Comment by mjn | Hacker News Reader

Comment by mjn | Hacker News Reader

Parent Full thread

If there are good references, Wikipedia has been moving in this direction. But as an encyclopedia, and one that claims no authority for itself based on any kind of expertise of the authors, articles really need to be referenced to good sources in the published literature.

If you do have good sources for something, e.g. there is even a relatively small biographical section on someone in a published book, or a journal article, or something similar, the separate "notability" requirement has been increasingly going away, so that "but I've got sources" trumps it. I wrote a bit on that last year: http://www.kmjn.org/notes/wikipedia_notability_verifiability... (I also discuss some of the history around these requirements, some of which were motivated by trying to do something about Usenet physics cranks who found Wikipedia and decided it'd be a great place for articles on all their pet theories.)

I've been systematically going through several references and adding articles on everything in them, and haven't run into people objecting to my articles or trying to delete them for several years, since the end of the more "notability" focused era. Now as long as my articles include some references, they seem fine. One of my projects is marching through the Allgemeine Deutsche Biographie adding articles on random people who were mayors of Prussian towns in 1850, and that kind of thing. There's even a Wikiproject trying to organize efforts to cover everything in that particular reference: http://en.wikipedia.org/wiki/Wikipedia:WikiProject_Missing_e...

I do think it's also valuable for there to be other projects, which catalog information that can't be referenced well. For example, Know Your Meme does a pretty decent job of doing original research in history-of-memes. There are several genealogy projects that attempt to catalog people more exhaustively as well. But those are pretty different projects from Wikipedia's goal of summarizing stuff that's published in the existing literature, with references. Wikipedia doesn't have to be the only wiki on the internet, so I don't see why those original-research projects can't develop in parallel.

I have not seen Wikipedia "moving in this directon." I had some editors delete a 5 year old page with multiple published sources and 10+ inbound wikipedia links on it. I am getting it reviewed now. There are some bad apple editors on Wikipedia who are high up on their horses.

I'm not arguing it never happens, but that it's been "moving in this direction", i.e. that notability plays an increasingly small role relative to verifiability. That's my observation over about 8 years of writing articles now and then, but I would be interested in a more quantitative test of that claim if someone knows of a way to do it. Nobody has given me any grief since around 2008 for new articles, even ones on quite "minor" subjects, whereas the question of "does this belong in an encyclopedia?" used to be a more common discussion in Wikipedia's early days, as everyone was trying to figure out what "a wiki encyclopedia" really meant. (In particular, people coming from a more academic background, like Larry Sanger, were advocating for a much smaller, more traditional encyclopedia, certainly not one with millions of articles.)

I do think there are specific areas that remain more controversial, mainly around recent pop culture and business. If people suspect you're self-promoting, then they may try to get an article deleted (there was a recent rash of hotels hiring PR firms to add articles on them, which wouldn't be so bad in itself if the articles didn't read like ad copy). Same if you write an article on a recent internet meme, or a website.

In the areas I've been working in, which are mainly geography, history, science, mathematics, and literature, "notability" as a separate requirement seems basically completely dead in practice. I used to have to argue against deletion of my articles on minor 19th-century Prussians. But nobody even thinks of deleting those these days, as long as they have solid citations.

edit: Looking elsewhere in the thread, it looks like your views of Wikipedia are generalized from one very self-interested example: an article about your dad. May I suggest that isn't the most unbiased way of forming an opinion on a complex subject?

A side note, how do you go through a reference and lift all the information from it while avoiding simply paraphrasing it? Or are you using it as a starting point then finding more sources on these mayors?

In the case of the ADB, it's public domain (published in the late 19th century), so could be used directly, like the 1911 Britannica is. Except it's in German, so some translation is necessary. :) I sometimes translate fairly directly, but I usually look for whether there are any more recent sources to expand on the bio, or correct any info that might now be obsolete.

With copyrighted encyclopedias, my preference is to use at least 2-3 sources for a bio. For example, with a physicist, I'd use a biographical dictionary / encyclopedia as a reference for basic biographical facts (dates, locations, awards, etc.), and then flesh out information about scientific importance from something like a textbook or survey paper commenting on his/her work.

If the main source is a non-encyclopedia, it's usually less of a problem, because the original text doesn't really read like an encyclopedia article anyway. For example, when writing articles on Greek archaeological sites, my source material is usually a discussion in a monograph or history book, which is sometimes scattered (it may be mentioned for a few pages in Chapter 2, then again in Ch. 8, sometimes as a main topic, other times in passing when discussing an event or person, that kind of thing). So it's a matter of going through, noting down salient facts and page numbers they came from, and then assembling the results into an article.

Reply on news.ycombinator.com