User talk:Jimbo Wales

(Manual archive list)

The future of Wikipedia

Hello Jimbo. As someone who is now dipping their toes into Wikipedia can I ask you how you think it will evolve in the future. As good as it is (and I am impressed by the depth and accuracy of many of the articles) do you see wikipedia continuing for a long time to come yet? As I understand it there is competition out there though not with the same success, but there is bound to be another wiki out there now or in the future that will one day compete. What do you think can be done to stay ahead of the competition? --Jonty Monty (talk) 09:44, 9 October 2012 (UTC)

Ps did you enjoy the bagpipes? :). --Jonty Monty (talk) 09:50, 9 October 2012 (UTC)

• I'll leave space, above, for Jimbo to answer, but Jimbo has made many statements about the future of Wikipedia. He has indicated that point-and-click updates to articles, as WYSIWYG edits, are likely to become more popular among a wide range of editors. When asked about millions of articles, he noted the major articles could be translated between the expanding 260 languages, and Wikipedia would have far more than 100 million articles. Also, he has noted that he intends to stay involved with Wikipedia for many years to come. -Wikid77 (talk) 20:53, 12 October, revised 12:26, 13 October 2012 (UTC)

Personally I see the future of Wikipedia being a fairly bleak place but I sincerely hope I am wrong and I continue to edit in the hope that I am. Every day we make it harder for people to edit, particularly those that are not intimately familiar with Wikipedia's thousands of rules, policies and guidelines. We block more and more users and IP's for smaller infractions and develop more tools to restrict editing like Pending Changes and ask ourselves why more people aren't editing. Why would they? It is human nature that if something is hard to do, we will do something else so as we implement new ways to make it harder for people to contribute by protecting templates, articles and even entire namespaces we damage the foundation of why Wikipedia was established and we erode the concept of "Anyone can edit". We need to get back to our beginnings of Assuming good faith and foster a spirit of cooperation and trust rather than implement new ways of showing our editors, future editors and readers how much we don't trust them and how we think that we need to protect Wikipedia and its articles and content from the people we should be hoping will be positive contributors. Kumioko (talk) 21:49, 13 October 2012 (UTC)

• Improvements everywhere as surveyed users said would edit if knew where to help: In one of those WMF surveys, the least worry of editors was the "editing technology" (9% concern). However, more users (41%) said they would edit more if they knew "where to help" and "avoid conflicts" or arguments with other editors. Our coordination tools have had "growing pains" due to time to learn what is effective, while some over-sized templates (or too-small template limits) have hindered quick tasking for editors to follow (1.6 million articles reformat 2-4x slower than needed). The monthly WP:GOCE drives have been fantastic to fix listed articles, averaging "99 copy-edit errors" each. However, I think we could switch other tasks to 2-week edit-drives, to gain people who could not commit for a whole month, and have more short drives. The slow Javascript tool-buttons or over-size templates (I mean S-L-O-W) have been disatrous to "quick-moving people" who do not want to wait "37 seconds" to edit-preview a major article. Hence, the new Template:Cite_quick can show that same article in 10 seconds, and allow 900 cites, not just 400. Tests with the new Lua script cite modules are showing similar results (scaled up because the test2 wiki seems 2x slower than enwiki). Plus, Lua modules can be more complex, and analyze article text faster than ever before, to allow modules to "suggest" other improvements during editing. The edit-window tool buttons are being redone faster. Meanwhile, some bots are getting better: DASHBot is updating citations to insert the archive-URL links, to have fewer deadlinks in future months. The copyvio bots are listing about 20 articles per day, some with "close paraphrase" text. Ironically, that was another area hindered by over-size templates, but the common link-menu Template:La was just updated 40% smaller to allow collaboration pages to be larger without hitting the too-small template limits. As more people discuss templates, from a scientific view, then template limits could be raised to reasonable limits. Just recently, the NewPP preprocessor limits were changed because some French Wikipedia templates tried to implement a huge database (knowledge base) in templates (perhaps like {cite_doi}'s 6,000 database subtemplates), and the French template crashed an Apache server (so the template limits were changed). I guess my point is that WP is coming out of the technology "dark ages" where better techniques are being used for templates, plus smart Lua modules, and WYSIWYG editing for newcomers. And those technologies should quicken our coordination tools, without S-L-O-W editing nor asking people to edit during whole one-month drives because the tools were too slow to fix much in two weeks. We could not seriously ask people to help edit major articles if everything stayed as slow as in recent months. Faster editing, faster templates, and smarter tools were needed, and they are being developed in the next 3 months. -Wikid77 (talk) 21:25, 14 October 2012 (UTC)

For-profit projects managed from within Wikipedia

Jimbo, we currently have at least two for-profit projects coordinated by Wikimedia consultants in Wikipedia project space:

• Wikipedia:GLAM/GibraltarpediA
• Wikipedia:Communicate OER

Roger Bamkin is a former trustee of Wikimedia UK, and Pete Forsyth is a former WMF staff member. Both are involved in these projects with their own for-profit consultancy businesses (Victuallers Ltd. and Wiki Strategies, respectively).

We have discussed Gibraltarpedia at length. Wiki Strategies' Communicate OER seems to be less beset by conflicts of interest, and a little more transparent (especially after this addition earlier today), but I still wonder if there should not be some kind of approval process and transparency ground rules for for-profit projects wanting to have a page in project space (i.e. a page whose name begins with "Wikipedia:" followed by the project name).

This could be something for the Affiliations Committee, for example. Are we not otherwise entering horribly grey areas, with projects like Pete's at one end, and (for argument's sake) a Nestlé-funded drive to improve Wikipedia's coverage of Nestlé products at the other, coordinated from a page like Wikipedia:Communicate Nestlé products? Or is this the future of Wikipedia? I am not sure many volunteers will want to contribute their time for free, while others are getting paid. JN466 12:38, 9 October 2012 (UTC)

PS: Discussions are currently underway to resume the publication of Gibraltar DYK hooks on the main page: Wikipedia_talk:Did_you_know#Outstanding_Gibraltar_nominations. JN466 13:07, 9 October 2012 (UTC)

It sounds like you've done a lot of hard work to address a very real problem, but I don't see a solution coming: one after another, the proposals come in to make paid editing subject to draconian standards or impossible approvals. I doubt if any will ever achieve majority, let alone consensus; they ignore the fact that if you make the procedure draconian, it will be bypassed. We should probably settle for notification to the arbcom-l list and leave it at that. However, you put your finger on what will be the most difficult angle: the interaction between editors who are getting paid, and those who are not getting paid.--Wehwalt (talk) 13:23, 9 October 2012 (UTC)

Thanks. The difference between covert operations and overt ones is that Wikipedia can condemn and dissociate itself from the former, and not from the latter. It is a question of reputation management. Open collaboration with commercial, social and political entities will be potentially disastrous unless these interests are ones that enjoy very, very broad support. There is also another angle: it's the interaction between the editors who are allowed to have a recruitment page for their paid project on-wiki, and those who are not allowed to have such a page. The current approach looks a lot like "If you are an ex-staffer, or an ex-chapter trustee, or have lots of GLAM friends, then you can advertise your for-profit project on-wiki. But if you are Joe Bloggs, we will delete your page and ban you." The criterion should not be who you know, but what kind of project you're running, and which masters you serve. And that, as well as the payment structure, should be very transparent on-wiki, on every page related to such a project; for the benefit of both editors and the general public. I am slowly seeing a dystopian vision of Wikipedia in 20 years, and it looks a lot like this: [1] JN466 14:07, 9 October 2012 (UTC)

Wehwalt, I think it is important to distinguish between straightforward paid editing and the more subtle practices of using Wikipedia to make money. Your comments may have some relevance to the former, but it is the latter that under discussion here. It is not surprising that people who are committed to the aims of Wikipedia find themselves drawn to areas where they can be paid while also advancing those goals. The difficulty is that when one's livelihood is based on Wikipedia, it is all too easy for one to rationalize conflicts of interest. After all, one has to eat. It may not be possible, or even desirable, to prevent these kinds of projects, but there should be a bright line drawn when it comes to support from within the projects. Delicious carbuncle (talk) 20:56, 9 October 2012 (UTC)

At the present time no one makes a living off Wikipedia, some WMF staff and Jimbo (somewhat) exempted. That being said, if someone was, I would expect them to act in an ethical manner if only to protect themselves. There is an old expression, you don't crap where you eat. I expect that a successful professional Wikipedian would be very careful about his standing on the project and very careful about the work that he did. After all, if it is for pay, someone may check to make sure they got their money's worth, if reviewers don't.--Wehwalt (talk) 22:39, 9 October 2012 (UTC)

Not so. I advised Pete of this conversation, and he can of course speak for himself, but he is actually on public record regarding this, telling The Signpost this April that "All my income relates to Wikipedia; it's my day job. I am comfortable with my income, but I won't be rich any time soon." I agree he is circumspect, and of course he continues to volunteer here like everyone else and is in good standing, but I am fairly sure that he is not unique in earning his living here – it's just that others like https://www.wiki-pr.com/ who work with throwaway sockpuppet accounts are far less open about it. JN466 23:31, 9 October 2012 (UTC)

Wehwalt you say "someone may check to make sure they got their money's worth". Given the choice between the paying client (who wants results if some kind) and volunteer colleagues (who have no particular expectations), which do you think most people will choose to please? I get the impression that you are thinking in terms of paid editing of articles. In the case of Gibraltarpedia, for example, the benefit was clearly intended to come in the form of positive media coverage (following the Monmouthpedia model). The ways in which Wikipedia were used as a means to this end are obvious, but there is no direct paid editing. Delicious carbuncle (talk) 02:50, 10 October 2012 (UTC)

JN466 writes: "I am not sure many volunteers will want to contribute their time for free, while others are getting paid."

I can think of a place where this happens all the time: open-source software. In the world of software, the users of the software benefit, so often have a clear motivation for improving the software. The Japanese telco NTT use and have contributed lots of code to PostgreSQL. The developers who work for NTT are probably (gasp!) getting paid to contribute. And yet, so do unpaid volunteers. This happens routinely in open source software. The contributors to the Linux kernel: lots of them work for big companies like IBM.

But in open source, there seem to have been ways for people to continue running community-directed projects with some paid contributors. The Subversion project was managed by a company—CollabNet—for a few years before eventually becoming an Apache project in 2009. While it was being managed by CollabNet, the community leads instituted a firewalling policy that said that new employees of CollabNet would have to earn commit access and other community privileges in the same way; they couldn't avoid going through the same processes community members would go through.

The problem comes not with corporate contributors or even corporate owners of projects, but when those corporate owners end up being bad open source citizens or, to use wiki parlance, they are dicks to the community. The reason LibreOffice, MariaDB and Jenkins exist is because quite a lot of developers aren't happy with Oracle. (As a Java developer, the continued idiocy over software patents, OpenJDK and the now deceased Apache Harmony frequently gives me cause for bad nightmares.)

It seems pretty simple to me that the line shouldn't be about money. I've suggested this thought experiment repeatedly: if Bill Gates suddenly decided to fund a programme to pay a group of Wikipedia editors to take a year off from work, live in subsidised housing and be paid a living wage to contribute to writing, translating and improving Wikipedia and other Wikimedia projects, that'd be pretty cool. They wouldn't be "getting rich" off editing Wikipedia (we all know that in the money/power/sex stakes, the historians, lexicographers and academic book authors are only a few rungs below Goldman Sachs investment wizards), there would be strict conflict of interest limitations so they'd not be able to write about the programme itself, nor about Gates or Microsoft or POV pushing, it'd all have to be universally agreed good things like science and history and medicine and so on. If your interest is improving the encyclopedia, and these people are improving the encyclopedia and not, oh, "profiteering!" or something, it shouldn't be a big deal.

That's not to say the Foundation ought to pay for it, or the chapters. No thanks. But if a billionaire philanthropist decided to, I can't honestly see a big problem with that.

As for outreach projects and Wikipedians in Residence? I do think it's probably time for some actual discussion about them. I mean, the easy thing to start with is: the Wikipedians in Residence we have had with GLAM institutions seem to do a pretty good job. GLAM seems to be a pretty damn good thing. There is a feeling among some that the GLAM people are separate from the community. This is unfortunately true, and we need to fix this gap. That the Smithsonian and the British Library and other such institutions want to work with Wikipedia shows we are doing something right. And we should work with them because we have very similar goals. The Smithsonian's mission very simply is "The increase and diffusion of knowledge". That seems eminently compatible with what we're trying to do on the Wikimedia projects.

While it might be quite difficult to work out the details, I think that we need to find a way to ensure that the community is involved with and aware of the good and productive GLAM outreach work that's going on, and making sure that the GLAM community isn't a separate thing from the community. It seems to me the problems with things like Gibraltarpedia and other things stem from not enough talking to the community. I think, in general, if we can get over the whole "paid editing is teh evil!" thing, productive outreach projects that are based on a shared mutual goal of sharing knowledge should have nothing to fear from community discussion.

Disclaimer: I may have had a few sandwiches bought for me by cultural institutions and I may have written some Wikipedia articles about objects contained therein. —Tom Morris (talk) 08:03, 10 October 2012 (UTC)

Thanks, I appreciate this and will mull over it. Just one brief comment on Gibraltarpedia and QRpedia: the big problem is that what was being "sold" there was Wikipedia's SEO value (see this screenshot of a Wikimedia UK slide entitled "Improving a city's Google position on the web", and then watch the entire video). This is something Pete Forsyth categorically says he will not do. Pete makes it quite clear to clients that he is not in the SEO business, and that if his clients are even thinking of SEO, or asking for SEO metrics, their approach to Wikipedia is probably inappropriate. You do not see Pete at DYK, trying to get his project's articles on the main page. The other problem is that QRpedia/Gibraltarpedia were advertised as "a phenomenally cheap, and very, very imaginative way to absolutely energize a city and put a city on the map" (time code 17:41), or "marketing ... done at the lowest possible cost", and "marketing Gibraltar as a tourist product through Wikipedia". At that point, when paid consultants make low cost a prime selling point for their project, the whole thing begins to look like cynical exploitation of volunteer labour, in return for kiddie awards like barnstars and DYK credits. I would go so far as to say that any suggestion of SEO benefits, and any promise of commercial or marketing value in project proposals, agreements, contracts, MOUs etc., should instantly disqualify a project from consideration. The rationale for any Wikipedia project should be educational value, and based on the common good, rather than a promise of competitive advantage. JN466 12:47, 10 October 2012 (UTC)

"I am not sure many volunteers will want to contribute their time for free, while others are getting paid." - Only if your volunteers are jealous dicks, I hope people contribute to Wikipedia because they want to make Wikipedia a better place, whether or not anyone is making money from it doesn't factor into it at all.

Some users seem more worried about the sales pitch and the perception than the actual work. Whether or not SEO is mentioned doesn't stop it from being true - if you energise a community to create Wikipedia articles on subjects on local attractions, it's going to look good on Google. This is obvious, should this fact be omitted from the record and instead communicated through nods and winks?

What's important is that organisations realise that their best interests are to act in Wikipedia's best interests. That ad-copy isn't in Wikipedia's interests, but energising the community is. Whether that's through Commons donations, a Wikipedian in Residence, or just opening yourselves up to editors, it's up to you. I have a really hard time believing that getting articles like First Siege of Gibraltar (1309) onto DYK is somehow harming the project. - hahnchen 13:55, 10 October 2012 (UTC)

I agree. What I don't think will happen is the one-dimensional responses like "our competition has a vanity article, we need one just like that" once someone above the level of a summer intern spends five minutes learning about Wikipedia. If a company wanted to contract with me for my expenses and possibly a small honorarium while supplying me with records, resources, and access to databases, I'd have a go at their article if it interested me.--Wehwalt (talk) 15:15, 10 October 2012 (UTC)

JN466, I broadly agree with that, but with one big proviso. Museums and cultural institutions are often measured on things like website visitor statistics as a metric for "engagement". Sadly, just like everything from healthcare to policing, in the UK, "target culture" rules. Funding for cultural institutions is often dependent on their meeting "targets". (Academics now have to meet 'engagement' metrics as part of the Research Assessment Exercise.)

And Wikipedia can help with that. You do some scientific research, and if there's a Wikipedia article on the topic, that's better public engagement than if there isn't. Wikipedia can help scientific, academic and cultural bodies better meet their targets and goals for engagement and so on. But that is only really problematic if the whole point of the institution differs significantly from that of Wikipedia. If the British Museum want to increase engagement on African art, say, and Wikipedia is the way to do that... and as a result, they get more money from the government for having met their target, is that such a bad thing? They'll spend it on some dastardly thing like preserving historical artefacts.

Telling the British Museum "hey, you know that Rosetta Stone. We get pretty good SEO on that. Just saying." is rather a different matter from the Gibraltarpedia situation which, well, wherever the line is, it has danced back and forth over it a few times. —Tom Morris (talk) 15:48, 10 October 2012 (UTC)

"I am not sure many volunteers will want to contribute their time for free, while others are getting paid." My income is not solely based on Wikipedia, but in addition to traditional marketing/writing work, I help companies contribute with a COI.

What I find is the greatest misconception is that helping a company contribute with a COI is just like volunteer work, but with cash. The truth is companies have dozens of stakeholders involved that all need to be educated on the rules, and it's an immense amount of work, pressure and delicate consulting.

Telling companies that their version of the truth is incorrect, that they must accept factual errors because they are not reliable, that their CEO is not notable or that what they think is neutral is not - it's all tough work. Meanwhile, contributing as a volunteer is fun and easy. It is not a good comparison to make. Corporate 18:42, 10 October 2012 (UTC)

I am roughly in agreement with you, Tom. But the sales pitch should be, "Wikipedia can help spread important knowledge". It should not be "Wikipedia can help you get one over on the museum next door". And it definitely should not be "Wikipedia can bring you tourist dollars that would otherwise go elsewhere." Getting people to learn about the Rosetta Stone is one thing, and consistent with the WMF mission. Getting them to spend money in Gibraltar, or any other "product", is another.

Speaking of Gibraltarpedia, Gibraltar has had two hooks on the main page these past couple of days (one running right now), with another ten or so lined up for the next couple of weeks. The product placement continues. JN466 12:21, 13 October 2012 (UTC)

• Speaking of projects that promote for-profit causes, what do you think of Wikipedia:GLEE? For example, User:BlueMoonset is vastly outperforming all of Gibraltar (so far as I can tell) in GA's (30) and has over half their DYK count (21), focusing exclusively on some TV show I never heard of. Wnt (talk) 22:49, 13 October 2012 (UTC)

Unless you believe that BlueMoonset or someone else associated with the Glee wikiproject is profiting from it, I'm not sure what point you are trying to make here, Wnt. I haven't heard anyone suggesting that wikiprojects are inherently bad or that wikiprojects dealing with for-profit subjects do not belong here. I believe the main issue is the distortion of Wikipedia's original goals for one's own interests. I see no indication that the Glee wikiproject is doing this. Do you? Delicious carbuncle (talk) 23:19, 13 October 2012 (UTC)

Well, why is it "spam" to cover a place in good detail, but ordinary editing to cover a current TV series in exhaustive detail? I still kind of wish that all the countries and towns would fight to get the most exhaustive coverage of their regions in Wikipedia, even if that means providing some indirect financial assistance to editors, but I'm not at all sure I'd say that about TV series. Wnt (talk) 23:53, 13 October 2012 (UTC)

Gibraltarpedia is not "providing some indirect financial assistance to editors". The people involved with Gibraltarpedia are not being paid to edit, they are being paid for other services as "consultants". It is that extra-Wikipedia part that has caused concern about for-profit enterprises that leverage Wikipedia's volunteer base for someone else's benefit. I guess you haven't been following the discussions. Delicious carbuncle (talk) 02:10, 14 October 2012 (UTC)

Biggest threat to innovation...

...is information technology. A very important read, confirming what we've known all along. Viriditas (talk) 06:39, 10 October 2012 (UTC)

Actually, it looks like 'lack of leadership' is the threat. Alanscottwalker (talk) 22:55, 10 October 2012 (UTC)

Well, in the context of the article, corporate IT doesn't "lead" on the innovation front, nor should they. They point is that the people they are serving and supporting know what works best for their business. So the leadership in this particular case needs to come from those at the bottom, not the top. This is a difficult concept for many to follow due to the hierarchical nature of many corporations. But, this is the reason IBM didn't invent the personal computer. Viriditas (talk) 00:34, 11 October 2012 (UTC)

I don't think any fan of Dilbert will see anything surprising here. Looie496 (talk) 01:45, 11 October 2012 (UTC)

PacBell wasn't as bad as some other companies when Adams was working there. In some respects, they were actually more innovative than others. I can't speak for Adams, of course, but some of the problems might have stemmed from an older management demographic that was transitioning to retirement. At this point, we've really moved beyond the corporate IT model. The MIS, top-down management approach is no longer relevant nor helpful and it is time to bury it. The bottom-up model exemplified by the Quantified Self movement should be the approach we take on Wikipedia. Start with getting rid of the admin model. Viriditas (talk) 09:10, 11 October 2012 (UTC)

Consider limits and mandatory breaks

• Consider edit-limits, term limits and mandatory wikibreaks: In some articles, the text is slowly slanted by hundreds of tiny edits to remove text phrases, or to slant text "one word at a time". If the system controlled per-article edit-limits, then such "death by 500 tiny cuts" could be thwarted. I have proposed that only suspected editors would be monitored, and any other editor could nominate an editor for edit-limit controls. That would make the concept feasible, rather than restrict the edit-count of every editor at every page. At some point, a person with "600" edits to one article needs to face a limit. Since 2006, the Swedish WP has had those 1-year term limits for admins, to be re-elected during one of 4 quarterly re-elections during the year. Wikibreaks, even for 1 week, can help people to gain a wider perspective. Other types of limits should be considered. -Wikid77 (talk) 21:13, 12 October 2012 (UTC)

  There are a lot of words I could use to describe this proposal. "Stupid" is perhaps the most polite. I especially like the bad faith assumption that people making a lot of edits to an article (e.g.: to maintain one with an ongoing history) are "slowly slanting" it - even if you try to soften your bad faith by claiming only some editors would be "monitored". Term limits for admins is pretty much WP:PEREN, and you have no right to dictate to anyone when they can and cannot edit. Resolute 00:11, 13 October 2012 (UTC)

  My thoughts exactly - not to mention it violates the prime directive of Wikipedia, i.e., anyone can edit. Mugginsx (talk) 15:50, 13 October 2012 (UTC)

Good god, I hope that's not the prime directive. I thought it was something about creating an encyclopedia which assembles the whole of human knowledge for the free use of humanity. Has the process obscured the product? OF COURSE not everyone can edit, see WP:COMPETENCE for an essay that is on the money... Carrite (talk) 16:25, 14 October 2012 (UTC)

"Imagine a world in which every single person on the planet is given free access to the sum of all human knowledge. That's what we're doing." is the closest thing we have to a Prime Directive. "Everyone can edit" is a means to that end, and very very very far down the list.--Jimbo Wales (talk) 22:49, 14 October 2012 (UTC)