User talk:EmilJ/Archive 4

Latest comment: 11 years ago by Kendall-K1 in topic Thanks!
Archive 1 Archive 2 Archive 3 Archive 4

Alexander Henry Buckley

/ˌælɪɡˈzɑːndə ˈhɛnri ˈbʌkli/ is correct, as you have stated. The IPA system that is used in the Wikipedia assumes that the final r on the ends of words is almost always pronounced and insists that it should be shown as the standard pronuncuation. Not only is it not pronounced in England, but it is not pronounced much anywhere else either. Their reasoning is that it is pronounced as a linking r when the next word begins with a vowel, such as , for example, in Herefordshire_apples. many editors and visitors to the encyclopedia find the reasoning is flawed and there will shortly be an RfC where the community can decide one and for all which system is to be used; I will keep you informed.--Kudpung (talk) 16:45, 28 April 2010 (UTC)

Pardon me for butting in here but, as I've corrected Kudpung about this, I'm concerned he may be inadvertently spreading a bit of a falsehood that he ought to know better about. The justification listed at the top of Wikipedia talk:IPA for English is that the coda r's are there to reflect the contrasts of rhotic dialects (e.g., between kappa and capper). We mark a phonemic contrast that speakers of rhotic dialects still make. If it were just about linking r, we wouldn't put any r's in words like derby with a pre-consonantal rhotic. — Ƶ§œš¹ [aɪm ˈfɹ̠ˤʷɛ̃ɾ̃ˡi] 20:13, 30 April 2010 (UTC)
Yes, I am aware of the rationale for the system in use and I don't see any problem with that, and I'm also aware of the discussion at Wikipedia talk:IPA for English (though I stopped following it in detail some time ago per tldr).—Emil J. 16:00, 3 May 2010 (UTC)

Merging article Bracelet to Necklace (combinatorics)

I'm recommending that the article on Bracelet (combinatorics) be merged into Necklace (combinatorics).

I saw that you had previously edited one (or both) of these article. You're invited to participate in the discussion here: Talk:Necklace (combinatorics).

Thanks, Justin W Smith talk/stalk 15:13, 14 May 2010 (UTC)

Browser

Hi EmilJ: In the article Pythagoras combinations like c/ show up that look like d in my browser, so I replaced such occurrences with c / . You undid that effort, with a comment that I should fix my browser. I am using Firefox, and I'd like some suggestions on how to do that. I have presntly used the settings under Tools/Options/Content/Advanced as follows: Proportional = Serif ; Serif = Times New Roman ; Sans Serif = Arial Unicode MS ; Monocode = Courier New ; Default character encoding = Unicode (UTF-32BE). I've been forced into these choices so that I can see serif fonts and still see some mathematics symbols like the wedge (∧). Can you afford some advice? Brews ohare (talk) 15:17, 20 May 2010 (UTC)

Your list of fonts suggests that you are probably using Windows, which I'm afraid implies that I won't be able to help you much, because I don't. I also use Firefox and it looks fine, which may indicate that the problem is with fonts rather than the browser, but on the other hand it is quite likely that Firefox relies on different font-rendering libraries on different systems. The default Wikipedia font is sans serif, so that's probably the setting to look into. Anyway, WP:Reference desk/Computing may help you more with Windows font issues.—Emil J. 15:33, 20 May 2010 (UTC)
Yes, I am using Windows. Absolutely no choice of fonts (there are scores of them) leads to separation of c/ so it doesn't look like d in Firefox. Using IE there is no problem. Brews ohare (talk) 15:51, 20 May 2010 (UTC)
Well, if the same font looks works in IE and not in Firefox, then it is definitely an issue with Firefox (or rather its font-rendering subsystem). Upgrading Firefox may help, if you are using an older version. Other than that, I don't know what can be done about it (no, I'm not going to suggest "use IE").—Emil J. 16:22, 20 May 2010 (UTC)

[dz]

Can you give an example? In what environments is /ts/ realised as [dz]? I speak Czech, but I can't think of a word where this occurs right now. —what a crazy random happenstance 08:39, 2 June 2010 (UTC)

Before any nonsonorant voiced consonant except /v/, just like voice assimilation works for other consonants. It's somewhat rare because of the historical change [d͡z] > [z] which more or less ensures that it can only occur before morpheme boundary or in words of foreign origin, but here are some examples (btw, http://bara.ujc.cas.cz/psjc/search.php is quite handy for searching for weird phoneme combinations): leckdo [ˈlɛd͡zɡdo] (and similar words, like leckdy), nanicdobrý [ˈnaɲɪd͡zˌdobriː], salcburský [ˈsald͡zburskiː], socdem [ˈsod͡zdɛm], and of course there is any number of cases like noc byla dlouhá [ˈnod͡zbɪlaˈdlo͡uɦaː].—Emil J. 12:46, 3 June 2010 (UTC)
My pronunciation of those swings between [ts] and [dz] depending on the speed of my speech, but I checked with my mum, and she says it your way. :) I only ever studied linguistics from an English perspective, so it's always interesting to find something like this about my ancestral tongue. —what a crazy random happenstance 03:58, 5 June 2010 (UTC)

I have marked you as a reviewer

I have added the "reviewers" property to your user account. This property is related to the Pending changes system that is currently being tried. This system loosens page protection by allowing anonymous users to make "pending" changes which don't become "live" until they're "reviewed". However, logged-in users always see the very latest version of each page with no delay. A good explanation of the system is given in this image. The system is only being used for pages that would otherwise be protected from editing.

If there are "pending" (unreviewed) edits for a page, they will be apparent in a page's history screen; you do not have to go looking for them. There is, however, a list of all articles with changes awaiting review at Special:OldReviewedPages. Because there are so few pages in the trial so far, the latter list is almost always empty. The list of all pages in the pending review system is at Special:StablePages.

To use the system, you can simply edit the page as you normally would, but you should also mark the latest revision as "reviewed" if you have looked at it to ensure it isn't problematic. Edits should generally be accepted if you wouldn't undo them in normal editing: they don't have obvious vandalism, personal attacks, etc. If an edit is problematic, you can fix it by editing or undoing it, just like normal. You are permitted to mark your own changes as reviewed.

The "reviewers" property does not obligate you to do any additional work, and if you like you can simply ignore it. The expectation is that many users will have this property, so that they can review pending revisions in the course of normal editing. However, if you explicitly want to decline the "reviewer" property, you may ask any administrator to remove it for you at any time. — Carl (CBM · talk) 12:55, 18 June 2010 (UTC)

OK, thanks.—Emil J. 13:54, 18 June 2010 (UTC)

Completely ridiculous

Your in-line justification of a revert here was unduly nasty. It is normal practice to put the year of publication of the source in the cite book template, which automatically places that in the footnote tag. I understand your preference for the original date of the paper reprinted, and I changed things to make that happen. However, a calm and reasonable tone suits collaboration better than slap-downs. Brews ohare (talk) 18:30, 30 June 2010 (UTC)

The publication year was all right until you touched it, there was no need to change the source in the first place. In other words, what you present as "make that happen" actually means that you fixed problems introduced by yourself. On a related note, I have no idea why you changed the publication year to 1957, but it is wrong. The paper was only published in 1959. One would thought that it is obvious that proceedings for a conference which only ended in 1958 could not possibly be published in 1957.—Emil J. 18:37, 30 June 2010 (UTC)
Could I get your opinion at Talk:Tarski's axioms#References? — Carl (CBM · talk) 15:05, 1 July 2010 (UTC)

svg

Hi. Thx for optimising src of commons:File:Circle 7 4.svg. Could you desribe whot you have done ? --Adam majewski (talk) 16:12, 2 July 2010 (UTC)

First, I have corrected the width and height attributes: units have to be written immediately after their numeral, with no intervening space. Second, I factored out common presentation attributes of all the circles: that is, I grouped them together in a <g> element styled with the relevant attributes (fill, stroke, stroke-width), then these attributes do not have to be repeated for each circle as they are inherited from their parent (<g>) element. I think that it should be fairly obvious from the source.—Emil J. 12:24, 5 July 2010 (UTC)

Thx. I thought that svg validator will find the first error. I have changed the source code also. Regards. --Adam majewski (talk) 15:42, 5 July 2010 (UTC)

Validators tend to be very strict about XML/SGML syntax rules, but often less so about things like permissible attribute values (because these have to be checked by ad hoc rules, whereas there are plenty of ready-made generic XML parsers available where one just drops the SVG DTD).—Emil J. 15:51, 6 July 2010 (UTC)

number systems

You are right, infinitesimals should not appear among the number systems. In this connection, it would be nice to have a page on whatever replaces the number system in smooth infinitesimal analysis, with its interesting nilsquare infinitesimals. Tkuvho (talk) 16:16, 6 July 2010 (UTC)

inext

Thanks for that, man. It's 1am and I can't sleep so I'm finding random articles to improve. It never occurred to me to check the source code for the page... Ha ha. - Richard Cavell (talk) 14:47, 16 July 2010 (UTC)

You're welcome.—Emil J. 14:47, 16 July 2010 (UTC)

definability of functions

Are you familiar with the following second terminology?

  1. Definability: A function f: AB, where A and B are subsets of the domain of a structure M, is "definable" in M by a first-order formula   in the language of M, if every a in M and every b in M satisfy:   iff a is in A and f(a) = b.
  2. Induction: A function f: AB, where A and B are subsets of the domain of a structure M, is "induced" by a first-order formula   in the language of M, if every a in A and every b in B satisfy:   iff f(a) = b.

If you are unfamiliar with the second terminology ("induction"), then how would you call the second property mentioned above? Additionally, how would you call the following property:

A function f: AB, where A and B are subsets of the domain of a structure M, is "?" by a first-order formula   in the language of M, if every a in A and every b in M satisfy:   iff f(a) = b.

HOOTmag (talk) 20:48, 17 July 2010 (UTC)

"Induced" is a word used a lot in different ways in various contexts, often even informally, and as such it is quite ambiguous. I don't recall seeing it used for the concept you describe, but I guess it could be a sensible name for the property, as long as you define it before use. I don't think the property has any widely recognized name. As a noun, however, "induction" is a very bad choice, as this word universally refers to induction in mathematics (and in logic in particular); if you really need a verbal noun to go with "induced", you'd be better off with "inducing", even though it does not ring so good.—Emil J. 11:24, 19 July 2010 (UTC)
"Inducement" could be better than "inducing", couldn't it? Anyways, I've described two new concepts (i.e. the second one and the third one), and I understand that you're talking about both of them, aren't you?
Regarding my question at the Ref Desk, have you seen my response to you there? I'd like to present there my solution for surjections/bijections, but just after I know that you're not going to give a solution of your own.
HOOTmag (talk) 15:05, 19 July 2010 (UTC)

Discussion of hidden responses on RefDesk

Thought you should know about this discussion. -- Scray (talk) 18:08, 18 July 2010 (UTC)

OK, thanks.—Emil J. 11:14, 19 July 2010 (UTC)

Second-order formulas

Ahoj,

I've built a second-order formula (whose variables range over a set of basic elements as well as of partial functions of basic elements), about which I want to state that: 1. every first-order variable ranges over the set of real numbers, and that: 2. every second-order variable ranges over the set of continuous (real) functions. Which way (of the three below) do you recommend (or see as the most common/elegant way) for fulfilling the requirement mentioned above?

  • To state that the model of the respective structure consists of: 1. the set of real numbers, and: 2. the set of continuous (real) functions (if so then how should I formulate exactly all of that about the model and about the formula, in a technical notation)?
  • To build - for the language of the formula - an interpretaion in which: 1. a special first-order predicate symbol (in that language) should be interpreted as "is a real number", and: 2. a special second-order predicate symbol (in that language) should be interpreted as: "is a continuous (real) function", and then to rebuild the given formula respectively, and then to state that the (rebuilt) formula is in a language interpreted as mentioned above (if so then I should I state all of that about the formula, in a technical notation)?
  • To use some combination of both ways mentioned above.

Thank you in advance. HOOTmag (talk) 08:03, 23 July 2010 (UTC)

The first one sounds reasonable. I don't quite understand what you mean in the second one, but it sounds too complicated for the problem at hand.—Emil J. 17:24, 9 August 2010 (UTC)
Děkuju mnohokrát.
So, can the model of a given second order formula contain second-order elements (e.g. functions of first order elements) as well? If your answer is positive, then how should the model look like, for the reader to know that the first order variable ranges over the set of first order elements only?
For instance, let's take the second order formula:  , that involves no first order variable other than x, and no second order variable other than F. I would like to state that: 1. the first order variable x - ranges over the set of real numbers, and that: 2. the second-order variable F - ranges over the set of continuous (real) functions. The first alternative (you've preferred) is to state that the model of this formula contains both the set of real numbers, and the set of continuous (real) functions. The second alternative is to use the second-order predicate symbol C which shall mean "is a Continuous real function", along with the first-order predicate symbol R which shall mean "is a Real number", then to rebuild the original formula this way:  , and then to state that the rebuilt formula is in a language which includes the predicate symbols R and C, interpreted as above.
After having read this clarification, are you still sure that the first alternative is better for stating that: 1. the first order variable x - ranges over the set of real numbers, and that: 2. the second-order variable F - ranges over the set of continuous (real) functions ? I ask all of that, because (as far as I know) the model of a given second order formula - should contain first order elements only, shouldn't it? If it contains second order elements (e.g. functions of first order elements) as well, then how should the model look like (notationally speaking), for the reader to know that the variable x ranges over the set of first order elements only?
HOOTmag (talk) 20:01, 9 August 2010 (UTC)
I don't see a problem, the model will be a two-sorted structure with one sort for reals, one sort for continuous real functions, an application operation, and whatever other predicates or operations are used in  . This is pretty much a Henkin model for second-order logic (except that it involves functions rather than subsets).—Emil J. 18:36, 10 August 2010 (UTC)
Děkuji vám. I was not aware of Henkin semantics, it's all new to me. Thank you again. So, what technical notation should I use for describing this two sorted structure, including the technical notation for describing the model and the language? Let's assume that the formula involves: identity, an application operation, a universal quantifier, one first order variable (over the reals) and one second order variable (over the continuous real functions), and no other signs. HOOTmag (talk) 20:43, 10 August 2010 (UTC)

comment on CH at reference desk

Hi, Thanks for your comment there. I copied the discussion to my talkpage before it disappears into some hard-to-find archive. I am still trying to understand what you said. Hope I get there :) Tkuvho (talk) 21:05, 7 August 2010 (UTC)

If you could elaborate on that material, I would appreciate it. It might also be of interest to a reader of a page such as continuum hypothesis. Tkuvho (talk) 12:42, 9 August 2010 (UTC)
How much do you know about descriptive set theory? The arithmetical hierarchy (first-order arithmetic) and analytical hierarchy (second-order arithmetic) can be naturally generalized to third-order arithmetic: here you have formulas with three types of variables and quantifiers: one type for natural numbers, one type for sets of natural numbers (which are identified with real numbers, as usual in set theory), and one type for sets of sets of natural numbers (= sets of reals). The formulas, normalized to prenex normal form where all third-order quantifiers are in front, are stratified into a hierarchy where you count the number of blocks of existential or universal third-order quantifiers, in particular,   which I mentioned there denotes formulas consisting of a block of third-order existential quantifiers followed by a formula with only first-order and second-order quantifiers. The claim is that CH is equivalent in ZFC to a  -formula.
First, CH is equivalent to the statement (*) "there exists a linear order on R whose every proper initial segment is countable": on the one hand, CH implies the existence of a well-order of R of type ω1. On the other hand, assume that < is any linear order on R whose every proper initial segment is countable. As with every linear order, R contains a cofinal well-ordered subset X under <. The condition on initial segments ensures that X has type at most ω1, and since R is the union of initial segments (which are all countable) bounded by elements of X, it has cardinality at most  .
What remains is to write (*) as a   formula. Since all three types have definable pairing functions, we can use quantifiers for binary, ternary, etc., relations instead of just subsets. Thus we can start with an existential third-order quantifier for a binary relation < on R. We can say that < is a linear order with a bunch of quantifiers over R (i.e., second-order in our setting) just by stating the usual axioms. Then we state that "for every a in R, the initial segment {x in R|x < a} is countable". The last part is slightly tricky and I'm not going to write it down explicitly, but the point is that we can encode a function NR (= P(N)) by a subset of N × N (and therefore by a real).—Emil J. 17:58, 9 August 2010 (UTC)
Thanks! Tkuvho (talk) 12:37, 10 August 2010 (UTC)

Few things

  1. I've applied rollback to your account. I trust this will be useful when reverting vandalism. Please read the link before using.
  2. Regarding P versus NP problem, the reason I didn't cite the Woeginger source is that it didn't appear to be reliable. I don't know what you think about that?
  3. Small point: not sure if it is a good idea to call other editors ignorant, especially when they are contributing in good faith ...

Best wishes — Martin (MSGJ · talk) 12:52, 12 August 2010 (UTC)

  1. Thanks.
  2. It doesn't quite meet the standard as it is self-published, but I don't think we have to be so fussy here. First, note that I have retained the original newspaper reference (exactly for this reason), so it only serves as an additional source of information. Second, pragmatically speaking, a page like this written by an expert is IMHO more trustworthy than an article by a journalist, and anyway the bulk of it basically consists of a list of trivially verifiable pointers to primary sources, so there is no real question about its correctness. I observe that there are other similarly questionable or even worse references in the article, namely Scott Aaronson's blog and lecture notes, and the Complexity Zoo (I know, WP:OTHERCRAP is not an argument, but still).
  3. OK, maybe I shouldn't have used the word in an edit summary, but if a user edits an article, sees a large boilerplate notice telling him/her not to do something, ignores it, and does exactly what the notice is telling them not to do (without any explanation either in edit summary or on the talk page), then in my book he/she is either ignorant or acting in bad faith.—Emil J. 14:09, 12 August 2010 (UTC)

Thanks

Ta for fixing Wikipedia:Village_pump_(technical)#Broken_Portal so promptly & efficiently. Cheers, Trafford09 (talk) 15:03, 18 August 2010 (UTC)

You are welcome.—Emil J. 15:46, 18 August 2010 (UTC)

Confirmation

This is a confirmation for the simple English WP bureaucrats that I'm requesting an account usurpation there.—Emil J. 17:48, 20 August 2010 (UTC)

I have renamed the EmilJ account on simplewiki. You can now SUL the wiki and join as EmilJ. Let me know if you have issues or need help! fr33kman -simpleWP- 04:16, 21 August 2010 (UTC)

Doplňování zemí uznávajících Kosovo na cs

Dobrý den, když doplňujete země, je dobré je doplnit rovnou i s referencí, protože později se reference někdy hledají už špatně. Stačí aspoň hodit tam odkaz, správný formát a užití správných citačních šablon už pak někdo doladí. Díky moc, Palu (talk) 10:50, 7 September 2010 (UTC)

To je zhola zbytečná ztráta času a prostoru. Země se během pár dní objeví na seznamu Kosovského ministerstva zahraničí, která je jako reference pro celý seznam uvedena nad ním. Zvláštní reference jsou potřeba jen v případech, kdy data MFA nesouhlásí.—Emil J. 10:55, 7 September 2010 (UTC)
Takhle to je ale dost obtížné na ověřování. Palu (talk) 18:42, 17 November 2010 (UTC)
Nerozumím. V čem je složitější ověřit jeden přehledný seznam na oficiálním vládním webu použitelný pro celou tabulku, než ~70 jednotlivých odkazů na náhodně zvolená media, které navíc mají tendenci po pár týdnech někam zmizet? Můžete mi to nějak blíže objasnit?—Emil J. 18:54, 17 November 2010 (UTC)
To máte pravdu, díval jsem se na něco jiného. Ale teď mě ještě napadlo, jestli jde o zdroj v dostatečné míře nezávislý a jestli by to neměly být zdroje spíše od jednotlivých států než od státu Kosovo. Ne že bych přímo říkal nevěřme státu Kosovo, to rozhodně ne, ale přecijen mi to přijde lepší a z hlediska neutrality a přesnosti správnější. Případně tam mít zdroje oba samozřejmě. Palu (talk) 16:13, 19 November 2010 (UTC)

Peer reviewed articles on P vs NP

In your edit comment, you say that there are several articles that are peer reviewed. However, I can find only one - which other you have in mind? All other are published in archive, which is not peer reviewed. If there are other peer reviewed articles, they should be mentioned, but there do not seem to be any (peer review is a filter, but this one slipped through). Why are you saying this is not notable - peer reviewed publication of a major solution to a problem is a significant event, even if it in the end turns out to be error in the paper. It is POV to put more weight to Deolalikar than to a paper that has already passed peer review - by Clay Institute criteria, paper needs to be published in peer reviewed journal first, and second, to survive two years of scrutiny after publication. Deolalikar didn't even publish the paper. Are you arbitrarily assigning importance that ignores what happens in India - IIT is known to produce results such as PRIMES in P, so it is something that should be noted. India has scientific community that counts too. Would you argue that Indian released films in Bollywood should not be mentioned, because only Hollywood counts as wikipedia worthy? If something is ignored by Western media, it does not mean it doesn't exist. Ckplato (talk) —Preceding undated comment added 17:16, 9 September 2010 (UTC).

  • Keep the discussion on Talk:P versus NP problem, it belongs there, not here.
  • For example, Woeginger's #38 is also published in a peer-review journal.
  • I have no idea what are you trying to say with respect to India, since (1) Deolalikar is also Indian, and (2) mentioning more Indian scientists in this cranky context is denigrating India, if anything.
  • Most importantly, immediately stop pushing your view into the article against consensus of other editors, as demonstrated on the talk page. The only outcome you can achieve by that is getting yourself blocked. You have to persuade others first.
Emil J. 18:14, 9 September 2010 (UTC)

Shell sort edit

It might not be a code repository however having several different implementations (because the implementations will vary between languages) is helpful. Not every person using this sort will understand the pseudocode and having it written in a "common language" is beneficial to the overall health of the article.

108.27.89.177 (talk) 14:50, 4 October 2010 (UTC)

Thanks

My paper was updated: http://arxiv.org/abs/1004.1808 Please, see my thanks for your help in Acknowledgments section.--Tim32 (talk) 11:20, 7 October 2010 (UTC)

Aligning superscripts/subscripts

A cleaner way to align superscripts and subscripts is to use the template {{su}}; for example X{{su|p=123|b=456}} produces X123
456
. r.e.b. (talk) 16:41, 21 October 2010 (UTC)

Last time I checked, there were about half dozen such templates, and a brief look at their talk page made it clear that they were broken in one browser or another (only each of them in a different way). Are you really sure that the current incarnation of {{su}} is cross-browser compatible?—Emil J. 16:44, 21 October 2010 (UTC)
As a matter of fact, I can already see that your change to the article is broken in my browser: in "Conservative over RCA_0 for Π1
1
sentences" in the table, there is a line break between Π and 1
1
.—Emil J. 16:49, 21 October 2010 (UTC)

Yes, I have a line break there too which I was wondering how to fix, but this is not a probblem with the template: one just needs to put in some sort of no line break instruction. On the other hand, your hand crafted versions do not always align properly in firefox. So we have a choice between an easy template that is sometimes broken and complicated html that is sometimes broken. The advantage of the template is that it only needs to be fixed once. r.e.b. (talk) 16:59, 21 October 2010 (UTC)

The line break is a problem of the template, because plain X<sup>a</sup><sub>b</sub> does not break. But that is a minor problem. The real problem is whether it actually works reliably in browsers other than Firefox, which you did not demonstrate. The template produces much more complicated code than the simple hack I used. The actual choice is between a complicated template which tries to be clever to align things automatically and may be broken in some browsers (and I mean totally broken, such as making the superscripts display on a line of their own far away from the base text), and a simple and robust CSS which relies on a manually adjustable and somewhat font-dependent shift (and therefore tends to be off a bit). I am perfectly aware that it is an unelegant hack, but the only reason I'm using it is that I have not yet seen a better reliable solution.—Emil J. 17:21, 21 October 2010 (UTC)

E (mathematical constant)

RE: It's standard practice to write such long numbers in groups of 5 digits, not 3.
How so? All I could find on the subject was this, from Decimal mark#Digit grouping:

...digits are usually in groups of three, that is, thousands. The most general name for this delimiter is digit group separator, because thousands are not always the relevant group. For example, in various countries (e.g., China, India, and Japan), there have been traditional conventions of grouping by 2 or 4 digits. These conventions are still observed in some contexts, although the 3-digit group convention is also well known and often used. Making groups of three digits also emphasizes that there is a base 1000 of the numeral system that is being used, which is not always a given (for example, in computer science)...

Perhaps pi and Euler–Mascheroni constant should be reformatted, too.
(: nagualdesign (talk) 21:09, 1 November 2010 (UTC)

The paragraph you quote is irrelevant, it only describes the practice when writing small numbers in everyday language, not mathematical conventions when writing long decimal fractions. The relevant guideline is MOS:NUM#Delimiting (grouping of digits). To give you an idea of the actual usage, a search for 90 digits of pi in groups of 5 gets ~189,000 hits, whereas the same in groups of 3 gets 231 hits (YMMV, but not by several orders of magnitude).—Emil J. 13:04, 2 November 2010 (UTC)

Ahh.. Fair dinkum. :) nagualdesign (talk) 07:59, 3 November 2010 (UTC)

IP, Franzen, and Goedel

Did the IP misunderstand Franzen, or did Franzen misunderstand Goedel? I am referring to your recent edit at Robinson arithmetic. It would be nice to clarify this. Tkuvho (talk) 13:02, 2 December 2010 (UTC)

I have no idea, since I didn't read Franzen. I was going to add a source for the second incompleteness theorem for Q, but I didn't get to do it yet.—Emil J. 13:05, 2 December 2010 (UTC)
I do have Franzen's book. The issue is the same as the one I was talking about on the article talk page: Franzen takes "second incompleteness theorem" to mean the one that assumes the derivability conditions, rather than Pudlak's generalization. — Carl (CBM · talk) 21:11, 2 December 2010 (UTC)

2nd incompleteness theorem

Thanks for fixing my mistake in the article there. I think it would be appropriate to add something about Pudlak's generalization as well, but that section is in need of rewriting and I hope to do that at the same time. — Carl (CBM · talk) 21:05, 2 December 2010 (UTC)

OK. BTW, Pudlák's theorem (the one referring to ordinary provability, not to restricted provability) can be further generalized by using arbitrary relative interpretations instead of cuts, giving this neat formulation: if T is a consistent theory [not necessarily in arithmetical language, and not necessarily extending any given base theory] and τ is a Σ1-formula defining an axiom set for T in N, then T cannot interpret Q + Conτ. The interpretability logic people use this version. Unfortunately they seem to treat it as a sort of folklore, so I'm not sure whether one can find a good source for the statement (for example, here[1] it's simply attributed to Pudlák's 1985 paper).—Emil J. 13:51, 3 December 2010 (UTC)

Everest AFD

Hi EmilJ. I wanted to drop you a note and let you know that I removed your comment from the closed and archived AFD on the Everest mathematical competition. Your comment I do think is apt, but could be more effective on maybe HJ Mitchell's or Fly By Night's talkpages...and adding it after the discussion is archived might give the impression that you had made the observation that the article was not a copyvio but that HJ knew that and deleted it anyhow. Thanks! Syrthiss (talk) 14:11, 7 December 2010 (UTC)

List of states with limited recognition at FLRC

I have nominated List of states with limited recognition for featured list removal here. Please join the discussion on whether this article meets the featured list criteria. Articles are typically reviewed for two weeks; editors may declare to "Keep" or "Delist" the article's featured status. The instructions for the review process are here. Nightw 15:32, 13 December 2010 (UTC)

Kosovo

Maybe you can add Qatar to File:CountriesRecognizingKosovo.svg --Vinie007 19:30, 5 January 2011 (UTC)
It's been there all the time, by accident. Something went wrong with the 15 Nov revert.—Emil J. 19:38, 5 January 2011 (UTC)

Someone erased France on the map. Maybe you can fix that. --GrandpaScott (talk) 11:02, 14 July 2011 (UTC)

first order logic

Hi, Could you comment at Talk:First-order_logic if you get a chance? Tkuvho (talk) 01:37, 10 January 2011 (UTC)

Merging Fraction to Rational number.

FYI. I moved your comments and mine from the Talk:Fraction (mathematics) page to the Talk:Rational number page per the guidelines for merger discussion at Wikipedia:Merging. — Preceding unsigned comment added by Clifsportland (talkcontribs) 20:46, 21 January 2011 (UTC)

Proposal to add "statistics" section to List of states with limited recognition

It has been proposed that a "statistics" section is added to List of states with limited recognition. Please contribute to the discussion at Talk:List of states with limited recognition#Statistics RFC. Alinor (talk) 08:16, 14 March 2011 (UTC)

Principle of bivalence

When you have more time to spend around here, that article could use your attention as well. It has improved, but there are still some issues. Tijfo098 (talk) 19:06, 4 April 2011 (UTC)

saturated model

while you are working on it perhaps something can be done about dumbing down the lede. Tkuvho (talk) 15:51, 13 April 2011 (UTC)

Barnstar

  The Original Barnstar
This barnstar is awarded to everyone who - whatever their opinion - contributed to the discussion about Wikipedia and SOPA. Thank you for being a part of the discussion. Presented by the Wikimedia Foundation.

Damage by Inkscape

Hi EmilJ, I just realized that my updating of File:CountriesRecognizingKosovo.svg damages the file. I have no clue why? Is it Inkscape or the way I handle it? Thanks for any hint you could give me. regards Gugganij (talk) 21:59, 19 March 2012 (UTC)

There is no way to tell Inkscape to colour countries by updating the list of recognizing countries in the CSS header, hence it leaves the file in an inconsistent state (some countries are in the list, some are coloured directly) which hinders further maintenance, so, please, do not use it. You can update the file as follows. Open it in a plain text editor (such as Notepad, Emacs, vi). At the beginning of the file, there is a longish CSS stylesheet. In the stylesheet, after line 120 or so, you'll find a section looking like this:
/*
 *  countries recognizing Kosovo
 */

.ad, /* Andorra */
.af, /* Afghanistan */
.al, /* Albania */
.au, /* Australia */
.at, /* Austria */
.bh, /* Bahrain */
.be, /* Belgium */
.bz, /* Belize */
.bj, /* Benin */
.bg, /* Bulgaria */
.bf, /* Burkina Faso */
.ca, /* Canada */
.cf, /* Central African Republic */
.ci, /* Cote d'Ivoire */
...
Add any new country which recognized Kosovo to this list (in alphabetical order, to ease maintenance), using its two-letter (ISO 3166-1 alpha-2) country code (usually, it’s the same as the name of its top level internet domain). (Unfortunately, some countries require special treatment, as you can see from the fake codes with x’s on the list.) Also, small countries and islands are marked with circles in the map; there is a list of the grey ones near line 25 (after a “but show circles for sovereign countries not visible enough by themselves” comment). If one of these recognizes, it may be necessary to remove it from this list. I guess all this sounds complicated, but it’s actually fairly self-explanatory if you look inside the file.—Emil J. 15:10, 21 March 2012 (UTC)
Thanks a lot. Next time, I hope, you won't need to clean up the mess I left ;-) Gugganij (talk) 17:29, 23 March 2012 (UTC)

Well-founded relation

Some 3 years ago, you had a long discussion on the above page, but somehow, it managed to miss an important point about infinite descending chains. I posted again, at the bottom of the talk page, on this. Perhaps you can clarify. linas (talk) 03:27, 21 April 2012 (UTC)

Never mind, my brain is lying to me. time to go to bed. linas (talk) 03:42, 21 April 2012 (UTC)

Thanks

for helping fix that old merge issue. It was way over my head, so I'm glad someone who knows what they're doing stepped in! cheers. --KarlB (talk) 14:23, 4 July 2012 (UTC)

You’re welcome.—Emil J. 14:24, 4 July 2012 (UTC)

Area of Kosovo wanted

Hello You are the creator of the file File:Kosovo outline.svg. Due to its precise data, one (you?) should be able to tell me its precise areas (in nature in square-kilometers). Can you? Achim1999 (talk) 19:18, 9 July 2012 (UTC)

Dear Achim, I didn’t create the file from scratch, I only redesigned the graphical presentation of the image. I took the data from File:Kosovo location map.svg.—Emil J. 10:25, 10 July 2012 (UTC)

IPA for Macedonian

Further to your comment (It makes no sense to change the key without also changing the transcription in all articles using the IPA-bg or IPA-mk templates.), perhaps this is exactly what needs to be done; all sources describe the Macedonian vowels as being: low central (in Wikipedia: /ä/), mid-front/back (/e̞/, /o̞/), close-front/back (/i/, /u/). The Bulgarian stressed /ɛ/ and /ɔ/ are noticeably lower than the Macedonian /e̞/, /o̞/, but it's /ä/ for both (and definitely not a cardinal low front /a/). --101.112.134.103 (talk) 02:07, 10 August 2012 (UTC)

You should raise these issues at Wikipedia talk:IPA for Bulgarian and Macedonian rather than here, I can’t do anything about it.—Emil J. 10:44, 10 August 2012 (UTC)

Removal of sourced material

It would be helpful for you to expand on your objections to my rendition of Davenport's definition. In particular I do not see that your objection about the principal character is valid. Please discuss at Talk:Dirichlet_character#Definitions_of_primitivity. Deltahedron (talk) 20:15, 30 August 2012 (UTC)

Hello

Do you have knowledge in Model theory? If you do, then please tell me if you think that: every model - uniquely determined (up to isomorphism) by an infinite class of second order axioms, can also be uniquely determined (up to isomorphism) by a finite class of second order axioms? HOOTmag (talk) 16:49, 8 October 2012 (UTC)

No. For every subset  , the expansion   of the standard model of arithmetic by A as an extra unary predicate is definable up to isomorphism by a set of second-order sentences: take e.g. the axioms of Robinson’s arithmetic, the second-order induction axiom, and axioms   or   for every   as appropriate. There are   such models (and they are pairwise nonisomorphic), whereas there are only   second-order formulas (or finite sets thereof), hence most of these models will not be definable by a finite set of second-order formulas.—Emil J. 17:58, 8 October 2012 (UTC)
Thanks. Additionally, can every first order proposition having a model, be contained in a finite class - of first/second order propositions - that uniquely determines a model (up to isomorphism)? HOOTmag (talk) 20:50, 8 October 2012 (UTC)
Yes, I think so. Let φ be a consistent FO sentence in a finite relational (WLOG) language L. We may assume that φ has no finite model (otherwise it’s easy, as the FO theory of any finite model is categorical, and if the language is finite, finitely axiomatizable). Let α(A,M) be a second-order formula (e.g., the conjunction of axioms of Q and the SO induction axiom, as above) defining the standard model of arithmetic   up to isomorphism, where second-order variables A, M are used instead of the arithmetical operations +, ·. The usual completion procedure shows that φ has a complete Henkin extension T in the language   such that the set of Gödel numbers of theorems of T is definable by a   arithmetical formula τ(A,M,x), where again, the arithmetical operations are replaced with A, M. Put
 
where F is a unary function variable,   denotes Gödel numbers, and kR is the arity of R. Then the only model of Φ up to isomorphism is the canonical model of T.—Emil J. 18:26, 10 October 2012 (UTC)
Oh, thank you Emil, sorry for being so late in seeing your interesting response!
Are you sure the only way to construct an appropriate Φ (i.e. such that uniquely determines a model up to isomorphism) is by using a richer language than L (if L does not contain the symbols you've added to it)?
Additionally, if φ (i.e. the original given FO sentence) is the FO two-sorted conjunction of ZF axioms, then do you think your new appropriate Φ (which will "contain" ZF of course) satisfies the Continuum Hypothesis? If it does and it's possible to prove that, then your procedure for building this Φ proves the consistency of the class containing both ZF Axioms and the Continuum Hypothesis (without having to use Gödel's proof), doesn't it? HOOTmag (talk) 08:54, 18 November 2012 (UTC)
Φ does not have a richer language (i.e., signature) than the original formulas. I only used the extra symbols to explain the construction, but in the actual formula, they become quantified second-order variables.
I’m not sure what you mean by “FO two-sorted conjunction of ZF axioms” (that’s kind of contradictio in adiecto), but in any case, the result depends entirely on the chosen Gödel numbering. For any finite set of sentences consistent with T, one can choose the enumeration in such a way that the resulting model will satisfy this set. The construction works by adding new sentences if they are consistent, so in order to prove that it includes a particular sentence (say, CH), you’d have to first prove that this sentence is consistent with the theory (and that it’s not contradicted by any sentences earlier in the enumeration which might have already been added), so it cannot be used to yield new consistency proofs.—Emil J. 12:58, 19 November 2012 (UTC)
Ok. As for your comment about "FO two-sorted conjunction of ZF axioms": Let's take the Axiom of replacement. If we want it to be formulated by one formula (rather than by one schema of infinite class of FO formulas), must it be a second order formula (rather than a FO two-sorted formula)? Additionally, I couldn't understand why you call it a "kind of contradictio in adjecto": Aren't there FO two-sorted formulas? HOOTmag (talk) 17:22, 19 November 2012 (UTC)
ZF is a theory in the language of ZF, and the language of ZF is single-sorted. If you add another sort and replace the schema of replacement with a single formula, the theory is no longer ZF, but some variant of GB (depending on what other axioms you include), which has rather different metamathematical properties. In any case, if you just want to apply the construction above to ZF despite that it’s not finitely axiomatizable, you can do it directly: the argument applies to any recursively axiomatizable (or even arithmetically definable) theory in a finite language in place of φ.—Emil J. 18:21, 19 November 2012 (UTC)
Back to your new Φ: Do you think it's constructive? I'm asking that, because AFAIK the usual completion procedure is not computable, is it? Btw, are you familiar with Libor Běhounek? HOOTmag (talk) 20:39, 19 November 2012 (UTC)
Second-order logic does not make much sense in constructivist context, I’m afraid, so the point is moot. The Henkin completion procedure is certainly nonconstructive. (Yes, Libor is a friend of mine.)—Emil J. 14:10, 20 November 2012 (UTC)
Yes, but when I asked whether "every [given] first order proposition having a model", can "be contained in a finite class - of first/second order propositions", I thought about a [given] proposition that was built constructively, and I also thought about a (finite) class that can be built constructively. Btw, is Libor a doctor or MA? HOOTmag (talk) 20:16, 20 November 2012 (UTC)
Oh, on a purely syntactic level, the construction of Φ from φ is perfectly constructive (only the argument that it works as desired is nonconstructive). Building the formula does not involve carrying out the Henkin completion procedure, the formula only describes the procedure. Why are you interested in someone’s personal data?—Emil J. 21:03, 20 November 2012 (UTC)
Are you sure? let's look at your Φ, mainly at the following part:  : Isn't it dependent on what one gets after one applies the Henkin completion procedure?
As for Libor: Just curiosity (after I read a few parts of his article); I thought it wasn't a secret, is it? Anyways, you don't have to tell if you don't want to. HOOTmag (talk) 22:15, 20 November 2012 (UTC)
  is something to the effect of “there exists a sequence w of 0s and 1s of length n + 1, such that (w)n = 1, and for every mn, (w)m = 1 if and only if the theory axiomatized by the formulas whose Gödel numbers are   and those i < m such that (w)i = 1 is consistent” (actually there has to be more stuff to deal with Henkin constants). This is an explicit arithmetical formula, whose only dependence on φ is that one has to plug in its Gödel number.—Emil J. 14:43, 21 November 2012 (UTC)
Yeah, I see, you're right. Thanks a lot (at least for the first part of what I'd asked). All the best. HOOTmag (talk) 18:04, 21 November 2012 (UTC)

(La)TeX templates

Please be more careful when trying to 'fix' templates that seem to be broken on your system. Your vector.css contains the following for .texhtml: font-family: inherit !important;, that is the reason the inline fonts don't work, not the template. Just remove the !important;; there is no need for it anyway, the inherit declaration will work without it, but still allow inine declarations. Edokter (talk) — 15:55, 9 October 2012 (UTC)

I put the "!important" there for a reason, the customization didn’t work otherwise. Is there any actual improvement your edit is intended to make, besides breaking the template for those like me who are forced to use funny CSS because of the template jungle that half of the math text on WP is wrapped in?—Emil J. 16:06, 9 October 2012 (UTC)
In fact, putting the TeX logo templates in class texhtml is wrong in the first place. The logos are not math formulas, and none of the styles applied by the class is applicable: the template does its own font selection to match the original logo, there is no whitespace to wrap, and size adjustments (if any) should be done in the template itself, as they are tied to the font being used (which is different for the template and for texhtml formulas).—Emil J. 16:24, 9 October 2012 (UTC)
These logos are set in TeX, then translated to HTML/CSS, so the class is appropriate and the styling should match that of formulae. The '!important;' should not be needed because there is nothing to override. I'm really curious what exactly does not work as intended without your !important; declaration. But the fact remains you should not fix templates that are broken by your own personal CSS. Edokter (talk) — 17:18, 9 October 2012 (UTC)
The fact that the logos are set in TeX when they are being set in TeX is both tautological and irrelevant. (FWIW, the logo is impossible to set in the subset of TeX supported by WP.) The texhtml class is to provide a uniform style to math formulas written in HTML (as opposed to <math> markup), whereas the TeX logo does not follow this style, so putting it in this class makes no sense. You didn’t respond to my main question, namely what is your edit to the template supposed to improve, and I repeat that question. I am not going to resist changes that will benefit the majority of users, but as far as I can see, you change does not do anything positive to anybody, whereas it can potentially harm users with custom CSS such as me, and as such it is strictly detrimental. IOW, you should not attempt to “fix” something that is not broken.—Emil J. 15:13, 10 October 2012 (UTC)
I did not try to 'fix' anything; I reverted a change that was completely unnecessary. Will you please tell me what it is that breaks when you remove the !important from your CSS? Edokter (talk) — 19:29, 10 October 2012 (UTC)
You introduced the change in January, I (partially) reverted it. So it is up to you to defend it, cf. WP:BRD. It’s been two years since I was tweaking my CSS, and I do not keep in mind for so long useless bits of information like what piece of software breaks under what circumstances, I have also work to do.—Emil J. 11:53, 11 October 2012 (UTC)
I’ve decided to test it now without !important. An immediately obvious problem (as seen in the basic usage of the template in Euler's formula#Using the limit definition) is that while I can override the font back to sans-serif, the undesired font-size enlargement is still in effect. Will watch it for further problems.—Emil J. 13:18, 11 October 2012 (UTC)
The font-size does reset properly for me to 100%. What browser do you have? Edokter (talk) — 21:08, 11 October 2012 (UTC)
Firefox 14.0.1.—Emil J. 11:51, 12 October 2012 (UTC)
In Opera (11.61), it’s the other way round: the size is (apparently) 100%, but with a serif font.—Emil J. 12:02, 12 October 2012 (UTC)
That is weird. Both Firefox (16.0.1) and Opera (12.02) on my end do behave correctly with your current code. Clear your cache and purge the page. Edokter (talk) — 20:21, 12 October 2012 (UTC)

Thanks for the tip on domains

Thanks for pointing me at domain theory here. I hadn't realized "domain" had such a specific definition; I thought it was just a set. (I was thinking more like the "domain" and "range" of a function.) --Doradus (talk) 21:13, 17 November 2012 (UTC)

You’re welcome.—Emil J. 12:44, 19 November 2012 (UTC)

Logical symbols

The Logical symbols template is for the symbols, not the operations that the symbols are being interpreted as. So even if we make the name "existential quantifier" (which is ok) it should still point to the article for the symbol. We already have a template for the logical connectives, so it would be redundant otherwise.

I have seen the "wedge" as representing "or" (it makes sense doesn't it? "either ... or" is to form a wedge between concepts.) So we need to see who calls it what.Greg Bard (talk) 13:58, 21 November 2012 (UTC)

What is the point of the template in the first place? I actually do think it is rather redundant. It makes no sense to create two-line stub articles for symbols which are only notable as denotations of the operations they represent. However, if there is to be such a template, then it should at least include the most commonly used connective symbols, not just those for which you decided to write the stub. The separate articles may be created later.
As for wedge, I have no idea where you got that usage. Wedge is a common name of  , see e.g. [2], [3], even its LaTeX name is \wedge (with \land as a synonym). The wedge product is also named after the symbol.—Emil J. 14:29, 21 November 2012 (UTC)
Independently I noticed that and fixed the article on vel (symbol) as far as the incorrect statements went. Wedge (symbol) redirects there, although it probably should not, but I can't be bothered to write it because of its low notability (as a symbol). Tijfo098 (talk) 18:05, 27 November 2012 (UTC)

A barnstar for you!

  The Writer's Barnstar
For the article on admissible rule. Tijfo098 (talk) 18:08, 27 November 2012 (UTC)
Thank you!—Emil J. 19:06, 27 November 2012 (UTC)

Set theoretic definition of the natural numbers

Sorry, about the edits. Thanks for explaining, I see how I was totally misunderstanding. --Atethnekos (DiscussionContributions) 20:53, 7 December 2012 (UTC)

No problem.—Emil J. 12:51, 10 December 2012 (UTC)

Population of the Czech Republic

You are obviously right, I have blindly copied from Population of France. Littledogboy (talk) 23:36, 20 December 2012 (UTC)

Another inaccurate Poincaré quote

I also would argue that the preceding quote in the Poincaré article (regarding Bachelier's thesis) is inaccurate and should probably be fixed. Sure, it's quoted from Bernstein's book . . . but it seems to be at best a paraphrasing rather than a translated quote from Poincaré's report, since the full translation given in the article "Louis Bachelier on the Centenary of Théorie de la Spéculation" (Mathematical Finance, Vol.10, No.3 (July 2000), 341–353) [4] does not seem to contain all that Bernstein writes in his book. — Myasuda (talk) 02:14, 11 January 2013 (UTC)

The Courtault et al. paper makes quite a different perception of Poincaré’s opinion than what we have in the article, and the quote appears to be taken out of context, to put it mildly. I agree something should be done about it. However, rewriting the section would need someone with at least some background in economics, I can’t do that.—Emil J. 13:22, 11 January 2013 (UTC)
Thanks for your response. I've taken and slightly modified my comment above and moved it to the Poincaré article talk page. — Myasuda (talk) 16:01, 12 January 2013 (UTC)

Continued fraction

[5] Of course, my edits may contain errors. But IMHO it is far more convenient to fix errors against my versions (with Show changes tool), than against wicked products of user talk:99.241.86.114 whose efforts are directed against the community consensus.

You could just notify me (or the talk page) about a mistake, if not to fix it yourself. But if you prefer to start an edit war… well, I'll make a war over the article. Incnis Mrsi (talk) 17:03, 14 January 2013 (UTC)

Watch your language, Incnis Mrsi. I didn’t “nuke a half of an article”, I undid your edit which consisted of the cosmetic change of reintroducing the math et al. templates removed by the IP user, bundled together with the introduction of loads of these awful frac templates which violate the MOS, with the end result being far worse than it was before, with or without the math templates. I did at first consider fixing it myself, but the sheer number of the frac templates you introduced in the article made it infeasible, and overwhelmed any other changes you made, so reverting the article was the only reasonable option. I do not take it lightly to be accused of trolling just because someone else cannot do their job properly, and I expect an apology.—Emil J. 17:13, 14 January 2013 (UTC)
Let's look on the facts:
  • 99.241.86.114 made numerous changes which encountered several objections from the community;
  • He never tried to discuss these changes on his own will;
  • I, Incnis Mrsi, edit-warred against 99.241.86.114, but during a day-long discussion no one claimed that Incnis Mrsi's changes degraded the article (although it does not prove that there were no degradation at all);
  • The discussion at WT:MATH was linked from an edit summary in the way that it cannot be unnoticed;
  • EmilJ sided with 99.241.86.114 in the edit war (as a bare fact, not as an assumption about his motives);
  • EmilJ did not try to discuss the problem previously;
    • Notably, EmilJ did not try even to notify either Incnis Mrsi or the ongoing discussion about the problem in any way but making a revert.
There is no doubt that such an action was highly inflammatory in the context of the discussion. Of course, I do not claim that EmilJ's intention was to aggravate the conflict (hence the question mark), but these are just actions which look to obey established policies, but intended to inflame and sustain conflicts, which constitute a trolling. Incnis Mrsi (talk) 18:11, 14 January 2013 (UTC)
I didn’t side with anybody in any edit war, and in fact, I cannot find the edit war you seem to be talking about in the article history. This is all your paranoid imagination. I am not going to respond further if you continue with this abusive tone.—Emil J. 18:33, 14 January 2013 (UTC)

Thanks!

Thank you for fixing Template:Sister project links. Kendall-K1 (talk) 17:57, 15 January 2013 (UTC)