User:Hillman/Wikipedia quality control

I am currently trying to collect my muddled thoughts as I ponder the reasons for my forthcoming disengagement from WP. CH 11:23, 18 March 2006 (UTC)

I am keeping this page mostly for my own use as I try to organize my thoughts on the problem on Wikicruft and how the Wikipedia community can or should respond to it, but if anyone wants to comment (with civility), please do so in the talk page.

Don't shoot the messenger!

TIA ---CH 05:13, 10 December 2005 (UTC)

See also User:Hillman/Media commentary on Wikipedia. ---CH 04:01, 7 July 2006 (UTC)

This is a draft. Things will get worse before they get better. Expect seemingly random insertions, redundancies, and general chaos until I begin to sort out the mess.

Balance and edit creep

Robert McHenry has pointed out that the wiki model appears incompatible with the goal of attaining any global balance of coverage, and I think he has a point. That is, because in order to further the goal of building a better Brittanica under the wiki model, to create good articles on topic X, we need to attract a fair-minded, wise, judicious, self-less, industrious, and dedicated pool of editors who happen to be experts on topic X. Amazingly, this unlikely combination of virtues has proven not impossible to attain in some areas, such as mathematical topics. However, I think we must concede that it is highly unlikely that any wiki could ever attract such a pool of editors competent to write good encyclopedia articles on every topic. The resulting global imbalance of coverage does not contradict seeking a local balance, however. Indeed, a resource such as this metalist suggests that the wiki model need not result in utter chaos within a well-defined subject area such as mathematics. Indeed, I suspect that if Brittanica itself is not already using an internal wiki (accessible only to editors and authors) to track and guide the development of the next edition, it soon will be doing so! Several astute observers of the software scene, recognizing that wikis work best when restricted to a pool of editors who share some common core principles, have predicted that such internal wikis is the path of the future, rather than Wikipedias somewhat notorious slogan, "which can be edited by anyone".

Similar considerations apply within individual articles, where I feel that the classical wiki model is incompatible with maintaining an overall structure and vision. Here, I believe that some kind of modified wiki structure, which establishes gradations of privileges, is required to ensure that articles are improved in coherent directions, and to encourage novice authors to improve their sensitivity to the sometimes challenging high-level intellectual task of seeing how to fit material they wish to add into the existing structure and vision of a given article. In an unstructured wiki model, all too often, novice writers prove unable to maintain consistent paragraph structure, verb tense, terminology, and notation. Or even worse, they often do not appear to even be aware of such issues! This is one aspect of edit creep, a lamentable phenomenon which tends to disprove the naive expectation that Wikipedia articles tend to naturally improve monotonically, at least "on average" (but on what time scale?, we should immediately retort). Or in an even more asinine version: Wikipedia is globally attracted to a state of perfection without requiring any leadership or concious organization.

Quality control problems in the Wikipedia

Like many Wikipedians, during the course of my year as an active contributor, I grew increasingly concerned with the problem of quality control at Wikipedia and what to do about it. Or, to put it more bluntly, with the misuse of Wikipedia in order to mislead readers by manipulating information in order to pursue a hidden agenda, sometimes motivated by the possiblity of financial or other personal gain. I was aware before I became a contributor of the potential for this kind of misuse – as well as the potential for good, by disseminating accurate and timely information about math/sci topics which might be hard for general readers to find elsewhere – and while I was here I tried to foster the latter while combating the former. Ultimately, I sadly concluded that the bad guys (the ideologues, hoaxers, linkspammers, crank physikers, undercover political "dirty tricks" operatives, and guerrilla marketeers, among others) are winning this struggle for control of the Wikipedia.

I tried several strategies to combat the spread of misinformation and disinformation in the Wikipedia, some of which were almost comically disastrous, but I do believe that I have proved willing to learn from my mistakes and to adapt to new circumstances. I think my later efforts, based on insights won at the cost of harsh experience, were wiser than my earliest attempts, but ultimately I was forced to sadly conclude that:

1. cruft control procedures at Wikipedia have not scaled well, and have become far too inefficient to prevent the widespread manipulation in this "encyclopedia" without imposing an intolerable burden in terms of time and effort upon editors of good faith,
2. the political process at Wikipedia is simply too cumbersome to admit any possibility of formulating and implementing more efficient procedures.

I say more about this elsewhere in my user pages. Here I will say only that I believe that while allowing anonymous edits from unregistered users may have played an important part in helping Wikipedia grow to its current size, I believe that continuing to permit anons and socks will prove a key strategic error which will prove fatal to the experiment if it is not addressed. I believe that the problem posed by anons and socks is only one of the potentially fatal problems which the Wikipedia community needs to promptly address in an effective way, but it is probably the easiest to fix. However, these are complicated issues, so I hope to be able to direct comments elsewhere in my user pages in the near future.

What is wikicruft?

I'll define wikicruft, via a gloriously mixed metaphor, as

wikicruft (noun): the accumulation of undesirable content which slips under the radar screen of the alleged army of watchful eyes, or otherwise (smack! twhack! whack-a-mole!) evades eradication from the WP.

Wikicruft can take various forms, including

• bad newly created articles (vanity, hoaxes, crackpottery, spam, nonsense) which do not quickly come up for AfD (or even worse, survive AfD due to nonscientists, including admins, being conned by seemingly plausible technical content),
• the same, for bad categories,
• the same, for bad internal links to or from existing articles,
• the same, for bad edits of previously existing articles.

For our purposes, content is considered to be bad if it includes elements such as these:

• blatant factual inaccuracies,
• intentional or unintentional misstatements,
• serious misrepresentation of past or present scientific knowledge and practice,
• unverifiable claims, or claims resting upon highly suspect "sources",
• highly slanted descriptions of scientific controversies,
• highly idiosyncratic personal opinions passed off as notions having more considerable standing in the scholarly research community.

OTH, good content is characterized by elements such as these:

• clear, concise and factually accurate description of current scientific thought,
• careful attention to use any technical jargon correctly (and to define technical terms in words comprehensible by laity!),
• citations to widely used textbooks, monographs widely cited in the relevant literature, published papers in respected research journals (especially survey papers, since this is an encyclopedia, not the arXiv), etc.,
• fair and accurate description of scholarly controversies (which are of course an essential and even healthy part of the scientific enterprise),
• overall, careful distinction between generally accepted knowledge and informed speculation going beyond the bounds of the body of generally accepted knowledge.

This page also attempts to discuss more insiduous forms of the degradation of content in WP and related issues. I focus on degradation of math/physics-related articles, but some of this discussion is no doubt relevant to WP articles and categories dealing with other parts of science or scholarly research.

Scholarly values versus Populist values

For purposes of this discussion, I will assume that in order to make good judgements in content disputes regarding encyclopedia articles on scientific subjects, one must neccessarily adopt scholarly values. Unfortunately, the populist values of many prominent Wikipedians are generally antithetical to scholarly values, which is a huge part of the problem in attempting to deal with bad content in the scientific categories. Needless to say, I didn't appreciate this crucial point when I first started participating in WP, and I doubt that the many other editors who have built some excellent math/physics-related content here did either!

Let me elaborate slightly. Populism has always formed a powerful undercurrent in American politics, and currently is enjoying a powerful revival in the U.S. To see why this affects the pursuit of science, at least in the U.S., it is crucial to recognize that populism tends to be hostile toward elitism of all kinds, because (to oversimplify) populists believe that the opinions of all individuals are equally valid. Scholarly values are on the other hand unintentionally but ineluctably elitist, because scholars address complex issues which often require much background knowledge to appreciate. It follows that in scholarly discussion, the contributions of well-informed experts who know the relevant scholarly literature well tend to be worth far more than contributions from people who know very little about the subject under discussion, however intelligent or otherwise well-informed they may be. This is particularly true for math/physics, where considerable mathematical ability may be required to closely follow a scientific discussion, regardless of the amount of effort made the participants to include laypersons in the discusssion. My point is that even those mathematicians and scientists who happen to prefer, in principle, populist ideals to elitist ideals, really cannot avoid behaving as elitists when the topic of discussion turns to their field of study, at least not if they want to function effectively as scholars.

Readers who haven't thought of populism in quite this way before might want to take a moment here to reread some of the many pages which describe official Wikipedia guidelines and policies. It should not be hard to spot populist ideals once you start looking for them.

(A quick aside concerning the Americanism implicit in the previous two paragraphs: I am tacitly assuming, on the basis of various public statements by Jimmy Wales, that a nightmare scenario in which a lawsuit effectively shuts down the WP overnight, whether due to some kind of injunction or to hefty fines and legal fees breaking Wikimedia's financial back, has driven much of the response to recent media scandals. My point is of course that Wikimedia is incorporated in the U.S., and some angry cranks and other malcontents have publically taken note of this fact.)

Jimmy Wales recently wrote: I am a firm believer in the validity of allowing anons to edit. Most anon edits are good, and done "on impulse". This seems innocuous enough, until you reflect that the result of "impulsive" edits of carefully constructed articles dealing with complex subjects is unlikely to lend support to the claim that wikis somehow encourage the monotonic perfection of encyclopedia articles. His claim also does not appear to correspond to the facts (see Zocky's study in the next section).

I suspect that much of the anguish (or anger) expressed by many Wikipedians regarding media stories involving the WP involves the inevitable collision of utopian ideals with some harsh realities of human nature. Our problems as scholarly wikipedians are compounded by the unintentional clash of the dominant political philosophies of the leaders of our community with some nonneogotiable requirements for the conduct of scholarly discourse (for example, in resolving content disputes concerning science-related articles). From this statement, I also suspect that the accidental conflict between populist values and scholarly values ultimately lies behind the departure of Larry Sanger from the WP (which happened well before I arrived).

Since I am not intrinsically interested in politics as such, I find it irksome that political ideologies should affect our work here, but it would be a grave mistake to ignore the role played by political ideals at WP simply because one feels that scholars should be left alone to sort out intellectual issues in the way which hundreds of years of experience has shown works best: scholarly discussion, peer review, consideration of evidence, and so forth.

Alas, it doesn't end here. Libertarianism is a younger but related political ideology which informs many RfC, AfD, and talk page discussions, and which is admired by many prominent Wikipedians. Unfortunately, Libertarian ideals also tend to be hostile toward deletionist practices, because (again oversimplifying) Libertarians tend to dislike any restraint upon individual freedom, and therefore are likely to be reluctant to adopt such measures as banning problem users, or in any way restricting what they regard (not always with due attention to legal nicities) as the right of free speech.

Some rough and ready statistics

User:Zocky tracked the fate of a sample of articles created on Fri 2 Dec 2005, from which he (if neccessary I will exercise my free will and redefine this pronoun to gender-neutral) made some rough estimates of the rate at which anons are creating bad articles (e.g. ones which are later deleted following an AfD discussion). He summarized his results in a self-explanatory table which he posted at the Village pump policy discussion forum on 8 Dec 2005. His results turned out to be quite similar to my own informal statistical study in Sept 2005.

Zocky's sample suggests these estimates:

• overall about half of new articles do not belong in WP,
• of these bad articles, about two thirds are created by anons and one third by registered users,
• for the good articles, these percentages are reversed,
• two thirds of all edits by anons are bad,
• about 12% of the bad articles are not nominated for AfD discussion within 3 days,
• about 6% of all new articles in WP are bad but are not nominated for AfD within 3 days,
• only about 1% of the bad articles appear to be honest mistakes by clueless newbies trying to create (for example) a personal user page,
• currently new articles are created at a rate of 1.2 million pages/yr,
• about one fifth of these are good articles created by anons,
• about three fourths of all bad articles are created by anons.

The sample size in Zocky's study might seem small, but statistical theory suggests that it might nonetheless be large enough to give some information about the magnitude of the problem.

The lesson seems clear: by forbidding new article creation by anons, WP can expect to

• greatly reduce the workload on the new page patrol
• reduce the rate at which bad articles duck under AfD,
• slow the rate of which new good articles are created (not neccessarily a bad thing, given recent strain server stress).

Here is Zocky's table:

New pages, Friday, December 2, 2005	pages I checked			calculated values for all pages created on Friday
created by:	IPs	logged in	total	IPs	logged in	total
new pages, still existing	95	145	240	773	1182	1955
good pages, properly done	63	125	188	513	1018	1531
good pages, didn't get attention from experienced users	16	10	26	130	82	212
all good new pages	79	135	214	643 (37%)	1100 (63%)	1743
on AFD	7	3	10	57	24	81
bad pages, should be deleted	8	3	11	65	25	90
mistakes (wrong language, namespace, etc.), should be deleted or moved	1	4	5	8	33	41
all bad pages still existing	16	10	26	130	82	212
deleted pages	264	89	353	1149	388	1537
all bad new pages	280	99	379	1280 (73%)	469 (27%)	1749
all new pages	359	234	593	1935 (55%)	1557 (45%)	3492

I should point out that one issue which Zocky didn't address is the commonly seen claim that most WP users start by editing anonymously, and eventually decide to create an account. I myself created an account right off the bat, so I would like to see this bit of WP mythology (?) put to the test. In any case, I don't think it neccessarily follows that forbidding new page creation by anons will slow the rate at which new editors are recruited into the WP community, but certainly this is a valid question to study by means of a proper survey followed by valid statistical analysis.

Studies of edits to particular articles

Recently I have carried out some surveys of anon edits in particular articles which are frequently the target of bad edits.

Zero point energy

Some edit rate statistics

This article was created 09/12/2002 14:11 by WkPark (talk · contribs) (single edit account!).

• To date it has received 385 edits, with a lifetime average edit rate of once every 3.9 days.
• It has been edited 284 times in the last year (to date), for a yearly averge of once every 30.8 hours, a three-fold increase over the lifetime edit rate.
• The most frequent contributors have been:
  1. Efilnickufesin (talk · contribs); I am not familiar with this editor but he has been extensively chided at User talk:Efilnickufesin for losing his temper in a related article
  2. Linas (talk · contribs), a WikiProject Physics member who IMO does not violate WP:NPOV
  3. Petri Krohn (talk · contribs), who IMO does not violate WP:NPOV
  4. Enormousdude (talk · contribs)
  5. Bbq332 (talk · contribs)
  6. 69.107.99.133 (talk · contribs), aka the dsl.pltn13.pacbell.net anon in the Bay area; this domain has been used extensively by both Jack Sarfatti and Bernard Haisch as anons, but these edits rather obviously are due to the latter; while this statement is disputed by Haisch, I regard his edits as tending toward fringe science POV-pushing,
  7. Yevgeny Kats (talk · contribs)
  8. ObsidianOrder (talk · contribs), who in my view tends toward pushing fringe POV (haven't examined his edits to this article, though)
  9. Lumidek (talk · contribs), IRL Luboš Motl, who certainly knows a great deal about string theory but who is (as I expect he would agree!) sometimes regarded as tending to push the string theory POV elsewhere on the web in a somewhat partisan fashion (haven't examined his WP edits though),
  10. YurikBot (talk · contribs), its a bot, how could it violate WP:NPOV? :-/
  11. Hillman (talk · contribs), that's me, with 4 edits; even my worst critics will probably view me as a defender of mainstream science against innovative heroes (their words) or assorted ignoramuses (my words)

Study of edits by anons over approximately 1.5 month period

In reverse chrononical order, covering the ten IP anons who made contributions between Apr 14, 2006 and Jun 3, 2006:

1. the Dial1.Chicago1.Level3.net anon (possibly somewhere near Cornell, MI) has used
   • 4.254.113.242 (talk · contribs) legit addition
2. the hstntx.swbell.net anon (registered to Southwestern Bell; geolocated near Houston, TX has used
   • 70.253.99.146 (talk · contribs) legit correction
   • 65.68.73.156 (talk · contribs) deleted merger suggestion
3. the dhcp.sprint-hsd.net anon (possibly somewhere near Kingston, NC) has used
   • 71.2.147.254 (talk · contribs) minor legit diction improvement
4. the Montreal area Bell Canada anon (near Dorval, Quebec) has used
   • 207.164.171.98 (talk · contribs) several rude vandalisms

• the Geneva College anon (near Beaver Falls, PA, which is indeed the location of Geneva College) has used

1. • 63.172.105.84 (talk · contribs) well-intentioned (?) but questionable edits
2. the hsd1.mi.comcast.net anon (near Grand Rapids, MI) has used
   • 24.11.20.99 (talk · contribs) added a cultural reference
3. the dynamic.dsl.as9105.com anon (registered to Tiscali UK Ltd; geolocated near Manchester, England) has used
   • 88.104.215.225 (talk · contribs) tried to introduce minor but easily overlooked vandalism
4. the exetel.com.au anon (registered to Exetel Pty Ltd.; geolocated near Sydney) has used
   • 220.233.16.174 (talk · contribs) added a cultural reference
5. the demon.co.uk anon (registered to Demon Internet's ADSL Service; geolocated near Eastbourne, England)
   • 80.177.20.202 (talk · contribs) added apparently legit paragraph
6. the btcentralplus.com anon (British Telecom; geolocated near Bournemouth, England) has used
   • 81.153.28.94 (talk · contribs) added inappropriate link to the website of one Robin Jones (curiously enough, there is a Robin Jones' Latin Jazz Sextet based in Bournemouth)

Here, geolocations/domains are used to disambiguate the various anon editors. By my count, this gives

• five who made legitimate edits,
• two who made questionable but probably well-intentioned edits,
• three who made bad edits.

Much better than any results I've seen so far in similar survey!

Black hole

Some edit rate statistics

This article was created on 11/20/2001 05:21 by Chenyu (talk · contribs) (inactive since Jan 2002).

1. Since creation this article has been edited 2406 times, and since creation has been edited on average once every 17.5 hours.
2. In the past year it has been edited on average once every 6.4 hours, a threefold increase over lifetime average.
3. Of the top five editors, Plautus satire (talk · contribs · block log) has been permabanned.
4. Of the top five editors, the most frequent editor, User:Keenan Pepper, edited once every 7.4 days between January and August 2006 inclusive.
5. The peak monthly edit rate so far was May 2006, when it was edited approximately once every 3.5 hours, almost a two fold increase over the baseline yearly edit rate to date (for 2006).

Study of time spent in vandalized state over 15 day period

(Note: copy of this study)

Just took a quick look at recent bad edits by anons.

1. 88.118.140.103 (talk · contribs · WHOIS) at 03:31, 27 August 2006 used the article as a sandbox. Self reverts in 2 min.
2. Several edits by Danras (talk · contribs · block log), already noted. Reverted in 11 min, 5 min, 75 min (too long!).
3. 12.148.189.25 (talk · contribs · WHOIS) at [1] by 12.148.189.25 (talk · contribs · WHOIS) is bad (this IP may have been recently reassigned from the toolmaker Black and Decker in Lake Forest, CA). Fortunately, Brygenwulf's subsequent revert of Danas also reverted this anon.
4. 124.47.166.87 (talk · contribs · WHOIS) at 06:19, 24 August 2006, henceforth the Bombay vandal, blanked most of the page. Reverted almost immediately.
5. 169.199.155.65 (talk · contribs · WHOIS) at 14:21, 18 August 2006 blanked several articles (reverted by a registered user). This is the Contra Costa County Office of Education vandal who has been observed on several previous occasions here. Made several successive vandalisms and was promptly reverted by other users.
6. 67.68.205.104 (talk · contribs · WHOIS) at 21:04, 16 August 2006 made a possibly intentional misstatement, promptly reverted. Several problem editors of physics-related articles happen to reside in the Montreal region. Reverted in 17 min.
7. 195.89.27.198 (talk · contribs · WHOIS) at 09:18, 16 August 2006 blanks most of the article and replaces it with a bungled and misleading citation of Ted Bunn's cosmology FAQ. Reverted in 1 min.
8. 84.77.143.12 (talk · contribs · WHOIS) at 03:11, 16 August 2006 replaces introduction with a version which exhibits substantially poorer English diction. Not reverted by time of next bad edit, by 195.89.27.198 (talk · contribs · WHOIS).
9. 24.77.216.252 (talk · contribs · WHOIS) at 00:57, 13 August 2006 incorrectly "corrects" [sic] article to state that no radiation except gravitational radiation can escape from from a black hole. Reverted almost immediately, repeats, then not reverted for 280 minutes (much too long!!!!!) Many bad edits to physics-related articles from this ISP (Shaw cable) have been previously noted.
10. 60.227.206.179 (talk · contribs · WHOIS) at 06:42, 12 August 2006 Silly blanking vandalism. Reverted immediately.

OK, so over the past 21000 minutes or so, this article was in an obviously vandalized state for something over 375 minutes, which would be suggest that this article typically exists in an anon-vandalized state just under 2% of the time. This is consistent with previous estimate from many months ago. Over this time period, only a handful of non-bad edits by anons were observed. This seems consistent with earlier estimates that less than one in five anon edits is acceptable. I didn't systematically check whether registered users made a comparable number of bad edits in this time period, but spot checks suggested that they did not. If we subtract out Danras, this article was anonvandalized for something like 285 minutes over about 21000 minutes, for a rate of about 1.4%. This also seems consistent with earlier estimates for anon vandalization rates, suggesting that contrary to the hopes of some, semiprotection does not seem to be ameliorating the anon vandalization problem; anons are just damaging easier targets, such as this article.

I think that finding this article in a vandalized state two percent of the time is too high a frequency to be acceptable. Even those who don't agree probably will agree that it is utterly unacceptable that this article should have existed in a vandalized state for 75 minutes on 25 August or 280 minutes on 13 August. Clearly, in such a popular article, that does not serve our readers well.

Once again, the inexorable conclusion is that Wikipedia must follow the model of all other large public forums (of which I am aware), by prohibiting all edits by unregistered users (by technical means, not social means). The figures above are consistent with my estimates from last fall suggesting that we could more than halve our problem with bad edits with no effort at all, simply by eliminating anon edits. ---CH 07:20, 29 August 2006 (UTC)

Study of damage done by one (word needed?) in one bad edit

The needed word would express the sense of: "this person has IMO absolutely no idea what he is talking about". Danras (talk · contribs) continues to actively edit this and other physics-related articles, with similar results.

(Note: copy of this study)

Alterations like this constitute utter hogwash, and any good student in an undergraduate gtr should be able to spot some specific howlers:

1. "For very large black holes the gravitational acceleration at the event horizon can be less than that at the surface of the Earth." Here, Danras has clearly confused the magnitude of typical components of the tidal tensor (measured by some free-falling or even accelerating observer who falls through the horizon), that is, the curvature of spacetime itself, with the magnitude of the acceleration vector, that is, the path curvature of a world line in said spacetime. It is true that the tidal forces on a human falling through a supermassive black hole's horizon can be mild, but this does not imply that photons can escape from inside the horizon (they can't, at least not according to gtr).
2. "Black holes are calculated from a hypothetical point that is distant from the black hole." This doesn't even make sense as stated. Danras probably misunderstood something like this: the easiest way to verify the physical interpretation of the parameter m which appears in the Schwarzschild vacuum solution as the mass of the central object is to compare the motion of distant orbiting particles with Kepler's laws.
3. "The calculated event horizon is not a fixed property of the black hole, but is dependant on the gravitational potential of the point from which it is calculated." Actually, "Newtonian gravitational potential" simply doesn't make sense in strong field gtr (where Newton's theory of gravitation breaks down entirely, which is why we need gtr).
4. "A black hole calculated relative to an observer at a high gravitational potential may not be a black hole to an observer at a low gravitational potential." Same comment; Danras plainly doesn't have a clue.
5. "By artificially positioning observers close to an alleged black hole, one can create successive frames of reference in which the observed black hole radius decreases to zero." This appears to rest upon some attempt to use special relativity over a large region in the strong field gtr, a procedure which is always doomed to fail badly (this is more or less what we mean by saying spacetime is curved; you know spacetime is curved because laws which would work if it were not do not work).
6. "Escape from a singularity would also only require a finite amount of energy, for while the gravitational acceleration there would be infinite, this fact would be counterbalanced by the fact that such a singularity would occupy zero volume." Actually, this is so bizarre I am beginning to think Danras is just trolling. He wouldn't be the first.

Well, you get the idea: absolutely every sentence Danras added is a howler (which is in itself evidence he is just trolling). Apparently this user was also responsible, at least in part, for the recently deleted "symmetrical relativity" (no such notion is recognized in physics).

The sad thing is, based on his contribs, I suspect his motivation might not have been trolling after all, but performing some kind of extremely misguided "service" to his church. See Robert Sungenis, another "apologist" who just doesn't know what he is talking about when it comes to gtr, but note that Danras's wild misstatements appear too naive to suggest that he might be some kind of Sungenis sock. BTW, for all I know, Danras and Sungenis might be fine fellows when they are not going on about physics; I am just saying that they don't know what they are talking about when they talk about physics.---CH 05:57, 29 August 2006 (UTC)

Albert Einstein

Forthcoming: studies of

1. edits by anons vs. registered
2. average time spent vandalized,
3. discussion of specific fanatical POV-pushing noted in this article:
   • attacks on Einstein's character, including but not limited to the Licorne edit war,
   • claims that Einstein has a learning disability or dyslexia or autism and that these "facts" (actually, fairly dubious claims) should be emphasized with what other users (including me) felt was greatly undue weight
   • arguments over how to describe Einstein's "nationality", with various proponents arguing for Swiss, Austrian, German, American or "Jewsish" nationality, ignoring the well-documented fact that Einstein consistently deprecated the notion of the nation state throughout his life,
   • arguments over how to describe Einstein's alleged "religious/spiritual beliefs", alleged "Zionism", alleged politics, etc.
4. discussion of how such nastiness dissuaded me and other knowledgeable editors from ever making any serious attempt to improve this article, to the detriment of our readers.
5. long and complex article affords many good illustrations of edit creep.

Will move to seperate page.

Some edit rate statistics

The article Albert Einstein, was created on 11/05/2001 18:26 by Zundark (talk · contribs), and eventually evolved into a featured article. Unfortunately, it is also one of the most vandalized "flagship articles" in the Wikipedia (at least compared with other science-related articles). As of 8 Oct 2006,

• this article has received 7207 edits
• over the past year, it has been edited on average 13.6 times per day, or about once every 26 minutes
• at the height (depth?) of the Licorne edit wars in Feb 2006, it was being edited at almost double this baseline rate
• since creation, it has been edited at a rate of about once every 5.59 hours, showing a more than ten-fold increase in the past year over the lifetime average rate,
• an edit by edit study over one month in early 2006 showed that approximately four out of five anon edits of this article represent vandalism or blatant POV-pushing; this rate is consistent with similar studies of other frequently vandalized pages
• the most frequent editors have been (in descending order, amalgamating known sockpuppets/anons):
  1. Reddi (talk · contribs · block log) and an anon known to represent Reddi edits exclusively or almost exclusively, 204.56.7.1 (talk · contribs · block log), aka the Linda Hall Library anon in Kansas City, MO, with 231 edits; in relativity-related articles, in my judgement this individual tends to push a decidedly nonmainstream POV,
  2. 24.253.120.206 (talk · contribs · block log), aka the lv.lv.cox.net anon near Las Vegas, NV (the same city where Friedtwardt Winterberg resides, although I have made no attempt to investigate a possible identification), with 172 edits; in my judgement, a quick spot check suggests that the edits of this user do not violate WP:NPOV,
  3. Fastfission (talk · contribs) with 153 edits; this user is the only one among the top five who has never been blocked and IMO clearly does not violate WP:NPOV,
  4. Licorne (talk · contribs · block log) and just two of his known IP anons 69.22.98.146 (talk · contribs · block log) and 69.22.98.162 (talk · contribs · block log), aka the cable.mindspring anon near Tampa, FL, with 143 edits; as just mentioned, this individual (who apparently once claimed to be one "Dean Mamas" IRL) has been permabanned for his anti-semitic remarks in various relativity-related articles; there appears to be very broad concensus in the Wikipedia community that this editor has consistently pushed a minority (and highly offensive) point of view,
  5. Jeff Relf (talk · contribs · block log), who has been blocked five times for editing this single article, with 102 edits; in the opinion of myself and others, this user is very obviously pushing a minority viewpoint in a highly disruptive manner.

See

• this tool, to verify these statistics
• User:Hillman/Dig/Reddi for documentation of evidence that the Linda Hall Library anon can be identified with User:Reddi,
• User:Hillman/Dig/Licorne for documentation of evidence that the two IPs named above have been used exclusively (or nearly so) by the same individual who used the account User:Licorne

These statistics clearly suggest that, at least for this one article, individuals who (at least in my judgement) tend to vigorously push minority viewpoints in their edits, vastly outnumber individuals willing to put in a comparable effort to maintain WP:NPOV, despite that fact that anti-relativists are such a tiny population compared to mainstream scientists and others who recognized that relativity theory is one of the best tested scientific theories ever known. Unfortunately, in my experience, this pattern holds true throughout the science pages at the Wikipedia; there is tiny handful of individuals who try to maintain WP:NPOV in science-related articles, but I have documented almost one hundred individuals (many of whom have created numerous sockpuppet accounts, and some of whom form more or less acknowledged meatpuppet clusters) who push specific minority viewpoints in the physics-related pages alone.

Some comments on the Seigenthaler defamation scandal

Media fallout from the Seigenthaler scandal apparently induced Jimmy Wales to make a statistical study of his own, whereupon he issued this thunderbolt (sent to a WP mailing list):

Today, as an experiment, we will be turning off new pages creation for anonymous users in the English Wikipedia.

Last Sunday at the wikimeet in London, there was some informal discussion in which David Gerard referred to the new pages patrollers as being very brave people (which they are) standing in front of a firehose of nonsense (which it can be).

If we diagnose what went wrong in the Seigenthaler case, this seems like a very opportune place to try a small change of policy.

1. First, the Seigenthaler article was created by an anon.
2. Then, a patroller simply corrected a spelling error and wikified the entry a bit, but did not recognize the spurious claim. This was a regrettable error, but one which we can understand in retrospect as being a byproduct of the sheer volume of work.
3. Because the entry was never well-linked from related articles, the subject-area sorts who would have spotted the dubious claim likely never saw it.

It seems to me that the first thing we can do is try to reduce the workload on the people doing new pages patrol. A fairly extensive monitoring and survey of new pages conducted by me over the past few days, coupled with discussions with several people who keep an eye on such things, suggests that we can have a substantial improvement here by eliminating the ability for anons to make new pages.

There are some potentially negative side-effects, which is why I call this an experiment:

1. Annoying anons may simply decide to create accounts and make annoying nonsense pages anyway. This will certainly be true in some cases, but it is an empirical question as to how many.
2. We will lose good new pages created by anons of good will. This may cause the growth of English Wikipedia (in terms of the number of articles) to slow a little bit. With 800,000+ articles, and ever-increasing traffic to the website, this seems to be a worthwhile cost.

Notice that anons can still edit. I am a firm believer in the validity of allowing anons to edit. Most anon edits are good, and done "on impulse". We would most of the good edits from anons if we did not allow anon edits, but we would probably not lose most of the vandalistic anon edits. So the net effect of forbidding anon edits would likely be negative.

But preventing anons from creating new pages is a different matter, and it seems a worthy time to make an experiment of it.

--Jimbo

Many users, including myself, don't subscribe to this list or (embarrased grin) read the paper, and therefore first heard about the change from not entirely accurate news stories. In the discusion of this policy change which ensued in various places on WP, including the Village Pump, several readers, including myself, commented on their first experience of the CNN Syndrome (in which the murrrrican Secretary of Defence complains that he learned from CCN that France has invaded New Zealand five minutes before he heard it from his befuddled minions. Not, of course, that France would really invade New Zealand...would she?)

Below I have archived some comments by Zocky and myself which orginally appeared at the Village Pump policy discussion formum on or about 8 Dec 2005. I have slightly modified the indentation for improved readibility and also removed a few remarks from my own comments which referred to pieces of the discussion which I am not trying to archive here.

I support the experimental policy change. And it should be clear to anyone who's read a newspaper—it turned up on my print copy of the Boston Globe yesterday—or performed a Google News search on "Wikipedia" lately--who made the experimental policy change and why. The scary part is that the reporters who wrote the news articles, and the public who reads them, probably don't understand just how little this policy does to prevent gross problems like the Seigenthaler issue. But it does something, and we'd better do something. The better Wikipedia gets, the more people trust it; the more people trust it, the angrier they are going to get at gross inaccuracies in it. Saying "you can fix it yourself" and "it's a work in progress" and "we never said you could trust it" are not going to be an acceptable answers forever. Dpbsmith (talk) 21:40, 7 December 2005 (UTC)

But they sure are acceptable for now. We are a work in progress. Any thinking that we're close to being in publishing quality is strongly misguided. We need hundreds of thousands of articles that far exceed our current FA standards to do that. We are a work in progress, which doesn't make us useless. People can use us to find links to primary and secondary sources and get informed there. We also give a good basic idea of great many subjects. But anyone using an often tertiary source like Wikipedia to check facts should be either told not to believe everything they read on the internet, or told to do their homework or job properly, as in the hillarious case of NYT reporters who need to be told not to use Wikipedia as a source for writing stories [2]). Shouldn't we be using them as a source, rather that the other way around?

We should stick to our guns and ignore what media says. Zocky 22:03, 7 December 2005 (UTC)

I strongly support measures, even experimental or draconian measures, to curb the creation of hoax articles and the vandalism of existing articles, and I strongly feel (on the basis of my own WP experience) that wikicruft poses a critical problem which urgently requires amelioration. I have no illusion that hoaxes or vandalism of WP can be eliminated entirely, but this is not the point. My point is that experience--- at least, my experience--- suggests that wikicruft is displacing high quality content. (In fact, even displacing content plus stubs.) I believe that as a community we urgently need to recognize this, and to take steps to curb the growth of the problem, which should enable us to catch our breath and consider how to eliminate the huge amount of vapid nonsense, crackpottery, vanitycruft, and even carefully crafted hoaxes which has already crept in.

Some months ago I spent a Sunday carrying out an informal survey in which I attempted to vote on every AfD (I could only vote on perhaps 10%, as it turned out) and also tried to monitor the listed articles with the list of recent edits. I concluded that the majority of new articles on that day were obvious hoax articles, or obvious vanity articles, or otherwise obviously crufty articles. In addition, in the categories I watch closely (math and physics), something like 1-2% of new articles are cranky or less obvious hoax articles.

As another illustration of the magnitude of the problem, I have also been tracking Albert Einstein for months. This is one of the most visited physics-related articles in WP, to judge from the fact that it is vandalized several times per day. Yes, most of this vandalism is reverted, but is not the point. Check the history page of this article, note the timestamps, and do the math. Despite the allegation that an army of honest folk quickly revert vandalism to such articles, by my estimate a random user has a chance of finding this article in a vandalized state at any given moment which I find unacceptable (higher than say one chance in a thousand). Yes, this article is vandalized much more often than some little known stub, but again this would miss the point. Random users, schoolchildren, etc., who visit WP to learn about a topic are much more likely to visit the most popular articles, but unfortunately, popular articles are more often vandalized. Surely this has something to do with why (according a recent and highly informal CNN internet poll) 66% of those voting believe WP is unreliable. (Too bad, incidently, that Wikimedia presumably lacks the funds for a properly designed Harris Poll or something like that.)

In addition, the community should recognize that it is a huge waste of the valuable time of users of good faith to force them to either spend the time required to revert with sufficient care so frequently, or else to let a core article (in my fields of interest, anyway) become corrupted by vandalism of various kinds. IMO, users like myself should be free to create new content, rather than spending all our time in bootless attempts to protect the articles we have already written or rewritten from vandalism or other degradation (e.g. the insertion of irrelevant political rants by certain registered users, which I think has been another recurrent problem with Albert Einstein, even when I happen to agree with the political opinions expressed!).

Deco, I notice that we agree that about "half" of all WP edits seem to constitute vandalism, but clearly we disagree about whether half is too much. Also, I guess you would agree that rather more than half of anon edits constitute vandalism, hoaxes, or other edits which (we would probably agree) are destructive to the stated purpose/goals of WP. Yes, I do sometimes see legitimate edits by anons to the pages I watch, but by my count, the ratio is roughly 5% legitimate to 95% illegitimate edits for anons.)

These are only a few reasons why, based upon my own experience trying to create and maintain high quality articles (in math/physics), I am much more pessimistic than some others who have spoken up about how bad the cruft problem already is, and how important it is to curb anonymous edits. When I see comments like those above from Deco and C. Parham, I think they must be living in some alternate reality (rather, unreality), but perhaps they are simply watching a very different set of pages from the ones I am watching. I would like to see Jimbo &c. acknowledge (by policy changes, not just discussion) that some pages (including Albert Einstein) need more protection than others.

I also feel that the rapid growth of WP has overwhelmed the current admin system, which simply has not scaled with sufficient grace to address the problem of administering such a huge and complex enterprise. In particular, I have doubts whether the current system supports experimental policy changes, because I have doubts whether there is an adequate system in place to track statistics and otherwise make a proper rational assessment of whether an given experimental policy change seems to be working well.

Banning all edits by anons is only one baby step toward ameliorating the wikicruft problem, but I strongly believe that this is an essential first step. I have the impression that the WP board is extremely reluctant to acknowledge certain regretable hard truths about human nature (or at least, the nature of some humans), but I strongly believe that it is only a matter of time before they will have take this step, however reluctantly. My point is that putting it off will only increase the pain, because WP will have been that much more degraded by the time Wikimedia gets serious about protecting high quality articles which have already been written (which in my view is the best way to encourage the creation of more high quality articles). Much better to do it now and allow those who fear it will somehow ruin the WP experience to learn better, while there is still sufficient ratio of signal/noise at WP to make it worth saving.

I would like to see (or learn of the prior existence of!) a convenient forum where registered users can discuss all these policy issues (and more) with a reasonable chance of being heard by those who are in a position to make changes.

But leaving aside the question of what policy issues still need to be addressed by further (perhaps provisional) policy changes, I have two immediate concerns about this particular new policy which echo points raised by others, including several who strongly disagree with me about the extent or seriousness of the wikicruft problem:

• why did I learn about this new policy from a friend who read about it on CNN, rather than from the Wikipedia welcome page? Shouldn't major policy changes be announced in a prominent place at WP? Sean said "Jimbo has assured everyone..." --- Sean, where did you see this assurance, pray tell?! I have been editing WP for months, so I am surely no novice, yet I after several minutes I couldn't find the official announcement which presumably exists somewhere on WP!
• Banning all edits by unregistered users is neccessary, but the new policy falls far short of even this tiny first baby step toward saving the WP from the rapid growth of wikicruft. The new policty strikes me as a pulsillanimous attempt to curb the creation of new hoax articles by anons, but its does nothing to begin to address the broader wikicruft problem, which I believe is already even more serious (and is rapidly growing even worse).

Where are such WP policy issues publically discussed? (Assuming they are publically discussed.) This page is too inconvenient a forum for user input, IMO.

To end on a slightly more positive note: I presume that this change means that the WP board is at least discussing the issue of wikicruft and what to do about it, and if so I am glad to know that hoaxing and other wikicruft is at least on the radar screen. I hope the board will remember a bit of traditional WP advice to newbies: go ahead, make changes, be bold!---CH 01:05, 8 December 2005 (UTC)

CH, I for one appreciate what you are saying: losing time with people who are not here to help is a problem. But you seem to be missing an important issue – all we know for certain is that allowing anons to edit is a part of the process that works. We don't know whether it's essential, but we do know that similar projects that tried to get quality through limitation of access rather than through sheer numbers of editors have failed. Zocky 02:54, 8 December 2005 (UTC)

Well, if we start thinking like that we might well wind up saying that WP is a utopian social experiment and therefore is doomed to ultimately fail, simply because all previous utopian social experiments have ultimately failed. In fact, I have often parroted this slogan myself, and mostly believe what I squawk, although I try not to think about it :-/ So perhaps we are simply discussing which mode of pstittacide we prefer :-/ But seriously, I would like to postpone the inevitable (assuming WP be not the exception which proves the rule). Others who are not as pessimistic about the long term future of WP as I might still agree that realizing the goal of providing the world with a free, on-line, universal, and high-quality encyclopedia is incompatible with permitting the rapid growth of hoaxes, vanity articles, blatant factual inaccuracies, crackpottery and other cruft in the Wikipedia.

Just to be clear: from time to time I have been pleasantly surprised to see some helpful anon correct my often careless spelling in some article I have labored over, and I am sincerely grateful to those editors. So the claim that some anonymous edits are helpful is not in dispute, at least not by me. I do not even dispute that from time to time anons have started legitimate articles which have sometimes grown into good articles.

But as I see it, we are discussing making judicious tradeoffs, in which we may choose to give up the genuine benefits from good anon edits in order to protetct the WP from further degradation by bad anon edits. And my belief, or rather my wild surmise, is that the best way of ensuring the rapid creation of more high qualtity WP articles is in fact to demonstrate that we care about the articles and categories we already have, at least enough to preserve them from careless or intentional destruction.

As I see it, dealing with anons is only one facet of protecting the good material already in WP and promoting the goal of reversing the figures in that CNN poll which claims that about 2/3 of respondents believe that WP is not reliable. (I venture to guess that surveying journalists would show that 90% think WP is not reliable, but that many or most of them would probably anonymously :-/ admit to using WP as a source, which raises the issue of what I see as our broader social responsibility, as citizens of the world, to clean out our cruft.)

For example, on several occasions since I came to WP, I have watched with dismay as articles reach a state of which WP can be proud, but then are gradually dismantled by careless edits, sometimes from well-intentioned registered users who are too hasty or inexperienced to take care not to shove in new material any old place, but rather to to try find some place where it fits neatly, or barring that, rewriting nearby paragraphs in order to correct any damage done to the previous flow of ideas. A dangerously naive WP myth holds that (apparently by some previously unknown law of nature) articles can only improve monotonically in quality. As I see it, this is rather like saying that we need not worry about energy resources because the laws of thermodynamics are too depressing to take seriously! I suspect that the typical evolution of a WP article is more analogous to statistical-mechanical fluctuations in discussions of Zermelo's objection to Boltzmann's so-called H-theorem than to the proverbial free lunch. In social experiments, as in physics, if a claim seems to good to be true, it probably is.

But while by no means all bad or otherwise problematic edits are made by anons, it is nonetheless true that anons are as a group responsible for far more than their fair share of bad edits. And the table presented by Zocky below says more than I could say in a thousand words about why we must, however reluctantly, ban unregistered editors from the WP.

It has not escaped my notice, incidently, that part of my frustration with the claim that we must at all costs preserve the right to edit anonymously, is that I still find it difficult to understand why anyone would feel (apparently) that this right trumps all other considerations in promoting the stated goals of the WP. Some users hint that they believe that habitual anons are too shy to register even under a pseudonym. I can well understand why users might not wish to use their IRL identity here--- in fact, from my own experience I'd probably even recommend editing under a pseudonym (too late, in my case!)--- but while WP is not without security flaws, the Wikimedia privacy policy is easy to find and the registration process is fast and easy. So, I tend to wonder why some users seem to have a huge problem with establishing even this minimal level of responsibility for their actions here. Speaking as one deeply concerned about current trends in my own country (the U. S.), I find it hard to understand why people concerned about supression of dissent or whatever would fail to recognize that free speech does not entail free denial of responsibility for what you do and say in a public forum.

I'd like to see a kind of round table discussion of all these issues (I'd guess participants would have to start by agreeing upon some list of issues to discuss, heh), in hope that as a community we can make wise decisions of this difficult and painful nature. If nothing else, in such a forum perhaps someone who truly believes that the right to edit anonymously trumps all other considerations for the health of the WP can explain their reasoning to me. Indeed, we might all perhaps learn that addressing the thorny problem of suppressing wikicruft is even more difficult than we had each previously realized! ---CH 09:59, 8 December 2005 (UTC)

You may be overlooking an important factor, that is recruitement. People burn out or just go do other stuff in life. We need a constant influx of new good editors. Most of started as anons, and many of us would have never tried editting in the first place in registration was required. Zocky 12:26, 8 December 2005 (UTC)

Zocky, Of course I agree that enlarging (not just maintaining) the community of active editors is essential to the continued growth of high quality content in WP. I suspect our only difference is this: I never edited anonymously before I became a registered user, but you did. I guess we both may have assumed that most other users are like ourselves. I'd like to see a survey of active WP editors to find out if an overwhelming majority agree with "I never would have started to edit the WP if registration had been required".

When I called for a public discussion here at WP in advance of major policy changes, I should have made it clear that I have in mind a "policy cyle" somewhat like this:

• public discussion,
• provisional implemention,
• statistical study of logs, survey data, whatever seems useful,
• assessment,
• loop

One reason for careful public discussion before making changes is to avoid creating a situation where Wikimedia rapidly makes a chaotic sequence of possibly unannounced or at least insufficiently publicized WP policy changes, which would obviously be frustrating for everyone here. I said above that I much prefer spending time at WP creating new content to spending time on cruft patrol, and as you would guess, this implies that I don't want to have to (hypothetically) spend energy trying to track rapidfire WP policy changes in real time so that I can be confident that I am not violating any policies with any given edit.

I should also stress that when I mentioned public discussion in advance of policy changes above, I did not mean to imply that the WP Board, sysops, or admins, should be prevented from taking emergency measures as needed, since occasional unanticipated circumstances seem to be unavoidable in a huge social enterprise like WP (or in maintaining any large website). Of course, in a well-run website any emergency measures should be reversible or amendable once the dust clears.

I also want to add that while I am frustrated with what I see as a weak and tardy response by the Board to the growth of the Wikicruft and accountability issues, we must all remember that one of the most remarkable aspects of the WP is that it has been built almost entirely by volunteers and has a comparatively tiny staff and operating budget, so we do need to cut the staff some slack in that respect. I am disappointed when I see Jimmy repeat shibboleths which do not accord with my own experience at WP, but I was struck by the comments of a poster at news.com, who drew a distinction between truth and knowledge, saying that while truth may ultimately be subjective, knowledge is something which can be shared. I also feel that we must avoid getting bogged down in disputes over moral philosophy or espistemology and focus on practical measures we can take as a community, and sharing seems to me to reside at the core of the WP experience. That might be worth keeping in mind.

I see another issue lurking in the background of this controversy. I don't much care for the possibility that the board might be more responsive to individuals who threaten legal action than to those who prefer to resolve content disputes by talk page discussion and if neccessary an appeal for arbitration, but we all need to recognize that as a practical matter it wouldn't take very many lawsuits to exhaust the WP operating budget (even if they have pro bono lawyers "on retainer"), and as I understand it, there is a real possiblity that a single adverse legal ruling could shut down the entire enterprise in a heartbeat.

Some outraged Wikipedians have suggested that Siegenthaler should have simply edited his own bio and have done with it. I think that is an absurd suggestion: no-one should have to waste time in daily edit wars with some anonymous crank to correct blatant factual misinformation, certainly not when the misinformation insinuates involvement in pedophilia, terrorism or murder (probably the most inflammatory charges I can imagine in the modern world). But perhaps these users were merely trying to express the sharp distinction between technology and social mechanisms which are inexpensive and involve volunteer labor versus those which are expensive and involve highly paid professionals. One reason I think the admin system needs overhaul is that currently I think there are some technical problems which make it more difficult than need be the case for newbies to discover and grasp the distinction between "wikifriendly" ways of doing things (like resolving content disputes over allegations of factual inaccuracies) and "wikihostile" ways. Naturally, we should do everything we can to encourage wikifriendly modes of discourse and dispute resolution.

Beyond this, we probably need to try harder to help technically incompetent newbies learn our ways, or at least seek effective assistance from more experience/competent users if they are having trouble accomplishing some legitimate goal here (like correcting factual misinformation). This is a huge challenge, but one thing is clear: layout is a crucial component of ensuring that WP is easy to use effectively. For example, while Wikipedians like to say that in the event of a quarrel, the sequence of events can be readily reconstructed from the history page, it only takes one technical incompetent to render a history or talk page virtually unreadable--- and now it seems that history pages are limited to the previous 500 edits, which isn't enough in some cases. Also, edit histories are currently almost impossible to reconstruct if a page underwent several moves. These usability issues combine technical and policy elements; technical problems and possible solutions can be best explored by the sysops (i.e. developers, not admins). From time to time I have noticed them playing with various changes, not always with good results. I'd like to see a "developer's sandbox" where they can test describe the rationale for and given samples of proposed layout changes and get feedback from users. ---CH 23:09, 8 December 2005 (UTC)

Problem users

There is no question in my mind that some users pose a particularly insiduous threat to the content value of the Wikipedia, because they are engaged in a persistent, determined, and often quite ingenious campaign to portray their highly idiosyncratic (and dubious) personal opinion as well-established mainstream scientific or historical fact. It is very striking that they use many of the same methods, including manipulation of users who subscribe to populist or Libertarian ideologies, which have been so successfully employed in the past decade to subvert the teaching of science in the U.S. Unfortunately for Americans, not even the will of the people can trump natural law, as they may yet discover the hard way--- unless scientists succeed in saving them from their own idiocy. (I'm an American, so I can say that, right? This is a free country..isn't it?)

Comments regarding the Carl Hewitt RfAr

Like EMS and Linas (I don't know Rudy Koot) I claim expert knowledge of some of the math/physics-related areas in which Carl insists on writing, and I also have extensive experience in expository writing both at WP and at elsewhere. See for example two expository websites which I created a dozen years ago and which still have mirrors:

• Relativity on the World Wide Web
• Entropy on the World Wide Web

Like the complainants, I have wasted a huge amount of time trying to reason with Carl. Like them, I have been led to conclude that the only way to stop his misbehavior at WP is to ban him. It is important to recognize that, if Carl is truly interested in promoting possible applications of his Actor model to physics, we feel that directing his energies to developing his (currently highly inchoate) ideas is in his own best interests, because he is the only person who can develop his ideas until they reach a more significant status than "vague speculation".

OTH, allowing him free reign here would be highly detrimental to the health of WP and the community of expert editors whose presence here, I think, should be particularly valued (as long as they don't lose sight of the fact that WP is above all an encyclopedia, not a stump for personal speculations).

I am not participating in presenting evidence because I am so sick of trying to deal with Carl that I have told him I want nothing to do with him, and I certainly don't want to revisit talk pages to collect links or study the appropriate pages to learn the rules of the arbitration process. However, I thought I'd drop by here long enough to try to provide a summary of why I have concluded that Carl is a problem user, indeed a problem user whose actions here are habitually manipulative, highly insiduous, and particularly detrimental to the reliability and fairness of the WP as an encyclopedia.

The problems begin with Carl's refusal to accept the premise that WP is not the place for unbridled personal speculation. Particularly not WP article space.

Unfortunately, rather than confining himself to writing fair and factually accurate descriptive encylopedia articles on the current state-of-the-art (as reflected in current professional practice and the current research literature) on subjects in his area of acknowledged expertise (certain parts of computer science, henceforth CS), he has insisted on writing articles which claim, at least by implication, to describe well-established interdiscliplinary theories but which in fact contain badly expressed and ill-informed speculations about (vaguely described) relations he thinks should exist between a CS concept he helped develop (Actor model) and relativistic physics (see WikiProject GTR and note that both EMS and myself are members). These articles are often followed by long lists of citations which consist largely of Carl's own CS publications. You might look at... arghghgh, now I can't find the link, but somehow Carl manipulated me into completely rewriting an article called something like relativistic information theory. I was going to suggest you compare his version with my rewrite to see what I am talking about.

I and others have pointed out to Carl the existence of a huge (if rather disorganized) research literature concerning relationships between relativistic physics and information theory, and pointed that there are apparently no papers published in physics journals or by persons other than Carl Hewitt himself (or, at a stretch, according to him, one section in the unpublished Ph.D. thesis of one of his former students in C.S.) which speculate on possible applications of his actor model to physics. I and others have suggested that he keep his speculative essays (which aren't even very well expressed) at the MIT website, or at his own personal website, and have urged him to spend his energies on developing his thinking rather than in tying up valuable Wikipedia users like EMS or myself in endless and bootless content disputes here. If you consult the talk pages of various articles Carl has authored, I think you will see how hard users like myself, CSTAR, and others have tried to reason with him.

His response has been especially frustrating because so many have tried have tried so hard to persuade him to develop his speculations to the point where they are acknowleged by some significant subset of researchers in both relativistic physics and CS before attempting to describe them in WP articles. Because of his background, Carl is surely well versed in developing ideas to the state where they can be published in the research literature, yet he apparently refuses to pursue this traditional (and far preferable) route, in favor of (in my view) spamming WP article space with his ill-expressed and virtually incomprehensible personal speculations.

Clearly, Carl thinks there should be an interdisciplinary field applying his actor model to relativistic physics and information theory, and clearly he wants to bring about this state of affairs. If that were the end of the story, I am sure we would all say, "more power to him!" The problem for WP is that Carl insists on manipulating WP, often in very subtle ways, to mislead non-expert WP readers into believing that such a field already exists, which is absolutely not true. There is as I said a large and rather disorganized literature on various relationships between information theory (a field on applied mathematics) and relativistic physics, but I have seen no papers in this area which even mention Carl's actor model.

Carl's refusal to pursue what we regard as the proper route to establishing a useful interdisciplinary field in science (namely, publication in repected research journals in all affected fields) is particlarly bizarre because as a faculty member (emeritus) at a major university, he has extensive experience in getting research papers published. As someone with expert knowlege of the classical gtr literature, I took considerable pains to give him some good advice on one possible direction in which he could try to develop his ideas, which I am confident would be of interest to researchers in this field if he were successful in publishing a good paper in one of the journals they read, but I was very disappointed to see that he was ignoring not only my suggestions, he was ignoring even the elementary step of studying the existing relevant literature in field B before asserting that he is creating, or even has already created, a new interdisciplinary field combining elements of A and B.

For me the straw that broke the camel's back was seeing Carl engage repeatedly in what I regard as gaming the system, manipulation of other users, and so forth. He is clearly highly intelligent and I am baffled why he would devote his intellect and energy to manipulating WP (as I eventually concluded), possibly as some kind of "social experiment", rather than trying to develop his ideas in the academic environment which has been so kind to him (and in which he has enjoyed considerable professional success).

Since this is a critical point, let me try to emphasize it: I happen to be a rare user with expert knowledge of at least two of the fields (relativistic physics, information theory) in which Carl wants to create an interdiscplinary subfield, and I offered to help him do that, by providing suggestions, references, and at one point even offering to discuss his ideas in another forum (I was open to suggestion), one more suitable for speculative discussions and scientific development work than the WP. So Carl cannot plausibly claim that I am somehow trying to "suppress" his ideas. To the contrary, I offered to help him polish his ideas to the point where he can publish them in respectable journals (other than CS journals; he has apparently published some speculation in CS journals, but the physics literature shows clearly that physicists don't read those journals and probably would need more help in explaining CS background than Carl tends to provide even if they did).

The issue is not whether his ideas have any merit (or rather, might one day turn out to have merit), the issue is whether he should be allowed to describe his speculations here as if they have the same scientific status as well-established theories such as information theory and general relativity, when in fact such a claim would be analogous to suggesting that some author's notebook of unorganized jottings is comparable to the Odyssey or The Great Gatsby. There is a huge difference, even if the author in question happens to be, say, J. D. Salinger. (Comparing Hewitt to Salinger is excessive, but no doubt you take my point.)

It is telling that another user with expert knowledge (in CS and physics), CSTAR, also offered to help Carl at every turn, and spent an enormous amount of time trying to help him clarify his thoughts (and to learn the WP way). And guess what has been the end result of CSTARs attempts to help Carl out? Carl's flagrant misbehavior has caused CSTAR to quit the WP! And now Carl is apparently on the verge of driving out EMS as well. If that happens, WikiProject GTR cannot survive, and I will have to leave.

If you somehow manage to add up the trouble he has caused here, I think you will be amazed at what you find, in terms of his driving away some of the most valuable editors I have encountered here. As a user loyal to the stated goal of creating an on-line, free, universal, timely, and accurate encyclopedia for the benefit of humanity as a whole, I find it extremely troubling that manipulative problem users like Carl are empowered by current WP policies to waste so much time of valuable Wikipedians like EMS, CSTAR, Linas, and now of course three members of the arbitration commitee.

Anyway, good luck, you'll need it. And be careful, in my experience, Carl is a master of gaming the system.---CH 02:01, 9 December 2005 (UTC)

Comments regarding the Roylee RfC

I came across this particular RfC (which lies outside my current areas of Wikifocus, if not my areas of interest) while researching social networks among anons who regularly engage in vandalism, trolls, hoaxes, or the promotion of cranky agendas. The RfC concerns the misbehavior of another user, one User:Roylee, who is also engaged in a persistent effort to slant the WP toward his own highly peculiar opinions, which seem utterly lacking in support in terms of verifiable facts.

An additional wrinkle is that Roylee, when challenged, apparently decided to abandon his user account and to start editing exclusively from anonymous IP addresses. This explains how he came to my attention, and it might support the suggestion that registered users are more accountable to demands that they cease making claims which they are unable to support with verifiable sources. Be this as it may, User:Mark_Dingemanse and User:BanyanTree (and possibly others) have created pages related to this RfC. In one of them, BanyanTree made some comments which I liked:

In my opinion, the editing highlights a flaw that goes beyond using new articles as nodes for bad edits. It attacks Wikipedia through the structure of Wikipedia user specializations. People choose specific areas in which to specialize and are thus have the knowledge to notice nonsense arguments and revert accordingly. Alternatively, the vandal watchers are generalists and, due to Assume good faith and Don't bite the newcomers, will not normally revert edits that aren't obviously vandalism and, even if they challenge an edit, will normally back off if the user seems to be more knowledgeable than they. Because Roylee makes authoritative-sounding edits across user specializations, the topic specialists will not follow the edit pattern out of their normal hunting grounds while the vandal hunters won't push too hard. I've often thought that, if it wasn't so destructive, I would admire Roylee for the way he exploits the social structure of Wikipedia. The dynamic IP address is certainly part of it, but the really troubling bit is how easy it is to pull one over on Wikipedians.

The good news is that it has been discovered and users have tried to repair the damage. The bad news is that I'm sure we've missed IPs and we don't know if there are other "Roylee"s out there who we don't know about. It took over a hundred edits before Mark began to reel Roylee in. Is there another user whose made 50 similar edits, who has not been discovered? Are there a hundred such users? I've certainly felt drained by the effort involved in tracking edits, even as peripheral figure, and can't imagine how exhausted the more active watchers have been. It's certainly made me have some pretty extreme thoughts in terms of solutions. I have actually considered that a massive range block would be an extreme, but appropriate, solution. Being entirely frank, it may also have convinced me that the perennial proposal of prohibiting anon editing may have some merit. I had previously thought that all of the good edits of anons outweighed the vandalism, but I had also assumed that the vandalism was being identified and removed. Roylee has shown that extraordinary vandalism can last months, and we don't know enough to say if this is closer to the best or worse case scenario. I think evidence towards either end of the spectrum will help me decide if I think Wikipedia is an encyclopedia or a game about being an encyclopedia.

I would like to hear your reactions. For me, reverting vandalism is either just an chore or sometimes interesting when the pace picks up, but Roylee just make me feel... unsettled, which probably means that it's important. With best regards, BanyanTree 03:56, 3 December 2005 (UTC)

I am sure there are more Roylees out there. The good news is that while the number of individual IP addies used is large, the number of guilty parties might be much smaller. (Possibly hundreds of individuals behind a huge number of sometimes very insiduous anon edits, compared to roughly six thousand active Wikipedians, of whom the majority are presumably well behaved.) Be this as it may, in response to a brief comment by myself in his talk page, BanyanTree remarked that he has seen valuable Wikipedians driven out by frustration over nasty RfCs and such like, saying in part:

As I stated, I don't know what it all means for me just yet. The two issues that seem to add users that I respect to the ranks of Wikipedia:Missing Wikipedians dispoportionately are too many ridiculous edit wars over minor issues and dealing with the constant onslaught of vandals and just plain schlock editors. It is not entirely clear to me if this is a steady state in which new users are replacing the users who are leaving fast enough to scale with the growth of Wikipedia as a whole or if Wikipedia's resilience is actually deteriorating. I think Larry has made some good points, but he does himself no favors by occasionally popping up to reminisce about how Wikipedia is all his idea and how much it began to suck once he left, at least if his intention is to help Wikipedia.

What the wiki desperately needs are some metrics, like someone threw together recently on Wikipedia:Village pump (policy) on the relative contributions of various categories of users, as right now all the discussion is taking place in a vacuum. Hopefully the recent hoopla will inspire someone to create statistics along the lines of "Percentage of new articles created by anonymous users that are deleted within one week", "...by users whose account is less than one week old...", etc. Once the article validation system gets put into place, something like this may be possible for individual articles as well to track individual editors.

A categorization of pages, going from "stub" to "growing" to "mature", combined with restrictions on user editing, is one idea that I've been thinking aobut. For example, any user may edit a stub, while only users who have made a few edits over the previous month may edit "growing" and "mature" articles can be limited to regular users. I'm sure there's something awful about this idea that hasn't occurred to me yet...

I'm still letting this all churn for a while, but I figure that there must be some way to address the issue of both lack of expertise and the "bad edit" onslaught that drives so many good users away. I have certainly begun taking a harder line on editing. I suddenly dawned on me that if a user inserts a racist rant into Rosa Parks, the last thing we want is for him to come back. But that's enough pondering from my end. Obviously, my opinions haven't coalesced on this. - BanyanTree 04:53, 8 December 2005 (UTC)

Bizarre question on Reference Desk

Question posted by Pce3@ij.net (talk · contribs) (IMHO) to Wikipedia:Reference_desk/Science 04:36, 19 June 2006 (UTC), apparently in reference to Half life

This may seem like a rare problem but I have noticed that some scientific and technical articles are being edited to prevent other users from obtaining a complete and full comprehension of a topic in the same manner as a member of a trade or artisan guild might try to hide techniques or methods or understanding of what the topic actually involves. Such articles are only permitted to have a highly technical version or explanation of the process being presented in the same manner as a tradesman or artisan might withhold simple explanations from a patron for the sole purpose of mystifying the topic and keeping the patron from knowing “too much.” What is the Wikipedia policy on such behavior where simple and accurate explanations are continuously deleted from an article on the false pretense that the article is not about the example although the example fully clarifies the topic?

— IMHO

This drew sceptical queries, but also this, from StuRat (talk · contribs) 17:11, 19 June 2006 (UTC):

The problem with reaching consensus is that there is a small group of committed academics who jealously guards some articles, immediately removing any attempt to make them accessible to the general public (which, while far more numerous, lacks the same level of commitment and is thus easily scared off by such actions). The only way I was able to find around this problem was to create a separate article for laymen. For example, the article Boolean algebra suffered from this problem, so I created the simplified version as Boolean logic, and added dabs at the top of each, pointing to the other.

— StuRat

IMHO then created his own article; see Wikipedia:Articles for deletion/Half-life computation

See also

Some WikiGroups whose members might generally favor the views expressed here:

• Forum for Encyclopedic Standards
• WikiProject Fact and Reference Check
• WikiProject Physics
• WikiProject Science
• Recent Changes Patrol
• Counter Vandalism Unit
• Association of Deletionist Wikipedians

Some WikiGroups whose members might generally oppose the views expressed here:

• WikiProject Countering systemic bias
• Association of Amoral Wikipedians
• Association of Inclusionist Wikipedians
• Association of Anarchist Wikipedians
• the Wikipedia Cabal, if there were one, which of course there is not!

The MetaWiki pages are humorous spoofs... I think. But still worth reading.

Also well worth visiting:

• the Scientific method portal
• my favorite perennial proposal
• The epsitemology of Wikipedia, by Larry Sanger, who unfortunately visited sci.physics.relativity soon after I quit that newsgroup (because of problems somewhat similar to those we are encountering here).
• Deferring to the experts, another essay by Larry Sanger expressing views similar to my own.
• Wikipedia and Why it Matters, by Larry Sanger.
• The art of Wikipedia weeding, by, er, Larry Sanger!
• Referees, a proposal being shepharded by User:Andrewa
• Foundation of Wikipedia
• The Future of Wikipedia
• A Three Year Plan for Wikipedia
• Heh, I thought that crack about Wikipedia being founded upon dirty (pics) money was just a scurrilious rumor, until I read this.