User talk:CBM

m:User:CBM	Please leave new comments at the bottom of the page, using the "new section" button at the top of the page. I will respond on this page unless you request otherwise.	Archives 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Maximum spacing estimation

Hello. I recently went back to the A-class discussion here Wikipedia:WikiProject Mathematics/A-class rating/Maximum spacing estimation and tried to implement the various suggestions. As User:C S is semi-retired, may I trouble you to give it a quick glance to let me know if you think it is ready for another A-class review? -- Avi (talk) 07:26, 27 July 2009 (UTC)

I am afraid the A-class process is somewhat defunct at the moment; once the academic year has started in September I want to propose something more lightweight. I would be glad to read through the article and the previous review and let you know if I see any issues in the present version. Please give me a day or two to get back to you. — Carl (CBM · talk) 13:23, 27 July 2009 (UTC)

I'd very much appreciate that; thanks! -- Avi (talk) 13:29, 27 July 2009 (UTC)

Thank you very much. I gave some specific responses at Talk:Maximum spacing estimation. Are you of the belief it does not pay to open an A-class conversation because outside of yourself no one will answer? Do you think that going to FAC directly would be preferable? I'd rather get the imprimatur of the math experts here before going on FAC, personally, unless the process is in indefinite hiatus. Thank you again! -- Avi (talk) 02:13, 29 July 2009 (UTC)

I think that recent nominations for A-class review did not get the sort of thorough review we had in mind when the system was put in place; this makes everyone reluctant to close the discussion and say the article is promoted, so it drags out indefinitely. However, you could always open the review with the expectation that it might not go anywhere. — Carl (CBM · talk) 03:04, 30 July 2009 (UTC)

I guess the problem is that in our project we simply don't have the critical mass for a review system that spreads over individual pages. Not enough people watchlist them. The German Wikipedia has the same problem for FA and GA reviews, so they have one page for all FA reviews and one for all GA reviews. They are soon going to vote on putting even FA and GA together on a single page, along with an open-ended review system that is not sufficiently active. If we had a single page for reviewing the classification of maths articles, I think it would soon be watchlisted by enough project members and therefore get a lot more activity. But it may be best to start such a system in term time to give it a chance to prosper from the beginning. Hans Adler 18:01, 30 July 2009 (UTC)

Well, we do have Wikipedia:WikiProject Mathematics/A-class rating that can be watchlisted. -- Avi (talk) 20:46, 30 July 2009 (UTC)

1) I wasn't even aware it exists, even though it's on my watchlist. 2) It's not the kind of page that appears on your watchlist at least once every 30 minutes. And if it did, it would still not make you notice if an article on a topic you don't care much about has issues related to a style or structure problem you care a lot about. Hans Adler 21:45, 30 July 2009 (UTC)

Here are some thoughts about the process.

• The process started off well, with 7 nominations and 4 promotions in 2007. But then there was only one nomination in 2008, which died before it could really come to closure. There have been 3 nominations in 2009 but 2 were just reviews of articles that are already A-class, and the third (maximum spacing estimation) also petered out.
• I only count about 10 editors in the 2009 archive, with a lot of overlap. I think this is the more serious problem. The question is how to get more editors involved.

My take on this is that the process is too heavyweight for the amount of editing time we can put into it. I don't see much point in having an A-class rating process that makes it a practical impossibility to promote an article.

I have not thought too long about details, but in general terms I think it might be better if we aim somewhere between GA and FA in terms of the rigor of the review. Here is one option I have thought about:

• The discussion is conducted on the talk page of the article, with an announcement posted at WT:WPM. Any three editors agreeing is enough for the article to be promoted to A-class, provided there are no significant objections to the promotion.

I don't know if that's quite the right change, but it shows the sort of thing I am thinking about. — Carl (CBM · talk) 21:51, 30 July 2009 (UTC)

Sounds reasonable. Per the A-class requirements, they are not as stringent as FA, and, interestingly, they are somewhat different than GA (A-class articles that have passed Ga are supposed to be nominated for FA - GA relates more to polish, A relates more to math content). I think I will open an A discussion again now that a few people have chimed in here, and we'll see what happens. If nothing happens after a week or so, I guess it should be closed and I'll consider running the FA gauntlet. Thanks again, Carl, Hans, and everyone, for your suggestions. -- Avi (talk) 15:25, 31 July 2009 (UTC)

• It would help to notify WT:WPM; I didn't know that this was going on. Part of the problem may be that few of us have serious knowledge of any given area of mathematics; we all know our own, and are reluctant to offer technical criticism outside it. Septentrionalis PMAnderson 20:02, 4 August 2009 (UTC)

For the record…

…As you kindly gave advice earlier, I would like to let you know that I have reopened Wikipedia:WikiProject Mathematics/A-class rating/Maximum spacing estimation. Thank you. -- Avi (talk) 15:38, 31 July 2009 (UTC)

Version 0.7

Hi CBM, if you're too busy to go through the 700 lines of output from Wizzy's script, can you email it to me so I can process it for Emmanuel? I think this is the last item we need to do before publication, assuming that the index is now fixed. Thanks! Walkerma (talk) 17:49, 31 July 2009 (UTC)

I have been slowly working on it this week, and finished this afternoon. The following revision IDs should be used instead of the ones selected before. Of course I might have missed something – manual review is only so good.

I'm going to be traveling on and off the next two weeks. I will have good internet access during that time, but not access to my home computer to do release work.

— Carl (CBM · talk) 19:21, 31 July 2009 (UTC)

• Alexandria,_Louisiana 257071494
• Barnacle 253921506
• Carlton_Football_Club 260909556
• Cucumber 260779941
• Cytoskeleton 260340380
• Film_producer 252804496
• Geography_of_the_United_States 257653908
• Ghetto 204335950
• Motocross 260870626
• Pear 259155615
• Swindon 260578526
• Uniform_Resource_Locator 260836131

TI

Hi Carl!

This response is long overdue. I thought I had responded, but apparantly I failed to save it properly.

I think the edit in TI is excellent. It also provided me with some new insight. If something is defined by transfinite recursion, then that something IS wellordered (by "creation order"). This is obvious, but I didn't see it until I read your edit.

Best regards, Johan Nystrom YohanN7 (talk) 16:19, 1 August 2009 (UTC)

PR tweak?

Hi Carl, I have noticed that there tend to be several peer reviews each month that are improperly archived in the same way - the person pastes in the the "This peer review has been closed" text, but does not replace the topic template with {{subst:PR/archive}}. Would it make sense to put in a hidden comment next to the topic with something like "To archive this PR, please replace the topic template with {{subst:PR/archive}}? I will also post this on Geometry Guy's talk page (with a reminder to do the August PR maintenance). I will not do any more SAPRs so the bot can stop linking them.

I also wonder what we should do with the SAPR archives - should some sort of inactive notice pointing to the new toolserver application be added? Thanks, Ruhrfisch _><>°^° 17:17, 1 August 2009 (UTC)

I forgot to "Notify Carl (CBM) that the current month's category needs to be listed, but the category from two month's ago has been stabilized. " Ruhrfisch _><>°^° 11:56, 2 August 2009 (UTC)

Thanks very much - I also noticed that the hidden comment <!--semi-automated peer review placeholder -- please do not edit or delete this comment--> is still in the PR template - removing this would save some space I think. I will also let Geometry guy know about this, thanks, Ruhrfisch _><>°^° 15:08, 5 August 2009 (UTC)

thumb captions

I was wondering if you had seen my response to Wikipedia:Village_pump_(technical)#Image_captions_can_be_too_wide. —TheDJ (talk • contribs) 12:43, 7 August 2009 (UTC)

Do you think that it would be reasonable to add overflow: visible to the appropriate declaration in common.css? — Carl (CBM · talk) 17:42, 7 August 2009 (UTC)

I think it might be. In my opinion, "hiding" content, is always less desirable, even when it potentially might collide with other content or be "visually broken". It is not perfect, but i don't see a "perfect" solution to this problem to be honest. —TheDJ (talk • contribs) 19:41, 7 August 2009 (UTC)

Guess what

Looks like some other well known computability theorist has an account here. See Special:Contributions/Pmt6sbc. His bio is a bit hyperbolic, but I can't complain about the book being added to articles because I like (and probably added it to some articles myself). Pcap ping 13:04, 14 August 2009 (UTC)

I have seen others as well. — Carl (CBM · talk) 11:02, 15 August 2009 (UTC)

User:VeblenBot/C/Wikipedia featured topic candidate main articles

Hey,

Featured topics now follow the FA model of having the nom page at "/archiveX" locations, so I imagine VeblenBot's functionality will need to be updated for this. Having said that, often the link to the nomination doesn't work anyway because at the moment, VeblenBot assumes that the topic's name matches the name of the main article, which often isn't the case - rst20xx (talk) 16:15, 15 August 2009 (UTC)

VeblenBot and policy guideline tag changes

Does VeblenBot still keep a cache of which pages are tagged {{policy}} or {{guideline}}? I looked for contribs to WP:VPP but didn't see any recent ones, last one was 25 October 2008. Hiding T 20:33, 18 August 2009 (UTC)

At some point it was disabled (probably mediawiki changed and I never fixed the script to follow). I fixed the script this evening and it should run once a day now. — Carl (CBM · talk) 00:02, 19 August 2009 (UTC)

Blimey, thank you. Hiding T 09:25, 19 August 2009 (UTC)

User talk:Brandy15

Why did you remove my comments on User talk:Brandy15? Who then was a gentleman? (talk) 19:41, 20 August 2009 (UTC)

I apologize if I offended you. Here is the analysis I made of the situation:

1. The editor was extremely inexperienced, having made fewer than 10 edits
2. The editor created an apparent test page, but the title of that page is the name given on the editor's user page, so perhaps it is an attempt to start a biography
3. The page was tagged for speedy deletion, but I found it due to it being listed as a manual of style page
4. It is a well known that we tend to overwhelm new users with multiple lengthy messages, full of dozens of links. The usual response is TL:DR.
5. You had substituted a CSD notice template on the page, and ClueBot had since then left a different template message.
6. So I deleted the page. At that point, lengthy descriptions of how to put a hangon tag on the page were no longer of much use to the user. Moreover, in light of #4, they were likely to just be confusing.
7. So I removed the instructions on how to place a hangon tag, pointed out the page had been deleted, and also pointed out that we discourage users from making their own biographies. I did this in a brief, clear way, so that the user could very quickly see what they need to do, which is avoid making test pages that appear to be autobiographies.

— Carl (CBM · talk) 20:34, 20 August 2009 (UTC)

Thanks for adding the example in Kleene's first recursion theorem

It made the article a lot more accessible than what I was suggesting! Pcap ping 03:11, 22 August 2009 (UTC)

By the way, you might know

Do you think there's some analogy/relationship between strictly non-standard fixed point combinators and Kleene's 2nd recursion theorem? Pcap ping 03:21, 22 August 2009 (UTC)

I don't know; I am not very fluent in λ-calculus. — Carl (CBM · talk) 04:31, 22 August 2009 (UTC)

History of Logic

Hi I am starting work on a complete revision of the above article, with the aim of making it a Featured Article. It is strange that such an important subject has been so neglected in Wikipedia. The current article covers the history up to the pre-modern period relatively well (though it needs some tidying). The parts on modern logic are abysmal however.

I have begun work here. Currently it is a collection of sources (Bochenski mostly - Kneale & Kneale are somewhat uneven on the modern period) which I will develop into an article when I am sure about the balance.

All comments appreciated - on the talk page please. Logicist (talk) 11:33, 22 August 2009 (UTC)

Algorithm footnotes

I know you're a fan of in-line cites, and so I used to be. But here's my reasoning for converting to "footnote style": Some of them have become excessively complicated because, to do the job well (and to be of best help to the reader), I have to cite not only the author+year but also the book from where I took the quote e.g. (Rosser 1939 in Davis 1965:225). It becomes even uglier when I have to cite the author of a commentary that appears before a paper reproduced in a compilation, e.g. here's one that I should amend: (van Heijenoort's commentary on Frege's 1879 Begriffsschirft in van Heijenoort 1967:1). (This gets weird because sometimes Willard Quine is writing commentary, so it's not a foregone conclusion that van H is doing it. This also happens in Davis 1965 when Davis is writing commentary). And then there's the problem of how to deal with a citation that includes parenthetic information such as the ones on the CASE and IF-THEN-ELSE; what I truly don't like is having to bop back and forth between articles . . . there should be a reasonable narrative flow.

I was persuaded to abandon my preferences for in-line cites during a dialog, and an experiment, with an editor re David Hilbert; he wanted citations, I wanted inline cites. I tried it my way (sounds like a bad song) and it was a mess -- there were zillions of citations, and they were annoying to dodge around -- so I converted it over it footnotes. But I admit that having to click on the footnote and go to the bottom is annoying too, and it makes fact-checking a lot harder (which is what I'm going to do next, no matter what the outcome of our dialog is here . . ..).

Converting "algorithm" back would be a piece of cake (I did the whole job in just a few minutes). But just eliminating the "ref" commands will now result in somewhat longer cites. Lemme know your thoughts . . . (if you want to convert them back 'tis okay by me). Bill Wvbailey (talk) 15:14, 22 August 2009 (UTC)

With one exception (#16 Kowalski 1979) I put every one of the citation in the article. I've fact-checked all but those in the history section, with a few emendations. Bill Wvbailey (talk) 17:47, 22 August 2009 (UTC)

Useless template ... or not

Not all articles, and I dare say most articles in theoretical computer science on this wiki are not being watched by some expert. Having an explicit tag that describes some issue with the article can sometimes attract contributors to fix it. It's nice when some "man in the sky" shows up here or here. I understand your perspective too, that for articles which are being actively maintained, to use software engineering term, like Kleene's recursion theorem, tags do more harm than good. Unfortunately the maintentance level/status is not easy to ascertain. I've fixed quite a few errors, most of them small, but some less so in the past few days in articles like kind (type theory), fixed point combinator, system F, pure type system (which I think still has an error). Pcap ping 00:03, 24 August 2009 (UTC)

Wikipedia:Use common sense no longer marked as a guideline

Wikipedia:Use common sense (edit | talk | history | links | watch | logs) has been edited so that it is no longer marked as a guideline. It was previously marked as a guideline. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:00, 24 August 2009 (UTC)

Offer to mediate in the Office Open XML article

I noticed your offer to mediate on the Office Open XML article. Thank you very much for that offer as attempt to get coments from the wikiproject and through the request for comments proces have failed so far. I would therefore like to take you up on that offer of mediation on the Office Open XML article. Especially the current edit conflict between me and user:Scientus should be resolved. hAl (talk) 20:13, 24 August 2009 (UTC)

VeblenBot/SPERtable updates have stopped

In case you aren't already aware... Celestra (talk) 21:16, 24 August 2009 (UTC)

There were grave problems with the toolserver today; see [1]. The bot runs automatically, and it seems that after they recreated its files from a backup, it has started again. — Carl (CBM · talk) 23:39, 24 August 2009 (UTC)

Computability logic

I'm not sure exactly how to categorize this. What I've done is probably better than how it was before I got to it, but you sure have a better idea of what that is about; the article is kidna vague and I'm not motivated to read any of the references. The reference flood doesn't quite help here. Pcap ping 18:50, 28 August 2009 (UTC)

The categories look reasonable. Many of the references are to mainstream, respectable mathematical logic journals, so I think the issue is just that the article here is written in a vague, somewhat promotional manner. — Carl (CBM · talk) 23:41, 30 August 2009 (UTC)

The finitary Hilbert and primitive recursion, a quote

I stumbled on this in van Heijenoort, thought you might like to see it: In his 1925 On the Infinite Hilbert is discussing the notion of recursion:

"Clearly, the elementary means that we have at our disposal for forming functions are substitution (that is, replacement of an argument by a new variable or function) and recursion (according to the schema of the derivation of the function value for n+1 from that for n). ¶ One might think that these two processes, substitution and recursion, would have to be supplemented with other elementary methods of definition . . . ¶ It turns out, however, that any such definition can be represented as a special case of the use of substitutions and recursions. The method of search for the recursions required is in essence equivalent to that reflection by which one recognizes that the procedure used for the given definition is finitary." (boldface added Hilbert (1925) On the Infinite in van Heijenoort 1967:385-386)

But then, in the following paragraphs Hilbert discusses other "recursions [that] would not be ordinary, stepwise ones . . . that is, a recursion on different variables at once" and ends with the hypothesis that "the function φ_a(a, a) is an instance of a function, of the number-theoretic variable a that cannot be defined by substitutions and ordinary, step-wise recursions alone if we admit only number-theoretic variables.⁹ (⁹ This assertion was proved by W. Ackermann [ [1928] ])." (loc. cit. page 386).

My interpretation of this is that you were correct, and Hilbert himself admits to it. Bill Wvbailey (talk) 19:55, 30 August 2009 (UTC)

There's even more later: "Or, to express ourselves with greater precision and more in the spirit of our finitist attitude, if by adducing a higher recursion or a corresponding variable-type [etc]" (p. 391), and "The final result then is: nowhere is the infinite realized; it is neither present in nature nor admissible as a foundation in our rational thinking -- a remarkable harmony between being and thought. We gain a conviction that runs counter to the earlier endeavors of Frege and Dedekind . . . certain intuitive conceptions and insights are indispensable; logic alone does not suffice. The right to operate with the infinite can be secured only by means of the finite. ¶ The role that remains to the infinite is, rather, merely that of an idea . . . through which the concrete is completed as to form a totality." (p. 392)

This is a surprise to me; I didn't realize that Hilbert considered himself to be a finitist. van H. says that "After 1927 Hilbert himself laid the problem aside. The scale of variable-types has not been further investigated" (p. 368). Bill Wvbailey (talk) 20:09, 30 August 2009 (UTC)

One executive summary of Hilbert's program is: find finitistic proofs that the infinitary methods of mathematics are consistent. Our article on Hilbert's program is, unfortunately, very brief. In the context of finitism, the benefit of primitive recursive functions is that it is extremely easy to prove that each one is actually a total function. This is in contrast to total recursive functions, some of which may be very hard to prove total.

For example, consider the function that does the following:

On input n, check whether n itself is the Gödel number of a proof of 0=1 from the axioms of ZFC. If it is not, return 0. Otherwise, go into an infinite loop.

Because ZFC is consistent, this definition yields a total computable function. However, because of Gödel's second incompleteness theorem, ZFC cannot prove that this is a definition of a total function. Such issues cannot arise with primitive recursive functions. — Carl (CBM · talk) 23:37, 30 August 2009 (UTC)

Wikipedia:Wikipedia talk:Naming conflict has been marked as a guideline

Wikipedia:Wikipedia talk:Naming conflict (edit | talk | history | links | watch | logs) has recently been edited to mark it as a guideline. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:00, 31 August 2009 (UTC)

Comparing various foundations of mathematics with bar charts!

You're probably going to find this paper amusing, if not interesting.

Freek Wiedijk, Is ZF a hack?: Comparing the complexity of some (formalist interpretations of) foundational systems for mathematics, Journal of Applied Logic, Volume 4, Issue 4, December 2006, Pages 622-645, doi:10.1016/j.jal.2005.10.011

Pcap ping 14:53, 1 September 2009 (UTC)

Somebody did point out an analogy with obfuscated C programs as not being that useful at the end of the paper. Probably a better analogy would have been Tarski's quest for minimal axiomatization, e.g. his axiom of an abelian group is not what you'd think of simple, even though it's the shortest by an objective metric [2] (can't seem to find it in an article here). Perhaps Robbins algebra would be a really good example why the simplest axiomatization in some metric isn't the most convenient to work with. It took some 60 years and an automated theorem prover to even show that it is a Boolean algebra. Pcap ping 15:14, 1 September 2009 (UTC)

Meh, that article (pdf) is kind of lame. I do remember some other article arguing that ZFC was close to optimal in some sense. But the guy who wrote that article Pcap cites claims to be a platonist, so he ought to know that "true" logic is infinitary... ;-) 67.122.211.205 (talk) 22:47, 3 September 2009 (UTC)

Wikipedia:Anarchism referencing guidelines has been marked as a guideline

Wikipedia:Anarchism referencing guidelines (edit | talk | history | links | watch | logs) has recently been edited to mark it as a guideline. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:00, 2 September 2009 (UTC)

Wikipedia:Manual of Style (France & French-related) has been marked as part of the Manual of Style

Wikipedia:Manual of Style (France & French-related) (edit | talk | history | links | watch | logs) has recently been edited to mark it as part of the Manual of Style. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:00, 2 September 2009 (UTC)

Wikipedia:Category deletion policy no longer marked as a policy

Wikipedia:Category deletion policy (edit | talk | history | links | watch | logs) has been edited so that it is no longer marked as a policy. It was previously marked as a policy. This is an automated notice of the change (more information). -- VeblenBot (talk) 02:01, 3 September 2009 (UTC)

Re: Move & Bot

I was unaware of the bot at the time: Wikipedia:Articles_for_deletion/List_of_mathematics_articles_(J-L). (if you respond, do it on my talkpage) --Cybercobra (talk) 13:38, 3 September 2009 (UTC)

I'll also add the page moves to my list of bones to pick with the bot author. --Cybercobra (talk) 13:42, 3 September 2009 (UTC)