Outsider's view re: contributions - more granular approach needed?

Outsider's view re: contributions - more granular approach needed?

I regret the length of what follows. I am the proverbial over-educated white male, and I've been reading Wikipedia for a long time, but I have contributed maybe 3 times over the last five years, and those were primarily corrections of obviously inaccurate information, or gross omissions. Here are some thoughts why it's been so little.

There is a wide range of quality and quantity of information from one article to another. One of the first determinations that I need to make when I am reading is how good that what I am reading is, and is this going to be sufficient, or do I need to look elsewhere, is it clearly written, and so on. I can roughly separate it a few broad categories:

- Complete articles on par or better than anything else out there. 5 of 5 or getting there.

- Large articles that are poorly organized but nonetheless useful.

- Large articles that have a lot of text are missing their point.

- "Junior" articles that are good overviews

- "Junior" articles that are not good for much, and stubs.

- Articles missing altogether

This is no news for many people, but there's a reason I put it here. Different types of contributors need to be attracted to different steps of "article evolution". For example, I'd like to have more information on robotics and motion control (my area of expertise). This is not easy task and I wouldn't take it up by self. But IF ONLY there was a process that make tackling tough subjects easier, I might contribute.

Something along the lines of:

- Identify the need for more articles/knowledge. An expert in a given subject area (e.g. robotics) may make a major contribution by filling a two-line text field at the top of an existing article with what's missing.(e.g. "Motion Control article needs links/info on Control Theory, particularly PID control."). The expert will most likely have no time to type anything else, but those comments are worth their weight in gold, complete with their self-rated proficiency rating. There has to be an editor that monitors this process. So this takes an editor and a few experts that will eventually identify the need. The addition requests should much simpler than starting a new page! Please. Three lines and a click, on the same page, no wading through discussions, comments or editing styles. Those who know a lot about something also know how hard it is to explain it clearly and succinctly.

- Once the need is apparent, the subject matter would move into "Early development" stage, where everyone is invited to contribute whatever, the more the merrier! The article should be labeled as such, tacitly, of course as not to discourage reading. The should be fewer (if any) "this article needs more citations" annoyances on it... it needs everything, citation and all, we know that already. Different editing strategy. It's valuable enough to have any contributions. This the stage where the most writing takes place, by the aforementioned 18 to 30 years old males with graduate degrees, in part because they have time to do so.

- Once enough is written, the article moves into refinement process, again should be labeled as such. This is when Wikipedia should contact the experts from step 1, as in "the people have written an article on the subject you deemed useful per your request, would you please be so kind to review the work and make suggestions". Most experts (myself included) would be willing to do this, although we'd have no time to do major writing (otherwise I'd publish a book). The "refinement" process would be a back-and-forth between contributors that wrote most of it, the experts, and the editors. If I review something I may be willing to associate myself as an "ongoing" senior reviewer - why not?

- "Maintenance" state is for those articles that are mature. This mostly requires editing, and, again, requests for what's missing.

I my humble opinion, a good part of Wikipedia needs to be reclassified according to something alike to this framework. There should be "knowledge developers" in addition to "editors". And they could benefit from training, as in real online classes that train people. The end result is ultimately as good as the core content; Wikipedia is running out of easy topics, harder/bigger topics require a more organized approach.

Again, going back to the very beginning - the determination of "what is this I am reading here?" also helps with "Oh, this is how it could use my help".

And another comment, just for reference: it took me about an hour to think and type this up. I am one of the "experts". My hour is worth about well into three digits of USD/hour, and I am working on a Sunday while it's sunny outside. I'd love to be able to write an article... but Wikipedia needs to recognize that there are more ways of contributing other than editing or writing.

Oh, and also, PLEASE build better editing tools! This 20-line textbox was good for 1995, and maybe some people love it... but surely does not help my productivity. I'd like to use the tools I am used to, how I'd like. The software should take care of the rest. In my case it's just MSWord (yes, whatever, it's just a text editor).

Thank you for reading!

Alexkai01:13, 1 February 2010

This is a fantastic analysis for someone who has only contributed a few times. You've clearly read a lot of Wikipedia.

I hope other people read this and get some ideas. You have many good ones.

The basic idea: different articles have different needs, and there are different editors to fill them. If we could only organize volunteers along these lines, there would be a lot more productivity, and a lot less friction.

Randomran05:28, 1 February 2010

+1.

~Philippe (WMF)05:32, 1 February 2010
 

Randomran - thank you for kind comments!

It's a 2-dimensional table: (kinds of contributors) x (kinds of articles). Each "cell" has to have the right process set up for it in order to work well. Right now Wikipedia is like a single cell table: (18-30 year old males) x (all articles).

On a more general note, it would appear to me that this is one of the tasks that are central to Wikipedia/Wikimedia core mission. If Wikipedia wishes to stay relevant it has to develop new tools and processes for knowledge management. Editable web pages were a great breakthrough -- 15 years ago! And no, adding embedded video players and color tags does not count. There is such a room for improvement - graphical visualizations, realtime collaboration, information theory based reformatting, etc. There is a similar level of untapped potential in human development - i.e. training. I understand that doing something like that would be a major software development effort (and expense), but given the other choice (stagnation and decay), I think it's worth the risk.

24.254.83.17310:31, 1 February 2010

Well, I realize that this is not the main point, but this assumes that the gradual-growth-by-spontaneous-collaboration-premise holds water. This has been part of Wikipedia propaganda from the beginning (just like the awful you-cannot-damage-Wikipedia-myth) but I have never actually seen it work in practice.

Of course I wholeheartedly support the idea that putting in meta-tags is not only pointless but counterproductive. - Brya 05:29, 2 February 2010 (UTC)

Brya05:34, 2 February 2010
 

I'm not sure what you're getting at. Spontaneous collaboration was very effective to build up the encyclopedia initially. Now that the encyclopedia is more mature, it's obvious that spontaneous collaboration is more effective for some tasks than others. Stubs definitely benefit from that kind of attention, whereas featured articles require something more deliberate. I think this idea is brilliant, personally: there is this wave of unfocused volunteer effort that could be much more effective if it were nudged in the right direction.

Randomran19:47, 2 February 2010

Yes, I am sure that there were places where spontaneous collaboration happened and was effective. However, I doubt that is all that frequent. I meant what I said "I have never actually seen it work in practice", and I mean that literally: in my experience 'collaboration' all too often means some editors getting together and deciding that the core values don't apply to them or 'their' topic ("consensus"!). Certainly, given how diverse Wikipedia is, the model is nowhere near the norm. - Brya 05:08, 3 February 2010 (UTC)

Brya05:08, 3 February 2010
 

Firstly, I think that this argument only supports the need for a better process. The only way to really find out if spontaneous collaboration works is to apply it to a problem that is suitable (such as a stub article, specifically, with proper guidance, etc.), get some data, and see what can be improved. Secondly, if there were to be no spontaneous collaboration, there would be no Wikipedia. Since Wikipedia does exist, some degree of collaboration did occur. How efficient was it - probably not superbly... but that's why it needs to be improved. It is a valid concern, no more, no less.

Besides, the spontaneous collaboration issue is just a part of the problem. In my opinion, the solution lies in establishing the framework that supports granularity, flexibility and careful refinement. Then, start addressing this smaller issue from within something that is moving in the right direction, as opposed to presenting the spontaneous collaboration as a "show stopper". That's not helpful. I can come up with 100 arguments as to why Wikipedia will never work. Heck, even if it is a show stopper - figure out a way to make it less of a show stopper and move forward from there. This is not going with the "gospel" - I am not suggesting ignorance, rather the opposite: take the problem into consideration, build a process that helps to isolate and study it. Do that while achieving the main goal, which is to grow Wikipedia, not to resolve the collaboration issue, although that may happen too.

Alexkai05:53, 3 February 2010
 

In my experience, spontaneous collaboration happens all the time.

  • There's the asynchronous pattern, where one person creates a stub, another person expands it, another person researches and cleans it up, and someone else brings it to good or featured quality.
  • There's the "we both like this article" pattern, where two editors both care about the article and feed off each other. Here there are more conflicts. But when people are working in good faith, a little discussion can resolve the issues.
  • There's the "hey everybody, let's do something about this" pattern, where a bunch of disparate editors all end up working on an article. Lots of WikiProjects have a "collaboration of the week" that looks like this.

The only thing you're talking about is the canvassing-type collaboration -- a bunch of editors with the same viewpoint working together to enforce that viewpoint. I agree it happens too often, but I think the other types are at least as frequent.

What Alexkai is proposing would not feed into that. It would match editor skills with content needs. That's much more effective than throwing editors into the deep end, where they might not have the skills to improve certain articles, and there is no guidance about what those articles need.

Randomran00:13, 7 February 2010