Skip to content

Comment on Deletionists slowing Wikipedia's growth

Comments

Currently the entries for Hacker News, Jessica Livingston, and Trevor Blackwell are all flagged as insufficiently notable:

http://en.wikipedia.org/wiki/Hacker_News

http://en.wikipedia.org/wiki/Jessica_Livingston

http://en.wikipedia.org/wiki/Trevor_Blackwell

Trevor's entry is a particularly interesting case of the pathology. Originally his bio talked about how he'd made the first dynamically balancing biped robot. That set off a firestorm about priority. One nutter got so mad that for a while Trevor's bio read as if the most significant thing about him was that he'd falsely claimed to have made the first dynamically balancing biped. Eventually the Wikipedians solved the problem by deletion, so now his bio simply says he's some guy who builds robots. Which of course wouldn't be notable, if that's all he was.

The rules for notability are straightforward. Subjects are notable "by dint of being written about". The notability tag is very easy to dispel: provide references to credible reliable sources. They clearly exist for Blackwell.

The misconception you're fostering here is that the {{Notability}} tag is somehow a black mark on the article. It isn't. The entire encyclopedia is under constant construction. The tags are there to direct the attention of editors.

Your complaint is particularly misleading because the Blackwell article is, in fact, badly sourced; it has "External links", but its "References" all point to Blackwell's own sites. The {{Notability}} tag is correct, not because Blackwell isn't notable (he again clearly is), but because the article doesn't properly establish why.

The rest of your critique may or may not be valid (I have misgivings about WP, too), but the main thrust of your comment here is bogus, and you should acknowledge that.

I don't think your argument holds up here, though. As I've pointed out elsewhere in this thread, there are already Wikipedia policies in place which cover verifiability of information and citations to reliable sources (in fact, these criteria form part of one of Wikipedia's "Five Pillars").

And there are already perfectly useful procedures in place for dealing with articles which fail these criteria: there are tags for indicating that particular articles, sections or individual statements are in need of citation, and there's a process for evaluating sources referenced by articles.

Given this, the notability guideline seems at best to duplicate matters already covered by full-fledged policies. And in real-world situations, its main function seems to be turning Wikipedia into a popularity contest -- prove that your topic has enough Google juice, and it stays!

If your whole argument is that {{Notability}} should instead be {{refimprove}} (or whatever that tag was), then, fine, but I think my point holds.

My argument is really that the notability guideline in general serves no constructive purpose on Wikipedia; everything useful that it purports to do is covered by other policies or guidelines, which leaves only the non-useful things it does, like cause flamewars.

The notability guideline supports the non-negotiable verifiability principle. In the absence of WP:N, the amount of random content on WP rises. With it, the amount of difficult-to-verify content. With that, the amount of blatantly false content. The burden of weeding out that content falls on people who could otherwise be improving articles on subjects of note.

I think something people miss about WP is the fact that, at the end of the day, all the articles on this massive free volunteer project are published under an encyclopedia's masthead. It really is an actual encyclopedia. It's not the Internet. If something survives in WP, it's supposed to be good. The project is fundamentally opposed to bogus articles; in fact, the project is about not having bogus articles.

Unfortunately "improving articles on subjects of note" (as well as arguing notability) requires more contentious editing/lawyering compared to "developing new articles on marginal topics". The lawyering is exciting for one kind of old-hand contributor -- but turns off many others.

So optimizing for "improving articles on subjects of note" may paradoxically serve to make total editor attention more scarce, and waste more of it on low-value disputes and incremental refinements. "Why bother," a newbie or loosely-attached casual contributor might ask, "when all these busybodies keep marking subjects of interest to me for deletion, and mangling prose with obsessive footnotes (or requests for same)?"

!#@!@^% deletionists are ruining Wikipedia. They'll be the first against the wall when the revolution comes.

Deletionist rational that makes sense. "not having bogus articles" that one statement is causing me to seriously doubt my inclusionist position.

At what cost are you prepared to hold the "no bogus articles" line?

Some of the deletionists are very far into 'burning the village in order to save it' territory. I.e., they're so obsessed over "quality" that they'll snuff out anything that might even be the slightest bit questionable, erring on the side of removing things.

That strikes me as stupid and needlessly destructive. If bogus articles creep in, the solution is to correct them and move on.

The obsession over Wikipedia's "reputation" is likewise misguided. Unless the entire concept of "an encyclopedia anyone can edit" is abandoned, it's never going to be a totally reliable source, and users will always have to be cautioned to fact-check before depending on the information. Outside of the Wikipedia community, this is pretty much taken for granted.

The best compromise solution I can come up with would be to periodically 'fork' the WP articlebase, and let the deletionists go to town on the fork, honing it down into some subset of the working version, which users could then choose to browse if they wanted something with a slightly higher barrier to entry. However, my guess is that very few casual WP users actually care.

I don't think your logic holds. The fact that it's "an encyclopedia that anyone can edit" is why deletionism is a healthy force. If the deletionists let up, and WP spiraled out of control with vanity articles, it would likely stop being an encyclopedia anyone can edit.

Again, I think people personalize this. The good deletionists don't care about you or your subject. It's the project they're sticking up for, not the non-notability of Trevor Blackwell. When the topic of debate is Trevor Blackwell, they'll lose. When it's Ketchup_salt, they'll win.

There certainly are bad deletionists. A lot of them. But I don't think that's a symptom of deletionism. I think it's a symptom of editing-as-sport and status-seeking, and that those are the problems that are really poisoning WP.

Deletionism nudges the project power more towards those with "editing-as-sport and status-seeking" motivations. Procedural games are what they like.

For new and casual contributors, deletionism forces them to engage on topics they aren't passionate about -- older topics and wikipedia lawyering -- rather than the marginal topics they're excited to get started (and which may become rigorously 'notable' in due time). Some of these people will just be driven away.

I share your concern, but this is an argument that applies equally well to all of WP's process. It's orthogonal to deletionism.

'Orthogonal' is the strong claim I'm disputing; other WP process does not create the same problem. For example, editing someone's contribution to improve its voice/NPOV or suggest verification can encourage casual contributors; it's positive attention. "I got something started, others are paying attention, progress is occurring. Fun!"

Deletionism -- whether the judgment that something should be deleted or following through with deletion -- is negative attention. It uniquely discourages contributors and often destroys content of small-but-positive value. (For example, it destroys the important 'first drafts' of topics that will someday easily pass 'notability'.)

Deletionism also shrinks the territory on which collaboration can occur. A deleted article can be neither corrected nor improved; it is a void. Perhaps there is someone somewhere who could add the citations... justify the importance... benefit from the partial information -- but deletion forecloses that possibility, even though cheap storage and cheap search means incomplete scraps of information can better find their audience/editors than ever before.

"In the absence of WP:N, the amount of random content on WP rises. With it, the amount of difficult-to-verify content."

Absolutely not. The difference would be as follows.

With the notability guideline, ten million monkeys bang on their keyboard:

"Delete, NN"

"Delete, NN"

"Delete, NN"

"Delete, NN"

etc.

Without the notability guideline, the same ten million monkeys who are already reviewing these articles at least enough to type a brief vote in a discussion will bang out:

"Delete, not verifiable."

"Delete, not verifiable."

"Delete, not verifiable."

"Delete, not verifiable."

While the latter requires more keystrokes (and thus in aggregate represents significantly more effort on the part of the monkeys), it is otherwise not distinguishable in purpose or effect from the former. At least, it isn't distinguishable so long as WP:N is only used for articles which cannot be verified by referring to reliable sources (hint: that's a bad assumption to make).

That argument isn't just false, it's prima facie false.

In the WP:V case, some articles are stricken for being nonverifiable. Others are kept despite being nonnotable.

Wikipedia is now charged with the task of maintaining and article about Thomas Ptacek. That article is fine right after the AfD, of course. But 3 months later, it claims that Thomas Ptacek is the inventor of ketchup salt. Some WP volunteer has to catch that and scrub it out.

That overhead doesn't exist in the WP:N encyclopedia we have today, because a Thomas Ptacek article wouldn't survive AfD for notability reasons.

This problem is the whole point of WP:N.

"But 3 months later, it claims that Thomas Ptacek is the inventor of ketchup salt. Some WP volunteer has to catch that and scrub it out."

And with the notability criterion, lots of WP volunteers have to spend lots of time trawling AfD and typing "Delete, NN" over and over and over again. Meanwhile, unverifiable information pops up and must be removed from articles about notable subjects just as much as non-notable subjects.

So how, precisely, does Wikipedia save effort under this plan?

Or, more accurately, why do you believe that more effort is involved in:

"Nominated for deletion due to: not verifiable/does not cite sources"

than in:

"Nominated for deletion due to: not notable/does not assert notability"

In both cases, research must be done by multiple volunteers to assess the claim and come to a decision. But in one case the criterion is relatively objective; in the other it's an invitation to highly-subjective flamewars.

Because WP:N cordons off a whole huge swath of human knowledge and says WP doesn't have to bother keeping articles about it accurate; it can just be removed wholesale from the encyclopedia.

WP says, Pokemon yes, Thomas Ptacek no. Pikachu must be yellow, not green, but it doesn't matter if Thomas Ptacek invented ketchup salt, because WP doesn't have to cover it.

I feel like you're kind of having to go through contortions to avoid this fact, James.

I just don't really understand why another redundant label -- "not notable" -- is needed to express the underlying concept of "not verifiable".

With only a couple of narrow exceptions, lack of notability is not sufficient for speedy deletion by an administrator, which means that all those articles still take up time and consume the effort of editors who participate in the AfD. In other words, I doubt that there's enough of a saving of time and effort to justify the myriad other problems the notability criterion is known to generate.

And so it seems to me that it'd be far simpler to do away with the notability guideline and keep the focus on what it allegedly aims to accomplish: ensuring that information in Wikipedia can be verified by reference to reliable sources. If an article cites no sources to back up the information it presents, get rid of it unless/until someone comes up with suitable sources, and throw the politics and the popularity contests out of the process.

Wikipedia notability tag:

If notability cannot be established, the article is likely to be merged or deleted.

How long does an article have to correct itself before the deletionists have their way with it? It seems to me that when plenty of reasonable people agree that someone is notable, and the problem is external links vs. reference citations, the {{Notability}} tag isn't the best way to highlight this and in fact is distracting in a destructive way.

How is a tag destructive? It is, for what it's worth, plainly obvious that the article needs improvement.

Stop taking tags personally. They're about the project, not about the subject.

The tag encourages users to delete the article rather than fix problems. That is the point. It is not personal.

That's precisely what the tag doesn't do. If you want to encourage people to delete the article, you put it up for AfD.

Then why not remove the warning saying "this article is likely to be merged or deleted"?

"Does not meet notability guidelines, please help fix" is not the same as "Does not meet notability guidelines, help fix or this article gets deleted." The latter frames the issue as content not permitted, the former frames the issue as content yet to come. The difference is significant.

I'm inclined to agree that they are not "generally notable"

The entire HN entry is 4 sentences long. All of the footnotes are sourced from either blogs or YC itself. If HN must be included in Wikipedia, it's probably more appropriate as a mention in the YC page.

The individuals are in the same boat. All the references about Trevor are either from his own site or his company site. The footnotes for Jessica are largely sourced from blogs and videos.

Wikipedia is an encyclopedia of general information. The fact that they are involved with YC or have a company that builds robots is not inherently notable or of general interest.

So, the question remains, how is the fact that those articles are on wikipedia harming you in any way?

Even before this article came up, I've searched for and read those articles, and found them helpful. It provides a succinct overview of the individuals, with links if you want to pursue futher. What difference does it make where the links are sourced from?

And who determines what 'general interest' is? It's of general interest to me!

Well, Wikipedia determines this by setting their notability guidelines.

At the end of the day, it's up to each individual community to determine what is and is not appropriate. It wouldn't hurt you if somebody started posting celebrity gossip articles on HN, but HN would probably downvote them into oblivion because it has decided those types of articles are not relevant to this community. I can say with a high degree of certainty that you would not object to this practice. With Wikipedia it's the same principle.

Except the notability guideline is at odds with the stated goal of Wikipedia -- to collect and provide the totality of human knowledge, for free.

Far better would be to do away with the notability guideline entirely. The purpose it allegedly serves is in keeping unverifiable (as in, cannot be confirmed by reference to reliable sources) information out of Wikipedia, but Wikipedia already has two full-blown and actively-used policies regarding that matter. All that's left for notability, then, is to turn Wikipedia articles into popularity contests, something that's definitely at odds with its stated goals.

You might be right, but I think you confused Wikipedia's goal with Google's.

I looked (casually) and did not find anything on Wikipedia that said its goal was "to collect and provide the totality of human knowledge, for free."

There's a whole huge article on Wikipedia that contradicts this:

http://en.wikipedia.org/wiki/Wikipedia:ISNOT

Well, technically it's the goal of the Wikimedia Foundation in general, rather than the Wikipedia project in particular.

See http://wikimediafoundation.org/wiki/Home which states:

"Imagine a world in which every single human being can freely share in the sum of all knowledge. That's our commitment."

It's clearly not the goal of Wikipedia itself.

Not the same though - in HN the rest of the community would be affected by celeb gossip links because we'd have to look at them on the main page.

But who really cares if these pages exist on Wikipedia? It's not as if you have to even see that they exist, much less actively ignore them.

But who really cares if these pages exist on Wikipedia?

Would it affect the speed of serving up the more desired pages? My heart is with the inclusionists, but would Wikipedia be as easy and as fun to use if its database size were much larger for the same budget?

How are these brief entries any less important than descriptions of the complete episodes of BSG, Doctor Who, The Wire, or dozens of other TV shows? Yes, I've read many of those episode articles and I'm sure they get more hits, but we're talking about making information available to anyone who might needs it. Who can say this is needed, this is not?

It seems odd that a real person can't get an entry but there are probably thousands of fictional characters written up.

You're really arguing not about whether topics should be included in the encyclopedia, but how their articles should be organized. Your argument is placated by reorganizing The Wire into one big article with all the detail from every episode. But the article was broken up not because of some separate notability for each episode (though some of them clearly are, ahem, notably notable), but because it makes WP's coverage of The Wire easier to read.

I don't think his argument is "placated" by that at all -- the point remains that lots of minutiae are apparently deemed worthy of inclusion, while arguably significant real-world people or organizations (which in many cases meet the letter of the law for notability) are not. The "notability" standard is applied very inconsistently, and in a way that seems to boil down to "whatever the Wikipedia brass find interesting." If that's the standard, fine, but be honest about it.

Again, a canard. Notability is far less subjective than it's made out to be. What episode of The Wire hasn't been written about in a reliable source somewhere? Zero, is how many.

Further, WP accomodates a myriad of non-notable facts. They're simply attached to notable subjects. The idea that every single sentence in WP must be notable is a straw-man argument. Once you accept that The Wire is notable enough for inclusion in WP, by nature of being the single best piece of long-form televisual drama ever created, then the question of how its articles are organized stops being about notability and starts being about information design.

And, I mean, have at it and all. But don't make it something that it isn't.

That may be, but you have to admit that there are thousands of fictional characters with larger impact on our world than the majority of real persons...

There is the possibility of analysis paralysis. In the extreme case where everyone had a Wiki page, an individual's name might return 100+ different results.

I would have to say I am in the inclusionists camp though. If someone took the time to create an article for some obscure topic, then they find it worthy enough to share meaning that most likely someone else out there would also be interested in the topic.

Spend some time patrolling AfD, and I bet you will lose that opinion quickly. You may not become a deletionist, but you will stop believing topics are worthwhile simply because "someone took the time to create an article". The sheer volume of dross people pump into WP every day is staggering.

tptacek in 3, 2, 1...

You're mean.

Sometimes my timing is impeccable.

If there's no reliable sources, what happens when somebody adds lies (whether intentionally or not)?

Well you see paper is expensive and shelf space is even more expensive. How does this matter to wikipedia you ask? Well imagine a website trying very hard to be like a book, I don't mean the ideas therein, I mean the physical limitations - Wikipedia!

I flipped Hacker News back, notability is already established by the included references, as per Wikipedia rules. I've already had this argument on Wikipedia, but Wikipedia pages tend to unravel if you don't check them continuously.

For Trevor and Jessica: shouldn't any of the feature length magazine articles that name them as founders of Y Combinator be enough to establish notability?

I fixed the articles of Trevor and Jessica.

Web reach should be enough to trigger notability. Any website in the top 100,000 in terms of traffic for a year is important enough to merit Wikipedia attention.

Notability should be determined by how useful the knowledge is, not whether or not the mainstream press bothers to pay attention.

The level of notability should be set by Wikipedia's ability to reliably incorporate information, not by some arbitrary standard. As the number and expertise of editors increases, the notability standard should decrease.

The goal of wikipedia is the "sum of human knowledge," (http://en.wikipedia.org/wiki/Wikipedia:Free_speech) not to "the sum of human knowledge that has been covered in the mainstream press."

I had the Alexa rankings for Hacker News in there the last time I edited it, but a Wikipedia editor has deleted that reference in his or her infinite wisdom.

Sadly, I was talking more about the wikipedia I would like to see than about the wikipedia that currently exists.

Personally, number of people using something is a lot more meaningful to me than number of journalists writing about it. Of course, as a very occasional anonymous editor, I suppose I don't have much right to complain.

Bio's of people have always been contentious on Wikipedia - especially in the realms of what is or is not notable.

I think the current accepted theory is that you ask the question: if I googled their name could I get all of the information I want from one pre-existing (non WP) link. If the answer is yes then in most cases there is no need for the article.

What if it takes 2 or 3 places to get all the information? Or 300 places that aren't the mainstream media? My problem with the current notability theory is that it isn't flexible enough for what should. For now, deletionists versus inclusionists isn't going to end, but the current notability policy could explicitly allow for the inclusion of articles with no mainstream media sources.

For example, I think every professor that has ever had an article published in Nature should have their own Wikipedia page. More professors than just that, but I don't think there is any argument against a scientist that is extremely respected in his field, even if no one in the mainstream media has decided he is worth mentioning left.

It is bad to depend on the mainstream media for what is notable because Wikipedia is already beyond that. I used the SFGate to establish notability on an article. Techcrunch has a better Alexa rank than the SF Gate, but Techcrunch doesn't count for some reason.

> What if it takes 2 or 3 places to get all the information? Or 300 places that aren't the mainstream media?

Exactly. In that case a Wikipedia article would be appropriate (I thought that was clear in my post, sorry)

In the example of Jessica Livingston this page: http://ycombinator.com/people.html provides all of the information found on wikipedia.

physicsforums.com the largest science question and answer site, a partner of Scientific American and sited as a source for a number of wiki entries has been repeatedly rejected as not significant enough.

It seems wiki won't let you create an article about another web site.

AboutSource Built by g1lg1l

Hackerly is an independent reader for Hacker News, built on the public HN API. Not affiliated with Y Combinator.