User talk:Hermitian/archive2

Welcome to my talk page. (archive)

Convergence of Bairstow's method

Hello. Thank you for writing Bairstow's method. However, I have a question. Is the method really quadratically convergent, even in the case of double roots, as the article implies? Do you have a reference for this? I would guess that it is only linearly convergent in this case, just as Newton's method. Cheers, Jitse Niesen (talk) 23:00, 13 January 2006 (UTC)

Bairstow's method loses its quadratic rate of convergence when trying to find a quadratic factor of multiplicity greater than one, although there are a couple of modifications to it which fix this. I do not believe it loses its quadratic convergence to a single quadratic factor based on whether that quadratic factor has distinct or identical roots, as it is searching for the quadratic coefficients, and not for the roots directly. I've never seen a reference which stated the latter specifically, but there is lots of literature on the quadratic multiplicity difficulty, and one would conjecture that if double roots also caused a problem, they would be mentioned in that.Hermitian 02:44, 14 January 2006 (UTC)

I was refering to the case of a quadratic factor of multiplicity greater than one. The article says

"Like Newton's Method, Bairstow's Algorithm will converge quadratically providing the initial guess is close enough to the zero. The algorithm can be quite slow to converge to quadratic factors of multiplicity higher than 1."

I read this as saying that the algorithm will also converge quadratically, albeit slowly, in the case of quadratic factors of multiplicity higher than one. What do you think about reformulating this as

"Bairstow's algorithm inherits the quadratic convergence of Newton's method, except in the case of quadratic factors of multiplicity higher than 1, when convergence can be rather slow.."

or something similar. By the way, do you know about the Durand-Kerner method? It was suggested at User talk:Bo Jacoby that the method described in Jacoby's method is in fact the same as Durand-Kerner. -- Jitse Niesen (talk) 03:47, 14 January 2006 (UTC)

I intended the comment about multiplicity to be an exception to the first line about quadratic convergence. You are correct that this is less than clear. Your suggested phrasing is fine. Durand-Kerner is Newton's method simultaneously on a vector of root guesses, with the derivative of the polynomial approximated by the derivative of what the polynomial would be if its roots were the current set of guesses. This permits a lot of redundant computation to be eliminated, and the algorithm parallelized. If f(x) is a 4th degree polynomial and p,q,r,s are root approximations, the approximated polynomial is (x-p)(x-q)(x-r)(x-s) and its derivative at p is (p-q)(p-r)(p-s) so a Durand-Kerner iteration for p would be p := p - f(p)/((p-q)(p-r)(p-s)) and iterations of identical form for q, r, and s follow immediately. You are correct this looks a lot like the example given in the Jacoby's method article.Hermitian 05:05, 14 January 2006 (UTC)

Thank you for your answer. I changed Bairstow's method as indicated above, and copied this section to Talk:Bairstow's method for future reference. I hope that you continue to contribute to Wikipedia; you clearly know this stuff and it is essential for Wikipedia to attract more experts. Cheers, Jitse Niesen (talk) 15:12, 14 January 2006 (UTC)

Jimbo Wales and Justin Berry

It seems the owner of Wikipedia, Jimbo Wales, has developed a close personal relationship with 19 year old Justin Berry, the former underage camwhore and adult porn webmaster recently profiled in a series of New York Times articles by writer Kurt Eichenwald. At Mr. Berry's request, Jimbo has deleted the entire Justin Berry article, along with its complete revision history, and its primary author has been indefinitely banned from editing Wikipedia for, get this, "pedophile trolling," which is apparently Wikipedia's description of dissent over its censorship. Jimbo has also gutted the talk page, calling any criticism of Mr. Berry a "personal attack," and demanded the article be completely rewritten from scratch, presumedly by persons who can only parrot the mainstream media's spin on the tale.

Jimbo has also claimed that pictures of Mr. Berry appearing in the article are "copyright violations", even though Wikipedia is filled with "fair use" pictures whose copyright is owned by others, including screenshots of copyrighted television programs. Mr. Berry is a news story. I greatly doubt there are any copyright issues with using a picture of him archived forever on a plethora of search engines, and available to anyone, to illustrate the article.

[Personal attack deleted by Theresa Knott | Taste the Korn] .Hermitian 04:47, 10 March 2006 (UTC)

I have removed the personal attack from the above rant. Being upset is no excuse for this kind of attack. You need to cool off. I've blocked you for a month.Hopefully this story will have settled down by then and so will have you. You are still able to edit this talk page so if you want to withdraw the remark you have my blessing to simply delete my "personal attack removed by TK" message and this message too. Feel free to critisize Jimbo or me as much as you like but you cannot make personal attacks like this. Theresa Knott | Taste the Korn 19:45, 11 March 2006 (UTC)

I'm not withdrawing anything. [personal attack removed by Theresa Knott | Taste the Korn Wikipedia is a large cache of conventional wisdom and urban legends which could collectively care less about where the real truth lies. I greatly regret having contributed to it. In Jimbo Wales' own words...

"If a viewpoint is held by an extremely small (or vastly limited) minority, it doesn't belong in Wikipedia (except perhaps in some ancillary article) regardless of whether it's true or not; and regardless of whether you can prove it or not."

So the truth is not an issue. It's what the majority wants to believe that is the "standard" for Wikipedia content. Wikipedia isn't a project to accrue and represent accurate information. It's actually a means for representing the most popular beliefs with blatant disregard for the truth. I'm surprised you even have a section on evolution.Hermitian 20:29, 11 March 2006 (UTC)

I'd also like to add that I think the indefinite banning of Rookiee for "pedophile trolling" because he politely complained about the deletion of the article he contributed to, comes very close to being a bigoted slur against someone based on their professed sexual orientation. What's next, bans for "Nigger Uppityness" or "Kike Craftiness?" I've formed a very negative opinion of Jimbo and the whole Wikipedia project at this point. There isn't a single numbered principle on Jimbo's user page which you people don't regularly violate. This is definitely a case of an organization having an entirely different agenda than the propaganda it writes and disseminates about itself.Hermitian 20:40, 11 March 2006 (UTC)

You have been warned: if you keep up your personal attacks on your talk page, it will be protected by an administrator and you will not be able to edit it. --TML1988 21:13, 11 March 2006 (UTC)

Your talent at invoking "personal attack" seems to exceed your talent at replying intelligently to any of my points.

If you can't take the heat, you shouldn't be editing an encyclopedia. It's pretty clear that Wikipedia's "personal attack" policy is simply a ploy to ban people with whom the administrators disagree, and prevent them from making contributions to Wikipedia, while at the same time preserving the illusion of some level playing field for intellectual discourse.

To illustrate this point, we need look no further than the items removed by Jimbo from the Justin Berry talk page, citing "personal attack", which included every mention of the term "camwhore," which is itself a Wikipedia topic, and the following paragraph. Please feel free to try and figure out who is the victim of the "personal attack" here.

"I'm dissapointed that the majority of this page was deleted, most likely because one of the editors is a self-admitted pedophile. Is it fair to declare that a pedophile can't be neutral on a page that involves pedophilia?. I guarantee that wikipedia wouldn't ban a Jew from editing a page on the holocaust, or ban a black person from editing a page on affirmative action. A pedophile is just as capable of editing with a neutral POV on a subject matter dealing with pedophilia, as a Jew or a black would be on an issue that involves Jews or blacks. Wikipedia claims to be a online encyclopedia that anyone can edit. If that doesn't include pedophiles, than how is that fair?. If we allow certain sexual minorities to be silenced, what's next?, allowing certain racial minorities to be silenced as well. As a frequent reader and occasional editor of wikipedia, I only want to read factual, neutral POV articles, and I don't care about the race, nationality, ethnicity, sex, or sexual orientation of the editor. Even if most people have moral objections to pedophilia, surely we can agree that we all have a right to post on wikipedia as long as we abide by the rules."

You people talk a good game, but what you say and what you do are two entirely different things.Hermitian 21:29, 11 March 2006 (UTC)

The phrase I just removed was a personal attack and is the reason I am protecting this talk page. Theresa Knott | Taste the Korn 07:13, 12 March 2006 (UTC)

In the Same Vein

All attempts to rewrite the Justin Berry article are getting wiped, with admins citing WP:OFFICE, and talking in circles out of both sides of their mouth. Not even information directly from the NY Times story is being allowed in the article. No one will state precisely what information in the original article is believed to be in error, or not correctly cited. At this point, enough people are pissed off, and enough people have been blocked without justification using contrived excuses like "personal attack," that unless an intelligent respectful answer to the questions raised by the people involved in this article is quickly forthcoming from Wikipedia management, this dispute is going to spill over outside of Wikipedia.

We already have a double digit number of people, some of them major contributors, who will never have another nice thing to say about Wikipedia unless this nonsense stops, and Jimbo is forthcoming with something that looks a lot like an explanation bordering on an apology.Hermitian 01:56, 12 March 2006 (UTC)

Oh look. Corax was just blocked for adding verifiable cited facts to the Justin Berry article.

All the Child Sex Hysterics in the Wikipedia administration have drawn a line in the sand, are doing whatever they feel like, refusing to explain, and are blocking anyone who has the balls to call them on it. Does Wikipedia really think it's going to get sued for libel because a teenage prostitute and pornographer who's a federal witness feels bad about his fact-based and thoroughly cited article? If Wikipedia's behavior wasn't so pathetic and dishonest, it would be laughable.

If Jimbo feels so sorry for Justin, why doesn't he just use his vast fortune to open a shelter for wayward teenage male prostitutes, and then we wouldn't all be forced to play along. Hermitian 02:29, 12 March 2006 (UTC)

The way the Wikipedia admin process works on the Justin Berry article now becomes apparent. Actions are pre-emptive. Nothing gets discussed in the designated forums on the Wiki. The admins sit on their Freenode IRC channel, chatting amongst themselves, and send a rotating succession of different people to revert anything that gets added to the article. All questions posed on the talk pages are answered with vague mumbling about unspecified policy, and when anyone is pinned down with a specific question they can't evade, the response is just to ignore it and stonewall.

Some notable recent lies and deception.

...there are no restrictions on the recreation of the article. There is just extra vigilence expected taht everything included by verifiable and meet the rules of the NPOV. Trödel•talk 01:30, 12 March 2006 (UTC)

Or how about this gem...

Everything you added back can be put in the article, but iff it is sourced from a real source (no, the old version of the article does not count). James F. (talk) 01:17, 12 March 2006 (UTC)

Of course, the reality is that stating so much as a single fact from the NY Times article gets instantly reverted, and if the person editing complains, they get a bunch of bullshit, and if they persist, they get blocked for any of a number of contrived reasons.

The Justin Berry talk page, and the talk pages of every admin involved in this ridiculous exercise ends with an unanswered question which was just ignored, or a direct refusal to explain further. Meanwhile, the list of people who will have nothing more to do with Wikipedia grows longer.

Good work, Jimbo. Hermitian 04:09, 12 March 2006 (UTC)