On 8/11/07, Lars Aronsson lars@aronsson.se wrote:
But the scientific method defines how to do original research, something Wikipedia has sworn not to do. I'm just saying this, toungue in cheek, to point out how hollow and vague the claim for being "scientific" can be. *We* should be scientific whenever we can, but the resulting *encyclopedia* can not always be. However, the word "scientific" with respect to encyclopedias is so loaded as a marketing buzzword, that it becomes practically useless.
I still don't know for any other method except a scientific (with limitations, of course) which may give us a basis to work on an encyclopedia. If you know for someone, please let me know.
The word encyclopedia, I'm told, is composed of Greek enkyklios (all-round) and paideia (education). These are the two areas where I think we should focus, but instead so much energy goes into requiring citations (verifiability) and weeding out irrelevant topics (notability). Of course, education implies solid, scientific knowledge, not superstition, propaganda or self-promotion. But it also means bringing out that knowledge in an efficient and useful way.
Analyze of words etymology doesn't say a lot because astrology would be a science then. Today, encyclopedia is a sum of knowledge which is gathered using scientific method. Yes, it is used for education and it is her primary goal. However, any religious or political pamphlet may be used for education and education may be its primary goal. So, the main difference is in a method.
I learned a lot from Wikipedia. Thanks to Wikipedia, I realized how much school system in my country was misleading me (it was socialist Yugoslavia, but the main problems were not related to socialism because it was easy to understand what was propaganda and what was not). And it is a nice feeling to imagine what amount of benefits from Wikipedia would have humans born at the beginning of this century.
However, we mustn't mislead them! And the only way to do so is be strict with verifiability.
For example, I may talk about linguistics. Outside of very well known linguistic theories and very well known linguistic features, Wikipedia articles are in a mass. One example is "distantive case". According to Google as well as according to one of the rare linguistic scholars on English Wikipedia (User:Angr), it is not a noun case, but a *verb form*. And such "fact" existed inside of [[Template:Case table]] for a long time.
If we were strict about verifiability, such stupidity would never pass. However, we are not and it is always possible that Wikipedia actively misleading its readers. This is not scientific nor educational.
* * *
Notability is completely other question. I don't think that notability is a relevant criteria for one scientific project. If something passes verifiability rules, it should go in encyclopedia.
Since raw size is no longer a problem for (the English) Wikipedia, I think we should start to ask how all-round it really is. The other day I found a "best-selling author" who didn't yet have an entry in Wikipedia (so I wrote one). And this author is American and contemporary, not a best-seller in Japan in the 1920s. And Wikipedia is still very good on popular culture, such as music, literature and movies. I think there are many other areas where we have far more left to do. (Jimmy Wales made a good comment recently about 19th century mayors of Warsaw.) We know the raw size in numbes: 2 million articles. But how can the all-roundness be expressed in numbers?
Yes, we need a lot more articles. I have a number of similar examples.
A couple of months ago, I had a couple of lecturers to teaching assistants and students of the Faculty for physical chemistry in Belgrade about work on Wikipedia. (Thanks to one professor who is Wikipedian, students of the final years have to write articles on Wikipedia as semester works.) I wanted to show them that English Wikipedia covered physical chemistry better then Serbian and that they are able to find some starting articles there. It was a kind of surprise when we realized that one basic concept of chemistry of proteins didn't existed on English Wikipedia. So, we wrote one little article in English even our main goal was to write on Serbian Wikipedia.
It seems that we came into position where we need people who know matter more in depth. In your case, it seems that we need someone who is well introduced in contemporary literature. In my cases, we need people introduced in linguistics and chemistry.
The next issue is education: How pedagogic or "efficient in teaching" is Wikipedia really? Are articles well written, well structured and easy to find? Do we care whether it answers peoples questions? This is where actual research can be applied. That is: Original research should not be documented in the articles, but original research can be used in WikiProjects as a support for improving articles. Paris is the capital of France. That simple statement is an efficient way to bring out knowledge. Loading the text with references to literature doesn't necessarily improve the encyclopedia's purpose in that case.
And after your statement, I may write that Sidney is the capital of Australia. Very simple, very logical, but false. Australia is a big country and every educated person knows that Sidney is not the capital of Australia. But, there are not so well known countries. Do you know what is a capital of random US state or a random Central Asian country? And do you think it is necessary to put a reference for that information? I explained that in Queen Victoria - Tolstoy - Broz case: for me there is no need to give a reference where Broz was born because it is well known truth to me; but it isn't to you. And, yes, there are people who use Internet and who are not familiar with the fact that Paris is the capitol of France.
About original research: I don't think that WikiProject space is a good place for that, but I think that something like that should exist as a Wikimedian project.
The biography I wrote for a best-selling author does improve the all-roundedness of Wikipedia. It's short, but well structured and easy to read. Unfortunately, we have no good system for indicating this progress. But now the article is flagged with {{unreferenced}}, as that is the main activity going on. Nobody questions the facts expressed in the article, neither its notability. It's just the lack of <ref> tags. It would be easy (and tempting) to fake that by making up a source somewhere. Just add a <ref> tag, and all is back to scientific again.
I am preparing one document which would deal with this issue: how to know what is a reliable source exactly. And Gerard Meijssen mentioned to me that OmegaWiki team is working on MediaWiki extension which would allow contributors to mark what is a reliable source and what is not.