Wikipedia:Meetup/Seattle3

This meetup has already occurred. You can see the notes from before the meeting if you care.

Date: Saturday January 14, 2006

Location: Suzzallo Library, University of Washington, U. District, Seattle, Washington.

Previous Seattle meetups

 * 1) November 6, 2004
 * 2) January 15, 2005

Report
In the nature of these things, none of the following has proper references, so these details may simply be the result of original research. Live with it. This isn't article space&hellip;

We met Saturday January 14, 2006 at Suzzallo Espresso, Suzzallo Library, a lovely Gothic Revival building with high ceilings and stained glass windows, just off of Red Square on the campus of the University of Washington, U. District, Seattle, Washington. To the disappointment of several, the espresso stand was closed, but at least the intended room was open and we were able to meet as planned. Afterwards, most of us went out for vegan Thai food at Araya's Restauarant on the corner of 12th NE and NE 45th. Even those who were originally daunted by "vegan" were pleased by the food, though some were quite disappointed that Araya is still working on getting her liquor license. Four who still felt that more talking was in order (Brassratgirl, Eclecticology, RobLa and GTBacchus) continued meeting over drinks at the Campus Inn [name?] until 11:30.

We started around 3pm, dragged together some tables and a lot of chairs and began by introducing ourselves:


 * llywrch
 * Brassratgirl
 * The Rod
 * Michael Snow
 * Ksnow
 * joshuadfranklin (with baby in pouch)
 * SchmuckyTheCat
 * Eclecticology
 * Shawn, a library sciences student who occasionally edits from an IP address
 * RobLa
 * Katy, a library sciences student who "lurks" (that is, reads), but has not yet edited (come over to the dark side, Katy!)
 * Jmabel
 * GTBacchus
 * DVD_R_W
 * JFPerry
 * WAvegetarian
 * Bumm13
 * Insaneinside

A couple of people wandered in late and at least one didn't sign up; if you were there & are not listed, feel free to add yourself.

As seems to be typical for these events, we ranged from newbies to grizzled veterans. I believe that only one or two edited Wikipedia as early as 2002; quite a few of us seemed to be of a 2003 vintage, but even some who are now very active had joined in the last six months.

With several librarians and library sciences students present, discussion focused mainly on Wikipedia (&larr; overlinking?) as a reference. We turned first to Katy (the one person present who had never edited&hellip;unless you count Joshua's baby) for her impressions. She was fulsome in her praise for Wikipedia: "one of the great knowledge tools that we've got", a development comparable to "the inception of the first encyclopedia."

But it wasn't all a mutual admiration society (&larr; Easter egg link). Katy and Shawn also went on to talk about why Wikipedia is often not trusted by librarians as a research tool. Part of this is simply that librarians, like most casual users of Wikipedia, are still not deeply sure of what Wikipedia is and how to evaluate its reliability. "It takes a lot of digging" to understand what makes a Wikipedia article any more reliable than a random website. Most Wikipedians learn by engaging in the process; those who have never edited are mostly unaware of the process. For example, relatively few librarians or other non-Wikipedian users understand how within Wikipedia you might question an assertion that you doubt.

This led to a suggestion that we should supplement Researching with Wikipedia with some more related project pages:
 * 1) How a Wikipedia article is built, which would explain the various mechanisms that facilitate collaboration (Recent Changes, Watchlists, etc.) and the policies (No original research, Verifiability, etc.) that support reliable content.
 * 2) How to read an article history, about what to look at in an article history to help evaluate an article. (SchmuckyTheCat said he might get this one started.)
 * 3) A researcher's guide to discussion pages: how to read discussion pages as part of determining reliability, and how to use them if you have doubts.

It was pointed out that we really need to help people understand our process better, because Wikipedia is created by a large group of uncredentialed and often anonymous individuals. A contrast was made to Encarta, which began with the content of several conventional print encyclopedias and whose (small, carefully selected) group of senior editors are all have degrees in library science, so it is much closer to the conventional model of why one would rely on a book from a known publisher.

We discussed several other interesting contrasts to Encarta. One is that, even independent of our direct editing mechanism, we have another advantage in getting useful feedback from our users. It is believed that the typical Encarta user is an elementary school or middle school student. These are not people likely to give expert feedback. Good feedback requires that you have a readership who, at least part of the time, are looking at things that they already know, or at least know something about. We also simply have enormously larger bandwidth to handle feedback (although it comes at the cost of a greater risk of errors slipping through, and perhaps remaining in the text).

Several of us were intrigued by one remark about Encarta: they describe points of view by means of sidebars. While the Encarta approach is to use this only for two contrasting points of view, there is no inherent reason such a concept could not be extended. This might be an interesting possibility for some controversial issues: instead of the enormous difficulty of devising neutral descriptions of individuals' or groups' opinions on the matter, one might place those in contrasting, frankly point-of-view sections.

From there, conversation turned to more general discussion. We discussed the proposal for Stable versions. Attitudes toward the proposal were generally supportive, although only a few of the people present seemed familiar with the details. There was some rehashing of discussion that has occurred at Wikipedia talk:Stable versions, but for the most part there seems little value to reproducing that here. One topic that came up was wiki purism versus user focus and the related issue of intellectual authority (&larr; our authority article barely touches on the issue). GTBacchus remarked that one of the issues is "Do we want to make Wikipedia more like a book or teach people not to treat books 'like books'?" Should it perhaps be part of our mission to teach people to question authority?

The general sentiment seemed to be that if stable versions of some articles are to be offered, they should be available to users via an additional tab. llywrch had quite a bit to say about the breakdown of article quality. (He observed, based on some almost-scientific sampling, that while 50-60% of Wikipedia's articles are stubs, about 4% of them could be considered "good articles".) There seemed to be consensus that somewhere between 30,000 and 100,000 English-language Wikipedia articles are far enough along that stable versions could make sense.

There were some (uniformly favorable) remarks about semi-protection; the Good articles concept also received praise in the abstract, but perhaps not in the concrete. In particular, Jmabel brought up that the article Jew was de-listed as a "good article" because one user unilaterally decided that it is too often vandalized (even though this vandalism is reliably, promptly reverted). It is an interesting question as to the degree to which vandals get to decide which articles may be designated as "good" or featured.

Discussion moved on to verifiability, original research, and related matters. All the veterans seemed to agree that two years ago the general attitude in Wikipedia was that "citations are for wimps." In this respect Jmabel pointed out that, to this day, 1911 does not have any indication of which 1911 EB article was used as a source. Clearly this has evolved, with some people arguing that absolutely nothing should be in Wikipedia without fact-by-fact citation (See Dalbury's remarks at for an example). Neither extreme seemed to be represented in the room.

Brassratgirl remarked that there is no current way to distinguish classes of uncited edits. If someone does not give citations, are they writing from genuine knowledge, from the results of unacknowledged Google searches and "uncitable" sources (such as personal web pages or blogs), or are they just making it up? The Rod suggested Wikisource Wikibooks as a repository for unreliable documentation of direct observations to cite. Someone (Brassratgirl, was this still you? If so please edit this accordingly!) remarked that there ought to be some way to identify even one's unreliable sources. That is, some way for someone to make it clear whether they are writing on the basis of what they heard in class today (or 20 years ago), or on the basis of a sub-par web page that they've decided looks (perhaps on internal evidence) like something they are ready to trust, etc.

Several users (including llywrch who writes a lot on topics like places in Ethopia &larr; llywrch agrees, but thinks it was said by The Rod, DVD_R_W or WAvegetarian) remarked on the problem of the attitude that can be summed up as "If it's not in Google, it doesn't exist." JFPerry, in particular, said that he has written a lot of articles about women's bicycle racing and that even major champions from the pre-Internet days often have no mention on the web; llywrch also pointed out that the same situation exists with skateboarding, where, despite the Gen-X and Gen-Y appeal, remarkably few of the reliable sources are on-line in a searchable form.

Jmabel suggested that we need a way to identify certain people within Wikipedia as genuinely expert in some areas, and have some (social, not technical) mechanism to trust their uncited edits in those areas more than those of a random user. Someone (GTBacchus? in any case, several people concurred) remarked that it's hard to see how this would be done, because so often the self-proclaimed experts aren't real experts and the real experts are modest about their expertise. SchmuckyTheCat also strongly concurred: "Not everything is on the web."

This brought us to the topic of the original intent of the policy against original research. SchmuckyTheCat, in particular, pointed out that this was originally largely about filtering out fringe science or worse; it was not intended to be a mandate against (for example) citing unpublished correspondence. The following is excerpted from remarks he made on this subject 27 October 2005 on the Village Pump ; his remarks at the meetup were along similar lines:

"The purpose of NOR is to prevent Wikipedia from being a publisher of new ideas. &hellip; If someone wrote in an article about their school that the school had won the state badminton championship 6 times in the last ten years since the hiring of Coach Belinda Munchabox, but that information isn't on the web - that doesn't make it personal experience. If the school has plaques on the wall that every student passes THAT is primary source material. Hiring records for Belinda Munchabox are available. If that is what make it verifiable then there it is." "Examining a primary source material isn't personal experience either. If you personally went to interview someone about some historic local event then take good notes, that person is primary source material and there is no reason not to include what you've learned. Your notes or recording would be verifiable of that persons claims" "Similarly, you and your memories can be a primary source. If you witness a bridge sink on a certain date - that's a fact. However, don't present new ideas or claim to know why it sank unless you've also examined data about that. A claim that it was shot down by alien missiles would, uh, need some verification." "For the example of Slashdot subculture, the slashdot forums are primary source material. Stating facts 'Slashdot has a lot of trolls' is verifiable by any examination of the forums. There are a few places that article verges on presenting new ideas, but the article is also backed up by a lot of example posts in the forums. A real sociologist writing a paper to a scientific journal about the Slashdot subculture would be absolutely awesome as a secondary source material - and I'm sure it's been written, go find it." "When interpreting NOR, don't stray too far from it's intent. The intent of NOR is to discourage crackpots not to discourage editors from contributing what they know. Don't let it get in the way of contributing what you know but be willing to back it up."

A problem with verification was illustrated by Jmabel, who mentioned that one of the most authoritative primary sources for non-mainstream (especially DIY) rock music history, Maximum RocknRoll, is available in a complete run to the general public only in one location in the Western United States (the San Francisco Public Library), requiring the active assistance of a local Wikipedian to use.

Finally, there was some discussion over whether we might need a different way to handle controversial articles. Semi-protection was generally praised; several people felt it does not go far enough, and that perhaps 1–2% of articles end up absorbing 10–20% of effort, often to little effect. There was some discussion of whether, once such articles reach a certain point, we might need a radically different process to move them forward, but there was no concrete proposal.

That was about it for the "business" part of the meeting.

Reminders:
 * Brassratgirl reminds everyone to come to Wikimania 2006, August 4-8, 2006 at Harvard Law School's Berkman Center for Internet & Society in Boston, Massachusetts. And you can submit a paper even if you aren't attending.

And we all wished Wikipedia a happy fifth birthday!

11 of us afterwards went to dinner at Araya, a Thai vegetarian restaurant in the University District, and four of those allegedly went on to another establishment and continued their discussions over spirits.



These notes are mostly by Jmabel; please, if you were there and feel something was incorrect or left out, including an attribution of who said what, fix it! This is a wiki&hellip;