CZ Talk:Formatting mathematics

I like this. It would be nice if the creator(s) would also create a [[CZ:How to use <math> formatting]] for those of us who don't know TeX formatting. Anthony Argyriou 17:28, 10 May 2007 (CDT)

I've been getting help from Help:Displaying a formula at Wikimedia. --Catherine Woodgold 13:16, 12 May 2007 (CDT)

scritpstyle

I edited somewhat the \scriptstyle section. I think the example with e^\int does not belong in, since it's as ugly with or without \scriptstyle (it's the baseline problem; in fact, we can partially avoid this by using \exp istead of e^ ). Secondly, I inserted suggestion that 'global resizing' is possible. BTW, it was good idea to create such a page! --Aleksander Stos 06:24, 11 May 2007 (CDT)

I oppose frequent use of "scriptstyle" for the following reasons: (1) That's not what it's for; (2) Greg Martin said it made the formulas look very ugly on his browser; (3) It's possible (in principle at least) to have the servers insert "scriptstyle" into all inline formulas (or tranlate them all into html) while serving up the articles, but not feasible to have the server delete "scriptstyle" since that would delete it in the rare cases where it's actually appropriate, (nor would it be feasible for the server to translate from html to math since it wouldn't know which things are supposed to be math formulas;) (4) if browsers are improved, the pages with "scriptstyle" will be obsolete. We might have to go through a process of re-approving a large number of pages that haven't been touched in a long time; (5) if it looks bad without scriptstyle on some browsers and bad with scriptstyle on other browsers, then by not using scriptstyle we can blame the browsers, but if we use scriptstyle it makes Citizendium look ridiculous; (6) I don't mind the big formulas much. An appropriate size would be better, but the big size also has advantages of being easily readable and highly salient. I can search for words on a page with my browser search function but don't know how to do that with math formulas, but I can find the large formulas by skimming. Also the large size helps counteract a tendency to skip over equations as "too hard" and just read the text; (7) it removes incentive to improve browsers; and perhaps not least (8) we would have to define clearly where scriptstyle was or was not to be used. In answer to Michael Hardy on my talk page, yes I see that the arguments in favour of scriptstyle or html formulas have some weight, but it seems to me that the argument that the server can translate things automatically pretty well makes those points moot.

I agree, it was a great idea to create this page! However, I suggest making it a guideline rather than policy, so that it will be easier to make exceptions in reasonable cases; or at least to delete "always" and sprinkle throughout the text phrases such as "usually", "normally", "unless there is an unusual reason otherwise", etc.; or put a disclaimer at the top or bottom to the effect that there will be exceptions when there are unusual or strong reasons to do otherwise in particular cases and that people need to use their judgement. (or that no one will be banned for life for breaking one little rule by accident, laziness, ignorance or because it seems to be an exceptional case.) Is it OK to write a draft in some other style out of laziness and then come back later and fix it (or let others fix it)? I think it should be OK though not encouraged. If exceptions do arise, things like comments in the wikitext can help avoid inadvertent reverting. --Catherine Woodgold 09:12, 12 May 2007 (CDT)

Now, I do not insist too much on \scriptstyle. I remarked on workgroup's talk page that general resizing would be better, partially due to reasons (1) and (2). But I feel I should answer some arguments. I agree that (3) would be good, but for this we should either enforce the standard formatting, i.e. indented <math> tag, or have --like in TeX-- two different tags for inline and displayed formulas (this is desirable, but will not happen soon, if ever). Essentially, I do not believe in (4) (explained on forum). But I'd be more than happy if Jitse would like to refresh his interest in technicalities of math formatting issues. I wonder if I could help... Number (6), personal perception while being a perfectly correct argument, is --as far as I can see-- not what is shared by many authors here. Moreover, does (m)any serious paper publication about math present math formulas like this (i.e. big inline formulas)? Concerning (7), yes, it is good to keep trying. But this reduces to (4): consider that with a much technically stronger math-related team the issue wasn't resolved on WP (see also some M. Hardy's posts on the matter; BTW, I did not see him advocating scriptstyle). Unless some of us declare we _can_ do it (how, who, when -- needs to be answered), I'd bet we'd have to live with what we have now for a long time. As for (8) the answer would be simple: everywhere inline, at least when the formula is PNG-rendered.

My favorite solution, arguably feasible, would be (1) setting the default size of the PNG-rendered formulas somewhat smaller than what we have now; (2) introducing two different math tags for displayed and inline formulas (this is probably harder and not meant to address directly the size problem -- considering that in a typical math publication the displayed formulas are of the same size as inline; this is just to address general markup and formatting text issues).--Aleksander Stos 17:11, 12 May 2007 (CDT)

suggested structure

I moved a couple of the new sections around - thank you for the contributions, by the way! For example, instead of a new section devoted only to fractions, I moved it to a subsection of "Issues for all <math> environments".

What I suggest is that this page itself contain policy (guidance), while any discussion of what our policy or advice should be goes on the subsidiary pages such as CZ:Formatting_mathematics/Scriptstyle. These are linked from the main article with (discuss this) links. That will make authors' use of this page easier, and it will allow for a thorough and structured debate over each contentious issue. - Greg Martin 12:23, 11 May 2007 (CDT)

re "obviously"

I think it's better to say "It then follows that..." than "obviously" or "it is clear that". "It is clear that" will be false for some readers, and both "obviously" and "it is clear that" imply that any reader who hasn't gotten it yet is stupid. Readers don't like that. In my opinion, "It then follows that..." gently implies that it follows fairly directly, without claiming that everybody who's worth anything has already understood it before they've even finished reading those words. Another one that's not too bad in my opinion is "It can easily be shown that..." Again, it allows the reader to take a few seconds doing the work of "showing" that, before being left behind. Better yet is to make the steps of the proof small enough that you don't have to tell the reader that it's obvious, because it really is. ("Obvious" is really only used when it isn't obvious, as a signal to the reader who hasn't gotten it yet, to let them know what kind of reasoning to look for, i.e. something simple. But it doesn't sound like it.) --Catherine Woodgold 13:55, 12 May 2007 (CDT)

Punctuation

I think punctuation should go outside the math tags, but with no space between the closing math tag and the punctuation. Could someone give me an example where it's necessary to do otherwise? I tried at User:Catherine Woodgold/sandbox. I resized the browser window many times and couldn't get the period to do on a different line from the preceding math formula, except in the 3rd line where there were spaces outside the math tags. I thought I had seen a formula with the period on a different line in one of the articles, but now I can't find it. The main reason I prefer to have the punctuation outside the math tags (even if it does sometimes look funny) is that people who aren't sure whether the punctuation is supposed to be part of the math formula can look at the wikitext to find out. Math tags are meaningful, in my opinion -- they mark something as a math formula, and putting punctuation inside would disturb that. --Catherine Woodgold 15:06, 12 May 2007 (CDT)

For example, look at this let us be bold and write $\Delta =\left(22{\sqrt {-1}}\right)^{2}$. (from Complex number, try to resize). The point is that, generally, a punctuation mark at the beginning of the new line does happen without special resizing. The reason is clear: in the browser's window the text flows freely and sometimes the punctuation mark does not fit in the line. I saw it a few times on WP/CZ. Our markup is not as perfect as TeX and I guess we just have to live with it. --Aleksander Stos 17:28, 12 May 2007 (CDT)

Punctuation needs to be INSIDE the math tags in many cases. You'll find about innumerable examples over at Wikipedia (Have you looked around there? Especially where they discuss these issues? You need to be aware of those discussions to be reasonably well-informed on these issues.)

Let's try one:

$\int _{0}^{1}e^{\sum _{i=1}^{\infty }a_{i}}\,dx$.

$\int _{0}^{1}e^{\sum _{i=1}^{\infty }a_{i}}\,dx.$

On the browser I'm using, the first form above looks incredibly stupid. For one thing, as one ought to expect, it's not properly aligned. Another problem is that as one adjusts the size of the window, the final period or comma sometimes ends up as the first character on the next line, like this:

The cat sat on the mat

, while wearing a hat

.

Citizendium needs to work on catching up to Wikipedia in many matters including this by studying and learning from Wikipedia's far greater experience.

(Of course, when TeX or LaTeX are used in the normal way, as opposed to the stripped-down way they're used here or on Wikipedia, then this doesn't matter.) Michael Hardy 22:12, 13 May 2007 (CDT)

Capitalizing theorem names

I believe that at times it is appropriate to capitalize theorem names. There is a difference in meaning between "the Prime Number Theorem", which is widely understood to mean a certain particular theorem, and "the prime number theorem," which could mean the theorem we just proved about prime numbers (as opposed to the other one we just proved about something else), even if it isn't the same theorem as the Prime Number Theorem. Mentioning "the Fundamental Theorem of Arithmetic" makes it clear that one is referring to a particular theorem widely recognized as having that name, while mentioning "the fundamental theorem of arithmetic" expresses an opinion that the theorem one is talking about is fundamental and is the only theorem of arithmetic which is fundamental, and does not make it quite clear which theorem is being talked about. I would suggest capitalizing theorem names which are widely recognized and commonly capitalized in the field, especially in cases where the meaning would be affected. --Catherine Woodgold 09:36, 21 May 2007 (CDT)

Can anyone help an ageing non-mathematician?

I can get by with basic maths and that's all I need for some simple economics articles. There is little in this article that is likely to be of any use to me (even if I could understand it). How about a paragraph on formatting maths for idiots? Nick Gardner 15:52, 8 February 2008 (CST)

The page http://meta.wikimedia.org/wiki/Help:Displaying_a_formula has some more information which is probably more helpful. But even that should probably be simplified. I'll put in on my todo list.

Of course, if you have a specific formula in mind, do ask me and I'll explain what to do. -- Jitse Niesen 12:16, 9 February 2008 (CST)

I've imported and started to clean up that page; it's at Help:Displaying mathematical formulas. Anthony Argyriou 14:38, 11 February 2008 (CST)