However, while the actual state of Wikipedia is much better than your alternate future, what we have right now is also extremely dangerous and misleading. Today it is composed of accurate, well-written sentences and inaccurate, self-aggrandizing, or just schizophrenic sentences. But they're all intermingled.
Pick any article, say block cipher modes, CTR in particular:
http://en.wikipedia.org/wiki/Cipher_block_chaining#Counter_....
It starts off ok, but then heads off the weeds with "Nevertheless, there are specialized attacks like a Hardware Fault Attack that is based on the usage of a simple counter function as input.[14]" This is a reference to a minor paper in a conference I've never heard of, likely added by the authors themselves.
"Fix it yourself", comes the cry. But without some sort of educated review board, you end up having to explain the background of the field in order to get around to why a given reference is not noteworthy. Often, the choice comes down to who spends more time on Wikipedia. Guess who doesn't?
Meanwhile, a generation is growing up with this as their only example of an "encyclopedia".
I think my kids are far better off with Wikipedia than they would be with "real" encyclopedias.
I suppose my friend Nate could be arguing the following: it's unfortunate that there is marginal content that survives wikilawyering while at the same time there's good content that doesn't. But he hasn't supported that argument with evidence. What's the "good" content that belongs under CTR mode that he'd have a hard time placing there? The CTR coverage in WP isn't bad because of deletionism; it's bad because there isn't enough crypto expertise in the WP editor community.
If Nate sat down and wrote a well-sourced article on CTR mode, it would survive.
(Sadly, I can't even look at the article, for the reason I already complained about: when Wikipedia decides to delete something, it is totally removed from the website in a way such that you cannot use edit history to get it back, making arguments about it impossible post-facto. I wish I could make a real judgement for myself.)
I do now see, however, why you made your Britannica comment: you are responding only to the final sentence of Nate's comment ("Meanwhile, a generation is growing up with this as their only example of an "encyclopedia"."), which I mostly ignored as it was just rhetoric ;P. (I am sorry that I did that, as it caused me to misunderstand your's.)
Regardless, looking at the actual argument thread, Britannica is entirely irrelevant unless it can be shown that either 1) they were of higher quality due to their more limited focus, or 2) that their eventual downfall was related to them being more likely to cover a niche topic than Wikipedia ;P.
Regardless--to now respond to your new comment--at some point you need to make a choice: either that content is notable, or it is not. If it is, you have to decide who is allowed to determine whether that fact. Wikipedia, FWIW, does not appreciate content that requires "expertise" much, as they want everything verifiable by the layman.
(Again, leading to what I consider to be a ludicrous situation with respect to the various articles surrounding RSA: there is much more coverage of the recent non-attack on embedded systems with bad random number generators than there is on actual modern mathematical attacks, seemingly because the former could be verified by newspaper articles and the latter requires citing academic papers, as the last time a review paper in this field was published, a requirement for Wikipedia's "primary sources must be verified by secondary sources" requirement, was 1999.)
"...similarly, a scientific paper documenting a new experiment is a primary source on the outcome of that experiment."
"A primary source may only be used on Wikipedia to make straightforward, descriptive statements of facts that can be verified by any educated person with access to the source but without further, specialized knowledge."
"For example, a review article that analyzes research papers in a field is a secondary source for the research"
-- http://en.wikipedia.org/wiki/Wikipedia:No_original_research
"All Wikipedia articles should be based on reliable, published secondary sources. Reliable primary sources may occasionally be used with care as an adjunct to the secondary literature, but there remains potential for misuse." (emphasis copied from original)
What you're doing here is cherry picking a very reasonable policy that Wikipedia has and generalizing it far past its actual application.
What you're saying is that WP frowns on journal articles because they are "primary sources".
What WP actually says is that you can't go take the results of journal articles and use them to synthesize conclusions that the author didn't actually make inside a Wikipedia article. In other words, you cannot use WP articles to ratify new science.
I feel like if there's one place on the Internet that should appreciate a policy forbidding unfounded extrapolation from journal articles, it should be HN, which routinely hosts debates that pick apart the media's broken analysis of scientific articles.
You do realize that I quoted Wikipedia's guidelines, right? That those links I added were not some kind of lie: they were direct references to validate my argument.
Wikipedia seriously goes so far as to state that "mainstream newspapers" are an example of "the most reliable sources", in addition to all of the stuff I already pasted from their guidelines about journal papers.
(OK, I have 2% battery, and I can't find the reference for this, so I'm going to have to temporarily cede this argument. Grrr. I had that bit on newspapers as direct quotes in an argument from a couple months ago on this same topic, though. :( It certainly explains common practice, which "you are allowed to cite primary journal articles without backing them up from secondary sources" does not, in addition to bring a contradiction of the various guidelines.)
It is simply not true that WP disallows citations to the academic cryptologic literature. The opposite is true.
Again, you've confused the actual policy, which says that you can't take a journal article, infer from it conclusions the author didn't make, and then draw those conclusions without any other support in the body of a WP article. The policy does not say that journal articles aren't sources.
"""None of the content in question is "wrong": we are only discussing things with "inappropriate weight". Are you saying that the game Space Empire Elite (the article that Jason Scott is angry about today) did not exist? Was there something else on the article that you believe was "wrong"? We are explicitly discussing "notability" here.
(Sadly, I can't even look at the article, for the reason I already complained about: when Wikipedia decides to delete something, it is totally removed from the website in a way such that you cannot use edit history to get it back, making arguments about it impossible post-facto. I wish I could make a real judgement for myself.)"""
Everything after those two paragraphs was "ok, of you insist on making a confusing non sequitur about Britannica based on that one rhetorical sentence from Nate's argument, I guess I can play along and argue", but is sadly the only thing you insist on arguing about, ignoring th deletionism issue. :(
That means you won, congratulations: it does not mean I don't want them rebutted, as if so I wouldn't explicitly cede; the implication there would be I don't want to lose, and yet here I am, saying you must be right, as I can't back my arguments up anymore. You won: don't be angry you won.
However, you simply ignored all of the main deletionist-related arguments in your quest to defeat the one unrelated thing you felt you could argue against (the verifiability of crypto papers on Wikipedia). "I'm mystified by why you feel the need to do that, by the way."
You first ignored Nate's argument, concentrating on the rhetorical ending note; you side-stepped my calling you out on that (so you could argue more about verifiability), and then you ignored my attempt to reconnect to the mainline argument... you are pretty much just trolling, and I fell for it :(.
(added:) Wait, define "quietly edited"? I didn't remove anything from them (normally for typos and grammar). I often redraft what I say during the couple minutes after I post it, but I don't alter the argument. The only edits I made to that post were 1) to remove a sentence "please argue your point better" which wasn't even in my original submission (and when I looked at it, I decided it was needlessly insulting; it was there maybe 30 seconds; I shouldn't have said it, and I apologize), 2) to add the paragraph at the top, which I explicitly said was an "edit", as I felt it not worth a second reply; and 3) to remove a hyperlink I had originally used as the reference for the thing on newspapers in a panic at 3% battery life based on a Google hit that was wrong, replacing it with me explicitly ceding the argument when I couldn't find the right hyperlink, which only puts you at an advantage.
That's completely ineffective. The paper by Jaffe shows how to perform DPA (passive, not active attack) against various counter modes of AES, despite the permutation in use for the counter.
http://www.iacr.org/workshops/ches/ches2007/presentations/S1...
However, in order for me to remove the first reference and add the second one, I'd have to undertake a war of participation on Wikipedia. The outcome would not be determined by anyone I respect in cryptography, but by people who happen to be dedicated to a particular a message board.
You have your own experience with such people:
http://en.wikipedia.org/wiki/Talk:StankDawg
Wikipedia is a cesspool, and I'm very sad to see "worse but free" winning out yet again.
Incidentally, while I strongly advise not reading the "StankDawg Debates", I'll note that they're a case of me, a deletionist, losing an argument to anti-deletionists.
1. Britannica choosing not to have an article on it since they can't do it justice.
2. Papers or books (you know, things that cost money and real effort to develop) on cryptography that give much more thorough, accurate, and unbiased information.
Perhaps the answer to this is to allow for "personal" wiki pages that can be approved for inclusion into the main body of Wikipedia.
From my perspective, I would rather have lots of extra fluff in Wikipedia that I will never see rather than whole subjects being deleted from the site on the whims of a few jerk admins.
Edit: And you seem to be defending the deletion of BBS history under the guise that "Else spam would be everywhere!" which isn't an argument I agree with.
A friend of mine recently created a Wikipedia article for one of her own economic theories. She is not an economist, and as far as I know has never studied economics. The theory is loosely based on the work of one economist, but the sources she listed included her own blog, that economist's work, and newspaper articles--not articles about "the XYZ Theory," but articles about events she believes are relevant to her theory. But the end result was a reasonably well-written article, with plenty of linked citations (which would appear relevant to anyone not paying close attention), about a theory that does not exist. And because the article got so little traffic, she was the only one editing it.
The thing is, when she linked to that article from her blog and social media profiles, and from other tangentially related Wikipedia articles, suddenly that theory seemed to have some validity that it hadn't had before. And googling "XYZ Theory," which would have gotten 0 results the day before, now got a handful of results. All of them were related to that article or her blog, of course, but someone casually searching for information might just click through to the Wiki article without noticing.
The point is, articles about non-notable topics aren't just irrelevant. They're of much lower average quality than the rest of Wikipedia because they have much less traffic and many fewer potential editors knowledgable about the topic. They're more likely to have factual inaccuracies, more likely to be biased, more likely to just suck. And it'd dramatically reduce the confidence one could have in any Wikipedia article about anything one hadn't heard of before, because you might be its first viewer.
The proper solution seems to be to let the articles stay, and learn that they don't have much weight.
And if deletionism is a reaction to SEO optimization, as the other comment states, it is even more of a tragedy. Really, we have to delete articles for fear of somebody getting an unfair SEO boost? Again the solution seems to be to just lower the impact of links from Wikipedia.
We all say "sure, we know not to trust wikipedia" but when it's the first google hit for many topics, can you really pretend you always read it critically?
Second, there are clear practical reasons why it would matter if the examples I provided survived as Wikipedia articles:
* The overall perception of Wikipedia's reliability --- which, in reality, is comparable to that of Britannica's for most subjects, and particularly for subjects that are important in the context of traditional encylopedias --- would be sharply diminished.
* The encyclopedia would be full of overtly false information, since no contributor to the project would be able to verify any of the information in those articles.
* Wikipedia would be ruthlessly gamed to get pages to the top of Google SERPs (it already is, and a lot of deletionism over startups and individual people is a reaction to that).
* The project would lose huge amounts of time to mediating squabbles over article name real estate as different garage bands or local volleyball teams argued over which should be the first listed name for their project.
The naysayers in 1999 said Wikipedia would be hopelessly inaccurate, not that it would contain lots of information. I don't think "This site will become a vast repository of information about all subjects great and small" would even have counted as naysaying.
You know what, though? For all of the breadth of your list of horrors that would suddenly be covered on Wikipedia: I'd love to see every single one of those on the site. I want to see them represented, in fact, so strongly that I am having a difficult time convincing myself that you aren't actually just being really sarcastic about this whole issue (and then I'm being too daft to realize "oh, he's being absurdist, listing things which we actually would want, to make a point").
The concern should only be whether or not the information, if found on the site, could possibly be accurate or trusted, and the realization to have there is that a better solution to that problem is to more effectively expose the edit state of the article (a la IBM's "History Flow" visualization), as the problem of accuracy and notability is already a serious problem in individual paragraphs of articles about topics whose conclusion is non-controversial even to deletionists.
Unfortunately for that rebuttal to my comment, AfDs on Wikipedia are archived:
Wikipedia:Articles_for_deletion/Log/#{ YEAR }_#{ MONTH }_#{ DAY }
For instance:
http://en.wikipedia.org/wiki/Wikipedia:Articles_for_deletion...
You will, perusing these pages at random, discover a couple things:
* There are a LOT of terrible articles submitted to Wikipedia every day
* The examples I provided upthread are representative of what Wikipedia actually deals with.
* An astounding amount of effort goes into diligently handling those articles on a case-by-case basis.
A few months ago, another "deletionism" freakout on HN happened when a famous SEO expert's friend found her article deleted, and the SEO expert wrote an angry blog post. In discussing the incident on HN, two things emerged: first, the erstwhile WP subject was not in fact notable, and second, in handling the AfD for the article, before any media attention had landed on that particular debate --- that is, while that AfD was just another routine debate --- a Wikipedia volunteer actually took the time to look up the library circulation numbers for a book the subject had ostensibly published.
My "rebuttal" was actually that I liked that entire list of things, and I can only see Wikipedia being better for including information on all of those topics (along with appropriate visualization to demonstrate the inability to as easily trust the content that has had fewer editors, something you need anyway to solve the "subparagraph of article 4" problem).
Honestly, I thereby cannot see one thing in your most recent comment that actually responds to either of those contentions; again, to be highly repetitive on purpose: 1) that those things are not inherently bad, and 2) that there are better ways of solving verifiability for both articles and paragraphs (the latter of which being important).
(Also, on the off-chance this wasn't clear to anyone: the actual articles are deleted permanently. What is archived publicly is only the argument regarding the deletion, which is often just "I don't think it has enough sources", "I think it counts", "no, it didn't", as in the case with the article Jason Scott is complaining about today: without the actual text of the article or what the references even were, you cannot judge for yourself. I maintain this is inexcusable for a project with Wikipedia's charter.)
* Those articles are extremely likely to be inaccurate, because nobody outside the very small number of people with firsthand knowledge about them can verify them.
* It's manifestly obvious that those articles would quickly overwhelm the encyclopedia with obviously bullshit content; again, just look to the absolutely enormous quantity of totally ridiculous articles logged, complete with deletion debates, on the AfD link I provided.
* The articles themselves would spark huge time-wasting debates about placement and weighting, which is something that already happens with verifiable articles.
* A huge incentive exists to push vanity content onto WP because of its prominence on the Internet.
I already made these arguments: I do not see you responding to them; I, and other people on this thread, do not agree with your personal assertion regarding how the site will be "overwhelmed", so that is not an argument unless you can provide actual evidence that an infinite number mostly-pointless articles will cause Wikipedia harm. The closest you come is jut asserting that people will overly trust it, not why or whether deletionism is a better fix than my proposal.
(edit:) You also should tie the response back to the game in question: even if many of these spam articles should be culled, maybe the barrier to culling should simply be higher, in order to decrease the false positive rate. That was the argument made at the very top of this thread and, you know what?... you seemed to ignore it as well, as you only seem to care about the one issue: whether crap could exist and whether the site could get a lot of it if it had no filters at all... that isn't even controversial (and to the extent that it is, it seems to mainly be surrounding whether the specific things in your list were crap, not whether one could imagine something that was truly worthy of being deleted, even "speedily").
Indeed, your tacit assumption is that deletion improves quality, on average. But (a) you don't consider the many false negatives and (b) you don't consider the effect of deterring knowledgeable contributors, without whom quality will decline. We know many knowledgeable people have been discouraged because of the hostility of Wikipedia editors towards newcomers. You've seen the stats on how Wikipedia edits are topping out, how a tiny core of editors rules the site.
It is not hard to have a good search algorithm determine which pages are crap and which aren't. Some obvious features: the number of editors, Flesch-Kincaid level, number of edits, size of the page, number of references, etc.
Automatic search will always outperform manual deletion at scale, especially for an audience as diverse as Wikipedia's.
If you think "search" is the answer, you don't need to do anything. Let the deletionists do their thing, and just put your content somewhere else. Search will find it.
But it's currently far, far too low.
That any small gang of weirdos can get pages deleted is bazonkers. And given the rise of deletionism it's easy to round up a deletion posse for damn near anything.
It needs to be harder to delete. The Library-of-Babel examples[1] you give would still be deleted even if the bar were higher.
[1] http://jubal.westnet.com/hyperdiscordia/library_of_babel.htm...
Also, do words mean specific things to you, or are you just trying to communicate how article deletion makes you feel? Because when you call Wikipedia a "lie" that has "inverted the premise" and "ratched the door shut" on contributors, you should know that's not only false, but trivially falsifiable.
Here's how:
1. Go to http://en.wikipedia.org/wiki/Special:NewPagesFeed
2. Sort by "Oldest"
3. Set a filter on articles by "new editors".
4. Notice how there are tens of thousands of them.
It seems likely to me that a lot of people who are up in arms about "deletionism" really do think individual admins randomly zorch articles for no reason. That's not generally how it works.
And because none of the defenders were wikipedia insiders there was nothing anyone could do.
Anyone who spends serious time on Wikipedia could tell you immediately what the site would be like if it required "hundreds" of people to delete a page: completely overrun with spam and vanity pages --- not about BBS door games or individual BBS's or open source authors or even bloggers, but about individual local little league teams, bands that have never played gigs or released tracks, marketing professionals, individual episodes of fake TV shows, attempts at defaming individual high school girls, 6th place finishers in city council elections, political parties with one member, "microbrew" beer pages for each of some guy's homebrew projects, made up cities in made up countries as part of some guy's counterfactual world history, repeat all previous entries in this list for each of Japan, England, Belgium, and South Africa, amateur open-mic-night standup comedians, individual bootlegged videos of concerts, "one of the 100 most influential Chinese-Canadians in British Columbia", individual unpublished short stories, as-yet-unreleased new energy drinks from as-yet-unstarted startups, 9000 pages about individual aspects of the Ron Paul campaign, film festivals from towns in Nebraska with fewer than 5000 people, repeat all previous entries in this list 10 times over the course of a year with slightly permuted names, fake professors at real universities, real professors occupying fake posts at fake universities, 46 different theories of cold fusion, fake cancer cure, fake cancer cure, fake cancer cure, fake cancer cure, fake cancer cure, world championship players from nonexistant Starcraft leagues, aspirational pages about made-up video games, individual authors of self-published ebooks, and of course, 1-5 pages about every single company incorporated in every US state.
In other words: exactly what you'd expect from an "encyclopedia that anyone in the world can edit". The site would be every naysayer's 1999 prediction for what Wikipedia would become.