From Wikipedia, the free encyclopedia

The future of the Arbitration Committee and its role in dispute resolution[edit]

As most people who see this page will already know, English Wikipedia is about to elect eight members of its Arbitration Committee. This election will consume an enormous amount of the community's time. More than twenty editors will become candidates, and will spend ten or twenty hours apiece writing candidate statements and answering dozens of community questions. Another twenty editors will review all the candidates' statements and answers and contributions and prepare detailed voter guides analyzing the candidates' qualifications and recommending whom to vote for. Several hundred more editors will then review the twenty-plus candidates' qualifications, some in more detail then others, and cast their votes. Administering the candidacy processes and running the election also aren't trivial tasks. Doing some quick math and making some rough-and-ready assumptions, at least a couple of thousand hours of community members' time will be expended in this process. That amount of time will be comparable to that involved in every election since 2004. And it's a necessary expenditure of time, I hasten to add; I wouldn't suggest selecting the arbitrators any other way.

The primary role of the Arbitration Committee is to resolve on-wiki disputes—primarily user conduct disputes. The Committee is not supposed to decide content disputes, and it is not supposed to make policy, and it generally avoids doing both of these things, except operationally by defining what types of editing or behavior are so beyond acceptable limits that they will result in an admin's getting desysopped or an editor's getting banned. The Committee also has other responsibilities, such as dealing with situations that can't be addressed on-wiki because private or sensitive information is involved, and selecting checkusers and oversighters. But the core of the role is meant to be deciding arbitration cases, and the candidates run and get elected primarily based on their qualifications for that task.

ArbCom receives lots of attention both on-wiki and elsewhere, in part because of what it (as a sitting arbitrator, I should say "we") actually does, and in part because for better or worse, there is no other real wiki governance body and so ArbCom is occasionally forced to fill in and do a small part of what such a body might do.

The Arbitration Policy stresses that arbitration is the last step in on-wiki dispute resolution. This makes sense: an arbitration case is an (overly) complicated process, whch typically takes weeks (and in too many unfortunate instances months) to resolve, is often an unhappy time for everyone, and frequently culminates in one or more editors being excluded from the project. Editors are routinely counseled not to file requests for arbitration if there is any possibility that the dispute will be resolved "by the community." I have certainly cast my share of votes to decline a case request because the issues are being, or could be, discussed somewhere else.

The result is that a lot of disputes that might have been resolved through an arbitration case in 2004 (when the Committee was formed) or in 2007 (the year I was a clerk) are now resolved through other means. Sometimes, such a dispute is resolved by an individual administrator directly with a misbehaving user, such as by an indefblock. (Indefblocking, while recognized as a very serious step, is done much more readily now than in the project's early days.) Sometimes, it's resolved through formal dispute resolution short of arbitration, such as on one of the dispute-type noticeboards (BLP/N, COI/N, DR/N, and so forth). Very often, it's resolved on one of the administrators' noticeboards (AN or ANI). Since the demise of the Community Sanctions Noticeboard circa 2007, AN is the designated location for community sanctions discussions.

The result of all these changes is that although the arbitrators' workload remains quite significant—one could keep quite busy with banned-user appeals alone, and we receive double-digit numbers of e-mails every day—in its core function of accepting cases and resolving them through decisions, the Committee is actually underutilized. According to Wikipedia:Arbitration/Index/Cases, in its first full calendar year of operation (2005), the Arbitration Committee considered one hundred cases. To be sure, some of these were very easy cases that would never get as far as arbitration now, and some of them involved questions that were unsettled in the early days of the project but are have well-settled answers now. Still, 100 cases. In 2006, 116 cases. In 2007, 91 cases.

And then, it suddenly plummeted. In 2008 (my first year as an arbitrator, though there's no cause-and-effect involved), only 35 cases. (I've written before about the reduction in the Committee's caseload, but I never realized until this minute that it was quite that sudden—a three-fifths drop in a year!) In 2009, 30 cases. In 2010, eleven cases (another drop of more than 60%!). In 2011, sixteen cases. And through the first ten months of 2012, eleven cases (including none in the past three months, although we have had a steady diet of clarification and amendment requests arising from prior decisions).

My point is not that editors should rush to bring disputes to arbitration. Of course, if a dispute can be resolved in a much simpler, more straightforward, less contentious, less faux-legalistic way then it should be.

On the other hand—have the Committee and the community gone too far in accepting that resolving any issue on AN is better than resolving it in arbitration? The virtues of having (for example) a community sanction discussion on AN instead of an arbitration case are that (1) any interested member of the community can participate in the decision-making, as opposed to arbitration, where anyone can comment, but only 15 people can vote; and (2) the decision, whether a good one rather than a bad one, is usually made in a couple of days (almost never more than a week) rather than in several weeks or even months. And some critics of the Committee (on- and off-wiki) would of course add (3) it takes the decision away from those muddleheaded arbitrators (every one of whom, of course, was elected by that same community within the past two years).

But community decision-making of the type that happens on AN has its drawbacks too, many of which are well-enough known that I don't have to list them here.

Another means by which the Committee has reduced its caseload (as well as the number of findings and remedies within individual decisions) is the increased use of the discretionary sanctions regime. As a result, what might have been a new arbitration case against a user in a topic-area that was the subject of a prior case, instead becomes the imposition of a sanction by an administrator, perhaps followed by one or more threads on the arbitration enforcement (AE) noticeboard, which is handled by self-selected AE administrators rather than by the arbitrators.

Questions for thought:

  • Should the Arbitration Committee be accepting and deciding more cases?
  • Should the community be bringing more cases to arbitration?
  • Should the arbitration process be changed to make it easier to bring cases and to get faster decisions in the cases that are brought? (An old perennial; undue delay in decision-making has been the Committee's biggest flaw since its inception, for reasons that probably warrant a post of their own sometime soon.)
  • Should the community sanctions process on AN, or the procedures for AN and ANI more generally, be modified to ensure the perception and the reality of fair, open, well-informed discussions and decision-making?
  • Is the net effect of transferring some of the workload from the arbitrators on RFAR to other administrators through discretionary sanctions and AE a positive one?

I welcome comments on these questions on the talkpage, and I also hope that they'll be discussed by candidates in this year's election. Newyorkbrad (talk) 01:32, 3 November 2012 (UTC)

Discuss this post

On having an article deleted[edit]

I'm about to have my first article deleted.

Although I've spent a too-high percentage of my wikitime for the past couple of years doing arbitrator and administrator work, I used to write new articles, and I plan to do it again in the future. The bulk of my new pages were law-related, including more than fifty biographies of judges, but there were others.

On a whim one rainy day in 2007, I wrote my least important article. It was about a very silly novelty song that I used to enjoy when it was played on the Dr. Demento Show, called "My Name Is Not Merv Griffin". (Enjoy that, as a bluelink, while you can.) It was a fun little article (I should say "stub") to write, because I got to use my best deadpan in describing the song: The song is based on the singer's lament that he supposedly resembles celebrity talk show host Merv Griffin, and that Griffin's fans are stalking him, with disconcerting consequences. Anyway, I wrote it, and someone categorized it and stub-tagged it, and it's sat there harmlessly for five years.

And I know it interested at least one reader, and as a lagniappe, I learned something I'd never known, when Scottandrewhutchins edited the article to add the source of the melody. I'd never known that under my breath, I was humming an old French song (that was used in a classic scene in a Charlie Chaplin movie, to boot!).

But even in 2007, I knew that My Name Is Not Merv Griffin almost certainly didn't meet the notability requirement for songs, and that it was living on borrowed time.

Obviously, we need notability criteria. Although I've written before that we probably put too much collective effort into policing the precise boundaries of notability in areas where having some borderline articles is harmless, a line must certainly be drawn somewhere. We can't be posting an article about every self-produced song recording that sold five copies. And even though I flatter myself that my instincts are generally sound, "Newyorkbrad really likes it" certainly cannot be an anti-deletion criterion.

My trivial little stub stuck around for five years, but now someone has spotted it and put it on AfD, and in a week it'll almost certainly be gone. I spent half an hour the other day looking for sources and references on the song, and came up pretty empty ... it's non-notable and therefore, under well-settled Wikipedia policy, out it goes.

The article will live on for awhile in any number of mirror sites and maybe a deletionpedia or two, but of course it won't be the same....

Will the Online Compendium of All the World's Knowledge really be a better encyclopedia without My Name Is Not Merv Griffin than with it? Newyorkbrad (talk) 02:02, 3 November 2012 (UTC)

Discuss this post

Help pick Newyorkbrad's FA project[edit]

A few years back, a couple of editors used to cast their RfA !votes on based primarily on a single criterion: "1FA". That meant that to gain their support for adminship, the candidate must have been the primarily contributor to at least one featured article.

That was, and would still be, a rotten criterion for adminship; lots of well-qualified administrators haven't written a featured article. But if it had been the prevailing wisdom when I had my RfA in 2007, it would have had at least one virtue, which is that it would have forced me to write an FA before I sought adminship, and today Wikipedia would have one more featured article.

In the past few years, I've spent much too much of my wikitime arbitrating and administrating. Too often I've said to myself that I need to spend more time in mainspace, and I've headed in that general direction, but typically to copyedit or to add a nugget of information. It's useful work, and nothing I'm ashamed of, but I can't say that I've exactly been helping to write the encyclopedia. And it hasn't gone unnoticed that I (along with a number of the other sitting arbitrators, though by no means all of them) have the authority to vote on sanctioning our best content contributors although my own best contributions lately have been of a very different kind.

So, whatever else I may or may not do on Wikipedia in the next couple of months, it's high time I got my first FA written. In order that it be on a topic that's familiar to me, I'm going to write about a deceased former Justice of the United States Supreme Court.

The question is which one, and I'm going to let my blogwatchers (since I now know that they are at least a non-empty set) decide for me. Should the mainpage someday feature someone of recent memory such as Warren Burger or Potter Stewart? Or someone well-remembered from longer ago like Oliver Wendell Holmes or the first John Marshall Harlan? Or a challenge such as Thomas Todd or Gabriel Duvall, each of whom was named as "The Most Insignificant Justice" in two famous articles in the University of Chicago Law review?

Post your preference on the talkpage, and I'll go with the flow as of a week from now and get to work. Newyorkbrad (talk) 02:57, 3 November 2012 (UTC)

Discuss this post

One approach to RfA reform...?[edit]

Ever since I started editing in 2006, the community has discussed how we might improve the requests for adminship process. There have been very few successful suggestions, and I haven't been able to come up with many myself.

But I do remember one time I thought one of the bureaucrats had done something bold....

We used to have a very good editor whose username was The Random Editor. (Unfortunately, he hasn't been around for awhile.) In 2007, he posted his RfA, which did very well, closing at 96/1/3.

So in due course, a bureaucrat closed the RfA as successful, and when he did it showed up on my watchlist, with the edit summary: "random editor promoted".

... And I thought to myself, "I suppose that's one way we could do it ..." Newyorkbrad (talk) 23:32, 11 November 2012 (UTC)

Discuss this post

A surprisingly missing article[edit]

Now that I'm struggling with the fact of having entered my second half-century, my thoughts were drawn to a poem on the perennial subject of intimations of mortality. (I first came across it in a book by P.J. O'Rourke of all people, if memory serves.) It's the poem whose last stanza concludes:

For health, wealth and beauty, wit, learning and sense,
Must all come to nothing a hundred years hence.

That couplet, which is all I remembered of the poem when I went to look it up, struck me as evincing a modern sensibility; if I'd been asked to guess the author from what little I remembered, I would have suggested Dorothy Parker. I was surprised that it dates to 1637 and was written by a seventeenth-century English poet named Thomas Jordan.

I was curious to find out more about Jordan and searched on-wiki, only to find that Thomas Jordan (poet) is still to be started. Someone else has noticed that we need the article—it's listed (as a redlink) on the Thomas Jordan disambiguation page.

I suppose I'll start it myself if no one else does, but I know about as much about seventeenth-century poetry as I do about how to take my own appendix out, so I'd prefer to wait for someone more qualified.

(And if anyone's curious, here is a link to the text of the poem. Hopefully no one will dispute that a text from 1637 is public domain. :) ) Newyorkbrad (talk) 00:14, 6 March 2013 (UTC)

Discuss this post

Clear remedies, arbitration decisions, and AE[edit]

As most of you know, I've been an arbitrator on this project for a while. As such, I've written or voted on dozens of ArbCom decisions, and I'm also called upon to comment on appeals from sanctions imposed under discretionary sanctions or on the Arbitration Enforcement (AE) noticeboard. (N.B.: This isn't a post about any particular arbitration decision or AE thread!)

Disputes sometimes arise on AE concerning the meaning of a remedy or sanction imposed in an arbitration decision or in some other wiki-forum. Sometimes the criticism is warranted, and the ArbCom (or whoever else wrote the remedy or imposed the sanction) could have been clearer. Sometimes a situation arises that couldn't have been anticipated when the remedy was originally written. And other times, there is simply the problem that some borderlines are fuzzy, and no attempt at precise wording can resolve in advance every potential situation that might come up.

It's important to realize that there is no way either the ArbCom or anyone else can craft remedies in a fashion that eliminates the need for those interpreting the remedies to use judgment in borderline or unanticipated cases. To stress this point, the classic law-school and philosophy hypothetical (Justice Breyer likes to use it) is an ordinance and a sign that says "NO ANIMALS ALLOWED IN THE PARK." That sounds clear and unambiguous, right?

Well, now suppose you are the park ranger or the town magistrate. Which of these people violated the ordinance?:

  • (A) Mr. A, who walked his dog in the park?
  • (B) Mr. B, who drove his car on a road through the park with his dog inside?
  • (C) Mr. C, who walked across the park carrying his small dog inside a container?
  • (D) Mr. D, who walked across the park carrying his hamster in its cage?
  • (E) Mr. E, who walked across the park carrying his goldfish in its bowl?
  • (F) Mr. F, who ran rushing across the park carrying his seriously injured dog to the vet's office on the other side?
  • (G) Mr. G, who walked across the park carrying a chicken sandwich?
  • (H) Mr. H, who walked across the park by himself, but knows that humans are animals?

I suppose it might be possible to write the ordinance and the sign in such a way as to provide a clear response in advance to each of these possibilities, not to mention dozens more that the enterprising and litigious might think up. But lawmakers and judges and sign-drafters, and even the most long-winded of wiki-arbitrators (he typed, looking in the mirror) are still usually going to write "NO ANIMALS ALLOWED IN THE PARK" rather than 17 pages of (wiki-)legalese. So we are always going to need interpretation. The suggestion that remedies and sanctions be clearly worded is well-taken; but any suggestion that any latent ambiguity that arises in applying a remedy means that we must have done a lousy job in writing it is not.

If anyone wants another example of how clear-cut rules prove ambiguous in the real-world, please read and enjoy the best law-review piece on statutory interpretation ever .

Discuss this post

A question about searchability, robots.txt, and NOINDEX[edit]

Most pages on the English Wikipedia are searchable by search engines such as Google and appear as results (often high-ranking results) in web searches. One key exception is that Google decided several years ago (I believe on its own; I don't know of any Wikipedia involvement in the decision) that article talkpages are Not Of Any General Interest and these pages are excluded from search results. Another exception is that editors, acting within consensus, have the ability to exclude particular pages or categories of pages from web searching. I'm not sure whether there is a list of the types of pages that are so excluded, short of working through the massive category of unindexed pages, but there has been consensus over the years that internal pages focused on individual editors, such as RfAs, RfBs, RfCs, AN and ANI, arbitration case pages, SSIs, and the like, are among those that should not appear in search engines. (Proposals to exclude user talkpages from indexing have not attained consensus, although I've never understood why anyone would think that article talkpages would be of less interest to the general public than userspace pages. But this post is about a different issue.)

According to the conventional wisdom and as summarized on Wikipedia:Controlling search engine indexing, there are two ways of excluding a page from searchability: either a robots.txt tag or a "NOINDEX" template. As far as I'd been led to believe, these two designations were equivalent, and applying either of them to a page would exclude that page from searchability by all currently significant search engines.

However, a couple of months ago, I learned of a page in one of the categories that was not supposed to be coming up in searches, but was in fact coming up. Upon investigating, it appeared that this page simply predated our introducing non-indexability and no one had gone back and tagged it. However, in further reviewing the situation, my attention was drawn to this page on Google's site, which appears to reflect that the "NOINDEX" marking is more robust and reliable that the "robots.txt" designation. (See the discussion in the first few paragraphs of the page, through the highlighted seventh paragraph.)

Given that I'm among the least qualified of English Wikipedia "functionaries" when it comes to technical issues, I'm raising here the question whether we ought to reevaluate any of our technical practices for how we tag pages that shouldn't be searchable. This is probably a question for posting on the Village Pump/Technical, but if what I am describing is an issue that was addressed years ago, or if my description of the problem is completely incomprehensible, I'll probably be told so more kindly here than there, so I'm starting here first. My thanks to anyone who has relevant knowledge or suggestions to share.

Discuss this post

Forgery and Wikiality[edit]

We've all read about people manipulating their, or their friends' (or worse their enemies') biographies on Wikipedia. We also all heard about people molding Wikipedia to fit the world as they wish it were, rather than the world as it is—what Stephen Colbert, in exaggerated form, calls "Wikipediality."

Alex Wilkinson reported an interesting example of this phenomenon in his article "The Giveaway" in last week's New Yorker. (A link to the New Yorker article is here — full text for New Yorker subscribers, the first two paragraphs for others. I recommend it.)

The article is about a 58-year-old man named Mark Landis. Mr. Landis lives in Laurel, Mississippi. For a time during his childhood, he attended St. Mary's Town and Country School in London.

According to the article, dozens of times over past 25 years, Mr. Landis has walked into a museum and donated what he described as a valuable but previously unknown artwork. He describes the pieces as the work of a reknowned artist, though not one of the very best-known artists (Paul Signac, Stanislas Lépine, Hans von Aachen, Alfred Jacob Miller are examples). And every time, it turns out that Mr. Landis created the artwork himself, and used classic art forgers' techniques to make the piece appear older than it was.

This sort of art forgery raises well-known questions ("is this work of art the less meaningful or beautiful because it was created by Shlabotnik rather than Renoir?"). But certainly the museum world sees quite a difference between the work of a great or near-great artist and even the most faithful re-creation or simulation of one, and for that reason, does not appreciate Landis's contributions. Since Landis never requested or accepted any payment for his donations, and apparently never even took a tax deduction for them, he hasn't been charged with any crimes. Wilkinson discusses Landis's motivations, but the workings of his mind remain unclear. What is clear is that Landis wants very much to be thought of as an art dealer, and as a philanthropist.

The relevance to Wikipedia? We have an article about this individual, Mark A. Landis, which details his history of art forgeries—but Wilkinson's New Yorker article doesn't mention that article. What it does mention is this:

One of the things [Landis] likes to do is check the Wikipedia article for Laurel, where he was described as a notable resident, and the one for St. Mary's, where he was an art dealer and a philanthropist. Late in 2010, he saw that the listing under Laurel had been altered, "to something derogatory," he said....

And Wilkinson's article concludes:

After lunch ... Landis was in good spirits. I'd seen him happier only once, a few days before, when we checked the Wikipedia page for St. Mary's. He hadn't looked for some time. He almost winced as he scrolled down the page. Then his face broke into a grin. "Hey, I'm still there," he said. "Art dealer and philanthrophist."
He turned the computer toward me so that I could read the entry, then he leaned over to be sure his printer was on so he could make a copy. "Otherwise, somebody might say something bad about me and change it," he said. "And then I won't be an art dealer and a philanthropist any more."
Discuss this post

The quality of Wikiscience[edit]

There has been a good deal of discussion of the quality of Wikipedia's science articles in recent weeks. In particular, the accuracy of one editor's contributions has been discussed both here and on other sites, and several reviewers have concluded that unfortunately, a fair number of the articles require corrections and clarifications.

As I said last week in an ANI post, errors in Wikipedia not only affect our own site, but too often propagate all over the Internet. Many of us enjoy creating and developing our encyclopedia as one of our primary avocations, but all our work is of little use, and may be actively counterproductive, if our article content isn't based on a firm foundation of factual accuracy.

Yet for all that we read about errors in Wikipedia, there are knowledgeable people who seem to think that we do a good job. The April 3, 2014 issue of The New York Review of Books contains a letter to the editor by a man named Michael Konrad, addressing Freeman Dyson's review of the book Brilliant Blunders by Mario Livio. It would be an understatement to say that Konrad didn't like Livio's book and he didn't like Dyson's book review. The reasons for his dislike are interesting, but not especially relevant here. What is relevant is that Konrad concludes his letter by opining:

I would advise a student of science not to read this book or the review by Dyson, but to look up the scientists in Wikipedia to get a far more insightful (and correct) history of their careers. References for full-length biographies are cited for those who want more detailed accounts.

To which Dyson replies in part:

I am glad to be corrected when I go wrong, and I have written elsewhere in praise of Wikipedia.

I haven't seen Dyson's earlier piece "in praise of Wikipedia," and I'd like to. But in general, this is someone who knows what he's talking about when he discusses the history of science and the biographies of scientists. For all the mistakes Wikipedia makes—and gosh knows we make too many, in science and every other field—if Freeman Dyson thinks that on balance we are doing a praiseworthy job, I think that is an interesting bit of information.

Discuss this post

Wikipedia and the law of computer misuse[edit]

The other day, I blocked a banned user who was socking around his ban through the latest in a series of IPs and throw-away account. Exasperated, instead of just labeling the block "sockpuppetry" or "ban evasion" or the like, I wrote in the block log and on the talkpage exactly what I was thinking: "banned user unlawfully accessing and interfering with the site in breach of the terms of use." I've been asked to explain the thinking behind that summary, which I am happy to do.

This being a lawyer's post, it comes with several disclaimers. The first is that I didn't come to Wikipedia to be a lawyer. Thinking about the law should (apart from articles about law and lawyers and judges, and with occasional exceptions for copyright issues) be remote from the Wikipedia experience of 99.9% of editors. The second disclaimer is that I haven't discussed anything in this post with anyone in the Wikimedia Foundation office; the Foundation has a highly qualified legal staff, but I have no affiliation with them, and am simply a volunteer like most of the rest of you. (I'm also not writing in my capacity as either an administrator or an arbitrator on this project.)

With all that being said ... does a banned editor who continues to edit, in breach of a ban imposed by the community or its dispute-resolution procedures, thereby act in a legally impermissible manner?

Merely breaching a website's internal rules or "terms of use" (TOS), without more, usually does not give rise to either criminal or statutory civil liability under statutes such as the US Computer Fraud and Abuse Act, although it may be civilly actionable by the website owner under other theories. There are several cases (some briefly summarized here) declining to find liability for "computer abuse" where courts believed that prosecutors or plaintiffs sought to overextend the concept, including to TOS violations.

The best-known of these cases is probably United States v. Drew. The facts of this case are sad. Two young teenage girls quarreled, and one asked her mother, Lori Drew, to help find out what her former friend was saying about her. The mother created a Myspace account under the fictitious name of a nonexistent 16-year-old boy, who purported to flirt with the girl for some time, but then abruptly told her that the world would be a better place without her—. Heartbroken, the girl hanged herself. After state prosecutors investigated but decided there was nothing they could do, the federal prosecutor indicted Drew for (among other things) violating the CFAA by accessing Myspace's computers "in excess of authorization." The basis for this charge was that Drew had breached Myspace's TOS by creating a fictitious account, which according to the MySpace TOS is not allowed. The jury convicted on this count, but the District Court reversed and dismissed the charge, on the ground that a reasonable person would not expect a simple violation of the fine print in a website's TOS to constitute a crime.

Another interesting precedent in this area will be made any day now by the New York Court of Appeals (the state's highest court) in People v. Golb. Raphael Golb is the son of Norman Golb, a scholar who espouses a particular theory as to the origin of the Dead Sea Scrolls. More established scholarly rivals of Norman Golb, including a professor named Lawrence Schiffman, support a different theory. Over the course of several months in 2008, Raphael Golb signed onto computers in the library of New York University (which as an alumnus he was permitted to use), and sent hundreds of e-mails accusing Schiffman and others of plagiarizing Norman Golb's work and denying Golb credit to which he was entitled. Most troubling, Golb created Gmail accounts in Schiffman's and some other targets' names, and sent dozens of e-mails in which (for example) Schiffman purportedly confessed and admitted to plagiarizing from Norman Golb.

Raphael Golb was indicted and convicted for numerous crimes under New York State law, including multiple counts of identity theft, criminal impersonation, forgery, "aggravated harassment," and one count of unauthorized use of a computer. The Appellate Division, First Department affirmed all but one of the convictions in this opinion. Among other things, the court rejected Golb's First Amendment challenges to his convictions, finding that Golb acted with criminal intent and that his e-mails could not properly be characterized as "satiric hoaxes or pranks." The crux of the decision was that "Defendant was not prosecuted for the content of any of the emails, but only for giving the false impression that his victims were the actual authors of the emails. The First Amendment protects the right to criticize another person, but it does not permit anyone to give an intentionally false impression that the source of the message is that other person."

The Court of Appeals granted Golb leave to appeal and heard oral argument on March 25. Anyone interested in computer law should stop reading this right now and take the time to read the transcript or, better still watch and listen to the argument. It's a fascinating 40-minute discussion (with Ron Kuby appearing for Golb), fully accessible to non-lawyers. The questioning, particularly by Judge Smith and Chief Judge Lippman, raised a host of cutting-edge issues in this field: the overbreadth of the harassment statute (see Volokh Conspiracy post here); the availability or not of a parody defense; and (the driest issue but the one relevant here), whether a person is guilty of "unauthorized use of a computer" under New York law where he was, in fact, authorized to use the computer, albeit not for the specific purpose he wound up using it for. Is it really the case, the judges asked the prosecutor, that if an employee has the employer's permission to use the computer for work purposes only but signs onto Facebook, the employee has committed a crime? The prosecutor's answer was yes. The court's answer is going to be no. (The New York statute criminalizes "unauthorized" use of a computer but not use "in excess of authorization," and in that respect is narrower than its federal counterpart, so this will not become a direct CFAA precedent, but depending on how the court writes the opinion, it may still be instructive.)

So, merely socking around a ban, by itself, does not seem not to violate the statute. And this is as it should be. No one wants a system where every frustrated editor who socks around a block thereby becomes a criminal, so it's good that neither the WMF TOS nor the statute seems to contemplate such a result.

Are there, however, some limits? Does there come a point at which screwing around with Wikipedia actually crosses the line into legally impermissible behavior? Frankly, my strong hope is that we never need to find out the answer in a court decision. But if someone does press hard enough, he or she may find that the answer is yes.

Suppose a user is banned. (I don't plan to debate here the bona fides of the ban in question, although it was fully justified; one small piece of the saga can be found in my talkpage archive here, where the user said he was scrambling his password and leaving Wikipedia.) And suppose the user continues to post after the ban has taken effect. And suppose the user continues to post on-wiki with the avowed purpose of disrupting the site, voluntarily describing his behavior in words like these:

  • "I have been creating accounts since September in anticipation. They just blocked about 60, but thats only the last few days. All I wanted to do was help the project and they threw me out so now I will be the most prolific vandal, troll and sockmaster in Wikipedia history."
  • "So now if they want me to be a sockmaster, then fine, I'll pursue that with just as much enthusiasm as I did editing. I know they'll catch me eventually but in the mean time I will be a drain on resources and divert them from being able to do anything else."
  • "[O]nly about half those 60+ [blocked accounts] are me. The rest were just helpless well meaning editors. Same with the ones [administrator 1] blocked and the IP that [admin 2] accused of being me. They don't all have to be me, all they have to do is be caught in the path. I don't really care anymore if Wikipedia likes me or not, they can delete every edit I did. They didn't want me there anyway and they made that clear as crystal. So, since they didn't want me there, I'll have some fun. And I haven't even tried to be a sockmaster yet. Pretty soon it will be 260+ accounts."
  • "Since my help wasn't wanted, I'll just distract them with socking and trolling as I find the time. Days or weeks might go by and it may come in waves but it'll be fun."
  • "I doubt they'll tremble of fear me and it really just amounts to a waste of time. But since they didn't want me to help, I'll just be a pain in the ass and a distraction. In the process though a lot of innocent editors will be blocked (several already have), time will be distracted form the project and I'll have some fun. The only way they'll keep me away is if they range block the whole t-mobile and Verison Fios networks. I doubt they have the desire to do that."
  • "Case in point, [admin 1] recently got so annoyed with my pings he disabled the Echo pings. Others probably did as well but didn't post it. That means they are disabling functionality because of me. Good! They have continued to block and accuse editors who aren't me or my friends as being me. Largely because the checkuser app is crap. More good news! And that's after only 24 hours of being banned. Imagine the impact after a month. Maybe they ban editing from the Verizon network or t-mobile. Its hard to say what the long term effects will be, but its not going to be pleasant."
  • "In less than 48 hours I have gotten 2 range blocks for Verizon Fios which means a lot of people coming from 172 or 208 will need to get an IPblock exemption to edit in which case most of them will assume its me and deny it. 1 for me, 0 for WP. I have also caused several users to turn off pings (Echo) and distracted several users. Childish perhaps but I am having fun."
  • "Well I am up to 88 socks and that's not even counting the ones [admin 3] identified here which are mostly mine (but there are a couple that aren't). It also doesn't include my Bots Kumi-taskbot or A bot called bob so including those that pushes me over a 100. Plus the folks who turned of Echo pings, the range blocks of 138, 172 and 208 preventing editing to a large number of Verizon Fios and Us Navy users and the editors who weren't me that got blocked as collateral damage or several of my friends who have joined in. Not a bad start for 48 hours of naughtiness. Can't wait to get to the month mark."

... and that's just from the first two days, and it continues in that vein for week after week. (It's not a good use of my time to hunt down more diffs from on-wiki to go with those quotes from Wikipediocracy, but anyone who's followed this saga knows that I could come up with dozens of them, not to mention e-mails.)

It is not the case that any website is helpless to seek a legal remedy against a user who insists that he is going to continue to edit without permission with the admittedly foreseeable and intended effect of (1) causing the site administrators to have to spend the time dealing with his unwanted edits, and (2) triggering rangeblocks and thus interfering with access to the site by other users with whom he is unconnected.

I can readily fashion an argument that this type of conduct is against the law. I will not advance that argument in detail here, because to the best of my knowledge it has not been tested in application (I'd be quite interested if anyone's aware of any precedents; the DDOS cases may be the closest, but I acknowledge that this isn't that), and because there are far more direct means of dealing with the problem than invoking a statute. (The Wikipediocracy thread on this topic, in particular, has completely missed that "unlawful" can refer to civil as well as criminal law.)

I very, very much hope that all of this will remain in the realm of academic discussion. And in that vein, I'm going to take off my lawyer hat and remind the banned user in question, and everyone else who is reading here, that almost all of us came to Wikipedia as a hobby. And ... when a hobby stops being fun for you ... or when for whatever reason you're asked to leave the club ... the rational thing to do is to step away and find another hobby. You don't stick around and complain that everyone else is doing a lousy job enjoying your old hobby ... and you certainly don't destroy the clubhouse or scrawl graffiti on it and leave a mess for everyone else to clean up. And you especially don't outlast your welcome to the point that even one of the notoriously most relaxed and lenient administrators on the site, who is not especially hard-assed even against banned users quietly returning and doing good work (those who doubt that should carefully review this thread) is thinking, even fleetingly and metaphorically, about calling a cop on you, or to wonder whatever happened to the old Abuse response process, a page I had hoped never even to have to read.

[Name of user redacted], I'm sorry you are so disaffected with Wikipedia. You had high hopes for the project and for your role in it, and for whatever reason (this isn't the place to find fault), your hopes were dashed. But you really, really, really need to step away now. Goodbye. Newyorkbrad (talk) 23:35, 21 April 2014 (UTC)

Discuss this post

The new Google book[edit]

Over the weekend I read How Google Works by Eric Schmidt and John Rosenberg. I found the book interesting, though the parts that interested me might not contain much that is novel for readers who start out with greater background knowledge than I did.

Much has been written over the years about the symbiosis between Google search and Wikipedia. In this regard, I found it noteworthy that this book doesn't mention Wikipedia at all. Is that simply because the book was about business insights rather than search content, or is there something deeper to be gleaned from this?

Discuss this post

"Wikipedia's notoriously gangsterish back channels"[edit]

Yesterday's New York Times Magazine contained the annual "The Lives They Lived" set of in memoriam essays remembering important people who passed away during the year. It included a tribute to our late colleague Adrianne Wadewitz, written by Virginia Heffernan. (Adrianne was also the subject of a Times obituary in April, written by Noam Cohen. I believe she may be the first person to be remembered in mass media such as the Times specifically for her work on Wikipedia.)

Heffernan has written a fine tribute to Adrianne Wadewitz and I urge everyone to read it. However, in the midst of commemorating and celebrating Adrianne's work and mourning our loss as I read the essay, one line caught my eye. Heffernan writes:

Wadewitz, against the odds, rose to command authority in Wikipedia's notoriously gangsterish back channels.

I'm the first to agree that site governance and administration here is far from perfect. But do we really have "gangsterish back channels," let alone "notoriously gangsterish" ones? And if we don't, why does a respected journalist think that we do?

Discuss this post

A reference librarian reviews Wikipedia[edit]

The level of Wikipedia's accuracy and comprehensiveness, or lack of them, is a perennial topic of conversation. When someone wants or needs information on a given subject, is Wikipedia a good place to look? How likely is Wikipedia to contain the information, and how likely is the information it does contain to be correct? Opinions on this subject vary widely. I previously discussed one author's largely negative assessment in the second part of this book review I wrote for the Signpost last year.

Dr. Thomas Mann is a senior reference librarian at the Library of Congress. He is the author of books including Library Research Models and Library Research Methods, and has just published the fourth edition of the latter, under the title The Oxford Guide to Library Research, fourth edition (2015). The book is a comprehensive, updated guide to research resources in a wide variety of fields, including both paper and online sources. Mann has spent more than 30 years as a working reference librarian and has thought deeply about how library patrons, both casual and scholarly, can best gain access to the information they are looking for. I recommend his book highly; it is available online from the Oxford University Press and the usual online bookstores—and, I hope, in bricks-and-mortar bookstores as well.

In his new edition, Mann discusses Wikipedia beginning on page 1—although it is not clear whether he does so because Wikipedia is ubiquitous and helpful, or because he just wants to get the subject out of the way. Here is the crux of what he says:

A major problem that researchers inevitably have with any Internet search engine is that finding something quickly within the first two or three screens of retrievals does nothing to give them any sense of "the shape of the elephant" of their topic. With Web search engines you can never tell what or how much you are missing, nor can you judge the importance of what is in front of you in comparison to (possibly) better retrievals from alternative sources. The best solution, or at least partial solution, to this problem provided within the Internet itself is Wikipedia. The latter, being a kind of universal encyclopedia, has the virtue of providing concise overviews of just about any subject. It's a quick way to see the basic facts about your topic—to gain what the Wikipedia contributors deem to be overview information.
I am not going to provide any elaborate criticisms of Wikipedia; I frequently use it myself in situations that do not require academic documentation or footnoting. The latter concern, however, is very important within college or professional environments; you cannot cite Wikipedia as a source in student or professional writings because its articles, being subject to continual modification, are not stable; what you cite today may not be there next week, let alone years from now. (While it is technically possible to call up earlier versions of the same article, the fact that the earlier versions had to be changed is in itself good reason to avoid citing them.) Moreover, those who make the modifications sometimes introduce particular biases for which more formal editorial procedures would compensate. Often the best experts on particular subjects simply don't bother to correct (or even notice to begin with) what Wikipedia may say on their subjects—which means that those who do bother may have peculiar interests or agendas to promote. Nonetheless, Wikipedia is indeed often a very good source for getting an initial overview of basic information on an unfamiliar topic, especially in comparison to an overwhelming retrival of search engine hits....
Other starting-point sources are available in libraries, however, that are frequently far superior [to Wikipedia]....

What do our editors think of this assessment of our encyclopedia, both in their own areas of expertise and more generally? What can we do to make Wikipedia a more useful resource, and how (if at all) can we simultaneously point them to the more specialized resources that can bring them to the next level in researching the topics about which they want to learn more?

Discuss this post

A troubling New York court decision about privacy[edit]

This is a decision issued yesterday by a state appeals court in New York. Here is the first paragraph:

In this action, plaintiffs seek damages and injunctive relief for an alleged violation of the statutory right to privacy. Concerns over privacy and the loss thereof have plagued the public for over a hundred years. Undoubtedly, such privacy concerns have intensified for obvious reasons. New technologies can track thought, movement, and intimacies, and expose them to the general public, often in an instant. This public apprehension over new technologies invading one's privacy became a reality for plaintiffs and their neighbors when a photographer, using a high powered camera lens inside his own apartment, took photographs through the window into the interior of apartments in a neighboring building. The people who were being photographed had no idea this was happening. This case highlights the limitations of New York's statutory privacy tort as a means of redressing harm that may be caused by this type of technological home invasion and exposure of private life. We are constrained to find that the invasion of privacy of one's home that took place here is not actionable as a statutory tort of invasion of privacy pursuant to sections 50 and 51 of the [New York] Civil Rights Law, because defendant's use of the images in question constituted art work and, thus is not deemed "use for advertising or trade purposes," within the meaning of the statute.

Without getting into whether the decision was legally correct under existing New York law, I find the result very troubling. Do others agree? For my part, I trust that none of my fellow Wikipedians would even consider obtaining images, for any Wikipedia-related purpose, in this manner, and I would block anyone who did.

I note that the decision may be subject to further appeal, and also that the outcome of the case could, and probably would, be different in many other states or countries.

Addendum: commentary by Eugene Volokh here.

Discuss this post

Books inspired by information found and connections made on Wikipedia[edit]

My topic for today is books that were inspired by information found on Wikipedia, or that were written as a result of connections made on Wikipedia, or that in some way came about because Wikipedia exists and would probably not have come about otherwise. (I am not speaking of books that simply reprint Wikipedia content, nor of books about Wikipedia itself.)

I first learned that this was even possible from a review by Nicholas Lemann in The New Yorker, called "Spy Wars" and published in the summer of 2009. [A]

The review concerns books about the espionage committed by people such as Alger Hiss. Where does Wikipedia come into it? Lemann's review is discussing the work of journalist Alexander Vassiliev, who did research on Hiss in the KGB archives in Moscow. Lemann continues:

In 2005, [Vassiliev] looked up Alger Hiss on Wikipedia, and through one of the links found out about John Earl Haynes, a historian at the Library of Congress, and Harvey Klehr, a historian at Emory University, who had jointly published several books about Soviet espionage. He e-mailed Haynes, and soon the three of them were talking about collaborating on a book based on the notebooks.
The result of this collaboration is “Spies: The Rise and Fall of the K.G.B. in America” (Yale; $35)....

This anecdote illustrates the use of Wikipedia as a meta-source. A valuable connection among knowledgeable people was made simply because one of them was able to use Wikipedia as a pointer toward other knowledgeable people and their work.

Perhaps there is nothing overly special about Wikipedia's role in this process. If Vassiliev had not looked up Hiss's article on Wikipedia, presumably he would sooner or later have come across Haynes' and Klehr's work in some other fashion. And there have always been people who say to someone working on project X, "you really should talking to so-and-so, who is working on related project Y." [B] Still, someone's browsing on Wikipedia helped lead to the creation of an important work of history. One point for the value of Wikipedia.

Or here's another example, closer to home. For many years, a group led by William Contento, Phil Stephensen-Payne, and others has been creating the FictionMags Index and related indexes. Their project is to compile a complete index to the contents of all of the major English-language fiction magazines, from the late 19th century to date, in a variety of genres—general fiction, mystery and detective fiction, science fiction and fantasy, Western and frontier fiction, sports, romance, adventure pulps, and even the "saucy" pulp magazines. This is an underpublicized project, and I urge anyone in a position to assist it to do so.

The indexes created by this project are searchable by author, and can be used to create or expand bibliographies or discussions of the authors and the magazines here on Wikipedia. An example, predictable here to those who know me, is Rex Stout (1886-1975), best remembered as the author of the Nero Wolfe corpus. A few years ago, User:WFinch expanded Rex Stout#Bibliography (now at Rex Stout bibliography) by adding about a dozen of Stout's earliest stories, which had not been reprinted since their original magazine publications between 1912 and 1918. These stories were unknown even to Stout's academic biographer and bibliographers, but they turned up in the indexing. Members of "the Wolfe Pack" (Stout's literary society or "fan club") asked me about the stories, and I undertook to track them all down. Helped by the leads originated with Messrs. Contento and Stephensen-Payne and their colleagues, but which would have remained unknown to me if not for WFinch in particular and Wikipedia in general, I succeeded. We were able to interest a publisher in the compilation, and The Last Drive and Other Stories will be coming out next month, making 11 of Stout's early stories readily available to fans, critics, and scholars for the first time. Do we owe the book to Wikipedia? Well, it might have happened sooner or later anyway, but the information my friends and I found on Wikipedia is what actually made it happen, and I think Wikipedia gets some of the credit for that.

Are there other examples of books that were written because of information that was found, or connections made, here on Wikipedia? I'd be interested to know.

  1. ^ Incidentally, there is a great article or even book to be written about Wikipedia as seen through the lens of The New Yorker. There have been quite a number of significant mentions, some flattering to Wikipedians and some much less so. I digress—but in my indecision about whether to keep this aside or delete it, I've just taught myself how to create a textual foonote in wikimarkup, so that is something, at least.
  2. ^ Another great book should be written—lots of articles already have been—on such examples of synergy and serendipity. Chapter 1 would be about Bernard Burke; chapter 2 would be about Arthur Fry. But again I digress. And now I owe Peter David another royalty check.
Discuss this post

A test of Wikipedia quality improvement[edit]

There's been lots of discussion lately about the quality of Wikipedia, including whether errors in the encyclopedia get corrected or not. This concerns not just deliberate hoaxes, but the far more frequent inaccuracies that wind up in articles due to good-faith inadvertence and misunderstandings.

There are anecdotal discussions of whether errors get fixed, and how well and in what time-frame, but it is hard to extrapolate from these, because they are generally based on errors that we know got fixed, rather than a universe of known errors, some of which were fixed and others weren't.

Here's an amateur attempt to update the research in whether Wikipedians are good about tracking down and fixing known errors in our encyclopedia, relating to topics that (unlike, say, Aboriginal Australian deities) are well within the realm of numerous English-speaking Wikipedia editors, in the country where we have the greatest number of editors.

On August 24, 2008, Steve Cuozo of the New York Post wrote an opinion column called "Urban Myths", in which he claimed that Wikipedia's coverage of New York City contained so many errors as to be a "joke." August 2008 is quite a long wikitime ago—the time since 2008 is one-half the time period over which Wikipedia has existed. That's plenty of time for any errors Cuozo spotted to be fixed, whether as the result of people seeing the Post article, or just through the normal process of what we hope is article improvement.

More recently, on November 16, 2013, Cuozo wrote another Post column called "Don't Trust Anything on Wikipedia", which is more of the same, although the focus is a little bit less on outright errors and a little bit more on slightly different problems such as failure to update.

Before writing this post, I deliberately haven't checked what percentage of the problems Cuozo wrote about have or haven't been fixed in the seven years (since 2008) or the year-and-a-half (since 2013) when he spotted them. Over the next few days, I will take a look, and I encourage others to do the same. The results, no matter what they are, may shed some userful light on the issue of quality control and the reliability of crowded-sourced improvement, which of course is one of the key issues facing Wikipedia—and Wikipedia's readers.

Discuss this post

Posner quotes Wikipedia on Weed[edit]

(Not that kind of weed, silly people.) I've mentioned before that Judge Richard Posner of the United States Court of Appeals for the Seventh Circuit has cited Wikipedia in his opinions more often than any other judge. Yesterday, his opinion in Discount Inn, Inc. v. City of Chicago contains a lengthy quotation from our article Weed. See page 7 of the opinion here. Congratulations to the editors on that article.

Discuss this post

Thoughts on the "monkey selfie" debacle[edit]

Adapted from a deletion discussion on Commons that I expected would be closed by now. My thanks to those whose comments there have helped me to improve my reasoning here. For additional factual background, see monkey selfies.

In 2011, the wildlife photographer David Slater traveled to the Indonesian habitat of a group of macaque monkeys. He set up a camera to photograph the monkeys, but then apparently stepped away for a few moments, during which time one of the monkeys pressed the photo button several times. Some of the resulting photographs of the monkey looking directly into the camera, playfully christened the "monkey selfies," have become iconic. Extensive discussion has followed both on- and off-wiki of whether legal and moral rights to the photograph belong to Mr. Slater, or whether he is without rights because he is not the one who pressed the button.

When the question "who owns the monkey selfies?" first arose, I perceived it as just an interesting parlor-game or law-school-exam type question, but I did not pay too much attention, because I doubted that the issue's practical importance were worth the energy being spent debating it. My view shifted more recently as I learned more about the person behind the photograph, who is deeply unhappy with Wikimedia, and who has asserted in the press, though not (yet?) in court, a very real claim of rights.

To date, Wikimedia and Wikipedia have thus far taken the position that the photographs are in the public domain and thus may, and will, be freely used on and off our projects for any purpose. I challenge that outcome and suggest that we change course. If we respect the intellectual property rights of photographers in general, as we must, then I believe our approach to this particular set of photographs cannot be justified.

It is undisputed, because Mr. Slater disclosed it (if he had not no one would know), that the monkey pressed the button on the camera. It appears to be equally undisputed that the monkey did not make any creative or substantive decisions relating to the photographs. In particular, it certainly was not the monkey who (1) acquired a camera, (2) studied photography and become a photographer, (3) decided to use the camera to take photographs of wildlife, (4) decided to bring the camera to a particular location in Indonesia on a particular date, (5) selected a particular image or creature to be photographed, (6) set up, configured, or adjusted any of the settings on the camera, (7) had any knowledge that if he pressed the button, the result would be a photograph, much less (8) had any knowledge that if he pressed the button, the result would be a photograph of himself, or even (9) knew what a camera is, what a button is, or what a photograph is. It was Mr. Slater who did, or had, or knew, each of these things.

Under the circumstances, I have little difficulty in recognizing Mr. Slater as the intellectual parent of these photographs, at least to the extent of not wishing to see them used on-wiki without his consent, and indeed over his express and deeply felt objection. Historically, Wikimedia projects have taken a fairly strict view of what constitutes potentially copyrighted material, which either should not be used at all or may be used only in limited circumstances and with a clearly stated fair-use rationale. Over time, this strictness has resulted in some unnecessary deletions, based on purely notional or theoretical copyright claims that were never realistically going to be pressed by any rightsholder in the real world. In that context of bending over backwards to honor borderline copyright claims, I think we should think long and hard before we insist on continuing to use a particular photograph over the express objection of an individual who, at a least, has a colorable and reasonable argument to be both the practical and the legal author and owner of the material.

Although I happen to be a lawyer, I do not find it helpful to view this primarily as a legal question. It was always perfectly obvious that the PETA lawsuit seeking a declaration that the copyright belonged to the monkey was going nowhere. (A defendant's brief put it beautifully: The words "a monkey, an animal-rights organization and a primatologist walk into federal court to sue for infringement of the monkey’s claimed copyright" belong in a comedy sketch, not a copyright action; and "monkey see, monkey sue" is not good law, even in the Ninth Circuit.) That does not resolve the more serious question of whether the image, under US law or any other law, belongs to Mr. Slater or is in the public domain. I do, however, think that Mr. Slater can make at least a reasonable and defensible claim that the copyright is his, and that it should be his.

A "free content" project should not be in the business of perpetuating bizarre and unnecessary disputes over rights ownership with persons who either are the owners, or in good faith make a defensible claim that they are the owners, of copyrightable subject-matter. Now, to be sure, this principle can only be taken so far: there are times, such as in certain freedom of panorama disputes, or where attempts to recapture large categories of PD images ("reenclose the common"), or other types of overbroad IP claims would significantly decrease freedom of expression, that taking a stand will be justified. But such cases have been rare, and this certainly is not one of them. The factual scenario underlying this image is a bizarre one, unlikely to recur, certainly unlikely to recur with frequency. Any loss of intellectual freedom that would be associated with our choosing to delete this image would be very slight.

It certainly would reflect questionable priorities for Wikimedia or Wikimedians to expend either scarce legal resources or our goodwill on litigating such an issue if, as is reportedly at least possible, Mr. Slater were to assert a claim. Sometimes, of course, we must spend money and resources and goodwill to defend an important principle, and we have collective resources devoted to precisely that endeavor, to be put to use if and when we need them. But defending the dubious principle that "if a photographer sets up a shot but by happenstance an animal inadvertently presses the button, we will ignore the intellectual property claims of the photographer" is not such a case.

I have read through much of what has been written about this subject over the past couple of years and I am disappointed by some editors' insensitivity to what is, at least, a good-faith and defensible rights claim. I am also troubled by some comments made over time that have the effect of, while I would not say intentionally bullying Mr. Slater, at least have the effect of rubbing his nose in the fact of his powerlessness to control the reuse of an image to which he reasonably believes he has or should have rights.

One of legal memoranda that Mr. Slater's lawyer submitted in the federal case is worth quoting as it summarizes his position concisely and well. I recognize that I am quoting an advocate's words rather than from a neutral source, but I find these points persuasive:

Snapping the shutter is something any human or monkey can do; setting up what became a world-famous, award-winning photograph is what professional nature photographer David Slater did.
As a non-human animal, [the monkey] is inherently incapable of setting up the photograph at-issue in this case. Slater set up what became the Monkey Selfie in the course of several grueling days in an Indonesian jungle. Developing a keen understanding of their subjects is a critical skill for any professional photographer, and it was vital here for Slater as he slowly built a trustful, friendly relationship with a group of crested macaque monkeys. Only a talented human photographer could have made the artistic choices involving camera lens width, positions and settings (e.g. predictive autofocus, motorwind, and flashgun.... [Slater] made the critical artistic decisions that resulted in a photographic work adored by millions worldwide.

I understand that there are those who think that this issue is a tempest in a teapot; as I said, I was until recently in that position myself. And frankly, I can imagine that someone else who found himself or herself in Mr. Slater's position might have chosen to go along and not make a fuss and be "a good sport" about the whole thing. He could have figured that the unique circumstances were not going to repeat themselves, and that the attendant publicity might be positive, for his reputation and business and for the causes of nature photography and protection of the wildlife that he adores.

But it is not for me, as a stranger to Mr. Slater and to his life's work, to say that he made the wrong choice by reacting instead as he has. I realize that this may seem a harsh judgment, but in my view, we dishonor the cause of free knowledge when, as the largest free-content set of websites in the world, we treat him as we have.

Discuss this post

Notes on notability[edit]

After I spent some wiki-time last weekend working on a controversial unblock request, I decided to turn to something more routine, and I took a look at the day's AfDs and DRVs. Enforcing our notability guidelines through these processes is a key part of what happens at Wikipedia every day. That day, I commented on three different deletion discussions in which the conventional, by-the-book answer might have been "delete," but in which my own encyclopedic instincts told me that "keep" would be a better result.

Does this miniature sampling suggest that I'm just an out-of-the-mainstream inclusionist? Were some of these unusual cases that would have justified pushing the boundaries of notability or even an "IAR" result? Or are there flaws in how we measure notability in some situations that would justify modifying a guideline? Let's see.


Wikipedia:Deletion review/Log/2020 August 24#Paul C. Gartzke seeks to overturn six AfD deletions of short articles about judges, past or present, of the Wisconsin Court of Appeals. Like most U.S. states, Wisconsin now has a three-tiered system of appellate courts with parallels to the U.S. federal system. In these states, the intermediate appellate court is the court of last review in the vast majority of cases; only a small percentage of cases continue on appeal to the state's highest court.

WikiProject United States courts and judges, an active wikiproject, covers this type of articles. It has developed a set of (obviously non-binding) specialized notability guidelines, under which all life-tenured federal judges (including at the district or trial court level) are deemed notable, and all judges of the states' highest courts are deemed notable. Intermediate state appellate judges, on the other hand, "are not inherently notable, but holding such a position is strong evidence of notability that can be established by other indicia of notability. In particular, state courts of appeals judges who serve for a comparatively long time, who preside over important cases, or whose opinions are often cited by higher courts in the state, by federal courts, or by state courts in other states, are highly likely to be notable."

Under this standard, these judges' notability is to be decided on a case-by-case basis. That's not an implausible rule, as some of these judges have more significance in legal history than others. But the rule has the disadvantage of being subjective, meaning that each of these articles is subject to a separate deletion debate and notability discussion, with the associated commitment of Wikipedia's most precious resource—our editors' time. It also has the disadvantage that if such an article is created in less than full bloom, and it doesn't happen to catch the attention of an interested and knowledgeable editor during the week it's up for discussion, then the article will indeed be deleted, which is what happened here.

In principle, anyone can come along later with greater evidence of notability for any one of these judges and try to re-create a better article. But there can be obstacles to such re-creations, including:

  • Incorrect G4 speedy nominations (CSD G4 is often mistakenly invoked in good faith by non-admins, who can't view the text of the deleted article to compare it with the improved new one) and sometimes re-deletions;
  • The reasonable fear that even if the new article isn't G4'd, the fact that it's been deleted once suggests that it's at risk of being deleted again, so why risk doing the work for nothing; and
  • The greater "barrier to entry" when instead of adding text to an existing, perhaps well-formatted article, one starts with the blank screen.

The editor who brought these six articles to DRV also made these interesting points:

It seems the same rationale is used for all [six of these Wisconsin judge articles] -- that because they're elected in four regional districts they are not "statewide" and thus not significant judges. The regional divisions are for administrative purposes, but the rulings of the judges have statewide effect and precedent (only a fraction are ever appealed to the state supreme court). Maybe this is an issue with the way the current guidance prioritizes state legislators over state judges, but any individual judge of this court in Wisconsin has far more significance to the legal and political landscape of the state than any single member of our 99-member state assembly. Several other judges of this court (who were left undeleted) are former state legislators, whose legislative career was nowhere near as consequential as their time on the court, and that legislative service will factor as a mere footnote in their obituaries -- yet it is that relatively inconsequential legislative service that preserves their notability for the purpose of this site.

Would we be better off simply having a rule that these judges are notable, as opposed to reviewing them case-by-case? Would we be better off saying that proof of notability is required, but being less quick to delete such articles because it's likely that such proof will eventually be obtained? Or did we do right to delete them, as long as we're willing to reinstate them if proof of each individual's notability is added? This isn't a situation in which notability is inherently subjective—we could have a rule that these judges are presumed notable, just as their federal counterparts are—I see pros and cons to either approach.

At this writing it looks like the DRVs will result in a relisting, given interested editors some more time to show which of the judges is notable—though that will still require a case-by-case analysis of each.


Wikipedia:Articles for deletion/George W. Blunt, No. 11 is considering whether to keep or delete George W. Blunt, No. 11, an article about a pilot boat that operated in New York Harbor during the nineteenth century. It's an interesting article that discussed the history of the vessel; reading the article gives an interesting perspective on what such a vessel looked like and what role it performed. I'd say we should keep that article, without hesitation. And yet it must be admitted that this wasn't a particularly important nineteenth-century vessel. As I put it in my !vote, responding to a "delete" commenter:

[T]he notability is a more serious concern. You are obviously right that if we made a list of the most important vessels of the nineteenth century, this boat would not be on it. Put differently, we wouldn't aspire to someday have an article on every nineteenth-century vessel with the same level of importance. Nonetheless, I think there are times when it's helpful to have an article on a typical example of a type of object or a phenomenon, where the information needed to create an article happens to survive. Although this keep rationale is not well-captured in the guidelines, it has come up before and at times has been accepted. The bottom line is that the letter of the rule-book aside, I think our encyclopedia is better for containing this article than it would be without it.

Is that a valid notability rationale? I don't see it in the letter of the guidelines, though I might have missed it someplace. Being a typical example of something, far from conferring notability, might be considered the very opposite of notability. And yet, historians realized long ago that it can be as important to understand the life of an era's "ordinary" person as the life of the king. I recall a set of articles about British women that a distinguished editor wrote a few years ago; there was no claim that any one of the subjects, as an individual, was among the most important people of the time—but enough information about them had survived to write credible articles, and certainly one knew more about the society of their region and era after one had read the articles than before.

This struck me as a somewhat parallel situation and a clear keep. But I'm not sure how one would enshrine this criterion in the rule-set. Should we? Should we just treat these as IAR cases? Or would deletion be the right result after all?

At this writing, a couple of keep !votes have come in saying that the article squeaks past the regular notability test (I'm not so sure)—and there is a deus ex machina, because it turns out the vessel was nationalized during the Civil War, and the additional information in USS G. W. Blunt (1861) will probably wind up with a merge-and-keep result. But the methodological questions remain.


My third example and perhaps the most interesting of the three.

Just Peck was an American comedy film released in 2009. You have never heard of it. Apparently it was a disaster, and went straight to video without a theatrical release. The film is available for purchase, and is listed in film databases, but there seem to be very few published reviews or other sources about it. Understandably, the article has been brought up at Wikipedia:Articles for deletion/Just Peck. Within the four corners of our policies on notability and sourcing, unless more sources are found, deletion would seem inevitable.

And yet. This film was not produced by a bunch of unknowns in their garage. The producer is a notable person, whose Wikipedia article contains a filmography that includes this film. At least six of the actors and actresses are notable people, whose Wikipedia articles contain filmographies including this film; one of them is a featured article. The music is by two notable composers and the film is mentioned in their articles also. Deleting Just Peck would either create one lonely redlink in all these people's filmographies, or one lonely no-link-at-all. (Or worse still, would the letter of policy say to remove any mention of the film anywhere?)

I'm not sure that's a good idea. As I explained in my so-far-sole keep !vote:

Keep. The article indicates that the film involved at least nine actors, and several other people, with articles. It would make little sense to turn all the links in their filmographies to redlinks, thereby removing relevant information about their careers and creating holes in the referencing.

I opined that given these considerations, it might be helpful and would certainly seem harmless to keep the article. A well-informed delete !voter cogently explained why my !vote is contrary to policy:

This article has no citations, meaning we cannot verify the information (WP:V) and it contains entirely original research (WP:NOR). Furthermore, we are having difficulty finding WP:RS that can be used for citations. As an encyclopedia, it's important that we hold ourselves to a consistent standard which we can showcase when someone challenges our work. The harm of keeping this article is that it breaks the core policies we gave ourselves for building the encyclopedia. The value of deleting it is stating, "We cannot verify the information in the article, so we at Wikipedia would rather have no information than a chance of publishing incorrect information." If a standard needs to be changed, perhaps an WP:RfC might be a place to continue that conversation.

(It may be a slight exaggeration to say there are no reliable sources that mention the film, but there certainly aren't the number one would ordinarily expect.) I'm familiar enough with our sourcing standards and am certainly not looking to dilute them—yet in this instance, I remain convinced that deleting this article would remove worthwhile content. It's not due to any partiality toward the film on my part—I'm not especially a film buff, and I'd never heard of it either. Absent some more sources turning up, the deleters have the better of the policy argument. But my encyclopedic instincts have some value too, and they tell me that deleting would be a mistake.

I don't particularly care about this one article, but I'm convinced there's a broader principle involved here, just as there should be in the two other discussions I've mentioned. However, I haven't figure out just exactly what those principles are yet. Maybe they're not really there to be figured out, and I'm just griping. Or maybe someone who reads this can help. If there are still refinements to be made in our inclusion and deletion criteria after close to 20 years, then that, at least, would be notable.

Discuss this post

For our readers on mobile devices: a request for input, a need for action?[edit]

Three caveats as I start this section. First, it probably belongs on one of the village pumps, where I may place it after I get some feedback here. Second, I am very much a "non-technically oriented user," and I am sure I'll be mistaken in how I describe some aspects of things, and that I'm ignorant of some relevant conversations I ought to be aware of. Third, I know that I'm far from the first person to raise the issues I'm going to raise, and for all I know they are being addressed at this very moment, and if so I'll be very glad to hear it.

That being said ...

The other night I was at a group dinner, and someone asked a question none of us could answer, and one of my friends pulled out her phone to look the answer up. The first Google hit was the relevant Wikipedia article, which contained the answer to our question, and a lot of interesting, related information in addition.

Except ...

My friend certainly knows what Wikipedia is, but she is not a regular Wikipedia user. She read the introductory section of the article, but did not know she could click on the subsections below for more information. When I showed her that, she was quite surprised and said she never would have realized it. And even then, it was still a pain for her to click on each individual subsection in order to read the entire article.

When I then showed her that by pressing one link at the bottom, she could convert the page from mobile to desktop view and be able to skim through and then read the entire article at once, she was stunned. She had never known that before. I then asked around the table—a dozen intelligent, well-read people, none of whom are Wikipedians but all of whom use the site at least occasionally in their daily lives, and none of them knew it either, and all of them were surprised.

We all know that an increasing percentage of our readership is on phones or other mobile devices. If what I've described is a typical experience, then a lot of our content is not reaching a lot of our readers. And this is to say nothing about the issues with editing on the mobile site, and about user notifications to IPs on the mobile site, both of which remain whole 'nother kettles of worms.

I'm sincerely not trying to pick an argument with the developers or with anyone else, and I know there have been discussions about all of this over the years, and community requests to the WMF, and yes, some improvements—but from my admittedly naive perspective, doing something about this situation ought to be a serious priority.

My questions for my readers here, please:

  1. Am I right that this is a problem?
  2. What is being done, or what should be done, about addressing it?
  3. Would it be helpful to create a group of community members—selected through some reasonable community-wide process—to communicate directly with WMF personnel on this type of issues? A few years ago, the ArbCom gained the ability to work with the WMF on issues of mutual importance, such as dealing with certain harassment and abuse situations. I'm not going to claim that opening that communication channel solved all problems, but it was certainly helpful. Ensuring that our article content, which our contributors work so hard to create, is readily accessible to all of their readers regardless of their type of device, would seem also to be of high importance.

I'd welcome any input that my TPWs might have. Thank you.

Discuss this post