Misplaced Pages

Revision as of 21:20, 19 June 2016

YouTube 'articles'

Was it really over a year ago since this talk page discussion? I was reminded of this when I came across this YouTube 'presentation' that looks like a spammy automated reuse (without any attribution that I can see) of Misplaced Pages articles (probably lots more like that). I don't frequent YouTube much, but is that relatively common? I remember some of the discussions around some reuses (some were correctly attributed, others were not). Misplaced Pages:Copyrights is confusing. Where is the place to mention this nowadays? Carcharoth (talk) 12:56, 3 May 2016 (UTC)

I think it's fairly common, given the thinly-veiled contempt the WMF displays towards editors nowadays; the original version of Qwiki had an automated process to "presentationify" every page on Misplaced Pages, before Yahoo bought it out and shut it down, while even on Misplaced Pages's much-vaunted official mobile site—which a dev will likely pop up shortly to explain is The Future Of Misplaced Pages and berate me for showing insufficient respect to—you'll search in vain for anything as prosaic as 'attribution to the authors'. The official process to report things like this is Misplaced Pages:Mirrors and forks#Non-compliance process, but I'm not sure the WMF still have the heart to pursue copyright violators any more, especially given that the strings of the WMF are firmly pulled by Google these days, for whom ripping off other peoples' work wholesale and passing it off as their own making use of the differential in licence arrangements between Misplaced Pages and Wikidata by encouraging people to mirror key elements of Misplaced Pages onto the CC0-licensed Wikidata to allow subsequent reuse without attribution is a fundamental part of the business model. If you care, Moonriddengirl will be able to advise if the WMF still takes things like this seriously and if so who to speak to. ‑ Iridescent 23:04, 3 May 2016 (UTC)

• Many moons ago, I started the article Three Wolf Moon. I was subsequently amused to find this used as the basis for episode 2 of The Things You Learn on Misplaced Pages – an Arthur Albert production on YouTube. I was quite impressed by the production quality of this and so feel flattered by the attention. I suppose the plan was to make money from the advertising on YouTube but it doesn't seem to have gone viral yet. The crowd at the London Wikimeet seemed to enjoy it yesterday though, when I showed it to them. Note that the Misplaced Pages page averages about 250 hits per day and so is still doing quite well compared to the YouTube version, which only averages about 5 hits/day. Andrew D. (talk) 16:47, 13 June 2016 (UTC)

• I do think it's a shame Qwiki lost their nerve with regards to their original model. I thought it was a fantastic idea—RexxS presumably recalls my prosletysing for its ability to convert a Wiki page into a spoken-word presentation without slavishly reading out the text word for word, or to create animated slides, with dots moving about maps and the like, all using software to extract the data from the article itself and a few key parameters like coordinates, without pissing about with Wikidata and the like—although I'll admit I never understood where they expected their revenue stream to come from. As you can see for yourself by heading over to Special:Preferences#mw-prefsection-betafeatures and ticking "Related Pages" and then flipping around a few random articles, for all Jimbo and pals' posturing about intelligent interpretation of embedded data, the WMF is still in the stone age when it comes to even the most simple data-centred work. ("Spoon: see also Maple syrup", "War: see also Germany", "Idiot": see also Friedrich Nietzsche", "Church: see also Liverpool" and my personal favourite, "Ugly: see also Kate Bush".)

  The popularity of Three Wolf Moon probably comes from it being on Misplaced Pages:Unusual articles, which generally drives a consistent fifty-or-so views per day to each article on it; if the article is interesting enough, this then gets magnified by orders of magnitude every so often whenever a celebrity stumbles across it and tweets a link. Of everything I've written, the most-viewed is consistently Tarrare, which gets well over half a million pageviews per year despite having virtually no incoming links (it's hard to get a field more niche than 18th-century French teratological case studies) owing to massive spikes whenever a celebrity or high profile blogger stumbles across it. (Incidentally, Three Wolf Moon and Tarrare give "see also J. R. R. Tolkien" and "see also Napoleon", respectively.) ‑ Iridescent 13:40, 14 June 2016 (UTC)

• I do remember it well, and I also thought it was a great idea. Was that really four years ago? It's easy to pull in content from Misplaced Pages articles and make a skin to display them in a nicer format; perhaps we should revive the idea? --RexxS (talk) 18:06, 14 June 2016 (UTC)

• Wikiwand does do this already, although I personally think their format is cluttered, confusing and almost unreadable. The problem in my view would be getting people to agree on what an advanced reskin which fundamentally changed how the articles are presented, as opposed to fiddling with fonts and image placement, should actually look like, given that (for better or for worse) most Misplaced Pages articles have been explicitly designed with viewing in Vector in mind, and that once Visual Editor becomes the default this will get even more pronounced. (While you're here, you may want to see if I'm unfairly maligning Wikidata and all who sail in her a few threads down.) ‑ Iridescent 18:14, 14 June 2016 (UTC)

Hate videogames?!

I'm usually the type of nosy person who likes to dive into reasons why people have strong emotions for or against whatever, and I'm a huge fan of games. I'm probably missing tons of conversations you already had about this thing, but what is it about games ya don't like? If you were recommended a game based on what you disliked about games (such as it gave you vertigo or you didn't like the complex controls and the recommended game did not have them or something like that) would you try it? I'm just insanely curious about things that don't matter :P Sethyre (talk) 20:20, 7 May 2016 (UTC)

Sethyre, just realised I never replied to this. I don't hate videogames; my comments regarding "my disdain for the topic" relates to the coverage of videogames on Misplaced Pages, rather than the topic per se. Misplaced Pages's videogame articles generally (with some honorable exceptions) tend to be drivel like this, and to combine an inappropriate level of detail with a jargon-filled vocabulary which is incomprehensible to non-insiders. I suspect it's largely due to the newness of the medium; the conventions for how to write for a general audience about books, films, paintings etc have been developed over decades (in the case of books and paintings, over centuries) and are widely understood, whereas most games journalism is in specialist magazines/websites aimed at aficionados, and thus those people who write on the topic tend to find it difficult to write for Misplaced Pages's general audience, where the target reader is "bright 14-year-old with no prior knowledge of the subject".

On the broader topic, my attitude towards games tends to echo that of Mel Croucher (whose autobiography ought to be required reading for anyone interested in the history of the industry, provided you can overlook the self-pity and self-promotion). I think the technical limitations of the gaming consoles of the 1990s-2000s stifled a lot of the creativity of the early Atari/Commodore/Sinclair days and pushed the industry off course into a rut of derivative shooting/jumping/running games for over a decade. While the industry is now starting to move away from this towards the genuine interactive experiences envisaged back in the early days of computing, the damage is done as to those outside the bubble, "video gaming" is synonymous with gratuitous violence and the over-use of primary colors so they quite reasonably feel "this has no relevance to me". That then creates a bunker mentality among the gamers themselves who feel (with reasonable cause) that something which is of great importance to them is being unfairly belittled by mainstream culture, and you end up with the kind of standoff and mutual hostility you see now. None of this is unique to videogames—the transition from "small group who see themselves as the cutting edge" to "accepted part of mainstream culture" has happened with every new art form in history. ‑ Iridescent 11:23, 8 June 2016 (UTC)

I have a tough time finding any reason to disagree with this. In fact it's pretty much spot on, moreso than I initially realized. I suppose the rut of games that we see now is comparible to how novels exploded in the fantasy sci-fi genre, trying to emulate JRR Tolkien, or how every movie needed to have that action movie star during the 80s. And with the description of the mainstream games, which are over the top to the extreme, I have to agree that I don't find it appealing to me. I'll take Banjo-Kazooie over Call of duty any day of the week. Sethyre (talk) 19:58, 9 June 2016 (UTC)

What I find striking is how little originality there tends to be; despite modern computer systems being literally millions of times more complex, I can think of very few games which aren't effectively prettier graphics and a larger game-world on top of a basic game structure dating back to the days of cassette tapes and 48kb-128kb memory. (I'm old enough to remember the first incarnation of Elite, which if you disregard the crude graphics is almost identical in both design, gameplay and complexity to all its spiritual successors like Eve Online and Wing Commander.) The aforementioned Mel Croucher autobiography has an amusing episode of him returning to the industry after a 25-year gap and Steam being unable to list Deus Ex Machina 2* as it didn't fit into any of their "war, sports, strategy, simulator, platform" pigeonholes. ‑ Iridescent 14:07, 14 June 2016 (UTC)

*a truly bizarre, intentionally incomprehensible and virtually unplayable game with an A-list voice cast ranging from Christopher Lee to Joaquim de Almeida—well worth buying for the Ian Dury soundtrack even if you only play it once.

TPS - I asked an acquaintance in the know about that Steam listing. Alledgedly the inside story is a bit different ;) And by 'in the know' I mean 'makes decisions about what goes on steam'. Functionally its different these days anyway, as they have moved to a more wikipedia-like category system where you can tag anything in completely made up categories. So if I want to search for 'Gothic noir match-3' genre I can. The rise in ease of self-publishing means all sorts of original stuff is on steam. Granted on a basic mechanic level you will not find anything new, but pen&paper & boardgames have that to blame, rather than previous videogames. Only in death does duty end (talk) 10:06, 15 June 2016 (UTC)

Probably, but Croucher's version of events (All we had to do was submit Deus Ex Machina 2 to a panel of mature, expert and savvy Steam assessors who would recognise the concept and the quality of our game and release it to an eager public. Unfortunately Steam changed the rules on August 31st and handed over decision-making to a bunch of bananas. Green bananas. Democracy is a very good thing, except when green bananas are enfranchised … Steam invited me to tick an appropriate box to categorise Deus Ex Machina 2, a product that could not be categorised when video gaming was in its infancy, let alone now. I was offered the choice of Casual, Platformer, Massively Multiplayer, Racing, Horror, Shooter, and similar formulaic pap. I toyed with the idea of ticking all the options, but I thought it best to be honest and accept the fact that no category fitted and the Steaming voters may not be wholly accepting of my game.) is more entertaining. FWIW the thing is on Steam now (I just looked)—the comments section on it reminds me of the reaction to the Sex Pistols circa 1977. ‑ Iridescent 14:43, 15 June 2016 (UTC)

Unfortunately he didn't really understand how to work (or game aha) the greenlight system. Greenlight was not about putting a product up there and letting random monkeys vote up your game - people who took that approach (and Croucher appears to have initially gone that route) often encountered the bananas approach. If you worked by growing your community fanbase, you would generally have enough to get *anything* greenlit. This of course infuriates more experienced developers because some really really terrible stuff gets greenlit while (subjectively) more 'worthy' products by less social media savvy devs languish. It was also a slap that if you were an established named developer (or in some cases, single party 'names') you could bypass the greenlight system completely. Only in death does duty end (talk) 15:16, 15 June 2016 (UTC)

I suspect he assumed that he'd get enough support from the retro-gaming community to give him instant acceptance; Automata does have genuine cult status in certain circles. (If you can rustle up a Spectrum emulator, iD is still one of the most peculiar games ever created, while iOS and Android ports of the 1985 version of Deus Ex Machina are decent sellers in app stores 30 years on and have aged surprisingly well.) In hindsight he should probably have marketed it to music and film fans and not treated it as a game at all—it's barely a game by any reasonable definition of the word, more of an interactive music video, and the presence of Jon Pertwee, Christopher Lee, Frankie Howerd et al would have created a decent buzz among people who normally wouldn't be seen dead near a videogame. ‑ Iridescent 15:43, 15 June 2016 (UTC)

Frankie Howerd, really? Martinevans123 (talk) 16:01, 15 June 2016 (UTC)

Really. He appeared mainly in the 1985 version; Croucher took a dislike to him, and largely edited him out of the 2015 version. You can pick up the soundtracks of both versions on iTunes or Amazon for a few bob if you're curious. ‑ Iridescent 16:17, 15 June 2016 (UTC)

How amazing. Perhaps his article needs attention. Martinevans123 (talk) 16:28, 15 June 2016 (UTC)

This is a (probably copyright violating, but I can't imagine they'd object to the free publicity) walkthrough of the whole 1985 version. There's an example of Howerd in action at 18:00 minutes exactly; Ian Dury's extraordinary turn as a singing sperm begins at 10:30. I would recommend playing the whole through start-to-finish (you don't have to watch it) as the soundtrack is genuinely astonishing, especially for its time. ‑ Iridescent 16:44, 15 June 2016 (UTC)

Infobox & Wikidata

At VPP, CFCF raised the example of Gout where it works well/as it supposed to. I actually tested it and found it worked *exactly* how I feared it would. I posted my results there but I think it got lost in the mass of replies below the actual voting. Pasted with some additions here for your perusal.

"So what actually happens when I remove Field = Rheumatology by editing the Gout article in the template section on wikipedia - because I have left it (field) blank, the infobox pulls the 'Specialty' from Gout at wikidata and populates it accordingly. You can see from the switch (in the infobox) from Rheumatology (wikipedia) to rheumatology (wikidata). However Gout's specialty field at wikidata is sourced to wikipedia, and the wikipedia article does not actually state in the article that Gout is in the Rheumatology field. Except in the infobox. Which is unsourced (unreferenced). (it probably doesnt need to be as its a sky is blue issue - for doctors anyway). In this case not an issue, but it clearly demonstrates the problem with the process. On a BLP this would be madness."

(I just checked, and someone has added a reference on wikidata since I did my test, so it is at least sourced there now to somewhere other than back to wikipedia)

The problem I have currently, is that the above example actually deliberately violates Misplaced Pages's policies on sourcing and verifiability. By automation. If I hadnt known what I was doing in advance this is what happens when I find dodgy info in an infobox.

1. Click edit

2. Remove info from parameter.

3. Save and sip Tea.

Now here is what I would need to do with wikidata adding content.

1. Click edit.

2. Remove info.

3. See info is still in infobox after saving.

4. Click edit again and check info is not there.

5. Go to template page to try and work out what the hell is going on.

6. See template is pulling data from Wikidata, but no explanation as to why or how to fix/prevent it.

7. Go to wikidata and attempt to change/remove info.

8. Get reverted by wikidata gnome who says it 'isnt wikipedia'.

9. Get bored of arguing, go back to article, remove infobox.

10. Pour away tea and open the rum. Only in death does duty end (talk) 16:24, 14 June 2016 (UTC)

{{{name}}}
{{{fullname}}}
{{{other name}}}
{{{native_name}}}
{{{caption}}}
{{{name}}}{{{map caption}}}
Coordinates: Unknown argument format
OS grid reference	{{{osgridref}}}
OS grid reference	{{{osgraw}}}
Location	{{{location}}}
Country	{{{country}}}
Denomination	{{{denomination}}}
Previous denomination	{{{previous denomination}}}
Tradition	{{{tradition}}}
Churchmanship	{{{churchmanship}}}
Membership	{{{membership}}}
Weekly attendance	{{{attendance}}}
Website	{{{website}}}
History
Former name(s)	{{{former names}}}
Authorising papal bull	{{{bull date}}}
Founded	{{{founded date}}}
Founder(s)	{{{founder}}}
Dedication	{{{dedication}}}
Dedicated	{{{dedicated date}}}
Earlier dedication	{{{earlydedication}}}
Other dedication	{{{otherdedication}}}
Consecrated	{{{consecrated date}}}
Cult(s) present	{{{cult}}}
Relics held	{{{relics}}}
Events	{{{events}}}
Past bishop(s)	{{{past bishop}}}
Associated people	{{{people}}}
Architecture
Status	{{{status}}}
Functional status	{{{functional status}}}
Heritage designation	{{{heritage designation}}}
Designated	{{{designated date}}}
Previous cathedrals	{{{previous cathedrals}}}
Architect(s)	{{{architect}}}
Architectural type	{{{architectural type}}}
Style	{{{style}}}
Years built	{{{years built}}}
Groundbreaking	{{{groundbreaking}}}
Completed	{{{completed date}}}
Construction cost	{{{construction cost}}}
Closed	{{{closed date}}}
Demolished	{{{demolished date}}}
Specifications
Capacity	{{{capacity}}}
Length	{{{length}}}
Nave length	{{{length nave}}}
Choir length	{{{length choir}}}
Width	{{{width}}}
Nave width	{{{width nave}}}
Width across transepts	{{{width transepts}}}
Height	{{{height}}}
Nave height	{{{height nave}}}
Choir height	{{{height choir}}}
Diameter	{{{diameter}}}
Other dimensions	{{{other dimensions}}}
Number of floors	{{{floor count}}}
Floor area	{{{floor area}}}
Number of domes	{{{dome quantity}}}
Dome height (outer)	{{{dome height outer}}}
Dome height (inner)	{{{dome height inner}}}
Dome diameter (outer)	{{{dome dia outer}}}
Dome diameter (inner)	{{{dome dia inner}}}
Number of towers	{{{tower quantity}}}
Tower height	{{{tower height}}}
Number of spires	{{{spire quantity}}}
Spire height	{{{spire height}}}
Materials	{{{materials}}}
Bells	{{{bells}}} ({{{bells hung}}})
Tenor bell weight	{{{bell weight}}}
Administration
Parish	{{{parish}}}
Deanery	{{{deanery}}}
Archdeaconry	{{{archdeaconry}}}
Episcopal area	{{{episcopalarea}}}
Archdiocese	{{{archdiocese}}}
Metropolis	{{{metropolis}}}
Diocese	{{{diocese}}} (since {{{diocese start}}})
Province	{{{province}}}
Presbytery	{{{presbytery}}}
Synod	{{{synod}}}
Circuit	{{{circuit}}}
District	{{{district}}}
Division	{{{division}}}
Subdivision	{{{subdivision}}}
Clergy
Archbishop	{{{archbishop}}}
Bishop(s)	{{{bishop}}}
Auxiliary Bishop(s)	{{{auxiliary bishop}}}
Abbot	{{{abbot}}}
Prior	{{{prior}}}
Subprior	{{{subprior}}}
Provost and rector	{{{provost-rector}}}
Exarch(s)	{{{exarch}}}
Provost	{{{provost}}}
Vice-provost	{{{viceprovost}}}
Rector	{{{rector}}}
Vicar(s)	{{{vicar}}}
Dean	{{{dean}}}
Subdean	{{{subdean}}}
Archpriest	{{{archpriest}}}
Precentor	{{{precentor}}}
Succentor	{{{succentor}}}
Chancellor	{{{chancellor}}}
Canon Chancellor	{{{canonchancellor}}}
Canon(s)	{{{canon}}}
Canon Pastor	{{{canonpastor}}}
Canon Missioner	{{{canonmissioner}}}
Canon Treasurer	{{{canontreasurer}}}
Prebendary	{{{prebendary}}}
Priest in charge	{{{priestincharge}}}
Priest(s)	{{{priest}}}
Assistant priest	{{{asstpriest}}}
Honorary priest(s)	{{{honpriest}}}
Curate(s)	{{{curate}}}
Asst Curate(s)	{{{asstcurate}}}
NSM(s)	{{{nonstipendiaryminister}}}
Minister(s)	{{{minister}}}
Assistant	{{{assistant}}}
Senior pastor(s)	{{{seniorpastor}}}
Pastor(s)	{{{pastor}}}
Chaplain(s)	{{{chaplain}}}
Archdeacon	{{{archdeacon}}}
Deacon(s)	{{{deacon}}}
Deaconess(es)	{{{deaconess}}}
Laity
Reader(s)	{{{reader}}}
Student intern	{{{student intern}}}
Organist/Director of music	{{{organistdom}}}
Director of music	{{{director}}}
Organist(s)	{{{organist}}}
Organ scholar	{{{organscholar}}}
Chapter clerk	{{{chapterclerk}}}
Lay member(s) of chapter	{{{laychapter}}}
Session clerk	{{{sessionclerk}}}
Treasurer	{{{treasurer}}}
Churchwarden(s)	{{{warden}}}
Verger	{{{verger}}}
Business manager	{{{businessmgr}}}
Liturgy coordinator	{{{liturgycoord}}}
Religious education coordinator	{{{reledu}}}
RCIA coordinator	{{{rcia}}}
Youth ministry coordinator	{{{youthmin}}}
Flower guild	{{{flowerguild}}}
Music group(s)	{{{musicgroup}}}
Parish administrator	{{{parishadmin}}}
Servers' guild	{{{serversguild}}}
]

break

I agree, which is why I support making importing from Wikidata opt-in only; Wikidata's information integrity is even worse than Misplaced Pages's, and there's no obvious way to see duff edits being made in Wikidata which filter through onto Misplaced Pages. (Even if you're willing to see your watchlist degenerate into incomprehensible gray goo, selecting "show Wikidata edits" just fills your watchlist with unhelpful comments like (A Wikidata item has been linked to this page.) which isn't any actual help; I do get the feeling the Wikidata people don't appreciate that their pet project is utterly incomprehensible to outsiders.)

Quite aside from the issues with Wikidata's dubious approach to referencing, there's also the bloat issue. There are some infoboxes like {{infobox protein}} which only have a limited number of fields and in which it's uncontroversial that all the fields should be populated if the data is available, and there certainly should be a means by which people can decide to activate automated importing in those cases. There are also a lot of infoboxes like {{infobox church}} which have an absolute bucketload of fields, most of which can very easily be sourced-and-referenced but which would make the infobox longer than most articles if they were filled in in full (see right).

User:Dank is the person you need to convince of this, as he's the one closing the RFC; on this page I can assure you that with a few exceptions, you're very much preaching to the converted if you're against the "Wikidata will solve every problem" cult. ‑ Iridescent 16:56, 14 June 2016 (UTC)

My gods, that's a monstrosity of an infobox! Ealdgyth - Talk 17:06, 14 June 2016 (UTC)

It's by no means alone. ‑ Iridescent 17:21, 14 June 2016 (UTC)

Lumme - they're even worse than that monstrosity at Winston Churchill! (No-one has ever been able to explain the sense in adding "Preceded by" and "Succeeded by" for each and every cabinet position, which makes the whole thing an unwieldy mess!) - SchroCat (talk) 17:24, 14 June 2016 (UTC)

Inspired by your comment, SchroCat, I have trimmed the Churchillian infobox, which still leaves it gargantuan. Kablammo (talk) 18:28, 14 June 2016 (UTC)

In some defence of {{infobox church}}, some of the fields are mutually exclusive (nothing which has the Churchmanship field completed is likely to hold any relics, for instance) and some of the fields like "demolished date" will obviously be blank in many cases, but it still doesn't explain why we'd ever need an entry in flowerguild = or organscholar =. Take it up with these people; I dare say you'll see a few names you recognise. ‑ Iridescent 17:29, 14 June 2016 (UTC)

User:Only in death (great username btw), add anything you like to the RfC. I have no problem with long comments. The deadline on the RfC is in about 3.5 hrs. - Dank (push to talk) 17:30, 14 June 2016 (UTC)

By an odd coincidence, I'm about to start work on making a Wikidata-enabled version of {{Infobox church}}. The RfC is predicated on the wrong premise, of course. The real question is no longer about how the infobox designers should decide to pull data from Wikidata, but about how we can devolve the decision to editors at the article level. I'm currently trying a solution where each article with an infobox can implement a whitelist and a blacklist, so that the information is only fetched from Wikidata for whitelisted fields, but fields that are blacklisted are never displayed for that article. We'll see if editors are interested in taking control of those decisions for themselves. --RexxS (talk) 18:21, 14 June 2016 (UTC)

It's more that there are two different issues which have become entangled. "Is Wikidata reliable enough that en-wiki should consider importing data from it without manually checking each instance?" and "What is the best mechanism for importing data from it?". The two questions are hopelessly entangled—if one doesn't accept that Wikidata is accurate, that has a major impact on which mechanism ought to be used. (If one concedes that Wikidata is untrustworthy, then the default position pretty much has to be opt-in; this is a separate issue from "do we want every field autopopulated?".) Plus, of course, there's the way Wikidata sources information from other Wikipedias which may be in disagreement with this one—to pick a not-at-all-random example, if you flip through the various language versions of President of Brazil you'll see there's no agreement between the various languages about who holds this post, so how do we decide which is the correct version to use? ‑ Iridescent 18:34, 14 June 2016 (UTC)

The first question has to be answered affirmatively before we even start to think about the second. If data from Wikidata is less reliable than our present article, then we shouldn't be importing it. I understand the result of Misplaced Pages:Requests for comment/Wikidata Phase 2 to mean that editors were content to import data from Wikidata within infoboxes, but you may be right that it's time to ask the question again. I've made a demo tool that allows anyone to see what data and references are available on Wikidata by pasting {{#invoke:Sandbox/RexxS/WdRefs|seeRefs}} into any article and previewing it. That might help folks decide whether the Wikidata information is suitable for inclusion in the article.

As for President of Brazil, if we're thinking about importing data from Wikidata, then we don't have to worry about how other Wikipedias treat a subject. On Wikidata, President of Brazil (Q5176750) shows: officeholder (P1308) as Michel Temer (Q463533) (start date = 17 April 2016); and five others with different start and finish dates. If we want to use Dilma Rousseff in our article President of Brazil instead, then it's simple to supply her name in the infobox ourselves. A local value will always override anything from Wikidata - at least in any infobox that I've had anything to do with. That keeps the decision in the hands of the article editors, and that's how it should be IMHO. --RexxS (talk) 19:57, 14 June 2016 (UTC)

Unreferenced statement added by a driveby IP; as I've pointed out on the RFC, even the man himself scrupulously avoids referring to himself as "President". When Sarah talks about the problem of Wikidata combining oversimplification of complex issues and a failure to check potentially problematic statements, this is a textbook example. ‑ Iridescent 20:07, 14 June 2016 (UTC)

I quite like the "(Tag: new editor changing statement)" auto-generated edit summary. It didn't take more than a few seconds to correct the Wikidata entry for President of Brazil (Q5176750), and one of the results of regularly importing data from Wikidata would be that we would have noticed any drive-by edit to Wikidata immediately, rather than months later. What Sarah hasn't got to grips with yet is that the type of problems potentially besetting her from Wikidata are identical to the problems she gets from Misplaced Pages editors. The only difference between "genre" being adding from Wikidata and being added by Misplaced Pages editors, is that it is trivial to stop it from Wikidata, and often difficult to revert stubborn Misplaced Pages editors. Wikidata doesn't edit-war with you. --RexxS (talk) 20:48, 14 June 2016 (UTC)

Hmm. I think Temer is the president of Brazil right now (he holds the office, even if it is only on a temporary basis), and I can find plenty of reliable sources that will refer to him as the interim or acting president of Brazil, if not the president (without any qualifier). So why did you remove his name, or at least not fix what appears to have been broken? The reason there's not a lot of edit warring at Wikidata is that there aren't enough people working there to care. But there will be, and then we will have epic battles that will affect not just one project but dozens. Mark my word. Risker (talk) 20:58, 14 June 2016 (UTC)

The reason I reverted, Risker (nice to hear from you btw!), was that the drive-by IP had overwritten the entry for Dilma Rousseff, rather than adding an entry for Temer. I made no arguments about who should properly be referred to as President currently (although the official Brazillian government website shows this: http://www.biblioteca.presidencia.gov.br/presidencia if folks want to argue about it). I merely restored an entry that had been deleted out-of-process. If you - or anybody else - care to add an entry for Temer to President of Brazil (Q5176750), that's fine, but please don't remove other accurate entries in doing so. Cheers --RexxS (talk) 21:10, 14 June 2016 (UTC) Update: I've added Michel Temer (Q463533) along with the correct start date and a reference. All's well? --RexxS (talk) 21:21, 14 June 2016 (UTC)

On Brazil, I'd consider "Acting President" to effectively be a regency rather than a transfer of office, but I concede it's not something about which I care a great deal. On Wikidata, I'm more inclined to side with Risker on this one—at the moment Wikidata is calm because it's something of a backwater, but once it becomes widely known that changes there will affect multiple-language Wikipedias, it will become a venue for huge editwars in which the warring parties don't even speak the same language so are literally incapable of discussing the matter and coming to a consensus. Sure, one can make the entries appear on your Misplaced Pages watchlist, but if you're not in the know then both the summary and the diff itself for an edit like Created claim: Property:P2604: 170941; ‎Created claim: Property:P2605: 7385 (the most recent Wikidata edit on my watchlist) may as well be a string of random characters. Yes, I do appreciate that Misplaced Pages itself is subject to editwarring, but Misplaced Pages pages tend to be watched more often, and even when they're not then serious problems are much more likely to be spotted and fixed by random visitors. This edit remained live and unfixed on Wikidata for over two years despite being in the high-profile description / en spot which is automatically shown in search results and the mobile site. ‑ Iridescent 14:40, 15 June 2016 (UTC)

Well, Risker seems to be admonishing me for removing Michel Temer, without understanding that I was actually fixing the problem that the IP had wiped out all trace of Dilma Rousseff from the record and nobody was showing as President of Brasil between 2011 and 2016. Surely you can't be on her side in thinking that's a good idea?

I don't think there's any disagreement that Wikidata has very few active editors to curate 18,000,000 items of data. The result is that vandalism is hard to spot and slow to be corrected. I suppose the question becomes "should we care?". The answer, IMHO, is "yes" mainly because of all the other small language Wikipedias that could potentially expand their coverage of many topics (or at least get a stub to start with), if only Wikidata could provide reasonably reliable data. At present, it's not an assumption we can make, sadly. The only way that I can see of improving that position is to engage the efforts of the editors of the largest Wikipedias to help watch over corresponding topics in Wikidata. I can't see that being a popular option, of course, because of the general parochialism of so many editors here. --RexxS (talk) 16:08, 15 June 2016 (UTC)

Not at all; I agree with you that we shouldn't be overwriting Rousseff with Temer since the situation is complex and she's still theoretically the office holder. (Note my comments on the RFC.) Regarding Wikidata, I'm not sure you'll ever be able to persuade the editors of the big Wikipedias to take an active interest. Along with Commons it's acquired a reputation, rightly or wrongly, as having taken over from Wikiversity and Simple English Misplaced Pages as en-wiki's dumping ground for problematic editors; the only way you'll overcome the "why the hell should I want to get involved with that pack of weirdos?" factor is if, as with Commons, you can make Wikidata so essential to the functioning of the other projects that even people who strongly oppose its very existence have no real alternative but to engage with it. ‑ Iridescent 16:30, 15 June 2016 (UTC)

But that's only if you can overcome the flaws in its very existence. WD is founded as a suppository repository suppository of base gobbets of "information" that it disseminates. This only works when the world can be pigeon-holed into neat boxes. Unfortunately that means some of these factoids—in many cases the banalest droplets of data—are ripped from their context and therefore people's understanding.

It's an even bigger problem when third parties try and use the data, and do so poorly. For example, the Google boxes for both the actors John Barrymore and Bernard Lee both have definitive information about their birth where there is no such clarity in real life (Barrymore's dates are uncertain and there are two possible locations for Lee's place of birth), and the sources are split in their opinions. Not Google, based on their use of WD: they "know" which one it is, and have stripped out the real-life uncertainty. That's only one of the problems of a computerised fact-stripping: WD mistakes data for knowledge and facts for understanding, without ever understanding the difference. It is the triumph of factoids over understanding, and a horrible, horrible concept. – SchroCat (talk) 16:45, 15 June 2016 (UTC)

Wikidata gives dates of birth for John Barrymore (Q95034) as 14 and 15 February 1882 (Gregorian), as does our article. It's not reasonable to blame a sister project for the misuse of its information by a third-party, even if that party is Google. The actual problem with Barrymore's entry in WD is that there is only one good reference supplied in the whole page, and what is needed is to supply references for all of the information there. John Barrymore is a Featured Article on en-wp, but only 1 in 10 of the different language Wikipedias have an article on him. I want my friends who work on the Welsh Misplaced Pages to have the tools to get started on adding all of those missing articles: if we could generate a set of key facts and a list of decent references, we'd be well on our way. If that information were available on Wikidata, I could do that for them, and I find it disappointing that more editors here don't share my passion for spreading our knowledge to other Wikipedias. --RexxS (talk) 19:40, 15 June 2016 (UTC)

And that's the problem right there. We can control what we put into articles and to some extent what goes into the abyss of WD. That's our responsibility and we manage it fairly well. With WD that responsibility is abrogated, which is something I'm not happy with. It's great to point the finger at the company that's largely funded WD, but if the a main funder of the project is using the data so poorly as to provide misleading information, it's something that leaves me extremely uncomfortable. You've raised the second major, major problem I have with WD, which is the utterly unsourced nature of what - 80, 90% of the information? (Are there any figures for the percentage of information that is unsourced in WD? And I don't mean that contains a dubious reference to Misplaced Pages, I mean an actual reference) And its proposed to fill in empty fields in IBs (or generate new articles) with these oceans of unsourced facts crap? You may be happy to fill shedloads of articles with unsourced nonsense rex, but it leaves me deeply concerned and reaching for the vandalism button to revert them all. – SchroCat (talk) 20:16, 15 June 2016 (UTC)

The signpost op-ed has some numbers. Only in death does duty end (talk) 21:37, 15 June 2016 (UTC)

Ouch.... And people think that it's sufficient to use that to generate non-existent articles, or to populate empty idiotbox fields? Please tell me that only the data that is supported by citations from external sources are to be used (unless common sense makes an unexpected but timely intervention and ensures this doesn't happen). Thanks OiD. – SchroCat (talk) 21:50, 15 June 2016 (UTC)

See my above testing with Gout (which thankfully now on Wikidata has a source that isnt Misplaced Pages). A wikidata enabled template just imports whatever fields are populated at Wikidata. (When the fields are removed here) To prevent a wikidata template from importing info from wikidata, you need to put the field in but leave it blank. As far as I am aware there is no way to tell a template here to only import wikidata if the info at wikidata is sourced to a third party. And even then it could be sourced to literally anything and it would still import it. Only in death does duty end (talk) 21:54, 15 June 2016 (UTC)

A wikidata enabled template just imports whatever fields are populated at Wikidata - a ridiculous generalisation that's like seeing an Appaloosa and claiming "all horses are spotted". Template:Infobox medical condition (new) does that because its users requested it to work that way. Template:Infobox book/Wikidata/Sandbox doesn't do that because the request was for an infobox that didn't autopopulate without the editor at the article specifically enabling it. Take any book article and preview {{Infobox book/Wikidata/Sandbox}} in it. Then come back here and try to tell me "A wikidata enabled template just imports whatever fields are populated at Wikidata ... To prevent a wikidata template from importing info from wikidata, you need to put the field in but leave it blank." --RexxS (talk) 22:54, 15 June 2016 (UTC)

I will stick with using actual live examples rather than your sandbox thanks. Only in death does duty end (talk) 23:03, 15 June 2016 (UTC)

@SchroCat: Where do you get the disinformation about Google being the major funder of Wikidata? The first year grant was 50% from Allen Institute for Artificial Intelligence, 25% from the Gordon and Betty Moore Foundation and 25% from Google. Since the first year, it's been funded by WMF. If you want to know what references are available in Wikidata for an article, preview {{#invoke:Sandbox/RexxS/WdRefs|seeRefs}} in it. I made the tool for folks to check in the hope that some will be responsible enough to actually try to improve the referencing. I can see that narrow parochialism wins out every time. --RexxS (talk) 22:54, 15 June 2016 (UTC)

What a bloody farce it all is. We all sweat blood to get articles to the highest standards possible, including hours of endless research into topics. We endlessly debate the smallest changes in punctuation and minor word alterations to ensure the meanings of our texts are clear, and the source reviews we have check minute details to ensure that what we write is adequately covered in sources. Now some bright spark has thought it would be a "good idea" to drop referenced, unverified and probably incorrect bollocks into articles with no verification process? What an utter clusterfuck! If enforced idiotbox data population is the order of the day, it automatically downgrades every graded article we have back to zero, because unreferenced and unverified information in an article is the quickest way to FAR, FLRC and GAR. At least drive-by IP vandals can only cock up one article at a time, not the few million that WD vandalism will affect! – SchroCat (talk) 23:01, 15 June 2016 (UTC)

"narrow parochialism"? No, rex. It's about having standards and adhering to them. And just because some people don't bow down at your altar of WD, it's not a sign of narrow parochialism, just a different way of looking at something. – SchroCat (talk) 23:07, 15 June 2016 (UTC)

And yet the vast majority of the sourcing on Wikidata that you're whining about is imported from English Misplaced Pages. The very same pieces of information that you're so proud of when viewed on Misplaced Pages become "unverified and probably incorrect bollocks" when seen on Wikidata. So which is it? I have no great love of Wikidata, so you can cut the ad hominems. My interest is solely how we can leverage it to improve other Misplaced Pages projects, as well as this one. You really make my case about parochialism for me. --RexxS (talk) 23:12, 15 June 2016 (UTC)

Break 2

Oh FFS, I'm guilty of "narrow parochialism", and having concerns about standards in Misplaced Pages is "whining", and yet I can cut the ad hominem comments? It comes as no surprise to see you revert to your usual 'style' once again rex. There is no "narrow parochialism" here, despite your oft-repeated PA: I'm not accusing foreign language wikis of providing dross which we here should guard against (you've erroneously leapt to that conclusion all on your own). I'm concerned about erroneous data from all wikis, including en, which we are going to rehash across the entire project and which is then repeated on third the party sites that don't even care about getting nuanced or confused details right, as long as they can shove some factoid in their 'one-size-fits-all' field. Try replying without the snide insults this time rex – they say much more about you than your targets. – SchroCat (talk) 05:46, 16 June 2016 (UTC)

OK, you're both talking past each other. Quick refocusing:

• SchroCat, RexxS makes a valid point that most of the unsourced data on Wikidata is actually sourced from somewhere and has just become detached from its source, and the issue is instilling a better culture of sourcing on Wikidata rather than jettisoning it altogether;
• RexxS (implicitly) makes an even more valid point that given that this is a WMF pet project, it's not going away so it's up to the individual Wikipedias to reach a modus vivendi with it, in the same way en-wiki has reached a grudging acceptance of Commons parking their tanks on the lawn. (It also occurs to me that it makes it considerably more difficult to set up a Misplaced Pages fork; whether you see that as a positive or a negative is up to you.);
• SchroCat also has a perfectly valid point; the intersection of "Assume good faith", "You can't copyright a fact" and "All language Wikipedias are equal" at Wikidata has unintentionally created a culture of poor sourcing in which the audit trails on sourcing and citation have become severely attenuated, and unlike Commons (or even the backwaters like Wikisource and Wikiquote) Wikidata doesn't have the culture of admins and gnomes to cope with the volume data they've absorbed, and it's not "parochialism" to point out that the big Wikipedias are likely to be better at spotting and fixing errors on-site than if they're exported elsewhere. (I'd add that, because of its radically different structure and steep learning curve, Wikidata doesn't even have the last-resort fallback which Wikivoyage, Wikispecies et al have, of being able to post begging letters at noticeboards on the big Wikipedias asking editors to temporarily move across to help clear up a backlog.)

All of this could do with some actual numbers. I'm not going to spend time writing up a grant application myself, but I could certainly make a case that it would be worthwhile if the WMF diverted a fraction of the funds they spend flying the party faithful to an annual knees-up, or producing site-wide banners inviting all-comers to suggest the best way to get rid of people to whom Jimbo's taken a dislike, on some actual figures for Wikidata's accuracy. I'd suggest a genuinely random sample of 10,000 Wikidata entries and 10,000 Misplaced Pages pages, and checking every claimed fact against reality. If the Wikidata model is working as its proponents claim, it should have an accuracy rate well above en-wiki's, and some actual data to that effect would go a long way towards silencing its critics. If Wikidata's accuracy comes out below en-wiki's, then per my comments somewhere near the top of this monster thread that's a powerful case for the "opt-in only" position. At the moment, from that RFC it's obvious there's a clear split between "Wikidata is unsalvageably bad" and "Wikidata is a miracle cure", and neither faction has much data to back their position up. ‑ Iridescent 13:45, 16 June 2016 (UTC)

Also, pinging Jayen466 to this discussion given that he's the one responsible for popularising the "Wikidata is inherently unreliable" concept. ‑ Iridescent 13:55, 16 June 2016 (UTC)

Iridescent, The problem is that "most of the unsourced data on Wikidata is actually sourced from somewhere" is way too vague to allow data dumping onto the various wikis. Don't forget many of the articles that could be infected by this data will be BLPs. We hold a high line to BLPs, and insist on sourcing even when we happen to 'know' that something is true. One of my utter bugbears is the addition of unsourced data onto any article, and it's second nature to me to insist on any addition to articles to be reliably sourced. WP:CIRCULAR is there for a very good reason ("Do not use articles from Misplaced Pages (whether this English Misplaced Pages or Wikipedias in other languages) as sources") and is another long step away from the accusations of parochialism - narrow or broad! - SchroCat (talk) 14:02, 16 June 2016 (UTC)

Sure, but the information that's sourced to Misplaced Pages could presumably in every case be sourced to whatever source that Misplaced Pages used, and the issue is that whoever initially imported it to Wikidata couldn't be fagged, rather than that the source doesn't exist. "Sourced to Misplaced Pages" isn't something you ever want to hear in an ideal world, but it's not necessarily the Mark of Cain some of the Misplaced Pages Review crowd make it out to be. ‑ Iridescent 15:58, 16 June 2016 (UTC)

But how—in the worst case BLP argument—do you ensure that the piece of WD that is imported across is one where there is a piece of 'separated' sourcing, as opposed to an unsourced piece, or a piece of vandalism (either on WD, or on some source article). That's a fairly big risk some people want to take based on way too little information. It may not be the mark of Cain, but it's a dramatic step backwards in terms of quality, verifiability and basic standards of sourcing. I really don't know what the rush is here. Getting WD up to scratch is a good way to start before we start dumping what I suspect is a lot of crap into a lot of places. SchroCat (talk) 16:27, 16 June 2016 (UTC)

The rush here is that a number of people with influence on the WMF operate businesses which are predicated on the ability to distil Misplaced Pages articles to a series of data points, and that the dead hand of the Knowledge Engine hasn't yet released its grip. I really don't think that's unduly cynical—regardless of the ultimate quality of the product, I don't think it's really in dispute that the reason it was rushed through was that Jimmy and pals are looking ahead to a day when Misplaced Pages can be written by source-analysing algorithms and editors can be dispensed with altogether. ‑ Iridescent 16:33, 16 June 2016 (UTC)

@RexxS, isn't an infobox that didn't autopopulate without the editor at the article specifically enabling it the very embodiment of what's being described at the RFC as "opt-in"? ‑ Iridescent 15:59, 16 June 2016 (UTC)

Not exactly. What the "opt-in" proponents have been asking for is an infobox where the editor has to write |isbn=FETCH_WIKIDATA or something like that in every field where they wanted to fetch the value from Wikidata, and then check that the result is sensible. Who is going to do that in an existing box when there's already a locally supplied value there? So it reduces the incentive to make use of Wikidata on the English Misplaced Pages. Also, if someone later adds, say, a value for date of publication to the Wikidata entry for the book, an editor on en-wp still has to enable it by adding |pub_date=FETCH_WIKIDATA to the article infobox, even though they have seen on their watchlist that the new value has been written to Wikidata and is sensible. It also means - if you take them at their word - that folks like Mike Peel will by told by some wikicop that they have just reverted all his changes to {{Infobox telescope}} because it's "opt-out".

If we replaced {{Infobox book}} by {{Infobox book/Wikidata/Sandbox}}, there would be no change in how the article displayed. In that sense, it's "opt-in". They would have to add |fetchwikidata=<list of fields to be imported from Wikidata> to enable the named fields. But if someone wanted to create a new article on a book that already has an entry on Wikidata, they could use {{Infobox book/Wikidata/Sandbox}} with |fetchwikidata=ALL and it would autopopulate with the values available on Wikidata - in other words it behaves as "opt-out". In addition, if Sarah wants to make sure that "Genre" never gets displayed in a particular article (perhaps because it's agreed that genre is not well-defined for that book), she can add the parameter |suppressfields=genre to the infobox and it will never display Genre, even if somebody specifically writes it into the infobox locally. By having both a blacklist and a whitelist (which can be a list of fields to fetch from Wikidata or "ALL" for all fields), we can be much smarter about how we decide to get our information from Wikidata. I've written a new Lua Module with the calls that should allow any infobox be modified for use as "opt-in" or "opt-out", or any desired combination, and also go some way to solving the problem of having to leave html comments telling editors not add "genre", or whatever.

I've created the tools (I hope) that will allow each side to have the Wikidata-aware infobox that they have asked for, but I'm faced at every turn by the "Little Englanders" who will make the point over and again that Wikidata doesn't offer much to a well-developed Misplaced Pages like the English one (which I don't dispute), but go deaf when I ask them to consider how much benefit the smaller language Wikipedias could make of using what may be developed here. --RexxS (talk) 19:00, 16 June 2016 (UTC)

But again, I think you're talking past each other. As I read the above, Wikidata will generate a lot of work for little gain on English Misplaced Pages (and ditto for the other well-developed projects like German and Polish), but will hugely benefit the smaller projects by allowing them to fill in gaps automagically. However, there's certainly a case to be made that that is just a variation of the "the primary purpose of Misplaced Pages is to be as machine readable as possible" argument. This is a view that does still have strong support in certain circles, particularly among the Silicon Valley types who control the WMF. However, as Lila Tretikov has recently illustrated fairly spectacularly, it's a deeply unpopular view among the editor base on which the big Wikipedias rely. (A disconnect and lack of understanding between a foundation that's increasingly run by people connected with Silicon Valley, and the volunteer community of editors who worry that may reflect a change in the WMF's focus from user-generated content to one led by automated data results, to quote our own article on the issue.)

Playing devil's advocate, I could make a perfectly valid and probably just as widely-supported case that the WMF should be actively closing down the smaller Wikipedias, or at least cutting them loose to sink or swim, and that the "little England" (and little Spanish, French, German) mentality that Misplaced Pages should be focusing on getting the widely-read projects in order is the correct one. (If the Siegenthaler controversy had taken place on Old Gothic Misplaced Pages nobody would ever have given a damn.) Why should donor funds continue to be spent on hosting and actively supporting projects whose recent changes feeds look like this, and which are clearly moribund? (The only three Wikipedias ever to be booted off the servers by the WMF are Klingon, Toki Pona and Siberian, all three of which are constructed languages which have never had a native speaker.) Sure, this WMF-created script can write articles from Wikidata and could inflate any given Misplaced Pages by a million new articles, but that would just leave a million new unwatched pages ready to degenerate into "Kyle is teh gay" and SEO spam.

TL;DR summary: The RFCs aren't asking the right questions, which should actually begin with "Does English Misplaced Pages have a duty of care to other language projects and to non-WMF commercial websites who reuse our data, or is our primary purpose still the English-language, non-machine, reader?", since the answer to that question will shape the answers to everything else. The only person who's really asked this question at a high level is Doc James, and look what happened to him. ‑ Iridescent 14:55, 17 June 2016 (UTC)

Aside: you need to maintain the indentation level on the otherwise blank lines if you want to be kind to screen readers (as we should).

Anyway, there's a lot of non-moribund small Wikis that could make use of some automation to help broaden their coverage. I'm sure you wouldn't want to tell my pal Robin that he'd be better off abandoning the Welsh Misplaced Pages, even though its recent changes look like this. I'm no advocate of automatically creating millions of unwatched articles as spam magnets (although Magnus's Reasonator makes a great starting point for any keen editor wanting to get started on a new article), but there's a case for creating list articles on current topics - Rihanna discography comes to mind - via a central database, rather than having to keep fans editors busy on each Misplaced Pages updating their articles every time a new album comes out. That's likely to be my next project.

You're right about the primacy of the question about En-wp's duty towards other Wikipedias, but I honestly believe in wmf:Vision and I take seriously the idea of "every single human", so the question is already answered for me. I know that others won't agree with me, but that's one topic where I don't share the fashionably cynical attitude. Cheers --RexxS (talk) 16:35, 17 June 2016 (UTC)

Data≠knowledge. There's nothing cynical about that rex, it's as bald a statement of truth as facts≠knowledge. And when that data (or those "facts") come from sub-standard or misleading sources, that's doubly the case.

Can I ask what will happen on one of the data dumps is there are two variables on WD for one field?? This must have been considered already, so how is it proposed to deal with it? – SchroCat (talk) 16:50, 17 June 2016 (UTC)

(edit conflict) Sure, I believe in the principle of "every single person on the world is given access" mantra, but I think there's a perfectly reasonable case to be made that duplicating Misplaced Pages 293 times is not the best way to go about it. For better or worse, a dozen or so languages are dominant and are becoming ever more so. One could certainly argue that "that girl in Africa who can save the lives of hundreds of thousands of people around her, but only if she's empowered with the knowledge to do so" is considerably better served if we teach her English, French or Arabic and thus give her access to the versions of Misplaced Pages in which the facts have at least a fighting chance of being comprehensive, reliably sourced and fact-checked and whatever she's looking up probably has an article (and also give her access to the sources, which are much more likely to be written in one of those languages than in her own), than by giving her a Misplaced Pages in her own language which consists of 5000–10,000 articles, most of which are unsourced and unmaintained stubs. And at least those are languages which actually have speakers; I fail to see why a penny of donor money should be spent on ensuring Misplaced Pages is available in Old Church Slavonic, Classical Chinese or Gothic. Yes, "focus on languages which are widely spoken" is cultural imperialism, but that ship sailed a century ago; regardless of whether one likes it or not, "English, Spanish, French, Chinese and Arabic are the world's lingua francas" is just a statement of fact.

Welsh Misplaced Pages isn't really a good example to use, since it can be reasonably assumed that every Welsh speaker over the age of five is also going to speak fluent English or Spanish, and will be used to flipping over to the English version of something if the Welsh version is unsatisfactory, so the grey-goo issue is a lot less pressing given that readers can easily copy references across from en-wiki. Welsh and Irish Wikipedias are also something of an outrider, as they have Robin and Alison keeping them in line; I suspect that Scots Misplaced Pages, Norfuk Wikkapedya and the like are far more representative of "minority language/dialect delete as appropriate in a place where everyone speaks English". ‑ Iridescent 17:22, 17 June 2016 (UTC)

@SchroCat: Yes, we all accept those truisms, but the data needs to used to help build articles, not as a substitute for article-building, and that's where I object to your mis-characterisation of my work. Again, let me remind you that the data which you describe as "from sub-standard or misleading sources" comes overwhelmingly from English Misplaced Pages and its peers. You should be directing your ire at the very articles you're worried about "contaminating", because that's where the data is coming from. Now if you want to complain about the poor quality of Wikidata references, that's up to you. I'll merely observe that I've been editing Wikidata to add references; I originated the idea of making an link in Wikidata-aware infoboxes to encourage other editors do the same; and I've created a tool to allow Misplaced Pages editors to preview the data and associated references available in Wikidata, in the hope that they'll spend a few minutes there improving the entries. You've made 81 edits to Wikidata. I'm sincerely pleased you added a reference for Bernard Lee's place of birth and I don't want to do anything to discourage you, but you are obviously well-aware of how much work is still needed to bring up those references to scratch. I wish I could convince you that complaining about it isn't the solution.

@Iridescent: But teaching the girl in Africa to read English and providing her with a decent quality article in her own language aren't mutually exclusive options. It's not as though I could make any impact on teaching an African to read English, anyway. Most of the cost of providing that article doesn't come from donor's money anyway; it's the cost of time and effort by many volunteers who write, translate or adapt the articles. That's a reason why I'm part of WikiProject Medicine Foundation which has a goal of providing good quality medical articles in as many languages as possible. Although I won't be asking for them in Old Church Slavonic, Classical Chinese, Gothic, or even Valerian. --RexxS (talk) 18:53, 17 June 2016 (UTC)

"You should be directing your ire at the very articles you're worried about "contaminating", because that's where the data is coming from": as far as I understand it WD harvests data from idiotboxes, where no references are held if the information is also held in the article body. My "ire" (actually despair) is directed towards the sub-standard method of factoid-gathering, and the misguided concept that somehow data=knowledge. Again, I'll ask: Can I ask what will happen on one of the smaller wikis when in one of the data dumps, there are two variables on WD for one field? – SchroCat (talk) 19:08, 17 June 2016 (UTC)

That, at least, can be answered definitively, as mediawiki:Extension:ArticlePlaceholder has started to go live on some smaller Wikipedias. This is what is currently output for John Barrymore on Haitian Misplaced Pages. ‑ Iridescent 19:16, 17 June 2016 (UTC)

Urgh.... I'm underwhelmed by the brave new dawn. I largely wrote the Barrymore FA, and I think I've just lost some of my understanding of him based on that! I see no knowledge there at all, just a series of random pieces of information that will leave Haitians going "huh?" That's an even bigger embarrassment than I feared it would be! – SchroCat (talk) 21:14, 17 June 2016 (UTC)

break 3

To be fair, those Wikidata-generated placeholder pages intentionally look like that, to make it clear they're autogenerated lists of factoids rather than anything human-written. It does serve its basic purpose, in that a Haitian coming across a mention of Barrymore somewhere and wanting to know who he was can at least find out the absolute basics, which is all these placeholder-pages are supposed to do. If bot-written articles ever get authorised, they will look more like this. (If you think that's bad, wait until this goes live; unless I'm seriously misunderstanding the proposal, this is possibly the stupidest idea I've ever seen from the WMF, as well as having the lamest name ever.) ‑ Iridescent 21:23, 17 June 2016 (UTC)

It depends what the basic purpose is, I suppose, but that doesn't provide knowledge or understanding as far as I'm concerned and is another example of the woefully misbegotten concept that factoids and the same as knowledge. It is also going to leave Haitians more confused than having no article. I suspect any non-Enlish speaking Haitian will ignore it and shove one of the written articles through Google translate, if they actually want knowledge or understanding. As to the "reasonator" version, it's disheartening to see that although they get both dates of birth lower down the page, the "lead" gives just the one. So there is a basic error of fact in the opening line, which is a concern. Still at least the lead tells us he is an "actor, stage actor and film actor", just so there is no doubt. (I'm mystified by the gallery too: I'm not sure why half those images are there, yet many of the related ones we have on Commons are not showing: most odd.) my eyes glazed over when I started to read "StrepHit": maybe I'll try again in the morning if I can raise my spirits to do so... – SchroCat (talk) 21:43, 17 June 2016 (UTC)

I've no idea where Reasonator is getting the images from—File:Louisa Lane Drew c1840-48.png appears on Barrymore's entry, for instance, but isn't in use on any page related to him on any project as far as I can see. Magnus Manske wrote the script, if you care enough to ask.

If I'm understanding StrepHit correctly, if you give it a subject it will scrape the internet looking for mentions of that subject, and parse whatever it finds into Misplaced Pages-formatted text complete with references. Anyone who thinks this is a good idea is presumably not familiar with the contents of the internet. Plus I need to repeat that I don't think I could come up with a worse name than "StrepHit" if I tried; it sounds like a battlefield medication for scarlet fever. ‑ Iridescent 21:55, 17 June 2016 (UTC)

At least Louisa Lane Drew is related: there's a pic of Robert Morley in there, which is odd, as I think they only appeared together in one 1938 film! – SchroCat (talk) 22:08, 17 June 2016 (UTC)

Am I to understand that we are going to have articles including BLPs compiled by a machine from information found anywhere on the Internet? Presumably this is at an early discussion stage given the number of issues this raises, including how the program will determine which webpages are reliable sources (and which ones are gossip rags, attack pages, or for that matter outdated mirrors of Misplaced Pages), and whether an actual person will review the material before it goes live. Is there somewhere this is being discussed? Newyorkbrad (talk) 22:01, 17 June 2016 (UTC)

Meta:Grants:IEG/StrepHit: Wikidata Statements Validation via References. To judge by meta:Grants:IEG/StrepHit: Wikidata Statements Validation via References/Timeline#Biographies, the plan appears to be to create a list of "approved biographical sources" which will then be data-mined, although since the whole thing is written in purest techie gibberish (After inspecting the set of verb lemmas, we found lots of noise, mainly caused by the default tokenization logic of the POS-tagging library we used. Therefore, we implemented our own tokenizer, to be leveraged by all modules) it's difficult to be sure. This is probably the wrong attitude to take, but I suspect if you point Greg towards that page and give it a couple of weeks he'll come back with an in-depth analysis of all its pros and cons which will be much easier to read than the proposal itself. ‑ Iridescent 22:09, 17 June 2016 (UTC)

Thanks. There already is a thread about StrepHit underway in Another Place, although so far they've only gotten as far as making fun of the name.

As I think about it, I suppose that to a very limited extent, this type of automated article-starting has assisted in creating one type of articles—that being (US) federal judges' biographies that originated with material taken from the Biographical Directory of Federal Judges. The "articles" so generated are initially hyper-choppy sub-stubs, but they are something; and they do have the advantage that the editor looking to write the real article is starting with some basic information rather than a completely blank page, which probably makes the real article more likely to be written. But the FJC database is organized in fields that are pretty much the same for every entry, and includes only basic information; I'm not sure to what extent that will extrapolate to many other sources. Newyorkbrad (talk) 22:20, 17 June 2016 (UTC)

If you really have an urge to dredge up ancient history, there was an attempt to do something similar once before, using a bot to create auto-generated stub biographies from the persondata on German Misplaced Pages where an article existed there but not here. It was not universally welcomed, and virtually everything created via the process ended up being deleted. ‑ Iridescent 22:25, 17 June 2016 (UTC)

(edit conflict) That's the bit I'm interested in, Brad. Getting a start on an article is the single best use of any of the databases we have access to. It's true that not all databases as as well-organised as the FJC, but there's the challenge: to create a system that can deal with a multitude of different collections of information and provide the aspiring article writer with a decent enough start for them to actually write something more than a bare collection of facts. It's easy for the snobs to sneer at stubs and start-class articles, but every article had to start somewhere. Of course, the obvious elephant-in-the-room is the lack of cast-iron reliable sourcing, unless the database itself can be considered a RS. That's one of the the reasons why I regard Wikidata as a failing project now - it needed to have sourcing "baked-in" from the very start. I could write Lua modules to import information from Wikidata only when it is referenced to something better than "imported from the English Misplaced Pages", but we'd end up with very little to import. If someone can come up with some good idea for attaching references to all of those unsourced statements, I might be prepared to change my mind, but I'm not expecting that any time soon. The best I have to offer right now is to use Wikidata in infoboxes wherever possible and try to encourage editors to see the associated Wikidata entry as part of the article, so that improving the quality of the Wikidata would be seen as improving the quality of the article, but that would require a real cultural shift. If we insisted that the Wikidata corresponding to an article had the be the same quality as that of the English Misplaced Pages article in order for it to become an FA, we'd soon have the badge-collectors falling over themselves to improve the referencing. One can only dream ... --RexxS (talk) 22:47, 17 June 2016 (UTC)

"All unsourced statements are deleted after six months"; I'm sure a bot could be written to do that (since, unlike on Misplaced Pages, on Wikidata it's always clear what fact a source is intended to support). Yes, people would scream blue murder at the start when the database size suddenly shrinks by 75%, but the culture there could do with some shock treatment, and unless it shakes off its reputation as Misplaced Pages's idiot stepchild SchroCat's attitude is always going to be the prevailing one. ‑ Iridescent 22:55, 17 June 2016 (UTC)

"Snobs"? "Badge collectors"? And you're trying to win people over to your side by insulting swathes of editors who spend their time trying to improve the encyclopaedia? Stacking barbed wire comments around things does not build bridges rex and isn't going to help people have the open minds you want them to. – SchroCat (talk) 08:15, 18 June 2016 (UTC)

Exactly. You take the attitude that only you are trying to improve the encyclopedia, and can't accept that others might be trying to do the same in another way. Unlike you, I'm prepared to accept that different people can make improvements that aren't just related to increasing the number of stars on their user page. You want to build bridges? Start by acknowledging that the contributions of others aren't automatically inferior to your own. --RexxS (talk) 11:47, 18 June 2016 (UTC)

Exactly what, rex? I'm afraid you have totally misrepresented my viewpoint there. I have never said that you are not trying to improve the encyclopaedia (show me the diff where I have clearly said that). You can also dig out the diff where I've expressed the opinion that the contributions to others are inferior to mine while you're about it (it's not what I believe, so good luck looking). It may help to try and find some common ground rather than drive people apart, which is what you're doing here, or to try and focus on your own arguments, rather than try and badly double guess my thoughts on my fellow editors. My view is—has always been—a lot less divisive than you are being here. Yes, disseminating knowledge is what we are about, but my view on what constitutes knowledge (and even better, understanding) differ considerably from the factoid driven approach. Go ahead, sneer at the people who enjoy writing articles, but it reflects rather badly on you rather than them I think, and is unlikely to win many over to your line of thinking. – SchroCat (talk) 12:08, 18 June 2016 (UTC)

break 4

Yeah, I have to agree with SchroCat there. I personally don't particularly like the badge collecting mentality on Misplaced Pages—I think the badges-and-barnstars culture can sometimes deflect editors away from what's necessary in favour of what will give them shiny barnstars or a higher placement on WP:WBE—but I recognise that this is a minority view and I don't see what's to be gained by denigrating those who don't subscribe to it. On a skim of the 23 signatories to the statement that "infoboxes are integral to the encyclopaedia" on the assumption that those are likely to be Wikidata's core supporters, and disregarding the six who are indefblocked and/or have blanked their user pages, 13 of 17 have some kind of high-score-table or row of shiny topicons on their user page. The MMORPG mentality is so heavily embedded in Misplaced Pages that you're never going to shift it, and sniping at those who participate in it is just going to alienate the people whom you need to convince that Wikidata isn't some kind of hostile force parking its tanks on Misplaced Pages's lawn. ‑ Iridescent 10:11, 18 June 2016 (UTC)
Note: intervening comments have separated this reply from the comment being replied to; the statement to which I'm agreeing is the one beginning ""Snobs"? "Badge collectors"?", not the later comment, regarding which I don't agree with SchroCat. I believe there are many circumstances when the factoid-driven approach is the correct one, most obviously languages where there's unlikely to be much interest in a topic and thus doing a full translation is not a good use of time, but where we still want to cater for those few people who do have an interest. (As a concrete example, very few Haitians are likely to ever want to read about Adam Gilchrist, and translating this 8500-word article into Haitian is not going to be a sensible use of anyone's time, but having the auto-generated entry means that some Haitian who sees a reference to him somewhere can at least find out at a glance that it's an Aussie cricketer who's being discussed. And yes, I know I'm picking on the Haitians here, but that's because Haitian is similar enough to languages I at least vaguely understand that I can tell at a glance what the information being imported actually says.) ‑ Iridescent 15:31, 19 June 2016 (UTC)

Precious anniversary

Three years ago ...

quality standards
... you were recipient no. 517 of Precious, a prize of QAI!

--Gerda Arendt (talk) 07:25, 16 June 2016 (UTC)

"Quality standards"? That doesn't sound like me. ‑ Iridescent 13:43, 16 June 2016 (UTC)

Sounds like me four years ago ;) - today I'd rather praise the precision of illustration above, but four years ago, I didn't even know what arbcom means, nor had I worked on Kafka, --Gerda Arendt (talk) 17:38, 16 June 2016 (UTC)

The World At Mind's End

Hi Iridescent,

I see that you deleted a page I created last week titled "The World At Mind's End", which is an LP/CD by the UK band Skywhale. The reason for deletion was notability. The A9 tag said that notability requires, among other things, a current wikipedia article about the artist who made the recording. Well, I also started a draft for the band Skywhale last week which is in the wikipedia review process, so am requesting that the LP/CD article be re-instated (if possible) until wikipedia has reviewed the Skywhale article, which links to the LP/CD article.

Thanks Cjcooper (talk) 00:26, 19 June 2016 (UTC)

Restored to your userspace in User:Cjcooper/The World At Mind's End for you to work on it; once you feel it's in a fit state, use the "Move" tab at the top of the page to move it back to its original title. Bear in mind that Misplaced Pages articles need to be sourced to reliable sources, and that the use of discogs.com as a source is explicitly forbidden. If you're having trouble finding sources, I'd strongly recommend posting at Misplaced Pages talk:WikiProject Progressive Rock, as there's a fairly good chance one of the participants there will have a big stack of '70s music magazines in their loft. ‑ Iridescent 07:32, 19 June 2016 (UTC)

About me

How do I turn myself around? I can't communicate well with others]. I want to be useful in ITN, but I've not produced good results. I want to raise issues, but people pity me a lot. I guess I am an issue, huh? I've come here to discuss myself here, not at ITN, which would drive away the purpose. I need help on being useful and influential at ITN, but I am uncertain on what it takes to be useful there. --George Ho (talk) 17:15, 19 June 2016 (UTC)

Per most of the comments you're ignoring on the AN thread you recently started on exactly the same topic, there comes a point when mentoring is no longer appropriate. If after a decade on Misplaced Pages, 80,000 edits, a set of unblock conditions which spelled out in extreme detail exactly what you were doing wrong and which issues of your conduct you needed to address, and Begoon painstakingly talking you through the issues, you're still not getting it, I'm not sure what exactly you're expecting anyone else to say. I have literally never seen you comment anywhere and not immediately launch into extreme WP:BATTLEGROUND behaviour the moment someone disagrees with you, regardless of how trivial the issue in question is. (For the benefit of the TPWs, this threat is prompted by George taking on all comers over the issue of whether WP:In the news/Candidates should use level 3 or level 4 headers on its subsections.)

To be blunt, if for any situation in which your participation isn't mandatory you need help on being useful and influential, my advice would be that that situation is not one in which you should be participating. I don't understand why you would deliberately aim to participate in any area where you don't consider yourself competent when you're not obliged to do so, and it's not as if WP:ITN is exactly an area of high importance which needs all the help it can get.

While it ended badly (to say the least), you might want to take a long hard look at User:Mattisse/Plan. Mattisse was another editor whose undoubted abilities in certain areas were offset by her tendency to pick fights for no apparent reason, and the rules drawn up to try to prevent her being banned are likely to be very similar to the rules imposed on you if and when someone's patience snaps and you're taken in front of an ANI ban discussion or to Arbcom. Even if you manage to avoid that, they're a good set of rules to follow in general, anyway. (Plus, you might also want to bear in mind what happened to Mattisse as a salutary lesson that even the most valuable and productive editors will be kicked out if dealing with their disruption becomes too much of a timesink.)

I'd also really strongly advise taking the "I'm considering retirement, please beg me to stay" header off your talkpage. It means that the first thing any visitor to your talkpage sees is a garish multicoloured banner which most (if not all) readers will interpret as overinflated self-importance or trolling for attention. If you want to stop editing Misplaced Pages, just stop editing Misplaced Pages; I think I speak for virtually everyone in saying that editors who constantly retire and unretire, posting variations of "without me you're nothing" in the process, are not a group which is generally held in high esteem. ‑ Iridescent 17:43, 19 June 2016 (UTC)

I did read the AN, and I didn't want to comment further there. I'm just "out of control" guy, which... saddens me. I didn't mean to get myself into messes that I have made. I guess I did. Now I'm feeling terrible about being a nuisance, but I don't which part I did. If at WT:ITN, I don't know whether I should feel guilty on those. If I'm unsuccessful at this time, then I guess I should owe people apologies and then try to make up to everyone, but I don't know how. If successful, then what should I do? George Ho (talk) 18:02, 19 June 2016 (UTC)

For the record, I cried myself all the time whenever I felt regret about walking on the desert path. And I don't know how to get out of that path without help (or self-help). If this one doesn't prove that I'm trying to make up for my errors, maybe I don't know which else is, but I'll try to prove to you somehow. George Ho (talk) 18:14, 19 June 2016 (UTC)

Re-reading the unblocking conditions, I must say that most of conditions have changed. User talk:George Ho/Mentorship discussions is inactive at the moment (marked "historical"). I now have only one mentor left. Right now, I'm in deep trouble. If help is unwarranted, I guess I'll accept that and need advice instead. As for listening to others, why can't people say "influence" instead? "Listening" sounds vague to me. I did listen, but I am not influenced by opposing sides... well, I hate to admit that the opposing sides may have a point on one thing or another. --George Ho (talk) 18:40, 19 June 2016 (UTC)

I am complained about being a poor listener. Therefore, I want to find a book that can help me effectively listen to people. I am torn between this and that. George Ho (talk) 18:49, 19 June 2016 (UTC)

Bluntly, just stick to improving articles. The Rambling Man (talk) 20:47, 19 June 2016 (UTC)

What TRM said—if you're not confident of your own competencies in particular areas, why are you getting involved in them in the first place, let alone trying to bully those people who are experienced in those areas, particularly when you've been warned about your aggressiveness and refusal to accept consensus for quite literally a matter of years? Per this thread you link to above, I get the impression you think Misplaced Pages editing—and writing in general—can be boiled down to a set of mechanistic, predictable rules, and the world really doesn't work like that. (Regarding self-help books, don't expect them to be much help in the context of Misplaced Pages. Misplaced Pages is a global project, and social interaction is very culture-specific; if you followed Carnegie's "Six Ways to Make People Like You" in Britain you'd be liable to be punched in the face.) ‑ Iridescent 21:13, 19 June 2016 (UTC)

Now this? Are you really trying to canvass an ANI or Arbcom case against me George Ho? Really? The Rambling Man (talk) 21:20, 19 June 2016 (UTC)

That moment where you're checking your watchlist and you think "Whoa, me?"

That. It was easily 50/50... Glad we're making a start on resolving this particular issue. Cheers, The Rambling Man (talk) 20:10, 19 June 2016 (UTC)

The more I think about it, the more I'm coming to feel that deprecating the Main Page and replacing it with www.wikipedia.org is starting to look like the most sensible option. Adding up the total editor time wasted in pointless discussions about TFA, OTD, ITN, DYK and TFP must be at "if all this effort had been directed towards something more useful we'd have cities on the moon" levels by now. Google has had a blank main page for well over a decade and they appear to be surviving. ‑ Iridescent 20:20, 19 June 2016 (UTC)

Indeed, retro-engineering it back to something useful would be just fine for me. Ironically though, worst thing about Misplaced Pages (beyond the idiots and trolls) is the search feature. Unless you get the subject title bang on, forget it. Google solved that problem way back. Nice doing business with you. Bon Sunday. The Rambling Man (talk) 20:38, 19 June 2016 (UTC)

Some of us love working on such a cutting edge internet project. Zoom!! Martinevans123 (talk) 21:07, 19 June 2016 (UTC)