Talk:Trust metric

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Well, I looked at the "trust metrics" wiki, and I was unimpressed by the quality, and unable to figure out how to edit a page, so I just gave up.

Anyway, the solution for Wikipedia is to add a "Rate this change!" feature. Each change could be rated on a -3 to +3 scale.

-3 : This user should be banned.

-2 :

-1 : Change has some minor problems, but nothing serious.

0  : no opinion

1  : slight improvement (grammar fix, small factual error)

2  : big improvement

3  : major contribution, verified to be correct

Then, users who themselves collect many "highly rated changes" would be able to bestow such status on others. However, to prevent people from just falsely rating each other highly, you just have to make sure that someone can only have a high rating themselves after earning it from someone trustworthy.

The important thing is to start collecting the data. After that, it should be very easy to calculate who is trustworthy.

Also notice that a "no opinion" rating would still be useful. It would mean that someone inspected the change and found no egregious errors or sabotage.

This page should really be used to discuss the article in the question. For the issue of trust metrics and networks in Wikipedia, see WP:TRUST. TheGrappler 18:25, 7 June 2006 (UTC)

Proposed Wikipedia trust metric[edit]

Moved this to the talk page, as it might be interesting for wikipedians, but irrelevant and self-referential in the article:

Levien also proposes a trust metric for Wikipedia to aid quality control and to minimize the effects of Wikipedia:Vandalism.

--Tgr 18:49, 28 June 2006 (UTC)

Unencyclopedic Tone[edit]

Is "kook" really an appropriate word used in a description?

The person in question is featured (twice!) on, he is famous for spamming discussion forums since before the web started. A gadfly provides a useful function despite being annoying. The person in question is just annoying. See for more details. I'm reverting this to 'kook', unless someone can come up with a better term. Gadfly is completely misleading. (talk) 06:03, 7 May 2008 (UTC)

I know Arthur Murray personally, and I think I can say we are friends. (Yup. He just walked in, is now literally at my elbow, and says I can say this. And is correcting my minor typos.) Aruther is nice, friendly, intelligent fellow. But in respect of "AI" or "theory of mind", "kook" is appropriate. He is also a notorious sock-puppeteer, very good at generating mutally reinforcing self-references. My initial impulse was (and even upon reconsideration is) to remove the link to his post on Advogato, on the grounds that this not a proper source. However, it is a good example of a kind of problem that trust metrics needs to deal with. (Arthur says he did not add the link to Advogato, so perhaps the person that did intended it as an example.) J. Johnson (talk) 00:47, 6 December 2008 (UTC)

Replaced the dubious "AI theorist" with "notorious individual". Although the original "kook" is fully justified. Personal note to Arthur: talk to me before you do any editing of this article. Keep in mind the possible consequences. J. Johnson (talk) 00:49, 16 May 2009 (UTC)

Google Page Rank[edit]

Google page rank can be manipulated. Gordo 10:16, 28 August 2006 (UTC)

See references at Search engine optimisation --Gordo 10:20, 28 August 2006 (UTC)
In particular Googlebombs and Googlebombing: Teh Rules --Gordo 10:22, 28 August 2006 (UTC)

Trust Coloring Demo[edit]

Has anyone here seen the demo at The UCSC Wiki Lab? I was hoping to find more Wikipedia discussion of this topic on this page, but, unfortunately, I'm probably looking in the wrong place. —Preceding unsigned comment added by (talk) 14:58, 2 September 2007 (UTC)

Questioned change (re "computer programming")[edit]

Xme's substitution of "computer progamming" as the context of trust metrics is questionable. Yes, there has been work to implement trust metrics in software, which is a use of programming, but I have never heard of trust metrics being used in programming. The proper and legitimate context of the concept appears to be as was originally stated, in psychology and sociology. Therefore I am restoring the original form. If it is desired to mention possible applications of trust metrics in the design of trusted networks or trusted computing, then the proper terms should be used; "computer progamming" is not the proper context. -J. Johnson (talk) 21:22, 15 September 2009 (UTC)

Repeated anonymous edits[edit]

There has been more anonymous editing of this article in regards of noted Internet kook "Mentifex", consistent with his past history of anonymous self-promotion and sock puppetry. (See references in article for details.) I am therefore requesting semi-protection of this article (and its talk page) from anonymous editing. J. Johnson (JJ) (talk) 19:03, 29 September 2009 (UTC)


The {{editsemiprotected}} template is for requesting changes be made to articles that are already semi protected. You're looking for Wikipedia:Requests for page protection. DoktorMandrake 19:32, 29 September 2009 (UTC)
Note however, that this page is unlikely to be protected on account of a single anonymous edit. Page protection is only for cases of persistent vandalism. DoktorMandrake 19:37, 29 September 2009 (UTC)
I figured as much when I saw the banner go up. I think protection is warranted here, as this is not an isolated case for this article, and individual suspected is notorious for this kind of thing; semi-protection would, in the end, save time and effort. But I gather we shall just have put up with this – Cassandra-like. - J. Johnson (JJ) (talk) 22:00, 30 September 2009 (UTC)
W.T.F. I am not mentifex. The section I removed about Mentifex being a crank is completely irrelevant to this article's subject. Why on earth are you re-inserting it?? (talk) 02:42, 1 October 2009 (UTC)
It ain't me, babe. (talk) 03:35, 1 October 2009 (UTC) Mentifex
Not irrelevant. The earliar version of the article cited "Mentifex" as an example of a problem that trust metrics has not adequately solved. This led to a question of whether Mentifex poses a problem that needs to be solved. I believe the URL given above adequately answers that in the affirmative. Just within this article we can see mysterious, mystical and quite anonymous edits that either try to enhance Mentifex's reputation (such as being a "noted" AI theorist, when he is entirely not), or remove references that are unflattering. So, at one level there is a question of whether any laudatory reference to Mentifex can be trusted, and as this goes to the heart of what trust metrics is trying to address I say it is very relevant to the article. At at another level, there is a question of not just who is doing these anonymous edits, but (to take the current case as an example of the general problem): how can Mentifex prove he is not doing these edits? - J. Johnson (JJ) (talk) 21:24, 2 October 2009 (UTC)
There have been other cases where the reliability of Advogate has been brought into question; there is no need to give Mentifex any additional publicity. (talk) 14:28, 19 October 2009 (UTC)
This IP belongs to the University of Cambridge; Mentifex is not here. So stop being a paranoid asshole. (talk) 14:30, 19 October 2009 (UTC)
Et tu, JJ? Have you joined the Mentifex-basher witch-hunt? On 6 December 2008, as you have described above, you let out a loud whoop when you saw me enter Zoka near Green Lake. I walked over and sat down to your right. Although I was correcting your spelling, I did not at first realize that you were posting an anti-Mentifex diatribe. Have you gone over to the Dark Side, JJ?. Back in 2003, you were so helpful in contributing Perl code to and in teaching me how to properly indent my JavaScript AI code. Are you not grateful for when my employer and I stopped by your bicycle in Ballard and had you come along with us to Ray's Boathouse on Puget Sound? Did you not appreciate it when we left a new bike tire for you hidden in the bushes near Zoka? JJ, JJ, why do you persecute me? Posted (talk) 22:58, 2 October 2009 (UTC) by Mentifex
Arthur, listen up: the purpose of the talk page (and this is not my diktat, but WP policy) is to discuss how to improve the article – not to debate whether you are the world's greatest AI theorist. Nor am I (or anyone else) persecuting you. (It is as your friend that I try to dissuade you from being a jerk.) You have a long, notorious, and documented history of being a real pest on various forums. Which is to say that you pose the type of problem that "trust metrics" would address, and therefore relevant to the article. (No?) But I would strongly suggest to you that this talk page is not the place to complain about your reputation, your feeling of being "bashed", or whether I am some sort of ingrate. - J. Johnson (JJ) (talk) 21:28, 3 October 2009 (UTC)
No fucking shit, the article is not a place to debate whether Mentifex is the world's greatest AI theorist. Why do you keep adding that shit in?? Get your head out of your ass. (talk) 14:26, 19 October 2009 (UTC)
The previous edit from (which I suspect will be purged soon on account of bad language) certainly does not advance the conversation. And is futher reason why anonymous edits should be banned. The gist of the comment ("the article is not a place to debate whether Mentifex is the world's greatest AI theorist") is correct, but only secondary, at most. The key issue here is whether Mentifex (and/or anonymous editors) can add laudatory (but untrue) material, or remove (censor!!) material that is critical. Which might be deemed as depending on a determination of what is truth (is he, or isn't he?), but I say that, based on reliable and verifiable sources, the matter is settled. No debate. So why do you continue to remove the links to those sources that settle the matter? - J. Johnson (JJ) (talk) 00:04, 21 October 2009 (UTC)
Oh shut up with the self-righteous bullshit. Your argument is totally unjustified. The key issue is the material that goes into the article, not your paranoid fantasies about who is editing it. I will point out again that you are adding irrelevant crap and removing sourced, relevant explanations of Advogato. (talk) 17:24, 30 October 2009 (UTC)

You really should review WP:civil, as your comments seem quite actionable on that basis. As to your particular points: 1) I do not believe (as previously explained) that the notes I added are irrelevant. If you disagree, then the expected WP:etiquette is to discuss it (not sling profanities and ad hominem attacks). 2) As to "removing sourced, relevant explanations of Advogato" — isn't that just what you are doing? Note that I have not removed any material, just reverted your deletions. But perhaps this is best continued below.) - J. Johnson (JJ) (talk) 21:48, 30 October 2009 (UTC)

Rebuttal by Mentifex[edit] is a statement by Mentifex in rebuttal of the cited Advogato Has Failed and in rebuttal of other attempts to use the Trust Metric to eliminate free speech. Posted (talk) 00:06, 7 October 2009 (UTC) by Mentifex

Arthur, "free speech" is not your personal license to intrude into adult conversations. Your babblings on AI are entirely devoid of meaning (except as indications of your own mind), you have no credibility, and just about anywhere you show up on the Internet people get real tired of you. The only significance of your "rebuttal" is to demonstrate just why people want some way of filtering out kooks like you. - J. Johnson (JJ) (talk) 22:53, 8 October 2009 (UTC)

Continuing reverts.[edit]

Undid a revert by an anonymous user that removed references about Mentifex, and carried a comment that "A page on Trust Metrics is not a relevant place to discuss Mentifex."

On the contrary. Mentifix is a prime example of the problem which trust metrics attempts to address, which in itself warrants at least a mention. That a mere mention gets dragged out into an extended discussion is because of his persistent efforts, in this article, to try to buff his totally discredited claims, and to anonymously remove references that document his notoriety. Any anonymous editing of that character is therefore suspect, so I will be reverting all such edits.

In this case the edit came from, putatively from Cambridge University, but 1) do not misdoubt Mentifex's demonstrated sock-puppet abilities, and 2) has already been flagged for incivility. If there is a real person (even Arthur) who would like to make a real contribution then I would strongly suggest getting a Wikipeida account. - J. Johnson (JJ) (talk) 21:17, 19 October 2009 (UTC)

I will not make an account just to satisfy your paranoia. Wikipedia is supposed to be a "free encyclopedia". I will be reverting all such edits to replace useful content with irrelevant crap. (talk) 17:21, 30 October 2009 (UTC)
Don't you mean, make another account? Look, I know who you are (and not Mentifex), and your Wikipedia account could be restricted on the basis of your anonymous behaviour, which is shaping up as disruptive, an actionable ground. (See WP:disrupt, and also about "good hand, bad hand" at WP:sock. This is an addition to your continuing and unuseful profanity and general uncivility – see WP:civil. Just like your comments on Facebook, eh?)
As to these continuing reverts – the topic of this section – well, you seem to have an odd idea of just what "free" means. It does not mean "anything goes", although you seem to think that "anything you can get away with" (because you think it can't be traced back to you) is okay. Sorry, not so. There are expectations and standards (see WP:etiquette, although I am beginning to feel you are quite oblvious to all this). And the WP expectation where editors differ is to discuss the matter. Which you are not doing. What you are doing – these repeated deletions of relevant material – is disruptive editing. How about not doing that any more? Or is it necessary to refer this for administrative action? - J. Johnson (JJ) (talk) 23:19, 30 October 2009 (UTC)

Don't you mean, make another account? - not going to have a discussion with someone who assumes, based on no evidence, that I am Mentifex. You should read WP:AGF. (talk) 08:24, 31 October 2009 (UTC)

You have repeatedly reverted my edits based not on its content, but because you suspect I am Mentifex. This is disruptive editing. Please cut the crap. (talk)

Re comments in the section 2 headings above, the argument that "the mentifex stuff is relevant" is in reality you twisting your paranoid fantasies into a non-reason relating to content. It has absolutely no relevance whatsoever. It's not even related to the topic of the article, and the sentence "shows how complex the issue of trust is" pretty much qualifies as WP:OR.

BTW, nice abusing of the rules of wikipedia to satisfy your ego, chump. I'm not going to pretend to be nice to some idiot who keeps on reverting useful content. WP:IAR. (talk) 08:47, 31 October 2009 (UTC)

Your statements are so contra-factual that I begin to suspect you are quite delusional. (And I begin to the doubt that you even can engage in civil discussion.) But on the prospect that I can settle what seems to be bugging you most I will state: I am NOT reverting your deletions because I suspect you are Mentifex, for the simple reason I do NOT entertain any such suspiscion. I know you are NOT Mentifex. (Language alone is a distinguishing characteristic.) You are ... well, I don't want to run afoul of the anti-outing policy, so just check your talk page for confirmation.
The reason I keep reverting your deletions is quite simple: because you are not adding anything. As to the edit war this has turned into, well, the first recommendation is to discuss the issue. Which are not doing. And attacking me, as you are doing, is expressly forbidden, as you could have seen if you had actually read WP:AGF. But trying to engage with you seems to be futile. - J. Johnson (JJ) (talk) 21:55, 31 October 2009 (UTC)

Actions speak louder than words, whatever essays you post on the talk page. But look, you say:

The key issue here is whether Mentifex (and/or anonymous editors) can add laudatory (but untrue) material, or remove (censor!!) material that is critical.

Which part of what I added is untrue? I added a fricking primary source ffs, the creator of Advogato addressing some of its criticism. Also, under what possible standard can you call Mentifex "critical" to trust metrics???? (talk) 10:23, 3 November 2009 (UTC)

Possibly I am overly optimistic, but possibly you have decided to be civil? And are dropping the bogus assumptions and name-calling? If so, good. Let's have a discussion.
Please note that I am not against you adding the bit about the video clip. (I think there are problems with it, but that's a different discussion.) What I am against is your deletions of the other material regarding Mentifex. The merits of that material are also a separate discussion (which I am willing to engage in), but for the moment let's stick to one point: your combination addition/deletion edit. If you want to add something, fine, do it. (Or even better, post it on this page, and I will be glad to help you improve it.) But add it, don't just take the simple path and do the revert that deletes the other stuff.
And if you want to have a discussion of whether the activities of Mentifex amount to criticism of Advogato's trust metric, fine, but let's try to keep the different issues disentangled. Perhaps that would be best done in a new section. - J. Johnson (JJ) (talk) 20:57, 3 November 2009 (UTC)