Wikipedia:Wikipedia Signpost/2010-05-17/WikiProject report

Most of the WikiProjects previously featured by the WikiProject Report have focused on improving articles in the main space. This week, we took a look at WikiProject Essays, which instead focuses on user written essays located in the "Wikipedia" namespace. The project started in March 2008 under the name WikiProject Essay Categorization and/or Classification and was recently revived with the new name taking effect in April 2010. We interviewed Noraft and SilkTork.

What motivated you to revive WikiProject Essays?
 * Noraft: When I found it, it was called WikiProject Essay Categorization and/or Classification, and I thought, "What a great idea," because there are over 1,000 essays in Wikipedia namespace now, and there were only a couple categories for them. This meant that they weren't very searchable. I had written an essay when I moved it into Wikipedia namespace, I felt it was just one in a thousand essays that few people will ever see. The idea of WikiProject Essays appealed to my Structurist sensibilities.


 * However, when I found WikiProject Essay Categorization and/or Classification, nobody had worked on it in months. The talk page had comments from two years ago. So I added myself to the members list, and asked a couple consensus-forming questions. Nobody responded, so I became a self-appointed Wiktator, and started moving the furniture around. Eventually some folks showed up and I was happy to give up my wiktatorship (as all benign wiktators are). We worked together to make the rest of the changes that have been made. I especially needed help for the technical aspects of getting the WikiProject Banner functioning so that the bots to work with it and auto-assign Impact ratings.

What are your short-term and long-term goals for the project?
 * Noraft: In the short term, I personally would like to see more discrete categories for essays. Basically I'd like to see a list of categories that have not too much overlap (i.e. essays that fit into more than one or two categories), that are specific enough that there are not more than 100 essays in said category (less would be better), so that the essays are easy to locate. Another short term goal would be to get three or more motivated souls to join the project. We could use extra help and the innovation that can come from increased participation. I'd also like to see Portal:Wikipedia essays get used more. Again, better participation will help with that.


 * In the long term, I think advanced assessment tools would serve Wikipedia well. For example, right now we have a system that ranks essays by their impact to Wikipedia, and that's much better than the giant puddle of undifferentiated essays we had before. But it would be great one day to be able to have a "Top 10" in every category, so if you want to see which essay in Category:Essays giving advice has the greatest impact, that would be available to you.


 * SilkTork: The project is a forum for people who have an interest, however casual, in the essays that Wikipedians write. Wikipedia Essays cover a wide range of themes and points of view - some are individualistic jerk reactions to an event, some are personal frustrations with other users or Wikipedia in general, some are jokes, others are serious examinations of Wikipedia process, some contain useful advice, and some have a status that is close to Policy or Guideline, but for whatever reason have not been promoted. Some essays - such as WP:BRD, WP:CREEP, What is consensus?, Quotations, WP:ATA - are widely used, and have a status that is clearly above that of an individual's moment of spleen, yet all the different types of essay were lumped together under the same category, and often with the same blanket template at the top saying that the contents were an essay and to "Consider these views with discretion", without an indication of how to make a judgement on the consensus that went into the essay, or how the community views it. It is clear that it would be helpful to look into better ways of organising and presenting these essays. It would be helpful to consider ways of allowing readers to make an assessment of the status of an essay that they may have been directed to. Which of these essays have wide support, and which are personal opinion - WP:WTAF, WP:COATRACK, A warning to concert organizers, Civility warnings, A navbox on every page?


 * People who have an interest, concern or a question about Wikipedia essays now have a forum where ideas can be raised and discussed, and - if appropriate - a collaborative approach can be agreed for working on improving organising and presenting essays.

'The project only includes essays in the Wikipedia:'' namespace. Why are essays in the User: namespace excluded? Are there any plans to include user essays in the future?'''


 * Noraft: Current consensus is that we don't have any business extending our scope into editors' userpages. When an editor decides to move his or her essay into Wikipedia namespace, they are "giving it up to the community," and opening it up to the collaborative editing/consensus process. While it is still in their userspace, we feel it is "theirs," so to speak, and I think that reflects larger community consensus.


 * SilkTork: This is a question which highlights some of the muddiness associated with essays on Wikipedia, and reveals a possible discussion point. Should all essays start as User Essays, and only be moved into Wikipedia space when there is some consensus that the point is worthwhile, so that readers can have some reassurance that a Wikipedia Essay has met some minimum criteria. Should we give Wikipedia Essays a higher profile? Give them greater respect? Should we examine the criteria in WP:NOESSAY to see if that needs stengthening?

'''Contrary to many other projects on Wikipedia, your project does not assign quality ratings. However, the project does designate an importance rating based on the essay's "impact on Wikipedia." Why does the project avoid quality ratings and how does the project determine an essay's impact?'''


 * Noraft: Actually, as of a couple days ago, the template has been updated, and now we assign "Impact Ratings," not importance ratings that are based on impact. That was creating some confusion. But to answer your questions, we avoid quality ratings because we don't think that it is a good idea to start evaluating the quality of editors' opinions. What constitutes a Top-quality opinion, really? Or a Low-quality opinion? That's an area we're not willing to get into, as it is too subjective. However, impact can be objectively measured, and we attempt to do so by analyzing an essay's number of watchers, pageviews, and incoming links. We use a weighted score that weights watchers most heavily (as we feel someone watching a page is a measure of impact), with pageviews next, and incoming links are weighted so lightly that they really only serve as a tiebreaker if the other two factors are about equal. This formula produces a weighted score, and all essays are ranked according to that score, at WikiProject Essays/Assessment/Links. An example of what this distribution looks like can be found here:




 * This graph plots each essay's impact score on the Y axis. This is a closeup of the top end of the graph. Remember that there are over 1000 essays. As you can see, just a small number of essays have a large impact, with over 85% having little to no watchers, readers, or links (i.e. little impact). We decided the top 10 essays would get Top-impact, the next 40 would get High- (i.e. top 5%), the next 50 would get Mid-, and the rest would get Low-impact. This means that to not be Low-impact, an essay must be in the top 10%. This is analogous to many other distributions, such as sports ability, or wealth. A few individuals have all the money, right? If you want to use basketball as an extended metaphor, Top- is the NBA, High- is the NCAA, Mid- is skilled amateurs, and Low- are everyone else. If you think about it, with 300 million people in the USA, the distribution (if plotted like the above graph, with "Basketball Skill" along the Y axis, and everyone in the USA on the X) is going to look a lot like our essay distribution. To even be Mid-level skill, that puts you above 250 million non-players. An easier way to see the numbers is like this:

'''Essays on Wikipedia range from serious policy discussions and how-to manuals to a variety of humorous mock-articles and self-deprecating quips. How does the project sort these various types of essays and how does an article's seriousness affect the essay's impact?'''


 * Noraft: We sort them by category. Impact is only a measure of how many people are reading, watching, and linking, so impact is not affected by seriousness or lack of it in some deliberate way, although we can see that less serious articles do not occupy the top spots. I think the fact that impact is so impartial is one of its strengths: it is hard to argue with a bot. Where there are some debates is on whether or not watchers, pageviews, and incoming links are the best measure of impact. If we find a better measure, we may use it.

'''What are WikiProject Essay's most pressing needs? How can a new contributor help today?'''
 * Noraft:
 * We need someone to take over coordinating Portal:Wikipedia essays and grow it. I think that one day it can really be a good way to showcase our essay talent.
 * We need people to sift through essays and categorize them. Because most of the categories are new, we still need help categorizing the essays. I've tried to lead by example: I went into Category:Wikipedia essays and chose a letter ("D") and over a week, read every essay in D and assigned it to one or more of the new categories. There are 25 more letters! We're giving a barnstar to anyone who completes a letter.

Share with us some of your favorite essays and/or essays you think are so important that every editor should read them.


 * Noraft: Wow. I feel a lot of responsibility in answering that question. Please clarify is the Top-impact essay right now, and I think that's important, because that essay gives instructions which improve the encyclopedia, which is what we're all ultimately here for. I think WP:BRD is something that all editors should read, if only because many people now cite it like a guideline. Even if you don't agree with it, you should understand how it works. I think that if every editor really took Don't-give-a-fuckism to heart, this would be a happier, more civil place to work, although I do think the title probably drives some people away. My favorite essay is one I co-wrote: Inaccuracies in Wikipedia Namespace.

Next week's Report will be heavenly. Until then, search for salvation in the archive.