The Article Feedback Tool v.5 and the current "Talk" tab are for discussing *the editing of the current article*, not for discussing *the topic represented by the current article*, although I think these two goals can coexist in a single discussion area. On Mon, Jan 23, 2012 at 5:51 AM, Tom Morris <tom(a)tommorris.org> wrote:

On 22 January 2012 21:43, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote: Similar to the "Opinions" tab on Wikinews. Could be interesting. Would need to be plausibly useful to the encyclopedia to be accepted by the communities - what was the rationale for the "opinions" tab on Wikinews? - d.

There is already a discussion page attached to every article. It's for discussing the article, though, rather than its topic. While we are more than a conventional encyclopedia, we are still an encyclopaedia and I don't think we should add job and product adverts to our articles. If people want to make friends, they can go to Facebook. If people want to find or contribute encyclopedic information (and, perhaps, make some friends along the way as an added bonus) then they should come to us. On Jan 22, 2012 9:43 PM, "Yao Ziyuan" <yaoziyuan(a)gmail.com> wrote:

On Mon, Jan 23, 2012 at 6:12 AM, Tom Morris <tom(a)tommorris.org> wrote: I forgot to mention, each user can show/hide the proposed comment section under every Wikipedia article. If you're not logged in, the comment section can be hidden by default.

On Mon, Jan 23, 2012 at 6:31 AM, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote: This merit is even more evident when the topic is very specialized, e.g. [[Phonological history of English low back vowels]]. I bet there isn't a forum on the Web dedicated to this very specialized topic, and even if there is one, it can be very hard to find it with Google (because it may use a slightly different term to describe this topic). However, if every Wikipedia article has a corresponding comment section (actually a forum), people with this special interest ("phonological history of English low back vowels") will know where to go to find each other, forming a "special interest group" (SIG).

On Mon, Jan 23, 2012 at 6:40 AM, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote: So this can mean very much for scientific research. For example, imagine if there are two mathematicians in the world interested in the same, very deep math concept, but they don't know each other. How do we let them meet and collaborate with each other? With a comment section under that math concept's Wikipedia article. Take another example. Imagine there are two medical researchers pursuing the same, very novel but very rarely known approach to a major disease, but they don't know each other. How do we let them meet and collaborate with each other? With a comment section under that approach's Wikipedia article. That's why I said this is of strategic interest to Wikipedia and the humankind.

On Mon, Jan 23, 2012 at 7:03 AM, Thomas Dalton <thomas.dalton(a)gmail.com> wrote: That's exactly an "egg first or chicken first" problem. Great discoveries almost always come from rarely known ideas.

This comment section idea can be an experiment. If it does more good than bad, we can keep it. Otherwise we can remove it. It's just as simple as enabling/disabling a MediaWiki extension. On Mon, Jan 23, 2012 at 7:20 AM, Thomas Dalton <thomas.dalton(a)gmail.com> wrote:

On Mon, Jan 23, 2012 at 7:27 AM, Thomas Dalton <thomas.dalton(a)gmail.com> wrote: If it's useful, we expect to see researchers claiming in the news that they got like-minded colleagues from Wikipedia's comment sections :-)

On Mon, Jan 23, 2012 at 7:39 AM, Svip <svippy(a)gmail.com> wrote: What about a Slashdot-like comment section moderated by users themselves? :-)

On 23 January 2012 00:43, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote: Slashdot's comment moderation system is my favourite comment moderation system, but it is not perfect. And it works for Slashdot, because it is usually read by computer literate people. We cannot expect the same expertise from people who are likely to be commenting on [[Cat]]. Which unfortunately would mean that a comment system like Slashdot's would become too confusing to most people, even if they did not have to participate in the moderation aspect. Then one might suggest a Digg/Reddit type system where comments can simply be voted up or down (or perhaps just up), but that is fine for a news site, where comments disappear as a new news story flocks to the top. But on Wikipedia, [[Cat]] will always be there and it will continue to have the same level of importance as it did yesterday, today and tomorrow. Hence the comments there will be carved in brine stone. And if *one* comment is elected to the top, it will continue to get more votes and continue to be the comment most people will see so there will be less ACTUAL new comments. Bash.org is an example of how this works (or lack thereof), as the same popular quotations remain in the Top 100 as they were 5 years ago. Then we come back to a no user moderated system, and then we run into my former problem. Where it will either way be a Lord of the Flies system where no actual interesting conversation is generated (because people with interesting comments worry their comments might get buried anyway), because either there is no moderation or no one willing to do it. What's further at issue is that a comment section on Wikipedia may also degenerate people's trust in Wikipedia as a source, because suddenly it would appear as every other Internet website where you comment on articles, forum threads and whatnot. Wikipedia *ought* to steep above that. It needs to be different. It needs to be information only. No discussion. And that is - in my opinion at least - the beauty of Wikipedia. In a world of a chaos, one remains committed for order. But if we really *need* a system where we can comment on broad concepts such as [[Cat]] or [[Solar calendar]]s, we could create a 'Wikipedia comment site', that would seemingly seem connected to Wikipedia, but at the same time not. And what's appropriate, it would be less obvious to find, which may gander some headway among people interested in actual conversations with others on the subject. And I am certain some IT news site out there will cover its formation. And if we can handle it, there might even be a subtle link from every Wikipedia article to this off-site comment site. In fact, I'm surprised wikicomments.org is still available.

Yao Ziyuan wrote: I think it's an interesting idea for an outside group to take on. It'd be trivial to re-use Wikimedia's content (page titles or page titles + page text) to create a discussion forum for all kinds of neat articles. But given the cost-benefit analysis for the Wikimedia Foundation getting involved, I don't see it as a very good idea. Lots of costs (implementation and development of the software, monitoring comments, page load increase, etc.) and not many benefits (see below). While I appreciate your optimistic view of comments on the Internet, I'm generally of the view that the Web needs fewer comment sections, not more. Has anyone ever found (for example) a YouTube comment insightful or useful? Ever? Occasionally a news site with some kind of noise filter (up-votes or the like) can produce an occasional comment that's not completely awful and useless. Most comment sections are filled with vile language (if people are paying viewing the content) or spam (if people aren't). It's certainly a reasonable idea for a MediaWiki extension, if such an extension doesn't exist already. MZMcBride

Hoi, A comment section under every Wikipedia article seems to be a very bad idea. You read a Wikipedia article to learn about the subject at hand, you can read comments on the talk page. Reading the talk page only makes sense when you are interested in learning more about what people have to say about the article or the subject. Typically I am not and I am sure that most of our readers could not care less. Having comments in your face at the bottom to me is not only something I would resent, it would also add more clutter that I have to download every time I read an article. Given that more and more people read Wikipedia on a mobile, it is added real cost with debatable benefit. Interesting idea, sure. Strategic hell no. Thanks, GerardM On 22 January 2012 22:43, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote:

On 23 January 2012 18:16, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote: Perhaps, but Wikipedia (and MediaWiki sites in general) are one of the few remaining websites that benefits from being Web 1.0 rather than Web 2.0, even if it has elements of the latter.

Since it's unlikely the foundation mailing list will agree to enable such a comment section on every Wikipedia article (although enabling it is quite easy: just choose a "comment" extension from http://www.mediawiki.org/wiki/Extension_Matrix/AllExtensions and enable it on Wikipedia), maybe we can talk about implementing it on a separate website. For example, create a website named WikiSocial.com. This will be a Web 2.0 version of Wikipedia, which lets you browse Wikipedia's content but also provides Web 2.0 social features such as the comment section, social sharing buttons (e.g. "Tweet this article"). Any interest? :-) On Mon, Jan 23, 2012 at 5:43 AM, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote:

Turns out, this "adding a comment section under every Wikipedia article" idea is not necessary, because if two people are interested in the same topic, they can find that topic's Wikipedia article and google that article's title (which is a globally unique term for that topic) and find each other. They don't need a comment section on that article to meet each other. On Mon, Jan 23, 2012 at 5:43 AM, Yao Ziyuan <yaoziyuan(a)gmail.com> wrote:

The difference is, we tread a narrow line here. We want talk, but of a contributory kind, high signal-noise, high proportion of information. There are three kinds of "discussion" that can take place: 1. *User feedback* - characterized by specific one-off posts left for others to uprate or downrate and (presumably as far as they're concerned) editors to hopefully do something with or listen to. Ideally this filters 3 ways - (a)* ignore, (b) pass to editors with thanks, or (c) note but no action taken, with explanation and thanks*. 2. *Editor discussion of the article*, high quality dialog specifically about the article, or good points fed back on it. 3. *Discussion of the _topic_, or general chat, forum-y stuff, random whatever*... this is what people also expect. Look at any popular blog or facebook page, the chat below is often just people discussing the subject, what's said about it... low signal to noise generally. The article feedback tool is working towards (1); when it's closer to complete I imagine articles will have a "give feedback here". (2) we already have. * What is worth asking is, is there a place in Wikimedia for (3)?* If so it can only be for social interest and "stickiness" (people who discuss may contribute or at the least will be made more aware). It could be very good for that. The downside is it attracts advocates, might draw attention away from content discussion, needs patrolling (distraction from content), etc. So here's the focused question -- is there a net return from the "plus" side, and if so is there a way to get that benefit that returns more than it costs? Where: - "returns" will be oxygen for the project generally and articles specifically, awareness, wider attention, stickiness, more public eyeballs, a way to get some more focus here of the kind social sites leverage, and maybe a start for more editors from (3), and - "costs" will be the distraction from working on high quality discussions (1) (2) and article editing as a result of patroling and other needs of (3) (And of course the standard of comparison could be "better of two evils". For instance if the crystal ball says a wiki project dies due to fading attention then maybe chat and patrolling is net harmful but less harmful than eventual loss) FT2 On Tue, Jan 24, 2012 at 12:09 PM, David Richfield <davidrichfield(a)gmail.com>wrote;wrote:

FT2: ... and loss of neutrality, when the comments on controversial topics (or even less controversial) are filled with arguments for one cause or another – and all comments *will* be considered to be, in some way, a part of Wikipedia as well. //Johan Jönsson -- http://johanjonsson.net/wikipedia

On Tue, Jan 24, 2012 at 4:50 PM, FT2 <ft2.wiki(a)gmail.com> wrote: Maybe implement a subreddit schema and some way to create a subreddit for each article? I don't know what Conde Nast's nastiness level is, though. Or maybe a slashdot portal? Just throwing ideas around here... -- David Richfield [[:en:User:Slashme]]

On Tue, Jan 24, 2012 at 10:31 AM, David Richfield <davidrichfield(a)gmail.com>wrote;wrote: I doubt relying on a third party service is going to fly. -- John

On 01/22/12 3:44 PM, David Gerard wrote: I don't care what you call it. The talk page is still the best place for wide ranging discussions on a subject. Ray

+1.... exactly There are 3 basic kinds of dialog - ** editors and participants actively hoping to improve the article; ** feedback that is intended to have a decent proportion of useful comments and can be sifted for them quickly; and ** "chat about the topic, article, and anything else people get into". However you label them, whatever means and venue we were to use, the third of those is where the question mark goes. FT2 On Tue, Jan 24, 2012 at 11:26 PM, Oliver Keyes <okeyes(a)wikimedia.org> wrote:

Yao Ziyuan is thinking along the right lines. Wikipedia can be more than it currently is. We need experiments, we need to be experimental again. FT2 makes the excellent case that we need a level of 'insulation' to protect our existing project from the potential negative effects of failed experiments. FT2 asks: My own answer is a resounding yes. The total set of the world's personal opinion is an exciting data set. Facebook, Google+, and Twitter capture "discussion" data in bulk. But Facebook and Google and Twitter don't have Wikimedia's Values. I don't trust Facebook with my data. I don't trust Google+ with my data-- they unilaterally change their privacy policies. For-Profit companies sell out their users and we all know that's a risk. At the same time, we know that Wikimedia respects its users. Wikimedia stops things like SOPA in the tracks. I trust Wikimedia. The more data we have, the more services we provide, the more our movement thrives and the more we prove that Wikimedia Values work. Whenever feasible, grow grow. Don't just be Wikipedia-- be more! :) AlecMeta