Wikipedia talk:WikiProject Mathematics/Archive/2007/Oct

Bailey Notation
The Bailey Notation article is up for AfD at Articles for deletion/Bailey Notation with concerns over its veracity, any light on this topic from a mathematician will probably be more than welcome and appreciated. Thanks. KTo288 01:43, 3 October 2007 (UTC)


 * This issue seems to have been resolved, with the consensus being delete. Please be aware that further comments on the AfD page may have the effect of biting a newcomer. Jim 03:27, 3 October 2007 (UTC)

What a mess!!
Degrees of freedom (statistics) is a mess. It will take time to fix. The statement that d.f. is always one less than the sample size is nonsense. the "n &minus; r &minus; 1" identity is just as bad.

I'll be back..... Michael Hardy 04:25, 4 October 2007 (UTC)


 * Would it help if we had a central generic decent treatment of degrees of freedom, of which the various current d.f. articles (Degrees of freedom (physics and chemistry), Degrees of freedom (engineering) and Degrees of freedom (statistics)) are then "spinouts". Currently Degrees of freedom (physics and chemistry) contains the statement: Note that "degrees of freedom" has a different meaning in the context of engineering and machines. I think, though, that it is essentially the same meaning, but applied in a different context. For Simultaneous equations it would not hurt to be able to refer to this notion either. --Lambiam 06:16, 4 October 2007 (UTC)

Unreferenced articles
Here are some unreferenced mathematical articles that have been tagged since June 2006.
 * Fixed-point lemma for normal functions ✅ Done
 * Free Boolean algebra ✅ Done
 * Geometric analysis ✅ Done

I thought editors might be interested in finding at least one general reference for each article. Thanks.-- Birgitte SB  19:20, 5 October 2007 (UTC)


 * I did the first two, and Geometry Guy did the third one. &mdash; Carl (CBM · talk) 20:09, 5 October 2007 (UTC)
 * Thanks!-- Birgitte SB  21:41, 5 October 2007 (UTC)

"Metric": Content syncronizing
Please see Talk:Metric (mathematics). `'Míkka 17:52, 8 October 2007 (UTC)

Germanium is back
Banned User:Germanium, also known as Archetype, is pretty much a single-issue editor whose repetitive claim is that 1/0 is the theory of everything, or some such nonsense. I cannot decide if he's serious, but it doesn't really matter. Need some eyes on theory of everything and 1/0 and any other contributions of this IP address. --Trovatore 23:33, 8 October 2007 (UTC)
 * Update: The IP has been blocked for three weeks by User:JForget. --Trovatore 23:47, 8 October 2007 (UTC)
 * Thank goodness, he also removes see also sections as part of his pattern. If you see him, report him immediately to WP:AIV.—Cronholm144 02:57, 9 October 2007 (UTC)

Another user to watch for
The username Ademh (talk · contribs) was created today, and immediately the owner started adding a long-winded and completely incorrect section to Gödel's incompleteness theorems. Keep an eye out for edits by this person to other pages, in case the account is not as single-purpose as it seems to be. &mdash; Carl (CBM · talk) 15:21, 9 October 2007 (UTC)

Wikipedia:Requested articles/Mathematics
A use manged to move Requested articles/Mathematics to QZ factorization. I tried to revert the move but its now living at Requested articles/Mathematics (wrong namespace). Could someone with admin privilages please put it back where it belongs. --Salix alba (talk) 15:33, 9 October 2007 (UTC)
 * Done. Kusma (talk) 15:36, 9 October 2007 (UTC)
 * And I deleted QZ factorization, which was basically empty. Kusma (talk) 15:39, 9 October 2007 (UTC)

Right angle reference added
I saw that the article missed references, so I added one (Swedis book, but valid, i guess). I needed to reformulate the definition to match the reference, and english is not my native language. Can someone just check up on it? Paxinum 16:44, 9 October 2007 (UTC)


 * I haven't seen the actual definition in the Lindahl book, but the version given now in the article seems unnecessarily roundabout. As a parody, I offer the following definition:
 * the angle R such that 7R'' is 13/4 turn, 630°.
 * For the purpose of verifiability, a definition from an English-language source is preferable. --Lambiam 16:47, 10 October 2007 (UTC)

Invariants of tensors
This article has been tagged for cleanup since Nov 2005. Could someone knowledgeable look at it and see if there are still issues or if has been fixed sometime in the past two year and the tag was simply left on.-- Birgitte SB  18:21, 10 October 2007 (UTC)
 * Also Maxwell stress tensor is in the same category.-- Birgitte SB  18:26, 10 October 2007 (UTC)
 * Both articles are miserable, but the latter is a stub: there's essentially no content to clean up there, so I have removed the tag. Geometry guy 20:06, 10 October 2007 (UTC)
 * I thought is was weird to tag stubs with cleanup. I am going to leave a note on he template talk about making note of that in under usage.-- Birgitte  SB  20:10, 10 October 2007 (UTC)
 * I can confirm that the articles still needs cleanup, or possibly merging into characteristic polynomial. (For what it's worth, is there any other context in which "tensor" means "rank 2 covarient tensor" without any further clarification?)  &mdash; Arthur Rubin |  (talk) 23:46, 10 October 2007 (UTC)
 * Actually, despite the first typo, those are rank three tensors. But I agree that it should be cleaned up; although an explanatory example is not a bad idea. Septentrionalis PMAnderson 01:55, 11 October 2007 (UTC)

Assessment
I have been told that this project is considering creating an assessment unit. If any of you would like, I have performed the requisite banner changes and category and page creation often enough to be confident in my ability to create functional counterparts here. I am myself not exactly the world's foremost expert on mathematics articles, so I doubt I would be of much use expect in regards to biographies of mathematicians and the like, but would be willing to offer what assistance I could. Also, having a banner which substantially parallels that of the WP:1.0 group would certainly make it easier for important mathematics articles to be included in the various release versions of wikipedia being created and considered. In any event, if anyone would want an assessment unit created, just leave me a note below or one my user page and I should be able to attend to it in no more than a day or so. John Carter 21:40, 9 October 2007 (UTC)


 * By "assessment unit", are you referring to something like this? If so, then WPM already has one, and has had for quite some time! Your help would be most welcome anyway. Geometry guy 16:03, 10 October 2007 (UTC)
 * Yeah, that's it. I didn't say here that I was real bright, and it looks like I proved that in this instance. I guess now the question would be which articles you all believe should be tagged with the banner. Generally, I've seen a few arguments elsewhere that it works best if the project says "any articles within Category X are within the scope of this project". I note the existing "scope" section, while very understandable, makes it hard to decide whether a given article is one you would consider within your scope or not. One way of doing it, particularly with the number of banners out there, would be to tag all the articles within a given category, but make the ones that are specifically mathematical, Automorphic number, for instance, among the higher importance articles to this project. If you could give me a clear idea of the scope, I could at least start tagging the articles, and assessing some of those which I think I might have any competence in. John Carter 16:12, 10 October 2007 (UTC)
 * Some time ago we decided against mass automatic tagging of articles, perfering the human touch instead, allowing articles to be graded and comments added. There are something like 17,000 mathematics articles so rather than tag the whole lot we have tended to concentrated on the more important ones mathematically, which are the ones most in need of improvement. In the large world of mathematics articles Automorphic number ranks pretty low, I'd probably give it a low priority. As for categorys we do have List of mathematics categories which include most of the mathematical categories. That could be a place to start on assessment work. --Salix alba (talk) 16:42, 10 October 2007 (UTC)
 * I'm also against automatic tagging. The list of mathematics articles (now about 17700 of them) is mostly autogenerated from the list of mathematics categories, and contains a large number of articles that are really not worth assessing. At WikiProject Mathematics/Wikipedia 1.0/Importance I suggested the goal of assessing approximately half of the mathematics articles. An assessment drive a few months ago (mostly by Cronholm and myself) made a modicum of progress and the number of assessed articles is now about 3000. We did not, however, touch the List of mathematicians, which I believe accounts for about 3000 of the 17700 articles above, but maybe Oleg Alexandrov has better figures. Given John Carter's comments above, this would seem to me to be a great place to start. Very few mathematician articles are assessed, and many of these articles are rather stubby. Some help with them would be of great benefit to the project. Geometry guy 17:24, 10 October 2007 (UTC)
 * I would not propose automatic tagging in any event. Also, regarding the article I mentioned above, I realize it probably isn't particularly important to this project. The down side, unfortunately, is that it is probably even less important to any other project. Would the members of this project be interested in tagging such articles anyway, given that, despite their relative lack of importance to this project, it is possibly the only project to which they are at all important? Adding a banner in such cases at least establishes a link to a group of people who might be able to assist in the event that there is any sort of difficult or even intractable argument about the content of the article. John Carter 14:27, 11 October 2007 (UTC)

Requested articles
A quick question for those "in the know". I've noticed a lot of cleanup of Requested_articles/Mathematics that consists of removing blue links. Unfortunately, some of those blue links appear simply because someone has created a redirect to another article that mentions the topic in passing. This does not, in my opinion, obviate the need for a separate article on the requested topic. I believe that these redirects are a good temporary solution; it's better that these link to some article where at least some information can be found. But how then do we maintain requests for articles that should be full articles in their own right? Or is Requested_articles/Mathematics not the right place for this? VectorPosse 02:15, 12 October 2007 (UTC)


 * I agree. I seem to recall having created such redirects recently, precisely because they were better than nothing but not as good as an actual article. Michael Hardy 02:18, 12 October 2007 (UTC)

An advanced mathematics FA?
A prominent contributor to the FA process has suggested to me that there should be no obstruction in principle to an article on advanced mathematics reaching featured article status. As there aren't any such articles at the moment, I am interested in following this up, but I need to know if there is any support here for such an initiative. I have in mind trying out Homotopy groups of spheres. I'd like to do this as much to test the premise of advanced math FAs as to promote a particular article, so I am neutral, but curious, about what the outcome might be.

I need to know if there are a few others familiar with the topic who would be willing to contribute to such an FAC. I hope this thread won't generate snipes at FA: I'm interested in whether there is any positive support for this experiment; I am well aware that it might be a waste of time, but even a negative outcome could provide useful information on the current relationship between FA and advanced technical articles. Hey, it might even be fun! Geometry guy 19:19, 7 October 2007 (UTC)


 * I know a reasonable amount about homotopy groups of spheres, but I know absolutely nothing about the FA process. I'd be happy to help in any way that I can. :-) Jim 19:49, 7 October 2007 (UTC)


 * Manifold (FA debate) was the last FA candidate I can recal on an advanced concept. There are some perennial problems with such articles as they are hard for the uninitiated to get their heads around. While the FA did not succeed I think the process was on the whole worthwhile as it lead to one of the best articles on the topic I've seen. I'm sure putting HGoS forward would be a valuable process as long as we are prepared to do a lot of work to address points raised, and work with the process rather than being confrontational. --Salix alba (talk) 20:04, 7 October 2007 (UTC)


 * Absolutely. My point of view is not to invest emotional energy in the outcome, but I am willing to put in a reasonable amount of work to give the article a fair chance. I also want to work with the process to give it a fair chance of being regarded as a positive thing. HGoS is different from Manifold because the latter article is a fairly introductory article to an advanced concept. In contrast, HGoS would not be encyclopedic if it were accessible to every educated reader: it involves fundamentally difficult concepts, which you really need to do a degree in mathematics to understand. I am interested to see how FAC responds to this. I'm glad I can count upon your help! Geometry guy 20:13, 7 October 2007 (UTC)


 * Would it help to also work on the linked articles explaining some of the related concepts? I took a look at Hopf fibration, for instance, and it's very technical; I added a less-technical lead but I think it still needs more help. The homotopy group of spheres article, in comparison, is amazingly readable for such a technical subject. —David Eppstein 21:53, 7 October 2007 (UTC)


 * Another idea might be to promote Integral to FA status. It has stagnated a bit after some substantial improvements due to a collaboration of the month. It is not that advanced, though, as Manifold or homotopy group of spheres. Jakob.scholbach 08:37, 8 October 2007 (UTC)


 * I think it would be great to get Integral (and Derivative) up to FA standard, but my interest here is really on pushing an article that is beyond the reach of a high school student or freshman. Thank you everyone for offers of help so far, and for contributions to the article. I agree that we need to make sure some of the related articles are in reasonable shape as well, and have added to David's edits of Hopf fibration. Geometry guy 22:56, 8 October 2007 (UTC)


 * So, as I understand the point of this experiment, we have a topic that is so advanced that only a tiny percentage of readers — even those with some undergraduate mathematics — will know about it or care about it or ever do the work to understand it. We want to polish the article so well that FA reviewers will love it anyway. If we succeed, we know FA can be done, and we may be inspired to try it again; if we fail, we might not be adequate writers or they might not be amenable reviewers (or both).
 * I reckon we'll benefit from the attempt.
 * This topic, however, presents a messy challenge, because of the state of what we know (and what I understand).
 * Algebraic topology developed homology groups (and their duals, cohomology groups) in part because homotopy groups turned out to be difficult and messy. Since one point of using algebraic methods is to help us better cope with the topology questions, we prefer easy answers! Homology groups of spheres are easy. So, fact one is the dramatically greater complexity when we look at homotopy groups instead. Fact two is that lower homotopy groups are still easy. Fact three is the most exciting, in my view: the existence of stable homotopy groups. But we are still left with complicated "random" cases, not yet computed, without apparent pattern.
 * To explain well, especially to lay readers, I have to really "get" it myself. But I don't really know why the homotopy groups are so difficult, nor what they're telling me. Sure, we can give definitions, and explain various tools that don't work. But I need an intuitive grasp beyond the fundamental group and the 2-sphere. Lacking that, I'm just going through the motions, regurgitating textbooks.
 * Ah well, if we knew all the answers it wouldn't be research. And some of us do think research is stimulating: the challenges, the setbacks, the successes. If we can bring out that drama, readers may take an interest. --KSmrqT 01:04, 9 October 2007 (UTC)


 * For some insight into these homotopy groups, I benefitted from reading John Baez's piece on the subject, which is linked from the article. Geometry guy 08:59, 9 October 2007 (UTC)
 * You mean the one I mentioned on the talk page some months ago? ;-D
 * Baez is a good explainer, but before he wanders off into 2-categories (one of his favorite topics), he admits he doesn't really understand these groups all that well either.
 * Spanier's Algebraic Topology (ISBN 978-0-387-90646-1) ends with a brief section on "Homotopy groups of spheres", pp. 512–517. His first tool is a lemma about the cohomology spectral sequence of a fibration, which gives some feel for the level of mathematical sophistication required of the reader. --KSmrqT 10:27, 10 October 2007 (UTC)

Some comments from recent experience with FA's, starting with Orion (mythology) (thanks again, Jim): The actual normal level of promoted FA's is ''Does it have any glaring flaws? If not, promote''. It is unlikely, and with homotopy groups almost certain, that any reviewer will be competent to evaluate content; what you will get instead is nearly random.

One possibility is that you will get loud and obscure complaints about violations of the Manual of Style. If I were doing this, I would determine exactly what complaints were being made, and decide if they made sense as demands for good English; if not, I would ignore them. I regret that WP:MOS is a gathering ground for opinionated and provincial cranks. Examples available at request, but looking at the talk page will show several efforts to lay down a law for Wikipedia as a whole, some by reformers who don't care about English usage, some by editors who are appealling to their own usage without realizing that the point is one on which British and American English differ. Septentrionalis PMAnderson 02:15, 11 October 2007 (UTC)


 * I'm happy to handle MoS issues. As for reviewers' expertise, well I hope that several editors here would want to comment at such an FAC. In some sense, this has already started. In particular R.e.b. has considerably improved the comprehensiveness of the article. Many thanks to R.e.b., and others, such as KSmrq and Arcfrk, who have made significant contributions since I posted this suggestion. Geometry guy 19:23, 12 October 2007 (UTC)

Logic task force
In the past, many topics in logic had been neglected because they were philosophical logic topics which had been disincluded from the foundations field of WP:MATH. Then there was controversy over which articles would be included after the creation of WP:LOGIC, a project distinct from WP:MATH. WP:PHILO has now reorganized so as to accommodate articles from all the branches of philosophy in its assessment program, similar to the way WP:MATH does it. There is a tag "logic=yes" in the project banner which allows us to organize these articles similar to the way "field=foundations" works for WP:MATH. WP:PHILO has designated WikiProject Logic as its "Logic task force." The Math and Phil. projects keep two separate assessment worklists so as not to pollute each other's assessment with articles that are not relevant to the larger project. The Philosophy banner allows an article to participate in more than one field in the assessment, so there is no problem with articles getting lumped into the inappropriate field.

I think it would be nice if the members of WP:MATH interested in mathematical logic, set theory and foundations of mathematics would officially designate WikiProject Logic as the "Logic task force" of WP:MATH and do all the things within its own organization consistent with that designation. This would mean calling WP:LOGIC a descendant project of WP:MATH jointly with WP:PHILO. For more information about "task force-ization," see WP:REFORM. I guess you could call it the "Foundations task force" or whatever. I think this move would make some of our members more comfortable since the logic project has recently become more closely integrated with the philosophy project. Our goal should be for there to be as small an overlap (of maths rating|field=foundations and philosophy|logic=yes tagged articles) as is appropriate.

Oleg, is there a way to produce an assessment table for the math foundations field that appears in a way similar to the way the philosophy logic task force assessment works? I would like to display them next to each other at WP:LOGIC. Be well, Greg Bard 21:49, 11 October 2007 (UTC)


 * What benefit is there to making a "task force" for logic? What outstanding problems is it meant to solve? &mdash; Carl (CBM · talk) 00:16, 12 October 2007 (UTC)
 * Generally, task forces are created to allow those editors who have a specific interest or aptitude in that specific subject to have a place to give focused attention to that subject. As I'm not personally sure what exactly you mean by the word "outstanding", I'm not sure what other reply is possible. John Carter 00:36, 12 October 2007 (UTC)
 * I just wondered whether there is actually a problem with the current setup (without task forces). As far as I can tell, the current organization of the math project is working well, so I don't see why it would be necessary to split out a separate logic task force. &mdash; Carl (CBM · talk) 15:52, 12 October 2007 (UTC)
 * You may well be right that there may not be any clearly obvious existing problems. The proposal of the joint task force seems to me to be primarily a way to ensure that articles which have relevance to both philosophy and mathematics can be recognized as such with the least possible complications and amount of "bannering" and greatest degree of cooperation in helping improve the articles. John Carter 15:59, 12 October 2007 (UTC)

Mathematics of bookmaking
The article titled mathematics of bookmaking is being worked on by various people who know that bookmaking is obviously a notable subject but are uncertain whether this stuff known under the name of "mathematics" (or something like that) is notable enough to get mentioned in an encyclopedia.

Perhaps some of those here can help that page. Michael Hardy 16:46, 14 October 2007 (UTC)

Linesearch vs. Line search
A discussion is going on in Talk:Linesearch about whether the article should be named linesearch or line search. I would appreciate any and all feedback. --Zvika 11:14, 15 October 2007 (UTC)

Entropy (disambiguation)
I'd be grateful if people could have a look at the Entropy (disambiguation) page, where I'm in a dispute.

It seems to me that that this previous version was a lot more helpful than the current edited one for users trying to find their way to the article they need.

In particular the older one
 * Helpfully grouped articles around the main two scientific meanings -- information entropy and thermodynamic entropy
 * Helpfully also contained links to the different Entropy categories

The latest edit -- removing Introduction to entropy from the page completely -- seems to me particularly user-unhelpful, verging on WP:POINT.

But I'd be grateful for third party input on this, as maybe I'm too close to what's been edited before. Jheald 16:23, 15 October 2007 (UTC)


 * Disambiguation pages exist to help people out if a link isn't precise enough, or if they search for too broad a term. They do not exist to provide overviews of topics, nor are they intended to be site maps for keywords (as the list of categories the previous revision included would seem to assume).
 * Introduction to entropy has its own problems. I removed it for the time being because if it's deserving of its own article at all it should be a sub-article of Entropy, as as such doesn't need its own entry on disambig (compare to, say, the number of links on the Macintosh presented on Mac). Chris Cunningham 21:01, 15 October 2007 (UTC)


 * It seems like this discussion would be better held on Talk:Entropy (disambiguation). I've coppied the above there. --Salix alba (talk) 21:40, 15 October 2007 (UTC)

Panjer Recursion
hi! could you please have a little look at Panjer Recursion. i wrote this article new and i dont know how these mathematical articles here are usually written. maybe i made other format/style/math errors thanks! --Philtime 17:39, 15 October 2007 (UTC)
 * You need to write out ASTIN at least once, with specific information to determine whether it is a peer-reviewed journal. If not, the references might be unsuitable under WP:RS, even if accurate.  I'm not sure I see the result as notable, but, if it's really used in actuarial science, it might be suitable for a Wikipedia article even if not mathematically interesting.  &mdash; Arthur Rubin |  (talk) 18:46, 15 October 2007 (UTC)
 * You need to write out ASTIN at least once, with specific information to determine whether it is a peer-reviewed journal. If not, the references might be unsuitable under WP:RS, even if accurate.  I'm not sure I see the result as notable, but, if it's really used in actuarial science, it might be suitable for a Wikipedia article even if not mathematically interesting.  &mdash; Arthur Rubin |  (talk) 18:46, 15 October 2007 (UTC)

Indenting formulas within list items
Nevermind. This really doesn't matter. Functor salad 19:13, 15 October 2007 (UTC)

this is not very important, but FWIW I think HTML is preferable for this. Example (wikilist workaround first, then HTML) :


 * 1) The finite direct sum of modules is a biproduct: If
 * $$p_k: A_1 \oplus \cdots \oplus A_n \to A_k$$
 * are the canonical projection mappings and
 * $$i_k: A_k \mapsto A_1 \oplus \cdots \oplus A_n $$
 * are the inclusion mappings, then

The finite direct sum of modules is a biproduct: If
 * $$p_k: A_1 \oplus \cdots \oplus A_n \to A_k$$

are the canonical projection mappings and
 * $$i_k: A_k \mapsto A_1 \oplus \cdots \oplus A_n $$

are the inclusion mappings, then 

The second aligns properly and the source is more readable. Functor salad 18:23, 15 October 2007 (UTC)


 * They look identical on my browser. Oh, wait.  I see.  The indentation of the text is what you're referring to.  Never mind. VectorPosse 18:30, 15 October 2007 (UTC)


 * Yes. Sorry if this topic is too insignificant to even write about, but it has bothered me a few times already, so I thought I'd post it. Functor salad 18:35, 15 October 2007 (UTC)
 * Actually now that I look at it, the first version has its merits. But it's probably dependent on the browser settings. Functor salad 18:37, 15 October 2007 (UTC)
 * I know of no happy solution without some HTML mixed in. This is an obstacle that rears its head once in a great while, and it would be nice to have advice for our editor resources.
 * The automatic conversion from wiki markup to HTML rewrites ";" (rarely used) with a "&lt;dt&gt;" definition term tag and ":" with a "&lt;dd&gt;" definition description tag — both inside a "&lt;dl&gt;" definition list; "*" with a "<tt>&lt;li&gt;</tt>" tag inside a "<tt>&lt;ul&gt;</tt>" unordered list; and "<tt>#</tt>" with a "<tt>&lt;li&gt;</tt>" tag inside a "<tt>&lt;ol&gt;</tt>" ordered list.
 * Incidentally, this problem is mentioned in documentation, with HTML as the only valid workaround provided. Dumb.
 * The parser is very sensitive to spaces and newlines. For example, do not put blank lines between list items; observe:
 * First item, followed by blank line


 * Second item, misnumbered
 * The spacing gets messed up even when the lists are unnumbered. For a real thrill, try nested tables. (See Talk:Boolean algebras canonically defined.) --KSmrqT 07:31, 16 October 2007 (UTC)

George W. Whitehead
George W. Whitehead was nominated for speedy deletion just after being created. R.e.b. 23:35, 8 October 2007 (UTC)
 * And apparently, has been speedily deleted already – is there a log page which keeps track of these? Arcfrk 00:34, 9 October 2007 (UTC)
 * the deletion log, not that it says much of use. —David Eppstein 00:46, 9 October 2007 (UTC)
 * Right. You can't even tell how long the article was when it got deleted, or when it had been created and by whom. No, I meant "is there a log page for speedily deleted articles" (similarly to AfD page)? So that at least we learn about their names, if not the content. Arcfrk 03:13, 9 October 2007 (UTC)
 * There's this one; it includes other kinds of deletion as well, but a large fraction are speedies. —David Eppstein 03:33, 9 October 2007 (UTC)
 * It may also be helpful to take a look at the articles in CAT:CSD, to get an idea of what kinds of articles tend to get tagged for speedy deletion and how many of them there are despite admins frequently going through and handling them. —David Eppstein 07:41, 9 October 2007 (UTC)

By the way, in its deleted state, the article consisted solely of the sentence "George William Whitehead is a mathematician working on algebraic topology who invented the J-homomorphism" and a link to his entry in the math genealogy project. If it had said something about his awards and distinctions (member National Academy of Sciences, Fellow AAAS, etc), or mentioned his text "Elements of Homotopy Theory" with some language about the significance of that book or of Whitehead's other works, and if it had also included some references to third-party sources about Whitehead and his works such as the review of that book in Bull. Amer. Math. Soc. 2 (1980), 237-239 or the MIT news office announcement of his death, it would have been much harder to delete. It's not so hard to build up a few details like that before putting an article online, I think. —David Eppstein 03:53, 9 October 2007 (UTC)

I've restored it and told the person who deleted it to use some common sense. Michael Hardy 04:11, 9 October 2007 (UTC)


 * Is that kind of confrontation helpful? Instead of insisting on restoring the version as deleted, why not just recreate it with enough detail to make it obvious that it's not eligible for A7 deletion? Because in its deleted state, I think one had to be familiar with the mathematics literature enough to do the searches to find the information showing that he's notable — it wasn't obvious from the article itself — and I don't think it's reasonable to expect the admins working the speedy deletion detail to have that level of domain-specific knowledge. —David Eppstein 04:16, 9 October 2007 (UTC)


 * Perhaps turning it into a confrontation is not helpful, but your suggestion only furthers the problem by "enabling" bad admin decisions to continue. There was a link to J-homomorphism and a hangon tag.  Plenty of admins would realize this was not a speedy candidate.  The admin goofed and Michael pointed it out.   --Horoball 09:55, 9 October 2007 (UTC)

Let the bureau-speak not dictate us how things are to be done. I agree with Michael. It is not too much to expect people to use a healthy dose of common sense, especially, while carrying out their hasty secret executions. The article was created at 23:11, tagged for speedy deletion at 23:30, the article's creator put the "hang-on" tag at 23:32, but it got deleted anyway at 23:46, contravening not only the common sense but also a number of common policies. I distinctly remember a discussion in this forum where almost identical sequence of events was reported to have occurred (articles being tagged for speedy deletion a few minutes after having been created). If this is a serious issue, then some policy level decision should be made to address it. For example, only experienced administrators to be given the authority to speedily delete articles. Warnings may be given to "trigger-happy" administrators, with their severity increasing if they exercised really poor judgement. Several warnings would lead to stripping them of the power to speedily delete articles. Arcfrk 07:02, 9 October 2007 (UTC)


 * It is not uncommon for stubby articles on topics that are completely unworthy of inclusion in Wikipedia (wannabe actors who haven't been in any movies yet, say) to be created, tagged for speedy deletion, tagged hangon, and deleted anyway. Hangon doesn't mean that the article is immune to speedy deletion, it just means that the admin should take a more careful look, but in this case there was not much for the admin to see in that second look unless he already knew something about the general area and how to search in it. It seems only common sense to me that we write our stubs in such a way that despite their brevity they convey even to outsiders some reason why the subject is notable, so that they don't get wrongly lumped in with the very many truly bad articles that are created and deleted every day. R.e.b's in good company, by the way — there was a bit of a kerfuffle a couple weeks back when one of Wikipedia Founder Jimbo Wales' stubs was treated the same way. —David Eppstein 07:26, 9 October 2007 (UTC)


 * I'm confused by this discussion. Doesn't the fact that Jimbo Wales had his stub speedy deleted mean that there might be a problem?  I mean, I know Jimbo does not personally dictate policy here, but his opinion should mean something, right?  But forget about Jimbo for a moment.  The original Whitehead article---as short as it might have been---mentioned his having invented the J-homomorphism.  The J-homomorphism passes the notability test since its Wikipedia article is well-referenced, whether you know what it is or not.  So shouldn't its inventor be notable by association?  Or at least, if a hangon tag is placed, shouldn't that require more caution than a random hangon tag in the case of the "unknown actor" mentioned above?  In other words, why should random admins get to make "close calls" like that without consulting experts first?  An expert could easily add references once a problem is made known.  VectorPosse 09:06, 9 October 2007 (UTC)


 * Exactly. David Eppstein points out it is "not too hard" to add additional details and create a stub that is immune to speedy deletions, but why should anyone need to do this anyway?  Can't admins speedily deleting articles use some common sense?  It's clear that one line articles on various pop culture topics do survive speedy deletion tagging.  David suggests that math articles should actually meet a higher standard than usual as we must make allowances for ignorance.  I disagree.  It is up to everyone else to educate themselves or at least allow themselves to be educated by those more knowledgable.  The difference in philosophies here is quite simple: be pragmatic about discrimination or fight it.  I believe positive change will only result from the latter.  --Horoball 09:55, 9 October 2007 (UTC)


 * Wikipedia has decided that anyone can get an account, and anyone with an account can create a new article. We conclude that among new pages will be deliberate spam, deliberate vandalism, blind foolishness, honest mistakes, and a few good articles. Only an admin can delete an article, and relatively few editors are admins. We expect admins to know their way around far better than the average user, and to mostly make reasoned and sound decisions.
 * Admins doing new article patrol must be able to work quickly through the volume of rubbish. On the other hand, article creators will not know Wikipedia's peculiar ideas (and demands!) about "notable" and "verifiable", and will rely on everyday experience and common sense.
 * If admins don't know a subject area (and they won't), and if new editors don't know Wikipedia's habits (and they won't), what is sure to happen? Well-intentioned deletions of well-intentioned articles. And the hapless newbie won't know what happened, nor why.
 * Something needs fixing. Shall we say admins should only patrol topics they know? I can't see that working. Shall we say newbies must pass a Wikipedia-awareness course before creating an article? I can't see that happening either.
 * I don't have a fix to recommend. places unrealistic demands on the newbies, while  places unrealistic demands on the patrol admins. David claims nothing is broken; he's wrong. Michael protests that something is broken, but he's wrong about what. Instead of blaming careless newbies and irresponsible admins, how about blaming the system that puts them in conflict? Get people thinking about how they'd like to fix it. --KSmrqT 11:23, 10 October 2007 (UTC)
 * New articles are often uncategorized. It's hard for field experts to find new articles about a certain field. And articles that look non-notable to non-experts shouldn't remain until an expert comes by coincidentally. How about a system where an admin who thinks an article looks like a speedy candidate can easily attract expert attention when wanted? For example through field subcategories of Category:Candidates for speedy deletion. PrimeHunter 12:19, 10 October 2007 (UTC)

For what it's worth, a similar case of an attempt to delete a brand-new and insufficiently-referenced article on a notable academic happened recently in chemistry with Richard H. Holm. Rather than going through speedy, though, it went to an AfD and was easily kept. —David Eppstein 18:43, 21 October 2007 (UTC)

Michel Kervaire
I've created a one-liner at Michel Kervaire as a kind of experiment. Let us leave it alone for a while and see what happens. --Horoball 10:09, 9 October 2007 (UTC)


 * I don't know why you'r all so insistent on making too-short inadequately-referenced articles, but to me the one line in the Kervaire article looks more clearly to be a claim of notability than the similar line in the deleted version of the Whitehead article. It has two different things he did, not one, and there's that "first to" language. —David Eppstein 14:48, 9 October 2007 (UTC)


 * If I ran into that article (Michel Kervaire) without knowing it was a trick, I would immediately delete it under WP:CSD. But rather than making a point by waiting for someone to delete this article, we should just expand it so that it actually asserts notability in a way that the average person could recognize. &mdash; Carl (CBM · talk) 15:24, 9 October 2007 (UTC)


 * Then you don't know what CSD A7 says, which is "An article about a real person, group of people, band, club, company, or web content that does not indicate why its subject is important or significant." How does the Kervaire article NOT indicate why Kervaire is important or significant?  My point wasn't to "wait" until it was deleted, it was to see if it would be deleted.  I'm shocked to see it would have been, by a math admin, no less.  Wow, what a crazy world!  --Horoball 17:31, 9 October 2007 (UTC)
 * The article doesn't state that what Kervaire did is important or notable in any way. I could write a parallel one-sentence article about any mathematician who has ever proved anything.


 * If I didn't delete that article, it would only be because I went out of my way to expand it so that it did assert importance. The article as it stands is one sentence: "Michel Kervaire is a mathematician who was the first to show the existence of topological n-manifolds with no differentiable structure and (with John Milnor) computed the number of exotic spheres in dimensions greater than four."  Why should someone who doesn't know anything about mathematics think this is an assertion of importance?


 * I was too sharp when I said I would delete the article - but I would at least consider it. The only thing that would make me pause is that I have some background knowledge in math. I would not expect a typical admin to have that sort of knowledge.


 * A7 is not about whether the person really is notable, but about whether the article clearly asserts that the person is notable. It's a criterion about the text of the article, and one consequence of the criterion is that it is no longer acceptable on WP to write one-sentence stubs that provide no context about the subject. We can easily do better. &mdash; Carl (CBM · talk) 17:46, 9 October 2007 (UTC)
 * It should be much easier to create properly-referenced stubs than to tackle the problem of changing the WP:CSD criteria, which are battle-hardened and optimized for rapid use. I have read that a thousand articles are deleted every day, to keep up with the cruft. If a change in the rules were granted, they might start asking for mathematicians to join in the new pages patrol, which is a less edifying activity than proving theorems. EdJohnston 20:36, 9 October 2007 (UTC)


 * Given the context of what Wikipedia consistently passes as notable, I find some of the arguments here favouring or understanding the speedy deletion of the Whitehead article quite surprising. Let's start with WP:NOTABLE: the guidelines in the section on Articles not satisfying the notability guidelines stipulates that one shoud (i) look for sources oneself or (ii) ask the author or place a notability / expert needed tag on the article. Only if sources supporting notability are not found, does the section propose merging into an existing article or deletion. Furthermore, the section makes it quite clear that speedy deletion is for extreme cases only, as the proposal process is still reserved for "uncontroversial deletion candidates", others being directed to AfD. Given all of the above, it is quite absolutely clear in my view that the speedy deletion of the Whitehead article of well beyond acceptable.


 * Next to the Kervaire example: the fact that "someone who doesn't know anything about mathematics" may not immediately recognise the notability is neither what WP:NOTABILITY asks for nor what Wikipedia tolerates elsewhere quite normally. Remember, this is the encyclopaedia which has articles describing in loving detail 493 different Pokémon characters (cf. Arceus for the last one on the list...). Now, while the whole Pokémon industry is quite sensibly notable enough for inclusion, how clear is that for all those individual specimens of what the mass entertainment industry manages to create? I would argue that for "someone who doesn't know anything about Pokémon" it is not that clear Arceus is notable. The only sources supporting notability there make the point that Pokémon are a big franchise generating significant sales. Like arguing that a particular mathematician is notable simply because mathematics is. Should we proceed to speedy deletion of Pokémon character articles? [Rhetorical]


 * Geometry Guy has some time ago argued nicely around here that Wikipedia is, rather than a single encyclopaedia, a system of partially overlapping specialist encyclopaedias, where in many cases it makes sense to include articles too specific for a pure general-purpose reference. This is indeed the case for most of the ardvanced maths articles we contribute to. As well as for Arceus.Stca74 12:37, 10 October 2007 (UTC)


 * If WP is a collection of overlapping specialist encyclopedias, it still only has one set of admins. The Pokemon editors don't create Pokemon articles that consist of only one sentence. (The link above is to a long list of characters. These lists are made so that each character does not have its own article.) With people like Whitehead and Kervaire, it would not be hard to assemble a one-paragraph stub that, even if one doesn't understand jargon, shows the subject is an important mathematician.
 * The issue here is that speedy deletion of newly created articles is based only on the literal text of the article, not on any deep understanding of the text. This is a result of the large number of newly created articles that are tagged for speedy deletion each day; the large volume prevents any deep analysis of each article. Rather than complaining about this system (I agree it's not ideal), we need to work within it. That means making sure there is enough text in each new article that it will not be speedy deleted. &mdash; Carl (CBM · talk) 12:59, 10 October 2007 (UTC)


 * Wikipedia has only one set of admins, but a user ignorant of how to do open heart surgery should not add material on how to do open heart surgery. Similarly if an an admin is so ignorant of mathematics that he does not know whether it's an assertion of notability to say that he doesn't know whether proving the existence of n-manifolds with no differentiable structure is or is not an assertion of notability, then he should leave it alone. Michael Hardy 23:29, 17 October 2007 (UTC)


 * I don't understand. How is "X is the first person to do Y" not an assertion of notability? If I did not know anything, I might of course think that making a space trip is like saying Ho hum twiddle fiddle poo pah pay, where did I leave my keys today?, so that the statement that Yuri Gagarin was the first human in space and the first to orbit the Earth is not an assertion of notability and I could speedily delete that article. But then I should probably realize at least that I don't know anything, so that I can't distinguish a genuine assertion of notability from the assertion that someone was the first human being to say Ho hum twiddle fiddle poo pah pay, where did I leave my keys today?. In general, if an article asserts some achievement or unique characteristic, it is imprudent to proceed to delete it on the grounds of CSD A7 unless you are in a position to know that that achievement or characteristic does not amount to anything. Totally ignoring a hangon request, not allowing the author some time to add more, is, additionally, rude. --Lambiam 15:36, 10 October 2007 (UTC)
 * "X is the first person to do Y" is not necessarily an assertion of notability if Y is not itself intrinsic notable. "Bill Anderson was the first person in Anderson, Indiana to wake up on the morning of August 31, 2002." Who cares? While there is a current discussion about the possibility of Inherent notability, that is only a discussion. An assertion of notability would generally be "Jack Kozlowski is a widely-recognized expert on the subject of the sociology of the Greys", for instance. John Carter 15:47, 10 October 2007 (UTC)
 * And that runs into this problem: I'd object to the Kozlowski stub as WP:PEACOCK. It may not be speediable, but I'd certainly consider AfD. Septentrionalis PMAnderson 02:02, 11 October 2007 (UTC)
 * Yes, if someone wants to write new a one-sentence article, with no references, they had better add some explicit language that says "this subject is important". Since we have 2,000,000 articles, chances are high that the subjects most people would recognize as notable by name alone are already created. So new articles (following that reasoning) are likely to be about subjects that the average person doesn't recognize. &mdash; Carl (CBM · talk) 13:06, 19 October 2007 (UTC)

May I gently suggest that this experiment is somewhat pointless: even the most trigger-happy admin is likely to check "What links here" before deleting an article. In this case Michel Kervaire is linked from 5 other mathematical articles. That would be enough to put me off from deleting it, even if I knew nothing about the subject. Furthermore, this discussion is linked, which rather gives the game away. So, how about fleshing out the article into a decent paragraph, adding a stub tag and a source, and moving on? Geometry guy 08:43, 18 October 2007 (UTC)


 * Geometry guy wrote in his edit summary: "This experiment won't work: time to move on." Actually it has.  George Whitehead was also linked from 5 other mathematical articles before it was deleted.  There was even a link from "User:Mathbot/Most wanted redlinks".  So you are incorrect in saying "even the most trigger-happy admin is likely to check "what links here" before deleting an article".  Obviously that admin did not check.  My "experiment" was to see if given a similar situation, and as you just pointed out it is indeed a similar situation, whether it would happen again or if it were just an aberration.  Only one admin (CBM) has stated he would have deleted it on sight.  Everyone else seemed content to just leave it, unreferenced as it is.  Although one difference though is that perhaps I made the "assertion of notability" stronger somehow.  For example, I did say he was the first to do something, and I also wikilinked John Milnor as a collaborator.  --Horoball 06:00, 19 October 2007 (UTC)


 * I don't think it has proved much other than p(chance of admin seeing a new article)!=1. It is however a case of Do not disrupt wikipedia to prove a point. The next step would be to actually expand the article. --Salix alba (talk) 08:34, 19 October 2007 (UTC)


 * Sigh. In what way is it a case of WP:POINT?  Are you discouraging people from creating stubs?  How can you interpret my creation of a much-needed stub as a "disruption"?  I asked people not to touch it a for a while.  That in itself is not a disruption.  Neither is the fact that some people chose not to work on the article immediately (implicitly following my request).  Not everything that tests how Wikipedia is working is a disruption.  --Horoball 10:40, 19 October 2007 (UTC)

Object theory
If anyone here is familiar with Object theory, that article could use some attention. My impression is that it is something in philosophy. It was recently expanded from a one-line stub to something longer, but I don't know that the something longer is actually object theory. &mdash; Carl (CBM · talk) 13:11, 19 October 2007 (UTC)


 * It might be better titled Abstract Object theory. See The Theory of Abstract Objects for something not too disimilar to what our article is discussing. --Salix alba (talk) 15:01, 19 October 2007 (UTC)

Announcement
That template was put on top of this talk page. I'm not sure exactly what its purpose is, but it looks like an announcement, so I'm moving it down here. Eventually it will get archived. If some sort of permanent notice is required, we can work it onto the WP:WPM somewhere. &mdash; Carl (CBM · talk) 13:39, 23 October 2007 (UTC)
 * Click on the middle link. It seems to be another method of navigation, which I've never seen before. Do we need it? Septentrionalis PMAnderson 00:10, 24 October 2007 (UTC)
 * The link to Contents has been in the sidebar on every page since March. PrimeHunter 02:25, 24 October 2007 (UTC)

Germanium again
Need more help at division by zero and any other articles edited by this IP. --Trovatore 22:12, 23 October 2007 (UTC)


 * I blocked him for 72 hours for unacceptable language in an edit summary. Probably not long enough to keep him from causing trouble, though. —David Eppstein 22:33, 23 October 2007 (UTC)

Article assessment - longest unrated articles, javascript rating tool
I have two announcements to help with the article assessment.

1. User:VeblenBot/Unassessed is a list of the 500 longest math articles that have not been assessed (from the list of mathematics articles). Not all of them should be assessed - the list of math articles is more inclusive than our assessments should be. But it's easy enough to page through the list to find a few articles you're interested in that are not assessed. Or find some long articles to work on. I will eventually start to have a blacklist to remove articles that shouldn't be assessed, so that fresh articles are listed.

2. I wrote a Javascript tool to make it easier to assess articles. To use it, you need to put the following code in your monobook.js:
 * importScript('User:AzaToth/morebits.js');
 * importScript('User:CBM/ratemath.js');

The script adds a tab called 'rate math'. Click that tab when you are viewing an article or its talk page, choose the rating, and the right template will be added to the talk page.

&mdash; Carl (CBM · talk) 14:34, 20 October 2007 (UTC)


 * 34k of text on whether zero is an even number... eat you heart out Britannica! Jheald 22:18, 24 October 2007 (UTC)

Waldhausen category
Waldhausen category has a line of TeX that fails to parse. Can anyone fix it? I don't know what the person who wrote it intended. Michael Hardy 19:16, 24 October 2007 (UTC)


 * Well, they're using the "xy" package, also known as "xy-pic", which given the limited support for real TeX we have here, I imagine is a no-no. One would need to TeX that line externally and upload a picture to fix it. Ryan Reich 19:35, 24 October 2007 (UTC)


 * Which leaves the question of what to change it to. Michael Hardy 23:11, 24 October 2007 (UTC)

A notation style issue
At studentized residual, this expression appeared:


 * $$\widehat{\sigma}_{(i)}^2={1 \over n-m-1}\sum_{j = 1}^n \widehat{\varepsilon}_j^2 \quad |\ j\ne i.$$

I changed it to this:


 * $$\widehat{\sigma}_{(i)}^2={1 \over n-m-1}\sum_{\overset{\scriptstyle j = 1}{ \underset{ \scriptstyle j \ne i }{ {} }}}^n \widehat{\varepsilon}_j^2.$$

This took a bit of TeXnical work that probably cannot reasonably by expected of lots of casual users. (The reason why \scriptstyle had to be used in the subscript is that without it the characters there were too small.)

What to people think of the TeXnique and the result? Michael Hardy 23:11, 24 October 2007 (UTC)


 * The result is exactly what I would expect to see in a journal or text. I would use \substack to achieve this in LaTeX. The underset method is a great trick to convince the WP tex system to do something useful. Thanks for pointing it out. &mdash; Carl (CBM · talk) 00:48, 25 October 2007 (UTC)


 * Given our hamstrung system, we do what we must. Some simpler alternatives are
 * $$\widehat{\sigma}_{(i)}^2={1 \over n-m-1}\sum_{ \begin{smallmatrix} j = 1\ldots n \\ j \ne i \end{smallmatrix} } \widehat{\varepsilon}_j^2 $$
 * and
 * $$\widehat{\sigma}_{(i)}^2={1 \over n-m-1}\sum_{\stackrel{\scriptstyle j = 1}{j \ne i}}^n \widehat{\varepsilon}_j^2 .$$
 * I tend to prefer the first (smallmatrix) alternative over the second (stackrel). I strongly prefer full AMSLaTeX over the crazy-quilt available through Wikipedia. --KSmrqT 06:41, 25 October 2007 (UTC)

IP vandalism and mathematicians
Hi, there's been a spate of drive-by teenager vandalism of bios of mathematicians lately... Sofia Kovalevskaya, Augustin Louis Cauchy, Évariste Galois and perhaps others. Y'all may wanna kinda take a look at your articles... Thanks... -- Ling.Nut 05:39, 25 October 2007 (UTC)

Internal consistency
The article titled internal consistency could use some attention, to say the least, from someone who knows mathematical logic. The section on "gaming" was apparently written by someone who thinks no idea existed until it was applied to computer games. Michael Hardy 01:34, 28 October 2007 (UTC)

category:networks
Should category:networks really be a subcategory of category:graph theory? It leads to many things that have nothing to with graphs, like category:bloggers. Arthena(talk) 16:10, 28 October 2007 (UTC)


 * Some of the more graph-theoretic articles in networks, such as average path length, Watts and Strogatz model, polytree, centrality, etc., already list graph theory as a category despite it being a parent of networks. So it seems it wouldn't cause too much damage to remove that link, if some care is taken to go more thoroughly through the articles there and in the subcategories and check that when appropriate they're also listed in some graph-theoretic category. —David Eppstein 16:21, 28 October 2007 (UTC)

MathWorld to be deleted?
At Articles for deletion/MathWorld it is proposed to delete the article titled MathWorld, to which nearly 1500 other articles link. The person who proposed it first proposed it for "speedy deletion" (with no discussion) on the grounds that (he said) it was created only to advertise Wolfram. Given that MathWorld and Wolfram were generally considered far more noteworthy that Wikipedia at the time that article was created, that seems hard to fathom.

Please express opinions on whether it ought to be deleted at Articles for deletion/MathWorld. Michael Hardy 17:12, 28 October 2007 (UTC)


 * Speedy deletion? What a hoot. CRGreathouse (t | c) 18:37, 28 October 2007 (UTC)

Can you help with average ?
In the article average, there's a section about annualization (as used in finance), and some discussion on the talk page. Geometric average is covered in the article on average, but in finance the geometric average is used to average sequential periodic percent returns. I know that if you lose 50% in one period and gain 50% in the next period, the average gain/loss is 0%, but the dollar loss is 25%. Mutual funds use geometric averages to report average returns. I think this might be related to compound interest or rate of return. Can you take a look at the math in these articles and make sure the math is ok? I've looked at some online math articles, and they don't seem to have any problem with averaging percentiles to come up with an average financial gain or loss, but the percentile averages don't seem to match the dollar gains and losses. I know this is basic boring stuff for you, but if you'd take the time to check the wikipedia finance articles, it's much appreciated! --Foggy Morning 01:39, 28 October 2007 (UTC)
 * Took a look at this article and have to say it needs some serious help. No references are cited, several sections need more explanation, and much of the explanation that is provided is written far too technically. Kaldari 19:55, 29 October 2007 (UTC)

Unbalanced data
Of all the Wikipedia articles that have been requested for more than 2 years, the only one directly related to Mathematics is unbalanced data. In an effort to wipe out the last of these missing articles, I'm posting notices to the relevant WikiProjects. Anyone brave enough to write a stub or article for unbalanced data will be awarded a barnstar for patching this long-standing hole in Wikipedia content. Kaldari 16:47, 29 October 2007 (UTC)
 * Actually, probabilistic potential theory is also related to mathematics. Not sure if this warrants an article, though. The term 'probabilistic potential' (as a term for not formally correct probability distributions) is used in some fields, for example Markov network. Samohyl Jan 17:47, 29 October 2007 (UTC)
 * Thanks for the clarification. Unfortunately, it's all over my head :) Kaldari 18:51, 29 October 2007 (UTC)

If I'm feeling ambitious this Wednesday or Thursday, I might essay a brief article on unbalanced data. It's one of those topics about which everyone hears in graduate school (except those in fields other than statistics) but that lots of people never actually learn. I'm more-or-less in the latter group but I can write a definition and something about why it's important and why it involves difficulties not found in balanced data. Michael Hardy 20:50, 29 October 2007 (UTC)


 * Here is a brief discussion from notes at the University of Minnesota that may be helpful. --KSmrqT 21:39, 29 October 2007 (UTC)
 * I'd be cautious with those notes -- I found them somewhat misleading in trying to come up with a good definition of imbalanced data. Not sure to what extent it's due to my limited statistics background (the material is presented very gently, really) versus the author's possibly misleading explanations. — <span style="font:bold 11px Arial;display:inline;border:#000066 1px solid;background-color:#E6E6FA;padding:0 4px 0 4px;">xDanielx T/C 05:48, 30 October 2007 (UTC)