User talk:RexxS/Archive 31

Wikidata & infoboxes
Hi -- I saw your comment to Curly Turkey about withdrawing from the discussion and from the development of the related infoboxes. Your decision, of course, but I would be very sorry if you do withdraw -- you and Izno have been extremely helpful in making the situation clearer, to me at least. The work you've done has been extremely valuable. People's minds can be changed by rational conversation; it doesn't happen all that often, but it does happen on Wikipedia. Sometimes. The two of you changed my mind, after all. Either way, thanks again for the work you've done so far. Mike Christie (talk - contribs - library) 00:00, 21 June 2016 (UTC)
 * Agreed; don't give up hope yet! It's good to have you around, since you both confirm and demonstrate some of the things that come out of fingers on the keyboard w.r.t. Wikidata (besides the occasional splat). --Izno (talk) 00:22, 21 June 2016 (UTC)


 * RexxS, I agree that it would be a great pity if you were to withdraw from the discussion, though of course I'll respect your decision. But please know how helpful and informative your posts have been. I'm very grateful for the patience you showed toward me, and for the extra time you had to spend, explaining things to me several times and negotiating your way through my confused vocabulary. I've learned a lot from you. SarahSV (talk) 18:38, 21 June 2016 (UTC)
 * Hi RexxS, some might be surprised with the fact that you and I are friends around here, given that we are on opposite sides when it comes to the dreaded infobox debate, but I won't ever forget your help here, and here - both of which, thanks to your advice, went together with this to make my first and only featured topic. I'm not familiar with the discussion you have been involved in, but your technical knowledge, as illustrated in my diffs above, is absolutley second to none.  For you then not to share this kind of knowledge would be a loss to us all.   Cassianto Talk   09:48, 22 June 2016 (UTC)
 * Hang in there! We need intelligent contributions from all sides. I'm still on the fence, but have been putting some work in at Wikidata on some communities from Anglesey, and just discovered QuickStatements, which has been a godsend. Robevans123 (talk) 13:31, 24 June 2016 (UTC)

Template to produce a citation from Wikidata
RexxS, I'm not sure if this is the sort of thing you'd be interested in, but I know you have the necessary skills. Do you think it would be possible to write a template that takes a statement in Wikidata and turns it into a citation? For example, Weird Tales has a statement asserting that Farnsworth Wright was an editor. That statement is referenced by a book, and a page number. The two references between them have all the necessary information to reconstruct this string: "Ashley, Mike (2000). The Time Machines: The Story of the Science-Fiction Pulp Magazines from the beginning to 1950. Liverpool: Liverpool University Press. p. 42. ISBN 0-85323-865-0". I was thinking that something like would be how the call would look: first parameter is the object about which the statement is made; the second parameter is the property in the statement, and the third is the value of the property for which the reference should be retrieved and converted. Perhaps something like this exists already and I don't know about it, but if not, do you think it's feasible, and would you be interested in building something like this, or do you know where I might ask? I think something like this would make the natural integration between Wikidata and en-wiki much more apparent to many editors. Mike Christie (talk - contribs - library) 00:28, 24 June 2016 (UTC)
 * I'm tempted to suggest you ask Curley Turkey to take on the job, as he seems to have a lot to say about the issues. But that wouldn't be productive, so let me say this: the template that you suggest would not be difficult to create, and the code already exists for each part you're interested in. However, you can't use a template like that in an infobox; although the QID of the article is known (you don't need to supply it - it will work it out from the page it is on), and the property will be determined by the infobox field you're using it in; but there's no way you can supply the QID for the reference. The reference may be a book or a url or any number of other things, If you have to look them up on Wikidata first before you can fill in the infobox details, you've defeated the object of using Wikidata, and it's probably quicker to just fill in a reference by hand and forget about Wikidata.
 * What you need to do is supply just a property corresponding to the infobox field; retrieve the value for that field and while doing that, check if there are any valid references; if there are, retrieve them, whatever they are, and then figure out how to construct a citation that you can return along with the value. The code to do a lot of that is in the getSourcedValue code in Module:WikidataIB, where I've already fetched the raw references to use in debugging. What doesn't help is that each property may have multiple values (e.g. two or more editors of a book), and each value may have multiple references. The code for that is in place, but you need to recognise that you may get quite a few things returned from just one call, and you have to manage the space in an infobox. To adapt getSourcedValue to do the job you want just needs the bit that constructs a citation - but that's quite messy and is different for each type of reference.
 * Right now, I can't summon up any enthusiasm for solving any more problems to do with Wikidata in infoboxes. I'm very grateful for the kind words from yourself and the folks above, but I'd rather wait until the destructive forces have left the scene before I consider returning to the area. --RexxS (talk) 01:17, 24 June 2016 (UTC)
 * I thought the new Infobox book was a splendid job—as I said, it was pretty much one of my proposals, and the direct links from each field was a nice touch. It's unfortunate that it's come to this. Curly Turkey 🍁 ¡gobble! 01:42, 24 June 2016 (UTC)
 * Of course the QID of the article is available; hadn't even thought about that. However, I wasn't particularly thinking about using this in infoboxes; I was thinking of using it in prose, whereever a citation might be used. I can see that means that the property isn't available any more, since you have no infobox field to tie it to, so you'd have to specify both sides of the statement the ref supports.  Even though this means you'd have to look up these items in Wikidata, I see two benefits from citing facts this way.  First, work done by en-wiki directly benefits other language wikis, and vice versa, since now there is a central repository for cited information about the article; and second, if the same fact is cited in multiple places in en-wiki, and the reference is improved (e.g. to a more reliable source), then the improvement appears in all the articles using that citation immediately.
 * I'm glad to hear it's doable. If you don't mind, I'll stop by in a couple of months and see if this is something you're interesting in tackling at that time?  Or do you know of another expert in templates who might be willing to try putting something like this together?  Thanks -- Mike Christie (talk - contribs -  library) 08:04, 24 June 2016 (UTC)
 * The reason I talked about using it in an infobox is that the enabling RfC Requests for comment/Wikidata Phase 2 showed community consensus for using Wikidata in infoboxes (but always subsidiary to local values), but also show consensus that explicitly forbids the use of Wikidata in article text. So until there emerges a new consensus, I'm afraid you can't use Wikidata in the way you want to. If you want to talk with other coders, I'm pretty sure and  have the skills and are familiar enough with the problems to find good solutions for you. Cheers --RexxS (talk) 10:44, 24 June 2016 (UTC)
 * I should probably get around to muddling my way through Lua to build something like this (RexxS: No, I don't have the skills--a handful of in-person and online courses to my name--but I should :D). Probably the format it would take would be something like, and in the programming of what would like be a module, supply the QID statements to Module:Citation/CS1 for subsequent output. (There would be a small issue of which I know already due to Wikidata's use of only a single "page" property rather than a split "page" and "pages" property, but that would be something to work on down the road.) That said, I think there would need to be some strong consensus for it besides the conclusion of the phase 2 RFC that RexxS points out, due to the deletion of the cite doi templates, which used a similar, if not precisely the same, mechanism. --Izno (talk) 11:46, 24 June 2016 (UTC)
 * (ec)Well, yes, that does settle things, at least for now; I didn't participate in that RfC and wasn't aware of that limitation. I might start a conversation at the Village Pump asking for opinions on using Wikidata in article text but strictly for citations, to see if anyone else thinks this would be worthwhile.  (post ec): I see from the cite doi RfC that the issue is at least partly that "templates should not contain article content".  That doesn't quite make sense to me, since templates have always contained citation data.  Izno, do you think a conversation at VPP is worth starting?  Or do you think this is something that the community has already shown a strong disinclination for? Mike Christie (talk - contribs -  library) 11:52, 24 June 2016 (UTC)
 * I think right now we should let the infobox stuff blow over before starting a discussion on citations, since there does already exist a consensus against having "separate" citations from the article content--there were other issues raised, which are just as pertinent for Wikidata as they are for cite doi, such as usability (how do you edit a Wikidata citation? how does a user know to go there?--this can be solved) and vandalism (which Wikidata doesn't have a good record at this time dealing with). --Izno (talk) 12:22, 24 June 2016 (UTC)
 * Fair enough. Maybe in a few months it'll be a worthwhile question to ask.  I was thinking about how one would edit a Wikidata citation; I think it would be fairly simple to have VE's citation menu expand to include the possibility.  Having the menu go from Cite -> Cite Wikidata -> List of items linked to from this article -> list of statements for selected item in Wikidata -> list of references for selected statement would seem to be one way to do it.  VE won't implement anything that's specific to one wiki, so there would have to be a handoff of the selected reference to local code to do the formatting.  If the resulting text were to be inserted as a string (substed, essentially) then some of the objections would vanish, though so would some of the benefits. Mike Christie (talk - contribs -  library) 12:34, 24 June 2016 (UTC)
 * Well, once you're at the point of substing it, might as well just use Citation bot... which actually might be a fair path as well to using the base data in a Wikidata citation. --Izno (talk) 12:41, 24 June 2016 (UTC)
 * I've never tried Citation bot; just tried it and read some documentation. I see what you mean; if it already understands citation structure then why rewrite it?
 * The base functional requirement for what I'm describing above seems to be "allow an editor to see Wikidata when adding a citation, and pick up citation information from references on relevant statements". If so, and thinking about Citation bot, how about having a parameter or parameters in ordinary citations that record the selected statement/reference pair?  E.g. "|QID=item|PID=property|RID=reference identifier" as part of the citation template text.  This wouldn't output anything; it's just a record of how the citation was generated. So adding a citation by this method would create a completely ordinary citation, plus the Wikidata information necessary to rebuild it.  (Denormalized data, I know, but bear with me.)  If the Wikidata reference is updated -- say to substitute another source, or just to refine the page range -- then the next time you use Citation bot it would tell you that the Wikidata reference doesn't match, and ask you if you want to delete the WD data in the citation fields, or update the citation to match, or leave it inconsistent (which would probably do something like place the article in a tracking category).  Ideally it would give a link to jump to the relevant WD page so you can fix the citation information there, if you want to.  Mike Christie (talk - contribs -  library) 14:01, 24 June 2016 (UTC)
 * I'll drop that in the category of "possible" and will not offer further opinion. :^) --Izno (talk) 14:33, 24 June 2016 (UTC)
 * I'll drop that in the category of "possible" and will not offer further opinion. :^) --Izno (talk) 14:33, 24 June 2016 (UTC)

MfD nomination of Wikipedia:Randy in space
Wikipedia:Randy in space, a page which you created or substantially contributed to, has been nominated for deletion. Your opinions on the matter are welcome; you may participate in the discussion by adding your comments at Wikipedia:Miscellany for deletion/Wikipedia:Randy in space and please be sure to sign your comments with four tildes ( ~ ). You are free to edit the content of Wikipedia:Randy in space during the discussion but should not remove the miscellany for deletion template from the top of the page; such a removal will not end the deletion discussion. Thank you. Maproom (talk) 14:05, 25 June 2016 (UTC)

Wikidata Infobox Tutorial
Hey! I just started out working on https://phabricator.wikimedia.org/T76229 here: d:Wikidata:Infobox Tutorial. I am asking around for people to share their experience with migrating infoboxes to Wikidata. The goals is to make a really accessible guide for people that want to build a new infobox, but also for people to understand the benefits of these changes. --Tobias1984 (talk) 13:32, 30 June 2016 (UTC)
 * Good luck with the tutorial! I've commented on Phabricator and added a section on d:Wikidata talk:Infobox Tutorial. Ping me here if you want to discuss any specifics. Cheers --RexxS (talk) 16:56, 30 June 2016 (UTC)
 * Good luck with the tutorial! I've commented on Phabricator and added a section on d:Wikidata talk:Infobox Tutorial. Ping me here if you want to discuss any specifics. Cheers --RexxS (talk) 16:56, 30 June 2016 (UTC)

Thank you little user
Thank you kind little user for fine fix of Zilla banner celebrating Xkcd and wonders of science! Good grassy colour, too! bishzilla   ROA R R! ! 17:27, 8 July 2016 (UTC).

10 July
Took only 300 years to restore a good name. - Thank you for your work on the article. --Gerda Arendt (talk) 06:26, 10 July 2016 (UTC)

Jane Austen
Hi RexxS, I would have been happy to help sort out the cites per MLA, but I don't want to help if I am to be challenged over every single point. The biggest problem is that the cites were copied over without their corresponding bibliographic entries, but sorting them out wouldn't have been that difficult and we really should be using MLA style. I've don't want to be involved in a war, so I'll leave you all to it now. The MLA manual isn't available online, but there are other sites that provide some insight. The most important thing is that authors are matched to the text they are reffing: i.,e citing a book with publishers isn't sufficient. A separate entry needs to be made for each chapter. I've done this in a number of literature articles I've brought to FA (as did Awadewit) so I'm sure you know where to look for inspiration. Victoria (tk) 20:00, 17 August 2016 (UTC)
 * Adding: all the information is here in this version from 2013. Note: in some instances the cites were bundled. If the text subsequently shifted, then that's a problem. But Awadewit's style really is consistent, can be brought up to date and there's no need to reinvent the wheel. Victoria (tk) 20:08, 17 August 2016 (UTC)
 * Thank you for the help you've given, . I agree completely that the reader is best served by each chapter having an individual full citation in the 'Bibliography' (as it's titled in Jane Austen). I had to do exactly that when I took Oxygen toxicity to FA where I found it best to use shortened footnotes for the books and cite journal for the medical articles. Hopefully we'll have the chance to improve the references further as we fix the concerns raised at the FAC. the problem has been that several editors as well as Adrianne have contributed to the article over several years, each adding citations manually in slightly different formats - and none coping well with the issue of collected essays. Nevertheless, has made substantial contributions to the article this year (now top editor by bytes added), and is keen to retain as much of the flavour of the earlier contributors' efforts as he or she can. He/she is quite rightly the sole nominator and for my part, I simply helped resolve some image copyrights before tweaking some alt text on the images. I would much rather help out the remaining principal author improve this article up to FA standard than argue the niceties of MLA format, but I don't feel that the opposes are fair on Fountains-of-Paris and I'll continue to make the case for the promotion of Jane Austen. Let me ask you, if the current article were brought to FA without any history, would you still be opposing its promotion? --RexxS (talk) 20:35, 17 August 2016 (UTC)
 * Not sure what you mean by history? I.e., that Awadewit wrote it? Assuming yes, I've had the FAC on watch since it was added (the article on watch since I started editing) and meant to post. Yes, unfortunately, one reason I hadn't yet posted there is that I don't believe it's yet FA quality. The issue with the citations needs to be sorted, but there are deeper issues (which have contributed to the citations being muddled). None of this is ever fair and unfortunately changing the citation style brought it to the forefront and badgering reviewers isn't right either. Especially when trying to help. Anyway, FA literature articles aren't difficult to find and if memory serves there aren't a lot them but I prefer not to go down this path any further before looking closely at the article (which I've not had time for and the reason I hadn't posted yet to the FAC). This week I'm supposed to be doing Other Stuff (ie. not embroiled in Wikipedia). Re MLA, since two people have suggested MLA style and it's what we routinely use in literature articles, there's really no reason not to, imo. Victoria (tk) 21:00, 17 August 2016 (UTC)
 * I don't want to appear callous about this, but Awadewit added 71 Kbytes of text in developing the article, which was six years old when she started editing it. has added 62 Kbytes of text to the article and has shepherded it through GA and peer review. If you believe it's not FA quality, then I think it only fair to give specifics to Fountains-of-Paris so that he/she can address your concerns. Re MLA: Geogre's Ormulum is an example of a literature Featured Article that is close to MLA style. And if you read its Featured article review, you'll see me defending Geogre's style of referencing because it was consistent. There is no such consideration for Jane Austen because it's never had a consistent style until two days ago - you only need look at Linghzi's comments for at least a dozen examples. The present version is a vast improvement. Why would two people's suggestion of adopting MLA format (that the article has never had) take precedence over five people who have agreed to a scheme that removes the inconsistencies while retaining a display similar to the earlier versions? --RexxS (talk) 21:28, 17 August 2016 (UTC)
 * RexxS I don't wish to continue this conversation. I had it on watch because I meant to comment. Now you've backed me in a corner and you've challenged my knowledge about citation formatting in an area where I've contributed (somewhat) to Wikipedia. I don't wish to continue a discussion if my opinion is only to be challenged and I must continually defend myself. Victoria (tk) 21:35, 17 August 2016 (UTC)
 * Victoria, you've backed yourself into a corner, and I'm sorry that you've done that to yourself. But you really mustn't assume that I must know nothing about citation formatting. It is one of my areas of technical expertise and I've been able to back up what I say about MLA with hard evidence, unlike your appeal to authority. I understand that you don't like citation templates, but you're going to have to be honest with yourself and accept that your personal preference is not a good reason to deny promotion for a perfectly good candidate for Featured Article. --RexxS (talk) 21:48, 17 August 2016 (UTC)
 * Your badgering and being patronizing. Please stop it right now. Victoria (tk) 21:50, 17 August 2016 (UTC)
 * And you're making a personal attack. I didn't ask you to come here and lecture me. You seem incapable of replying to a reasoned discussion; you have no idea what MLA citations look like; and you've no idea of English grammar if "Your badgering and being patronizing" is an example of your writing style. If you don't like the reception you get here, there's an easy solution - and I don't mean yet another insincere GoodBye. Clear enough? --RexxS (talk) 22:32, 17 August 2016 (UTC)
 * And you're making a personal attack. I didn't ask you to come here and lecture me. You seem incapable of replying to a reasoned discussion; you have no idea what MLA citations look like; and you've no idea of English grammar if "Your badgering and being patronizing" is an example of your writing style. If you don't like the reception you get here, there's an easy solution - and I don't mean yet another insincere GoodBye. Clear enough? --RexxS (talk) 22:32, 17 August 2016 (UTC)

Is there any documentation for creating wikidata fed infoboxes?
Hi

I'm trying to work out how to create Wikidata fed infobox template but I can't find any documentation, do you know if any exists? I'm starting to learn a bit about templates but its still very much baby steps so I can't work out from looking at an existing template.

Thanks

--John Cummings (talk) 21:41, 15 July 2016 (UTC)


 * Hi, if you can't work things out from looking at an existing template, then I'm not sure the documentation, e.g. at Module:Wikidata/doc will be any help, but you can have a look and see. Otherwise you'll see three sections above that Tobias1984 has been making a tutorial at d:Wikidata:Infobox Tutorial. If that fails, the best thing is to Skype me when you have an hour to spare - you have my details - and I'll talk you through some techniques. Cheers --RexxS (talk) 21:52, 15 July 2016 (UTC)


 * Thanks very much, I'll try my best and hopefully I can help improve the instructions. John Cummings (talk) 01:37, 16 July 2016 (UTC)

Hyperbaric Medicine Reply
Something about that lead section "didn't / doesn't feel quite right" but I wasn't (and I'm still not) 100% sure what's wrong with it.

I will try to be more careful in the future. (I had no idea that "Tag Bombing" was disruptive...)

Darklight Shadows 20:27, 20 July 2016 (UTC)
 * Thanks for your understanding. You're right - the lead of Hyperbaric Medicine isn't quite right because the article itself isn't really stable, so a summary is difficult. I mean there are quite a few "fringey" or "off-label" uses of HBOT, and the elephant in the room is the belief that it helps children with autism, for which no solid evidence exists. That's why there's no structure to the article; it grew from some very unreliable beginnings and really needs a re-write from scratch using the best WP:MEDRS sources. That will be a big job.
 * Thank you for prodding me into re-writing some of the sections to turn a few of the lists into prose. I hope if you look again at the article, you'll find it improved a little. Cheers --RexxS (talk) 21:14, 20 July 2016 (UTC)
 * Thank you for prodding me into re-writing some of the sections to turn a few of the lists into prose. I hope if you look again at the article, you'll find it improved a little. Cheers --RexxS (talk) 21:14, 20 July 2016 (UTC)

Virtual Synchrony
Virtual synchrony has a TON of problems. Do you think it can be saved or is it better off being deleted.

Darklight Shadows 01:50, 4 August 2016 (UTC)


 * , I suggest that you tag the specific problems in the article where they occur, and discuss them in detail on the talk page. That way the article's editors can see exactly what you perceive as the problems, and will be able to address them. Also, any visitor to the page will know where and what the problems are, and will be encouraged to help if they are able. If you think it should be deleted, you should give specific and valid reasons in more detail than "a TON of problems" The subject at first glance appears to be sufficiently notable, it is not a controversial BLP, there is a lot of information, however badly it may be formatted. What would the reason for deletion be? &bull; &bull; &bull; Peter (Southwood) (talk): 09:01, 4 August 2016 (UTC)


 * The most important lesson I learned in Computing II was "knowing when I've spent enough time on an assignment." (We were allowed to turn in a write-up about the problems and solutions for the that assignment for full credit)

I looked closely at how to best tag the article. However, from my perspective, it seems like trying to fix it /could/ eventually end up going nowhere. I might be wrong though.

Darklight Shadows 14:04, 4 August 2016 (UTC)
 * Having read the talk page, I agree that there have been complaints for years without any resolution.
 * To improve the article, I would recommend concentrating on the one aspect that is most in need of improvement - in this case the real problem is lack of in-line citations, so that it makes the job of anyone trying to verify the content much more difficult. The first step, IMHO, would be to read through the body of the article (not the lead) and tag statements that are bald claims with citation needed. If parts are unclear, then tag them with clarify.
 * I've now placed the connected contributor template on the talk page to set the context of Ken Birman's contributions, and made clear my position in the talk page section Talk:Virtual synchrony, along with the policy I'm relying on for the clean-up. If you're interested in detailed clean-up, I've done the first section, Virtual synchrony , and you may wish to try some of the other sections yourself. Cheers --RexxS (talk) 16:42, 4 August 2016 (UTC)
 * I've now placed the connected contributor template on the talk page to set the context of Ken Birman's contributions, and made clear my position in the talk page section Talk:Virtual synchrony, along with the policy I'm relying on for the clean-up. If you're interested in detailed clean-up, I've done the first section, Virtual synchrony , and you may wish to try some of the other sections yourself. Cheers --RexxS (talk) 16:42, 4 August 2016 (UTC)

Wikidata thoughts
Your talk page edit notice hypnosis is very annoying! It almost made me forget what I came here to talk about. :-)

I hope the following won't be too rambling or too long, but I'd appreciate some thoughts on this from someone willing to discuss this.

On Andy's talk page, I asked this question. The key bit is this: "Why can't the Wikipedia editing page actually display the Wikidata value being pulled over, so an editor knows what is there?"

More general thoughts. I can see the value of Wikidata when working with massive databases. The Commonwealth War Graves Commission is actually a good example of such a database (though there are plenty of others as well). Have a look at Commonwealth War Graves Commission for an overview of the topic if needed. There are two main templates that call on CWGC ids, both set up a long time ago: (i) CWGC for casualties in their database; and (ii) CWGC cemetery. The latter has an associated Wikidata property (CWGC burial ground ID (d:property:P1920)), but the casualties one has not(?) been set up properly yet (the sidebar link takes me to Q7745463).

There are some 1.7 million casualties in the database (the vast majority will never be needed for anything other than tagging photos taken of the gravestones; it was total luck that I noticed that we had a photo of a particular gravestone on Commons and I added it here), some of which may warrant mention on Wikipedia pages. A couple of examples from my editing are: ; ; ; ; ; and so on). I wish I had been more consistent in the use of my edit summaries so I could find all the examples and replace them with the template. According to the external links special page, there are 1,271 uses of that link with that URL and 1,708 using the old URL format (total of 2,978), so with only about 86 transclusions of the casualty template, there is the potential there to use the template a lot more. The uses are in lists (of Victoria Cross people for example) in articles on actual people notable enough to have an article (or for example on lists of generals killed in WWI), and finally a probably much larger subset that could be used on the articles of those people who lost children or other relatives in the wars (that is mainly what my examples above are).

There are over 23,000 burial sites, though many of these sites are existing cemeteries in the UK and other countries. The number of purpose-built cemeteries built worldwide is about 2,500. Most are unlikely to have articles, but lists would be useful. Most of my work has been with the memorials (though I am gradually becoming more familiar with the cemeteries). The list articles are in the template I linked above.

One thing that would be incredibly useful is a way to update the casualty commemoration figures when that is updated each year by the CWGC. The numbers change as new information comes to light and reburials are made and so on. This list is particularly out of date. It would be really useful to be able to use Wikidata in some way, but I suspect that in most cases downloading the data from the CWGC site and sorting through it is easier than trying to do the same through Wikidata. The key seems to be to use Wikidata to call on the most useful elements of the data.

Photos are another area. There are thousands and thousands of photos on Commons (mostly in Commons:Category:Commonwealth War Graves Commission cemeteries; see also Commons:Category:Commonwealth War Graves Commission gravestones and Commons:Category:Monuments and memorials of the Commonwealth War Graves Commission). I am part-way through a rather big project (not sure how useful it will be) that has ended up with a gallery of a particular architectural feature (the Cross of Sacrifice) over at Commons:Crosses of Sacrifice. There are currently around 300 photos there (I estimate there are well over 2000 Crosses in total). What I would love is to be able to associate each photo with its cemetery, and to pull data like when the cemetery was built (i.e. identify the post-WWII cemeteries).

That's probably enough for now. I think one thing that probably deep down annoys editors who work with information from large databases like this, is the feeling that their knowledge of and use of such databases is in someway threatened by machine-like mining of the databases, generating instantly lists that they might have laboured for many hours to build (with great care). Sure, it is good that it is possible to generate such things quicker, but careful checking is still needed (and the learning of new skills it seems). I would hate to see less technically proficient editors being left behind or feeling forced to learn how to use Wikidata when what they really want to do is integrate and present the information in useful ways (or be told how to do that, by people who understand how to explain things).

And I am still very queasy about anything that will make citations more opaque. I actually think that whenever a citation is made, it should be preserved in the article history in full as it looked at the time (with all data calls output in full). The risk is too great that remote manipulation of citations from another site (Wikidata) will mess things up and those who know the subject area will be left with a mess to clean up. Carcharoth (talk) 00:12, 8 August 2016 (UTC)


 * RexxS: Three's a related discussion on my talk page, please feel free to reply there, to avoid a fragmented discussion. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 11:19, 8 August 2016 (UTC)

Henri Fouquet
Hi RexxS,

Some days ago I created this page from the French fr:Henri Fouquet (médecin) ; unfortunately I made a mistake when I wanted to link the new English article to its French model and I don't know how to repair. I know there's a place to ask for help but really it's too complicated for me to reach the right page for that. Your intervention would be much appreciated. Thanks in advance to take the time to solve the matter. LouisAlain (talk) 09:51, 12 August 2016 (UTC)

Thanks, LouisAlain (talk) 12:59, 12 August 2016 (UTC)
 * Hi Louis, I looked at the template on the Talk:Henri Fouquet talk page and it looks fine to me. The only thing I can see is that you linked to a version of the French page dated 18 March. Was that the mistake you mean? Would you like me to change that to the version of 9 August, perhaps? --RexxS (talk) 15:39, 12 August 2016 (UTC)
 * The problem wasn't with the talk page but with the interwiki links. I first linked to the French disambiguation page Henri Fouquet whereas it was meant to lead to Henri Fouquet (médecin). It has now been fixed, probably by someone I asked some days ago but since there was no answer I thought the user must have been on wikislow, or on vacation and fear I would forget this tiny thing, I came to you. Now, I have to remember who was the user I addressed to in the first place to thank him. That person intervened on my TP some months ago. Hmmm... shall I retrieve his name? Sorry I made you lose your time RexxS ; LouisAlain (talk) 16:42, 12 August 2016 (UTC)
 * I don't think that any time was wasted, . Interwiki links are now stored on Wikidata and after your post, Andy (User:Pigsonthewing), who often keeps an eye on this talk page, merged the two Wikidata entries on and tidied up the interwiki links, so all should be working well. That's probably not coincidence! Cheers --RexxS (talk) 17:53, 12 August 2016 (UTC)
 * I don't think that any time was wasted, . Interwiki links are now stored on Wikidata and after your post, Andy (User:Pigsonthewing), who often keeps an eye on this talk page, merged the two Wikidata entries on and tidied up the interwiki links, so all should be working well. That's probably not coincidence! Cheers --RexxS (talk) 17:53, 12 August 2016 (UTC)
 * I don't think that any time was wasted, . Interwiki links are now stored on Wikidata and after your post, Andy (User:Pigsonthewing), who often keeps an eye on this talk page, merged the two Wikidata entries on and tidied up the interwiki links, so all should be working well. That's probably not coincidence! Cheers --RexxS (talk) 17:53, 12 August 2016 (UTC)

Glad to learn this option exists: it will spare me the embarrassment to knock at some random doors to beg for help LouisAlain (talk) 19:44, 12 August 2016 (UTC)

Suggestion at the ANI talk
RexxS: Robert McClenon has made a point that your use of the word "lie" was too strong. I have pointed out to him that he was in error when he said you did not try to discuss the decline and he has acknowledged that error but would like you to retract the word "lie" as it obviously offends him. See this edit. Would you be willing to apologize for using the word "lie" and soften it to "mistake" in the interest of smoothing ruffled feathers? I think it would be a good idea. Koala Tea Of Mercy ( KTOM's Articulations &amp; Invigilations ) 15:07, 13 August 2016 (UTC)
 * Thanks for your concern, . McClenon's "mistake" wasn't just once, but twice:
 * The original poster didn't even try to discuss the decline with LaMona. Robert McClenon 03:06, 11 August 2016 (UTC)
 * As noted above, there wasn't any effort to discuss with her prior to coming here. Robert McClenon 13:20, 11 August 2016 (UTC)
 * Nevertheless, I'd be happy to retract the word "lie", as soon as McClenon acknowledges that he was "mistaken" in implying that I hadn't previously raised the issue with LaMona, and/or that he hadn't read my original post, which made it clear I had. --RexxS (talk) 15:54, 13 August 2016 (UTC)
 * Nevertheless, I'd be happy to retract the word "lie", as soon as McClenon acknowledges that he was "mistaken" in implying that I hadn't previously raised the issue with LaMona, and/or that he hadn't read my original post, which made it clear I had. --RexxS (talk) 15:54, 13 August 2016 (UTC)

Hello RexxS
If I may be bold here, may I ask that you help me to ensure that if ever we edit from opposing camps, as apparently can happen, that disagreement does not become disassociation; that would be a terrible end by any means. I hope you will always remember that I hold you in very high esteem. I have told you so in the past. I tell you so now, regretting that we have hit on some disagreement. Does this sound "fair enough"? I will abide. Best.--John Cline (talk) 21:58, 20 August 2016 (UTC)
 * , I apologise. Thank you for your gracious admonishment. I've been too invested in the problem (as I see it) of hidden text forbidding something that hasn't been discussed. That has led me a merry dance over multiple venues and worn my patience too thin for my own comfort. Please rest assured that I do appreciate your work on Wikipedia, and when I become snarky in debate, it's not my intention to be personal; it's because I foolishly allow myself to become too combative in discussions, often because I genuinely can't believe the positions sometimes taken by others. As an olive branch, I've self-reverted my last comment. I agree it was not the sort of posting that on reflection I would be proud of. Cheers --RexxS (talk) 22:20, 20 August 2016 (UTC)
 * , I apologise. Thank you for your gracious admonishment. I've been too invested in the problem (as I see it) of hidden text forbidding something that hasn't been discussed. That has led me a merry dance over multiple venues and worn my patience too thin for my own comfort. Please rest assured that I do appreciate your work on Wikipedia, and when I become snarky in debate, it's not my intention to be personal; it's because I foolishly allow myself to become too combative in discussions, often because I genuinely can't believe the positions sometimes taken by others. As an olive branch, I've self-reverted my last comment. I agree it was not the sort of posting that on reflection I would be proud of. Cheers --RexxS (talk) 22:20, 20 August 2016 (UTC)

Citation styles


On talk:Jane Austin, you wrote:

"I'm considering the idea of forking the CS1 templates to allow an editor to set the display of the long cites as MLA or Chicago or what you will, so that citations would be very easy to reuse between articles using different display styles, just by changing MLA to APA or whatever to match the style used in the destination article."

I have a slightly different idea (and somewhat less technical ability that you): Change the templates so that the output is a user-selected preference.

In other words, from the same (templated) input, I could select to see Harvard style, while you could select to see MLA, or whatever. We currently do that for coordinates, emitted by Coord, using CSS, but that, and the citation preferences, should be in a user script, later a gadget, and eventually the standard user preferences dialogue.

I touch upon this in Citations - the future. Work being done under the auspices of Wikicite should facilitate such improvements. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 16:56, 21 August 2016 (UTC)
 * Thanks, . Nice as it would be to be able to select whatever style of output we desired, 95% of our readers don't have user-selected preferences (and another 4% don't know how to use them). So while we could write javascript to re-arrange the outputs of CS1 templates into pretty much whatever display we want, I'm not sure I'd want to invest so much time into a feature that hardly any reader would use. You could suggest it to the WMF devs; they have a good track record with ideas like that. Well, OK, in a perfect world, I'd definitely be in favour of the feature. By the way, the coords scheme of outputting both decimal degrees and dms and hiding one or the other via css classes would get really cumbersome if we had half-a-dozen different citation formats (APA, Chicago, Vancouver, MLA, Harvard, CS1 default off the top of my head)
 * What I think would be awesome would be to be enable an editor to add a "magic word" like __APA_REFS__ to an article reference section and have that "automagically" cause all the refs in that section to display in pseudo-APA style (or whatever style we have a definition for). In that case, we'd still need to have the Lua code written to switch output into one of a number of predefined formats, so it's worth developing CS1 that takes display_as (which needs placing in each long citation) because however we would read the magic word, it would still require the relevant code to be triggered, either internally or via an external parameter passed to the function. I'll have a good look at Module:Citation/CS1 to see what the size of the job is. To be continued ... --RexxS (talk) 17:39, 21 August 2016 (UTC)
 * Well, as with, there would have to be a defaults style for those who haven't selected one, or who are not logged in. I'd prefer we opt for one across the project, but if necessary it could be specified in the manner described by you in the above quote - the two ideas are not mutually incompatible. Your idea for a once-per-page setting would fit best in Reflist. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 20:01, 21 August 2016 (UTC)
 * Of course. For default, I'd settle for the current CS1 default style (there's an option to display author names in Vancouver-style format already there) as it would yield the least disturbance to what most of us are used to. In other words, making this sort of change should be transparent in every article until an editor activates the extra features for that article. Would you agree? --RexxS (talk) 21:31, 21 August 2016 (UTC)
 * When RA was around (still around with a rename at User:Tóraí), he did some work with the cite extension as proof-of-concept which would enable the entire page to display a certain style; take a look at phab:T24134. --Izno (talk) 13:16, 22 August 2016 (UTC)
 * Yes, that seems reasonable. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 14:22, 22 August 2016 (UTC)
 * I was just discussing that very extension with Iridescent yesterday. I remember Rannpháirtí anaithnid had set up a test wiki to try it out. The major improvement was the speed of page rendering; doing all the work in php was so much quicker than via the mediawiki parser that it made some articles usable again. I showed SlimVirgin at the time that Israel (575 citations) could generate an editing preview (which forces a cache flush) in a few seconds if it didn't have to use the old citation templates that took over a minute to process. Once Scribunto was implemented, most of that advantage disappeared, and the proposal was forgotten about. Looking back at Centralized discussion/Citation discussion shows how something as obviously advantageous could be talked down by the usual Luddite reactions and failure to understand. I see that we exchanged views briefly on that very page at Redux. Looking back at that section shows that nothing has changed in exactly six years, and really, I should realise that it's a waste of time trying to appease those who feel that hand-written citations are the only possibility for "their" best articles. Templates already offer so many advantages over hand-written citations that it shouldn't even be a debate. "There's none so blind as those that don't want to see." --RexxS (talk) 16:30, 22 August 2016 (UTC)
 * For what it's worth I don't think the average reader cares what format the citations use as long as they work, so reverting to a default for IPs and logged out should logically not be a problem. In reality there will probably be the not invented here contingent. I would like a single method of adding citations which is easy to use but allows all the options, (along with world peace, justice for all and a solution to the population problem - why not aim high?) Cheers &bull; &bull; &bull; Peter (Southwood) (talk): 17:07, 22 August 2016 (UTC)
 * Yes, that seems reasonable. Andy Mabbett ( Pigsonthewing ); Talk to Andy; Andy's edits 14:22, 22 August 2016 (UTC)
 * I was just discussing that very extension with Iridescent yesterday. I remember Rannpháirtí anaithnid had set up a test wiki to try it out. The major improvement was the speed of page rendering; doing all the work in php was so much quicker than via the mediawiki parser that it made some articles usable again. I showed SlimVirgin at the time that Israel (575 citations) could generate an editing preview (which forces a cache flush) in a few seconds if it didn't have to use the old citation templates that took over a minute to process. Once Scribunto was implemented, most of that advantage disappeared, and the proposal was forgotten about. Looking back at Centralized discussion/Citation discussion shows how something as obviously advantageous could be talked down by the usual Luddite reactions and failure to understand. I see that we exchanged views briefly on that very page at Redux. Looking back at that section shows that nothing has changed in exactly six years, and really, I should realise that it's a waste of time trying to appease those who feel that hand-written citations are the only possibility for "their" best articles. Templates already offer so many advantages over hand-written citations that it shouldn't even be a debate. "There's none so blind as those that don't want to see." --RexxS (talk) 16:30, 22 August 2016 (UTC)
 * For what it's worth I don't think the average reader cares what format the citations use as long as they work, so reverting to a default for IPs and logged out should logically not be a problem. In reality there will probably be the not invented here contingent. I would like a single method of adding citations which is easy to use but allows all the options, (along with world peace, justice for all and a solution to the population problem - why not aim high?) Cheers &bull; &bull; &bull; Peter (Southwood) (talk): 17:07, 22 August 2016 (UTC)
 * For what it's worth I don't think the average reader cares what format the citations use as long as they work, so reverting to a default for IPs and logged out should logically not be a problem. In reality there will probably be the not invented here contingent. I would like a single method of adding citations which is easy to use but allows all the options, (along with world peace, justice for all and a solution to the population problem - why not aim high?) Cheers &bull; &bull; &bull; Peter (Southwood) (talk): 17:07, 22 August 2016 (UTC)