News for hackers. Stuff that matters.
Re:In Defense of Inclusionism by StavrosK on Monday December 12, 2016 @11:43AM (#13159408)
Is this one of those situations where someone uses a proxy metric for something that is itself easily measured? "We must protect verifiability, non-notable articles aren't easily verified, therefore non-notable articles should be banned"?

Is there a name for this? I've been noticing (and being frustrated by) it a lot.

Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @01:29PM (#13160455)
I don't understand how this is a "proxy metric".

I'll give you the shortest defense of deletionism I can write:

There are two problems with non-notable articles in Wikipedia.

The first is definitional. An encyclopedia is a tertiary source: a roadmap and synopsis of existing sources. Non-notable sources are by definition those for which existing reliable sources are absent. All such articles are, in a very simple sense, "original research". One of the most basic founding principles of Wikipedia, as old as NPOV, is "no original research".

The second is pragmatic. Wikipedia is committed to minimizing error. Entropy exerts tremendous force on the project, which occupies extremely valuable Internet real estate. Every article on Wikipedia is in a sense a commitment by the Wikipedia community to mount a defense against bias, advertisement, promotion, vandalism, and basic inaccuracy. The notability requirement, which is at bottom based not on "fame" but on the quality of sources available for a topic, is an extremely reasonable boundary to put on the expectations Wikipedia can have of its unpaid editors to combat that entropy: to wit, that we will do so only when available sources make the job possible to do in the first place.

Re:In Defense of Inclusionism by Tomte on Monday December 12, 2016 @02:21PM (#13160974)
> Wikipedia is committed to minimizing error.

Try to tell that to the guys who feel like they own the /dev/random article on Wiki-DE:

https://de.wikipedia.org/wiki/Diskussion:/dev/random#.2Fdev....

An "IP" (which is basically a kind of online racism, because ostensibly "IPs" are not addresses but a certain kind of people) tried to correct the usual /dev/urandom misinformation. It was reverted, called vandalism and the user bakunin finally stated that he reverted because "the discussion was getting on his nerves" and "even if the change was factually correct", it was obviously vandalism.

I know why I don't try to contribute to Wikipedia. When the half-time of donations to the project can be measured in minutes, it's just not worth it.

Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:29PM (#13161064)
It will probably not surprise you that this is also the reason I no longer contribute to Wikipedia. When you're an expert in a subject, writing an encyclopedia article about it is very tedious. It's easy for me to write an explanation of how the LRNG works, or what its faults are, or what the urban mythology is about it. It's much more painful for me to do so while backing every point I make up with some secondary source --- none of which can be related to me in any way, lest I be accused of injecting my own bias into the encyclopedia.

But while I'm not going to say that the way Wikipedia deals with expertise is perfect (it is deeply imperfect), there's a sense in which this is the way it's supposed to be.

If I have important things to say about the LRNG, why would I write them in Wikipedia? Should I not write my own articles, or a chapter of a book, or an academic paper, and let people who want to write encyclopedia articles cite them? In my own work, I can obey whatever guidelines I want to, and write in whatever fashion is most gratifying and effective for me.

Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @01:59PM (#13160749)
(EDITED)

> An encyclopedia is a tertiary source: a roadmap and synopsis of existing sources.Non-notable sources are by definition those for which existing reliable sources are absent. All such articles are, in a very simple sense, "original research". One of the most basic founding principles of Wikipedia, as old as NPOV, is "no original research".... we will do so only when available sources make the job possible to do in the first place.

Why not just incorporate that into verifiability then? As I understand you, you're arguing that (1) non-notability generally implies un-verifiable, (2) Wikipedia info should be verifiable, and so (3) Wikipedia info should be notable.

Isn't it easier to demonstrate objectively that an article contains unverifiable info than it's non-notable?

Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:06PM (#13160814)
I don't understand the question. That's what the standard in fact is.
Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @02:11PM (#13160862)
I expanded my comment for clarity. Let me know what you think.
Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:29PM (#13161074)
The notability guideline is part of the verifiability requirement.
Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @04:04PM (#13161888)
Yes I know. Your argument sounds like you're justifying the notability guideline primarily by appealing to its implication for verifiability, and I'm asking why you don't just require verifiability (and possibly making it more stringent).
Re:In Defense of Inclusionism by StavrosK on Monday December 12, 2016 @05:43PM (#13162722)
> I don't understand how this is a "proxy metric".

I didn't say it was, I asked if it was.

> Non-notable sources are by definition those for which existing reliable sources are absent.

This would also make them non-verifiable, no?

> Wikipedia is committed to minimizing error.

That's fair.

Re:In Defense of Inclusionism by luso_brazilian on Monday December 12, 2016 @12:22PM (#13159787)
It reminds me of the XY problem [1] that plagues both our field and any other field where one can devise a solution to a problem without formalism or thoroughly understanding what's being solved.

> The XY problem is asking about your attempted solution rather than your actual problem. This leads to enormous amounts of wasted time and energy, both on the part of people asking for help, and on the part of those providing help.

> - User wants to do X.

> - User doesn't know how to do X, but thinks they can fumble their way to a solution if they can just manage to do Y.

> - User doesn't know how to do Y either.

> - User asks for help with Y.

> - Others try to help user with Y, but are confused because Y seems like a strange problem to want to solve.

> - After much interaction and wasted time, it finally becomes clear that the user really wants help with X, and that Y wasn't even a suitable solution for X.

I'm sure that's not the name of the phenomenon you are asking but it is certainly a closely related one.

[1] http://xyproblem.info/

Re:In Defense of Inclusionism by StavrosK on Monday December 12, 2016 @12:44PM (#13159983)
Yep, that is very closely related, just not exactly that. Mine goes something like:

- We need to improve X, something that is easy to measure improvement in.

- I've noticed that Y somewhat correlates with X.

- Let's make an effort to improve Y!

Re:In Defense of Inclusionism by resfirestar on Monday December 12, 2016 @01:36PM (#13160515)
It's not a "proxy metric", it's the only possible metric for inclusion in an encyclopedia as I understand encyclopedias. I saw the example of a mildly popular band somewhere else on the thread, so I'll run with that.

What can an encyclopedia say about a band that has a website and a Spotify page, but isn't big enough to attract attention from the press or music critics? We can certainly say that they are a band, anyone can see that. We can say they have 10,000 plays on Spotify, and list their current members. Maybe we can dig up a newspaper article saying that they performed someplace. But beyond that, what is there? They've probably got a biography on their website, but it's in no way verifiable, and the only person required to believe it is the guy putting it on the Wikipedia page. Everyone after that will read it on Wikipedia and assume it's 95% likely to be true. And so on. We're left with barebones facts that are of no help to anyone looking for reliable information about the band. All of this applies equally well to little-known authors, random Indian corporations, and your cat. If all we have to go on is a couple of primary sources and maybe a passing mention in the press, what is there to say without giving undeserved credence to the primary sources?

When notability is assessed, don't think "is this topic something that will be googled by someone looking for more information?" Think about the role of Wikipedia and what we can offer by providing an overview and references to secondary sources. The notability guidelines are about ensuring that there is a baseline of quality facts that we can provide to readers.

But really, I'm just rewriting the ideas of the consensus based notability guideline, which is much better written and specific: https://en.wikipedia.org/wiki/Wikipedia:Notability

Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @01:50PM (#13160654)
> We're left with barebones facts that are of no help to anyone looking for reliable information about the band.

The fact that the band exists and has a newspaper article about it is helpful. Otherwise it's article will be very short. So what?

> All of this applies equally well to little-known authors, random Indian corporations, and your cat.

Yes to the first two. (My cat has zero verifiable information about it.) They will have very tiny stub articles. So what?

> If all we have to go on is a couple of primary sources and maybe a passing mention in the press, what is there to say without giving undeserved credence to the primary sources?

So don't say anything else.

I really, really don't understand what's motivating your position. I'm trying!

> But really, I'm just rewriting the ideas of the consensus based notability guideline, which is much better written and specific: https://en.wikipedia.org/wiki/Wikipedia:Notability

That page is almost exclusively about defining notability (which, because of the intrinsic vagueness, necessarily takes a long time and is mostly unsuccessful). The only part that's directly about the reasons for the requirement is this subsection:

https://en.wikipedia.org/wiki/Wikipedia:Notability#Why_we_ha...

And the only argument it brings up that I didn't address is this: "We require multiple sources so that we can write a reasonably balanced article that complies with Wikipedia:Neutral point of view, rather than representing only one author's point of view." This, again, is something that can easily be addressed by just requiring verifiability to include coverage by multiple sources. (Even in highly notable articles, there are a large number of facts that are only verified through a single source; this isn't a reason to delete the fact.)

Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:10PM (#13160854)
The problem Wikipedia sees for these very short articles that state nothing other than the fact that they were mentioned in passing in a newspaper article is that there is no potential for them to ever be good articles. The band is defunct, no additional reliable sources will ever be created for it, all available sources have been exhausted to produce an article of almost no value.

You're right that the article has some marginal value. I think many deletionists would even agree.

The problem is risk/reward. Every WP article is a commitment to defend a topic against entropy and malice. Every article is part of the attack surface exposed to vandals and spammers.

So in the view of most of Wikipedia, articles that are fated to forever remain "stubs" have negative net value.

Reasonable people can obviously disagree about this, but I find the deletionist argument against zombie stub articles to be very compelling.

Re:In Defense of Inclusionism by Tomte on Monday December 12, 2016 @02:14PM (#13160889)
> Every WP article is a commitment to defend a topic against entropy and malice.

But at this point you've basically given up any attempt of civil community processes and you're doing things not because they are right, but because not doing them might give the enemy a speculative advantage.

Re:In Defense of Inclusionism by resfirestar on Monday December 12, 2016 @02:26PM (#13161030)
If that were true, we'd refuse to have articles on Middle East topics and tell people to look elsewhere, because it's an absolute nightmare and NPOV routinely loses to people with agendas. But it's worth it in the mission to be a comprehensive encyclopedia.
Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:23PM (#13160992)
I don't understand. My paraphrase of Wikipedia's position on (what I'm calling) zombie stub articles is that the risk/reward for them doesn't pan out. They have marginal positive value but pose significant risk of overhead and error.

But that's very much not the case with other stub articles, or even other zombie articles (where sources have no doubt been permanently exhausted, but the sources we have today make room for a solid 100 word article).

Re:In Defense of Inclusionism by Tomte on Monday December 12, 2016 @02:28PM (#13161056)
I my have read too much into the word "defend", but before my inner eye I saw all-out war against "them".

All over some moderately important issue, at best.

Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @02:18PM (#13160936)
If this is really a problem, why not just lock the article at a stub with a few verifiable facts? How is this more editorial work than deleting, and fighting a subjective battle over notability?
Re:In Defense of Inclusionism by resfirestar on Monday December 12, 2016 @02:25PM (#13161015)
Because it's obviously (well, to me) better to leave the article in the default state of not existing and have the option of allowing anyone to quickly add coverage, rather than needing a committee to unlock a topic if it does become relevant. This is good adherence to the "Wikipedia Is Not Paper" idea.
Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @04:09PM (#13161921)
Except that there already exists a whole special mechanism for patrolling frequently created-then-deleted pages.

You only need to lock stub articles that repeatedly have people add unverifiable info to them.

Finally, if the pages doesn't exist, then anyone who wants to exercise the "option...to quickly add coverage" doesn't have access to the info that would have been contained in the stub!

Re:In Defense of Inclusionism by resfirestar on Monday December 12, 2016 @02:12PM (#13160864)
In my opinion, there is absolutely no reason to have a tiny stub article with information of no interest to anyone. Keeping the band example going, people come to Wikipedia to read about a band without diving into everything that has ever been written about them; if there's nothing to say, we can't help. But I guess your question is more "why not?" Editors will spend more time per year reverting vandalism and removing original research than readers will ever spend using it. And that's really, really optimistic: what actually happens with stubs on little-known topics is that they fill to the brim with original research and info from bad sources, and anti-vandalism editors don't stop it because either they don't know if the sources are good, or they're rightly afraid of being yelled at for removing content, since anything looks like a potentially constructive addition to a shell of an article. If you require a decent amount of sourced information, someone adding a wall of text without adding references looks suspect and is (somewhat) more likely to be removed in a timely manner.
Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @02:22PM (#13160982)
If this is really a problem, why not just lock the article at a stub with a few verifiable facts? How is this more editorial work than deleting, and fighting a subjective battle over notability?

I can't understand why you think one can't defend against this stuff using mechanism that don't lead to effects like the one described by rsync https://news.ycombinator.com/item?id=13159057

Re:In Defense of Inclusionism by resfirestar on Monday December 12, 2016 @02:40PM (#13161175)
Addressed the locking below. I'm sympathetic to rsync.net, and maybe I spent too long in the Wikipedia bubble, but what exactly is the argument for it having a Wikipedia article?

>rsync.net has been mentioned in the press widely over the course of over 10 years - everything from articles about our warrant canary to articles about our ZFS support. In fact, other wikipedia pages mention and discuss rsync.net.

So it's had some coverage related to first commercial use of a warrant canary and is discussed in that context. It commercially implemented a feature of rsync (the application) that should be talked about on the page about rsync the application. But what can be said about rsync.net? Maybe someone read about their warrant canary and wants to know more about this company...but no independent source has written about who they are and what makes them so savvy with opsec and rsync. Anything outside those narrow topics is original research or primary sources. I think things are working as intended.

Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @04:18PM (#13162003)
You're arguing that because the info on rsync.net is contained in other Wikipedia pages, there's no point in having an rsync page. But the page discussing ZFS support doesn't link to the discussion of the warrant canary, i.e., there's nowhere to go to find all the rsync.net info on Wikipedia. And the reason to not have this page is...because it's too much of an exposed surface for bad actors?
Re:In Defense of Inclusionism by tptacek on Monday December 12, 2016 @02:34PM (#13161122)
This is answered here, in the second place you asked this question:

https://news.ycombinator.com/item?id=13161015

(better if replies go there too)

Re:In Defense of Inclusionism by jessriedel on Monday December 12, 2016 @05:56PM (#13162834)
OK. I explained there why I think resfirestar's answer is unsatisfactory, i.e., why there are easier, cleaner, and less destructive ways of dealing with that problem than using notability. Would still be interested in your take there.
Re:In Defense of Inclusionism by saint_fiasco on Monday December 12, 2016 @03:03PM (#13161403)
>But beyond that, what is there?

The hypertext. The names of the band members can be links that lead to the articles on individual members. There can be a section describing the music style, with quotations of review sites or just the band homepage and the part that says they are a neo-core-techno-metal or whatever can have links to articles that explain what those things are.

Re:In Defense of Inclusionism by pwdisswordfish on Monday December 12, 2016 @01:27PM (#13160431)
Can you tell me how do you 'easily measure' notability in a robust, non-gameable way? I'm not aware of any such measure existing. 'Substantial coverage in reliable, independent sources' seems the best possible at the moment.
Re:In Defense of Inclusionism by warkdarrior on Monday December 12, 2016 @12:46PM (#13160004)
> Is there a name for this? I've been noticing (and being frustrated by) it a lot.

Laziness?