Here are some thoughts I've had about Wikipedia. Comments are highly encouraged; please post them on my talk page. Most recent WikiThinkings are at the bottom.

Comment added 04:39, 25 July 2005 (UTC)

This is a seriously cool endeavor. I've been involved with the Internet since before the Web, and I've been happy to see how the Web as a whole is evolving into a storehouse of easily available information. But some kinds of information have been hard to get on the Web as a whole. Much of this is finding its way into Wikipedia (and similar projects). The result is an extraordinarily useful reference. It's great. I love it.

Actually, I'm surprised how well Wikipedia works. Allowing absolutely anyone to edit, without even requiring an ID, seems ridiculously naïve, a recipe for disaster. And yet it works, largely due (I guess) to the efforts of people who evidently have no lives, and spend all day reverting vandalism. Realistically, though, I imagine it will stop working soon. Automated vandalism is on its way; I predict that Wikipedia will begin requiring an ID (at least) to do edits within a couple of years (for the record, it is now 04:39, 25 July 2005 (UTC)).

Another thing that makes it all work is thousands of people that are utterly obsessed with some small fragment of human knowledge. These people make the details of Wikipedia accurate and authoritative. But when they do their work, completely oblivious of the world of knowledge around them, the results can be humorous. E.g., the CAPS page used to be a redirect to the Cirrus Aircraft Parachute System, of all things. Is that what "CAPS" means to you? Of course it is.

It seems customary to have a brag list of pages one has written. But I find I much prefer editing to writing. As of this writing, I have 134 edits in 103 days (is that a lot? I dunno), but I've only created one page (not including this one, talk pages, etc.), and it was on a topic I know nothing about, just a place for a broken link to go. But as Wikipedia grows, it will need a vast amount of tweaking. Pages that continually get extended will evolve into swampy messes. Someone has to clean these up. And I find I like doing this. Odd? Maybe. Flashy and heroic? Certainly not. But people like me, the hyenas (earthworms? dung beetles?) of the Wikipedia community, make it all work.

Comment added 04:39, 25 July 2005 (UTC)

Many pages are part of a general cloud of pages being worked on by a sub-community of the larger Wikipedia community. How does one find these people? If there is a general discussion going on with guidelines decided, how does one discover what they are and help flesh them out? Talk pages are great for a single page, but, again, many pages are part of a coherent subcollection. Maybe this question has a good answer (if so, what is it???), but I can't figure it out. So even if you tell me, other new folks also won't be able to figure it out. What do we do about this?

Comment added 08:04, 9 August 2005 (UTC)

In my brief time on Wikipedia, I have uncovered a fair amount of vandalism (or, more generally, clearly inappropriate editing, whether deliberate or not). Each time I have found this, I have looked at all the other edits made by the same person, to see if there are other things that need fixing. I have observed some patterns, most expected, but one rather surprising.

• Vandals tend to make few edits.

We're all thankful for this. The record so far, for a vandal I've discovered, has been 14 edits total. [News flash: Just found one with 30 edits, but it's an IP, so maybe multiple people are involved. 21:26, 9 August 2005 (UTC)] This makes sense; the people who spend lots of time working on Wikipedia will be those who support it. Of course, vandalism is still a problem; both physical and Wikipedial vandalism is such a pain because it is so easy to do and such hard work to fix.

• Vandals tend to edit inappropriately more than once.

Sadly.

• Vandals tend to make positive contributions as well.

This is the surprising one. So far, every vandal I have discovered has, at some point, made an edit that was clearly beneficial to Wikipedia. There is definitely some interesting psychology behind this observation.

Comment added 11:30, 13 August 2005 (UTC)

I recently gave myself the following challenge:

Look at 100 random Wikipedia pages, and see how many of them you can constructively edit.

I'm happy to say I met the challenge, but the result was surprising: 70 of them.

To make a short story long, I clicked on the Random article link 100 times and examined what I got. In 45 of the articles, I was able to make improvements that required no knowledge of the subject matter of the article. I corrected punctuation, spelling, grammar, or English usage, and I changed formatting to conform with the Wikipedia Manual of Style.

In another 25 articles, there were improvements that were clearly needed, and which I was able to make after a small amount of research, despite being unfamiliar with the subject matter of the article. I added internal and external links, restated sentences in NPOV form, etc.

In the remaining 30 articles I saw no changes that I could be sure were beneficial.

The scary part of all this is that 100 random pages makes a pretty good sample, for statistical purposes. We can conclude that about 45% of all Wikipedia pages can be improved by anyone who is familiar with English punctuation, usage, etc., along with Wikipedia editing procedures and the Manual of Style. An additional 25%, for about 70% total, can be improved by anyone who, in addition, knows a little about doing research. If you want to be a little more rigorous, you can toss in the fact that the margin of error in this number is about 10% (at the 95% confidence level). Thus, it is very likely that at least 60% of Wikipedia pages can be constructively edited by non-experts.

As of right now, there are 681,706 articles on Wikipedia. That means there are likely upwards of 409,000 articles that can be improved by anyone who can proofread and do a little research, who knows how to edit in Wikipedia and is familiar with the Manual of Style.

That's a lot.

There is work to be done, folks.

Comment added 22:57, 15 August 2005 (UTC)

In my last comment ("The 100-page challenge"), I noted that some 70% of random Wikipedia pages are in a state such that nearly anyone, regardless of knowledge of the topic, can improve them. I closed with a call to action.

But there is of course another spin one can put on this information: Wikipedia is full of awfully low-quality stuff.

To be fair, the average page viewed by a typical visitor to Wikipedia will tend to be of higher quality than those I looked at. This is because someone who is looking for information will not be viewing random pages. And pages that people are interested in, as well as pages with lots of links to them, will tend to be of higher quality (actually I'm just guessing here; I should check this out).

Still, Wikipedia is full of awfully low-quality stuff. I don't expect this to change. As the frontiers of Wikipedia move outward, people will expand it by adding poorly written articles. This is part of the price we pay for such great breadth.

In this, Wikipedia mirrors the Web as a whole. There is much that is useful on the Web. But the average Web page is of utterly abysmal quality. This has been true for over a decade. Possibly it will be true for centuries.

On the Web, we have solutions: filters, ways to find good stuff. Google is the quintessential example. Other search engines, pages of links, sites with reviews, etc., all help us find high quality content on the Web.

Let us now consider how to do something similar for Wikipedia: how do we help people find the good stuff? Again, solutions are already in place. Frozen articles, feature articles, Wikiprojects, and the simple fact that articles we actually look at tend to be of higher than average quality, all help us find the good stuff here.

Now, what else can we do?

Comment added 23:25, 15 August 2005 (UTC)

I wrote a little something on Wikipedia talk:Help desk that I thought was worth enshrining here. It was asked why people go to the Help desk to try to get Wikipedia to market their stuff. The general consensus was "stupidity". I wrote:

Some possibilities:

1. The idea of "open development" is a new concept, one that many people have not really grokked. So they see a huge thing like Wikipedia, and they figure there must be a large profit-motivated corporation involved somewhere. Thus: advertising.
2. They are stuck in their own little world, which they think everyone everywhere is terribly interested in. If that world revolves around some company or product, then they figure everyone is interested in it.
3. They have learned that if you are always a nice little boy/girl and always follow all the rules, you never accomplish anything or get any recognition or money or anything. So they learn to Take Risks and Break The Rules. Lots of people do this. After all, if certain people hadn't Broken The Rules a few years back, we wouldn't have Wikipedia, would we? Some people Break The Rules by founding an open encyclopedia. Others Break The Rules by saying, "Hey, wanna buy my cool product?" to everyone they meet.
4. They remember that there is a Real World. And in the Real World, a "help desk" is a piece of wood or metal furniture with a representative of some company or organization behind it. So they approach as a representative of their company/organization. And they do what organizational rep's do when they meet: they talk about collaboration between the organizations. (And then they wonder why we all hate them. "But these people said it was a 'help desk'," they think, "Why are they mad at me?")
5. They learned about the Web by watching TV commercials and reading Business Week in 1998. They think, "The web is a place where people sell things." Wikipedia is on the web. Logic dictates only one possible conclusion.
6. Or, yeah, maybe they're just stupid.

I can add a couple more now:

• They know what Wikipedia is all about, but they don't care. They just want to sell stuff, no matter who they offend or what communities get messed up.
• They know what Wikipedia is all about, but they disagree with the consensus on using Wikipedia as a marketing tool. Further, they feel that, as a Wikipedia editor, they have the right to point Wikipedia's purpose in new directions. They do this by talking about marketing their products.

Wow, that last one sounds almost noble, doesn't it?

Somewhat related, User:Dmcdevit has a personal page called Help desk funnies. I'm a little down on ridiculing people like this, but, well, these questions were public already, and some of them are worth a chuckle.

Comment added 02:02, 16 August 2005 (UTC)

After being a Wikipedia editor for some 124 days (and reading Wikipedia for much longer), I finally wrote myself a full-fledged article: Haber's rule. That's my edit #376 (or #333 if you only count edits of regular articles).

So now my dear little baby is out there on its own, and <sniff, sniff> all you mean people are just going to tear it up.

Please be kind to my baby. Make it your baby if you want, but be kind. Thank you.

Comment added 08:37, 17 August 2005 (UTC)

I wrote the following on Wikipedia:Village pump (policy). It concerns the proposal to add a new tagline or disclaimer to all Wikipedia pages.

I think there are issues behind the tagline discussion that need some closer examination.

I note that people are speaking in terms of a "disclaimer". Disclaimers are legal devices. The idea is to cut through the PR nonsense and (using legal nonsense) indicate precisely what is being claimed. Usually contractual issues and protection from lawsuits are the relevant concerns. I think this is an issue for Wikipedia, and it needs some serious thought; however, this does not seem to be what is driving this discussion.

What is driving the discussion is the problem that people come to Wikipedia, read it, use it, maybe even edit it, without a clear understanding of what it is. Some of them end up using information from Wikipedia in inappropriate ways due to these misunderstandings. Others, when they discover what Wikipedia really is, feel deceived and angry. Others get angry due to their misunderstandings. Quite rightly, we want to address these issues.

So, first, I want to point out that we cannot be responsible for other people's actions. Many, many people are in the habit of grabbing some source, getting info from it, and leaving, without considering reliability or other important issues. Many of these people use Wikipedia. What can we do about them? Nothing. If someone does not want to give any thought to the source of his information, then all the explanatory text in the world will not help. Let us remember then, that some things are the reader's responsibility, not ours.

Second, Wikipedia is a new thing; the world has never seen its like before. We call it an "encyclopedia", and it is, I suppose. However, it is clear that many ideas that people associate with encyclopedias are not applicable to Wikipedia. But there is no word or phrase in any language that will concisely and thoroughly indicate to newcomers what Wikipedia is. So: how can we quickly give people a clear understanding of all the principles and process behind Wikipedia? We cannot. It is a waste of time to try.

Third, there is an annoying tradition, especially in the U.S., that every time there is an issue with some product, we tack on a notice in its documentation somewhere. I bought a soldering torch. It came with pages & pages of lists of things I should be careful of. And I read and thoughtfully considered every one, of course, wouldn't you? </sarcasm> This approach was invented by corporate lawyers as a way of stopping lawsuits. It is not about communicating information, and so it is not going to help us here. In short, don't think that tacking on gobs of little notices is going to eliminate everyone's misunderstandings about Wikipedia.

Fourth, there are people who are interested in checking their sources. Many of them do not understand Wikipedia, and could make better use of it if they did. Taglines & such are not going to help them. What might help is a short essay about who writes Wikipedia, and what approval processes an article needs to go through to be published in it. (Yes, I know, the short answer is "none", but we should still talk about the approval process, since that is what people want to know about.) The hard part is helping people find this explanation.

And that is what I think it is important to address. So, how about an actual concrete proposal: Instead of a tagline intended to communicate what Wikipedia is all about, how about a tagline that tells people where they can find such information, aimed at newcomers. Here's an off-the-top-of-my-head line: "Who writes Wikipedia?" Then make this a link to that short essay I mentioned earlier (or to a list of bullet points, or whatever). I'm sure someone can improve on this idea. Please do.

As I said earlier, I think disclaimers should be discussed as well, but that is a separate issue. Disclaimers are about contracts and lawsuits, not introducing newcomers and explaining things.

Comment added 15:12, 18 August 2005 (UTC)

It is amazing how thoroughly people can disagree on some issues. Anyone who does much browsing in Wikipedia will eventually run into some pretty nasty disputes. Look at pages like Book of Genesis, Macedonia, or Israeli-Palestinian conflict, and you find people who disagree strongly, not merely on what is good and what is bad, but also on how to write a factual, NPOV article.

An interesting solution to this, when there are clearly delineated factions, would be to let each side write its own article. We could still have the same nominal requirements: no insulting the "other side", no arguing for a position, no weasel words, stick to the facts, strive for NPOV. But still allow the separate groups to write separate articles.

Probably there would still be a main page on the topic, protected from normal editing, which would hold the Wikipedia article. The main page could link to the factional articles, but they would be marked as having somewhat lesser status. This would fit well with a larger trend, in which Wikipedia heads toward the recognition of varying levels of quality or importance.

Of course this idea may cause more problems than it solves. Once we let factions write their own articles, every little group with an axe to grind may start clamoring for the right to produce its own version. And of course this departs significantly from the common idea of an "encyclopedia", as well as many Wikipedians' vision for what Wikipedia should be like. Even worse, it effectively says that we simply cannot get along; we might as well not even try. So this may not be a good idea.

But I think it is an interesting idea.

[News flash: It's been done. Essentially this idea is standard policy on Wikinfo (external link), a Wikipedia fork. 11:59, 20 August 2005 (UTC)]

Comment added 10:15, 20 August 2005 (UTC)

Wikipedia seems to be going through its adolescence now. Will it emerge as a grown up encyclopedia-or-whatever-it-ought-to-be-called? I really, really hope so. I think it has the seeds of something wonderful in it.

But it might die. And here is what might cause this.

A fight at the top

For all the accusations that Jimmy Wales and his pals get, based on my experience here I think they run this project very well. This is in part because they don't really seem to "run" it. They are largely a hands-off group. Wikipedia really is created by you and me. Nonetheless, these people are crucial to Wikipedia's continued existence. For example, Wikipedia runs on specific computers. Someone maintains these. If they stop, Wikipedia stops.

Now, if that happens, it is possible that someone else (some company, maybe) will take over the administration of Wikipedia. But suppose the folks at the top have a feud, and there is no clear winner, and other people start taking sides, and lots of people walk off together in a huff. Then Wikipedia, as a movement, splits, which probably means that Wikipedia is dead.

Lack of improvement

In recent months, a number of articles critical of Wikipedia have been published. For example, one noteworthy critical article was The Faith-Based Encyclopedia, by Robert McHenry, former Editor in Chief of the Encyclopædia Britannica. McHenry, and others, have argued that Wikipedia has serious problems. Predicably, Wikipedians seem to have responded mostly with rebuttals and self-promotion. But self-examination might be more appropriate.

Wikipedia does have serious problems. Its reliability and quality of writing are, to put it charitably, inconsistent. Its aims and policies are widely misunderstood. Some people are reluctant to use Wikipedia in public settings due to the possibility of page vandalism involving pornographic images.

We must make progress toward solving these and other problems, or Wikipedia's reputation will decline. And for projects like Wikipedia, reputation is everything. A bad reputation means the end of Wikipedia.

The GFDL

Wikipedia content is licensed under the GNU Free Documentation License (GFDL). And this may kill Wikipedia.

Don't get me wrong; I think the GNU General Public License (GPL) and its cousins, which include the GFDL, are good licenses. I have released software under the GPL. I will probably do so again. But consider: is the GDFL the license that will be exactly what Wikipedia needs for the years (decades?) to come? Maybe, maybe not. I don't know. Neither do you. No one does. But the GFDL is the license we are irrevocably stuck with.

This is because the GFDL allows anyone to make derivative versions of works, but requires that, if they are distributed, they must be licensed under the GFDL. Now, any work can be released under other license terms with the consent of the copyright owners. But once two people have edited a GFDL document, they are both copyright owners for the final version; they must both give consent, for the final version to be released under other terms. The copyright for Wikipedia is owned, individually, by all the contributors. We can only choose a new license for Wikipedia as a whole if we can get the consent of everyone who has ever edited it. Obviously, that is not going to happen.

I am not the first to notice this problem. Various movements are afoot to get Wikipedia contributors to dual-license their contributions, usually using one of the Creative Commons licenses. Thus, one sees dual-license notices on many user pages (including mine). I support the aims of most of these movements, but I think there is no hope whatsoever of getting everyone to dual-license.

So we are stuck with the GFDL. But years down the road, the world will be different. If the survival of Wikipedia depends on changing the license terms, then Wikipedia will die.

Something better

This is the "good" way for Wikipedia to die: something much better might come along. I can't imagine what that would be, and I hope that, whatever it is, it doesn't make all our effort here wasted.

Wikipedia is one of a number of high-profile open development projects. Other examples are the GNU Project and the Linux kernel. However, these are software projects; Wikipedia is not. GNU and Linux now appear to be leaving their adolescence, becoming grown-up world citizens and proving that open development can and does produce high-quality software. It remains to be determined whether it can also produce an enduring, constantly evolving, respected collection of useful information.

Will Wikipedia grow up or die? We shall see.

Comment added 07:58, 22 August 2005 (UTC)

In December 2004, Larry Sanger, former editor-in-chief of Nupedia (and thus, essentially, of Wikipedia), wrote an article (Why Wikipedia Must Jettison Its Anti-Elitism) in which he argued that Wikipedia fails to value sufficiently those contributions that come from genuine experts. He wrote that Wikipedia should "both prize and praise its most knowledgeable contributors".

I agree, but getting the experts here may be much harder than people think.

Consider: where are the experts? Some of them are out turning their expertise into fortunes, of course. Experts in law, management, engineering, or finance have lucrative careers available to them. Experts in, say, Victorian literature do not, however. Most of them can be found at universities. This is true of many, many fields. In particular, I think it is true of the kinds of areas in which Wikipedia tends to be lacking. In short, Wikipedia needs university professors.

Now, how does a modern university work? Certainly there are financial issues; universities have to pay the bills after all. But universities are not run like for-profit corporations; it is generally understood that many things a professor does will not bring in money. What they are supposed to bring in is credit for work certified as well done.

For example, there is journal publishing. Academic journal articles are "peer reviewed", meaning experts in the field evaluate each article to determine whether it is good enough. If an article passes the test, then it is published with the author's name on it (credit). The act of publication indicates that the article meets the standards of the journal (certified as well done).

Traditional encyclopedias meet both of these standards. Articles are published with a byline (credit), and hopefully, the encyclopedia has standards that must be met for an article to be included (certified as well done).

These criteria are why traditional academic journals continue to exist, despite how much easier it is — for both authors and readers — if an article is simply placed on a web page. The latter gets the author credit, but does not allow for certification.

An interesting flip side is a software project following an open-source development model, such as the Linux kernel. This has the certification, in the form of code reviews, but usually not the credit. However, the Linux kernel project has no need for the services of experts in Victorian literature.

Wikipedia does need them, but it fails both tests. Authors get their username associated with their work, but then others edit, move, and generally remake things until they are unrecognizable. No one person is The Author of a Wikipedia article, and so no one person gets credit. Further, the certification process is lacking, to say the least; certainly there is no such process that meets the standards of universities

What does Wikipedia have to offer academics? As of now, nothing but a bit of fun, really. How do we get real experts here? I am not sure, but Sanger has a suggestion. He intends to solve other problems, but I think he solves the certification problem as well. He writes:

... if the project participants had greater respect for expertise, they would have long since invited a board of academics and researchers to manage a culled version of Wikipedia (one that, I think, would not directly affect the way the main project is run). But because project participants have such a horror of the traditional deference to expertise, this sort of proposal has never been taken very seriously by most Wikipedians leading the project now. And so much the worse for Wikipedia and its reputation.

If done properly, this would solve the certification problem. Unfortunately, it does nothing for the credit problem, unless a new kind of article were invented, one which could only be edited by its creator and others the creator approves. And that would change Wikipedia a great deal, perhaps far too much.

This problem is one to think about. Thoughts, anyone?

1. The whole world, except the U.S., uses the metric system. Wikipedia is for the world. Learn to live with it, Americans.
2. Context, context, include context. I just edited a page that informed me that all I needed to know about a fellow named "Dave Lister" was that "Rimmer usually calls him this, but Holly usually calls him Dave." No links, nothing. My goodness, is it that hard to say "Red Dwarf"?
3. The first line of an article should be a definition or categorization of a term. It should be aimed at people who have no idea what the term means. For example, an article I edited said that Dr. Aloysius X. L. Pendergast is a special agent with the United States Federal Bureau of Investigation. Except that he isn't. Pendergast is a fictional character in various novels, and in those novels, he is an FBI agent. The first line needs to say he is a fictional character.

Below are the multi-license conditions I agree to.

	Multi-licensed with any Creative Commons Attribution Share-Alike License
I agree to multi-license my text contributions, unless otherwise stated, under Wikipedia's copyright terms and the Creative Commons Attribution Share-Alike license version 1.0 and 2.0, and the Creative Commons Attribution Non-Commercial Share-Alike license version 2.0. Please be aware that other contributors might not do the same, so if you want to use my contributions under the Creative Commons terms, please check the CC dual-license and Multi-licensing guides.

Minor edits multi-licensed into the public domain

I agree to multi-license my eligible text contributions marked as minor edits, unless otherwise stated, under Wikipedia's copyright terms and into the public domain. Please be aware that other contributors might not do the same, so if you want to use my minor edit contributions in the public domain, please check the multi-licensing guide.

Licensing rights granted to Wikimedia Foundation

I grant non-exclusive permission for the Wikimedia Foundation Inc. to relicense my text contributions under any copyleft license that it chooses, provided it maintains the free and open spirit of the GFDL. This permission acknowledges that future licensing needs of the Wikimedia projects may need adapting in unforeseen fashions to facilitate other uses, formats, and locations. It is given for as long as this banner remains.