November 2017 (UTC)

This was pretty obvious vandalism. The upside it that it was pretty obviously vandalism, so unlikely to have endangered anyone's health. The downside is that obvious vandalism on a medical topic persisted so long. · · · Peter (Southwood) ^(talk): 16:41, 18 November 2017 (UTC)

It appears to have lasted the whole of 10 minutes before being reverted on Wikidata (i.e., not that long). So presumably this was a caching issue somewhere - not sure if that's in the mobile app or elsewhere. Thanks. Mike Peel (talk) 18:24, 18 November 2017 (UTC)

10 minutes is not so long, agreed, but almost all of the people who would fix a thing like that on Wikipedia did not even know that it had happened, because they were not able to see it. If they had seen it, a large number would not know how to fix it, as they know how to edit Wikipedia, and it was not on Wikipedia, it just looked like it was. Even if they had seen it, tried to fix it on Wikipedia, found that it wasnt there, spent the effort to find where it actually was, went to Wikidata and fixed it, it seems that it would not have helped, as you say it had already been fixed but was displaying the old vandalised version, and no indication of why and how it could be remedied. Not actually any better. · · · Peter (Southwood) ^(talk): 09:27, 20 November 2017 (UTC)