Layout of numbered equations? - Page 5

•  Subject
• Author
• Posted on

Re: Layout of numbered equations?

In article
<dorayme-C48D1B.18000209092011@c-50-133-29-211.hsd1.mi.comcast.ne
t>,

...

I should look up Stan Brown's fonts and make the example
surrounding text the same as the screenshot. Be easier if I had
typed up the maths in the first place and then imaged it. Anyway,
when there is more time.

With the inline set at 1em as you suggest (but 'top:...' rather
smaller) it looks fine for me on Windows. It is relatively easy
to do images displayed on a separate line or block but care is
needed for the inline, if one creates the maths lines in the
first place in a particular font, one then has easy access to the
font used.

--
dorayme

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j4cb66$pgu$1@dont-
email.me>, Fri, 9 Sep 2011 09:19:52, Jukka K. Korpela

The old French beta in <http://www.merlyn.demon.co.uk//euler327.htm#ATP
para 4 sentence 2 scales to text size for all practical purposes using
<img src="graphics/old-beta.png" style="width:0.55em;" alt="old-beta">.
Granted, that's scaled for Times New Roman, and I suppose will be a bit
short for Verdana.  It could there be better to scale the height, in ex.

--
(c) John Stockton, nr London, UK. ?@merlyn.demon.co.uk  Turnpike 6.05  WinXP.
Web  <http://www.merlyn.demon.co.uk/ - FAQ-type topics, acronyms, and links.
Command-prompt MiniTrue is useful for viewing/searching/altering files. Free,
DOS/Win/UNIX now 2.0.6; see <URL:http://www.merlyn.demon.co.uk/pc-links.htm .

Re: Layout of numbered equations?

On Sun, 11 Sep 2011, Dr J R Stockton wrote:

Probably U+03D0.

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <Pine.LNX.4.64.1109121827
520.30659@zen.rrzn.uni-hannover.de>, Mon, 12 Sep 2011 18:28:20, Andreas

Yes.  But is that reasonably safe across all likely reader-preferred
font systems?  While it may be shown as a glyph in a correct or
substitute font, it looks dangerously like a 6 in my preferred font.
That would be no problem in a document containing only words - it would
be recognised immediately in context  - but it would be entirely
unacceptable in mathematical expressions, particularly in this case
where it there are no genuine 6 characters in expressions to compare it
with.

OTOH, your remark is not germane.  I have an image of the character,
taken from Euler's work, and it shows the size-tracks-ordinary-text
property which is commonly unknown. That is the point.

--
(c) John Stockton, nr London, UK. ?@merlyn.demon.co.uk  Turnpike v6.05  MIME.
Web  <http://www.merlyn.demon.co.uk/ - FAQqish topics, acronyms and links;
Astro stuff via astron-1.htm, gravity0.htm ; quotings.htm, pascal.htm, etc.
No Encoding. Quotes before replies. Snip well. Write clearly. Don't Mail News.

Re: Layout of numbered equations?

On Thu, 08 Sep 2011 15:15:29 +1000, dorayme wrote:

<<blush>> Thank you!  I try to apply what I learn here and in
c.i.w.a.s.

I believe that I use an image only when I think the reader's
comprehension will be significantly harmed by an inline version.  Of
course, that means that when I do use an image the alt text is
horrendous. :-)

--
Stan Brown, Oak Road Systems, Tompkins County, New York, USA
HTML 4.01 spec: http://www.w3.org/TR/html401 /
validator:      http://validator.w3.org /
CSS 2.1 spec:   http://www.w3.org/TR/CSS21 /
validator:      http://jigsaw.w3.org/css-validator /
http://diveintomark.org/archives/2003/05/05/why_we_wont_help_you

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <dorayme-
FC3F46.15152908092011@c-50-133-29-211.hsd1.mi.comcast.net>, Thu, 8 Sep

HTML <canvas>, written with JavaScript, will do any maths for which
Unicode and the computer's fonts provide, including creating characters
by overlaying Unicode characters and other lines.  It could, for
example, write equations in Korean even in a computer lacking all Korean
resources.

will show an example expression, except in IE <= 8.

And, if you want to use fonts or characters that the reader's machine
might not have, or you want to support IE<=8 and similar, you can always
save the <canvas> to an <img> and transmit that.

SVG might share those properties; but I don't know how to use it.

--
(c) John Stockton, nr London, UK. ?@merlyn.demon.co.uk  Turnpike v6.05  MIME.
Web  <http://www.merlyn.demon.co.uk/ - FAQqish topics, acronyms and links;
Astro stuff via astron-1.htm, gravity0.htm ; quotings.htm, pascal.htm, etc.
No Encoding. Quotes before replies. Snip well. Write clearly. Don't Mail News.

Re: Layout of numbered equations?

9.9.2011 23:59, Dr J R Stockton wrote:

It's a drawing board integrated into the DOM of the page. That's often
great, but it has its problems, and it won't do any math for you - you
need to tell it how to draw the equations. Presumably some libraries for
that will emerge, but even then, what will you win? Will the libraries
get even close to what one can get using good typesetting and layout
software for math?

Images have their limitations, and so do approaches like MathJax. But
using <canvas> without any libraries becomes either very tedious or poor
in terms of presentation quality. (<canvas> also has the IE issue, and
although this can be addressed using the explorerCanvas library, you
would need to check things carefully, since the library has many
limitations.)

So what's really the benefit of using <canvas> for math? Doing so to
learn things and to test what can be done is great. But it's hardly

The example expression has integration limits set to the right of the
intergral sign, not above and below. The upper limit has "x" in upright
font, should be in italics. The operator "d" IN "dt" is in italics,
should be upright. The "t" below the line touches the line.

--
Yucca, http://www.cs.tut.fi/~jkorpela /

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j4fdop$ucn$1@dont-
email.me>, Sat, 10 Sep 2011 13:22:18, Jukka K. Korpela

It gets you not needing to acquire and learn additional software; it
gets you a page which can in full be inherited edited by anyone else.

Yah boo sucks to IE <= 8 users.  If they are smart enough to read my
pages, they're smart enough to get alternative browsers.

You persist in considering things from the point of view of a
professional whose job is to convert some sort of approved draft or
submission into an optimal Web rendition - rude mechanicals.  That's of
no interest to me.  I'm interested in the production and transference of
real-world knowledge without always wasting time on superfluous
presentational polish.

It's an example of the ability to position characters, and the layout
closely resembles what is found in books from real publishers.
Standards may be standards, but they are not necessarily right.  Read,
for example, ISO 8601; then taking that as gospel, read the JSON/ISO
date parts of ECMA 262 5.1.

--
(c) John Stockton, nr London UK.  ?@merlyn.demon.co.uk  DOS 3.3 6.20 ; WinXP.
Web  <http://www.merlyn.demon.co.uk/ - FAQqish topics, acronyms & links.
PAS EXE TXT ZIP via  <http://www.merlyn.demon.co.uk/programs/00index.htm
My DOS  <http://www.merlyn.demon.co.uk/batfiles.htm - also batprogs.htm.

Re: Layout of numbered equations?

11.9.2011 21:21, Dr J R Stockton wrote:

Using <canvas> means that anyone who edits the page needs to know both
JavaScript and the Canvas API. That doesn't like something that a person
editing a math page should need to know.

You need to get better books then. I would never expect to find such a
combination of flaws in mathematical typesetting even in a book.

Or maybe I need a better troll detector.

--
Yucca, http://www.cs.tut.fi/~jkorpela /

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j4k3l7$des$1@dont-
email.me>, Mon, 12 Sep 2011 08:00:23, Jukka K. Korpela

Other than professional compositors, those who want to edit mathematics
should know mathematics - and for such people, understanding JavaScript
and <canvas> sufficiently well is but a small step.  Remember that we
are considering editing rather than composing; the editor of existing
maths will have the existing code as an example.

--
(c) John Stockton, nr London UK.  ?@merlyn.demon.co.uk  Turnpike v6.05  MIME.
<http://www.merlyn.demon.co.uk/ TP/BP/Delphi/&c., FAQqy topics & links;
<http://www.merlyn.demon.co.uk/clpb-faq.txt RAH Prins : c.l.p.b mFAQ;
<ftp://garbo.uwasa.fi/pc/link/tsfaqp.zip Timo Salmi's Turbo Pascal FAQ.

Re: Layout of numbered equations?

J R Stockton on 1 Sep 2011:

J R Stockton on 11 Sep 2011:

I think an author should make every effort to get himself

--
In memoriam Alan J. Flavell

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j40q5e$kkc$1@dont-
email.me>, Mon, 5 Sep 2011 00:22:19, Jukka K. Korpela

Without CSS, most of my pages do far worse things than leaving no gap -
it's more important that the equations will not be on lines of their
own.  And that page (currently not public) has, at the top, "Needs
styles-a.css in the same directory.".

In that page, italics and upright have different meanings,  That happens
to be from a book probably written before you were born, which uses
italics for variables.  Granted, the 'cos' should be upright.

The ones I use work in IE8, FF3.6.21, Opera 11.51, Safari 5.0.5, Chrome
13.  That's good enough for me.  The longer I spend making my work
available on all systems, the less time I have for other work that the
majority will be able to read.

If a browser does not show a reasonably well-established space character
correctly, then the reader should complain to the browser supplier, use
a different browser, or both.

There are fat bigger problems with browsers, though.

<http://www.merlyn.demon.co.uk/js-datex.htm#S503 seems to show that
Apple Safari (at least, for UK settings in WinXP pro sp3) has used the
digit 3 in implementing the JavaScript Leap Year rules, though there
could be some other explanation.  It would be of interest to know
whether it appears in other circumstances.  I set my laptop to Helsinki
: no difference.

--
(c) John Stockton, nr London, UK. ?@merlyn.demon.co.uk  Turnpike 6.05  WinXP.
Web  <http://www.merlyn.demon.co.uk/ - FAQ-type topics, acronyms, and links.
Command-prompt MiniTrue is useful for viewing/searching/altering files. Free,
DOS/Win/UNIX now 2.0.6; see <URL:http://www.merlyn.demon.co.uk/pc-links.htm .

Re: Layout of numbered equations?

6.9.2011 1:41, Dr J R Stockton wrote:

[...]

And so should the digits (superscripts or not) and operators. That was
my point. Only the variables should appear in italics. This requires
more markup, but there is no way around it:
<i>c</i><sup>2</sup> = <i>a</i><sup>2</sup> + <i>b</i><sup>2</sup>
&minus; 2 <i>a</i> <i>b</i> cos <i>&gamma;</i>

They "work" via font fallback. In theory, it should work well if _any_
of the fonts in the system contains the EM SPACE, and its width should
independent of the font (1em). In practice, older versions of IE, still
with us, have many issues with fallback.

Quoting my document on Unicode spaces:

"In my tests, when Times New Roman or Arial is used, IE 6 shows only the
first two (space, no-break space) and the last two (ideographic space,
zero-width no-break space) [of the Unicode space characters] correctly,
presenting the other space characters as small rectangles that indicate
unrepresentable glyphs or as spaces of incorrect width. When Arial
Unicode MS is used, all but the narrow no-break space (which was added
to Unicode in version 3.0) are shown correctly."
http://www.cs.tut.fi/~jkorpela/chars/spaces.html

Most importantly, the risk is quite unnecessary. You can use a simple
CSS instead and get much more flexibility.

I'm not saying that you should work on modifying existing pages - that's
usually waste of time, unless there are essential problems and tangible
benefits to be gained. For new pages, I would suggest a more natural and
more flexible approach to spacing.

This is not an HTML problem at all, is it? It seems to be strictly about
implementations of the JavaScript Date.toUTCString() method (something
that I'd avoid anyway, since it is by definition
implementation-dependent and tends to create a string presentation using
English abbreviations, not adapted to the language of the page).

--
Yucca, http://www.cs.tut.fi/~jkorpela /

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j448ea$7fi$1@dont-
email.me>, Tue, 6 Sep 2011 07:43:53, Jukka K. Korpela

That would be effort without benefit.  I have enough to do needing
effort that will, or should, provide real benefit.

It is sufficient that they work; it is immaterial whether a user sees a
space in his favourite font or in another font.  And, as I use them, the
exact amount of space given is unimportant too.

That's rather old, going no further AFAICS than IE6.  Its list of spaces
is incomplete - no 1680, no 180e.  My js-valid.htm shows what I believe
to be a full list of the 28 "JavaScript" spaces, and shows something
about which have been seen with which browsers.  And, if you view the
page in IE <= 8, you can see that JavaScript RegExp \w there recognises
one more character than in other browsers.

could conveniently read it in his choice of available font, and see what
the spaces then looked like.

That would waste time.

Agreed.  As I wrote, it was to show that there are far bigger problems
in browsers.

Test 1 does use toUTCString for reporting.  But Test 2 does not use
toUTCString, and neither does Test 0 (I'm adding that <H5>).  The error.
as the <H4> says, is in Safari new Date(String).

If you read the Test 1 code, you will see that it does
new Date("YYYY/07/22 12:00 GMT")
for successive years, and reports when the value of getDate() on that
changes.

The test in Test 2 compares the actual number of milliseconds since
epoch.

New Test 3 just uses new Date("...") and (implied) .valueOf()/864e5.

--
(c) John Stockton, Surrey, UK.  ?@merlyn.demon.co.uk   Turnpike v6.05   MIME.
Web  <http://www.merlyn.demon.co.uk/ - FAQish topics, acronyms, & links.
Proper <= 4-line sig. separator as above, a line exactly "-- " (SonOfRFC1036)
Do not Mail News to me. Before a reply, quote with ">" or "> " (SonOfRFC1036)

Re: Layout of numbered equations?

7.9.2011 22:12, Dr J R Stockton wrote:

It's your privilege to ignore advice, but I wonder why you don't
consider it as a waste of time to quote the advice and say it taking it
would be a waste of time, instead of just not commenting on it at all

The principle I mentioned is an age-old principle of presenting
mathematical formulas, recorded by typography authorities and confirmed
by international and national standards. Why take the trouble of
converting printed material to HTML format if you don't want to follow
even the very basic standards for it and reproduce the original?

(I has suspected that the original might deviate from the conventions,
for some strange reason, in which case it might have been adequate to
follow its style _if_ the intent is to illustrate the history of
mathematical typesetting rather than the history of mathematics. But
your reaction apparently says this is not the case.)

--
Yucca, http://www.cs.tut.fi/~jkorpela /

Re: Layout of numbered equations?

In comp.infosystems.www.authoring.html message <j49ghu$47r$1@dont-
email.me>, Thu, 8 Sep 2011 07:33:03, Jukka K. Korpela

But not much time; and using time in an attempt to change your mind may
not be inevitably a waste of time.

Euler wrote, or had typeset, his equations in order to convey particular
meanings to his readers.  My wish is to convey the same meaning to
modern readers, with modern character generation rather than the
relatively crude typesetting and printing of Euler's day; for that,
there is no need to make a faithful copy of his layout, warts and all.
So, for example, by using <sup>2</sup> I avoid the legibility problems
of &sup2; and of Euler's font, in which ordinary-size 2 was no bigger
than o rather than matching A, and superscripts 2 & 3 were likewise
tiny.

But you are welcome to offer better HTML for the equations in my
euler327.htm, provided that you do so with reference also to the actual
PDF of E.327, and provided that the overall storage is not significantly
increased.  But improving the English rendition of the Latin would be
__MUCH__ more useful.

Referring to another post : I found no difficulty at all in getting
proper vertical alignment between ordinary text and an image of a
descender-less character - just crop the character's image so that the
character's base is at the bottom of the image.  I guess character-
images with descenders could easily be dropped with <sub>, though there
should be a better way.

But can Windows XP Paint produce images with a transparent background,
and if so how, step by step?

--
(c) John Stockton, nr London, UK. ?@merlyn.demon.co.uk  Turnpike v6.05  MIME.
Web  <http://www.merlyn.demon.co.uk/ - FAQqish topics, acronyms and links;
Astro stuff via astron-1.htm, gravity0.htm ; quotings.htm, pascal.htm, etc.
No Encoding. Quotes before replies. Snip well. Write clearly. Don't Mail News.

Re: Layout of numbered equations?

On Mon, 5 Sep 2011, Jukka K. Korpela wrote:

Another point is that the author can specify *several* typefaces
whereasReader=92s font=94 is just one or perhaps two when you
allowserif=94 andsans-serif=94.

--=20
From the New World: