Jinx wrote:
>>> suggesting 'm' for both milli and micro doesn't work well
>
>> milli 10^-3 m
>> micro 10^-6 m (Greek mu)
>
> Gerhard, I will backtrack on what I said. A bit. On the web page
> and what I copied/pasted I see the "mu" character. In the sent mail
> it has changed to "m". You must have seen "m" in your browser. In
> the source for that page
>
> m (Greek mu)
Here we go... this is non-standard HTML. It is relying on platform-specific
(Windows in this case) non-standard fonts.
Of the three browsers I have available (on WinXP), IE and Mozilla use the
specified platform-specific font to display the intended mu, whereas Opera
doesn't use the font specification and displays an m instead. I used
Opera... :)
The proper way to program this in HTML would be to simply use the
appropriate entity "μ" instead of the sequence "m". This is not platform-specific and would work
across (standard-compliant) browsers and platforms -- and be correct HTML.
Come to think of it, it may be that the unlucky use of the Symbol font is
the source of many intended 'mu's appearing as 'm's. Just don't use it if
you want to share your documents. It's a left-over from the early 80ies and
before, when Unicode wasn't widely available. Want to insert a mu into a
Word (or other) document? Use the appropriate character from the standard
(Unicode) font you are already using, not the one from the Symbol font. If
the font you're using doesn't have a mu (AFAIK that's rare), use another
Unicode font that has it. Don't use the Symbol font...
As usual, the problem lies with too many programmers and documenters not
being aware of 'i18n', the existence of various platforms and related
problems. ('i18n' is short for 'internationalization', coming from the Java
camp AFAIK. The '18' symbolizes the 18 left-out letters.) Many just don't
care enough to get familiar with these issues on their own and only do it
when forced.
"Don't ever use anything platform-specific on a public web site. Use
standard HTML whenever possible." These are two basic web design rules, and
whoever wrote this page never heard of them. I would expect a little more
diligence from somebody writing a manual about of all things notation in
scientific online publications.
Just for fun (and some education), look at the w3.org validator output for
this page:
(Long link, you may have to recompose it.)
- The first comment is "No Character Encoding Found! Falling back to
UTF-8." A clear indication that the author didn't care about any font
encoding questions. If it were meant to be utf-8, there would have been no
need to use the Symbol font in the first place.
- The second comment is "No DOCTYPE found". A clear indication that the
author didn't care about HTML standard compliance. Adding the appropriate
DOCTYPE to define what kind of document you're writing is the first thing
you do when creating a standard-compliant HTML document.
- The error 2 shows a second problem with the way they want to display the
mu: the font tag is not allowed where they used it. (The validator doesn't
know about the characteristics of the Symbol font, so it doesn't say that
it's a bad thing. It doesn't check these things; it only checks for HTML
compliance.)
Concluding... a completely screwed-up page that should have never passed a
minimum of quality assurance :)
For a counter-example, run http://en.wikipedia.org/wiki/Unicode through the
above validator... The page starts with a proper !DOCTYPE tag, and has as
first header tag the Content-Type definition "". This is
one of the reasons why it works here... It's not impossible; you just have
to care about it.
Gerhard
--
http://www.piclist.com PIC/SX FAQ & list archive
View/change your membership options at
http://mailman.mit.edu/mailman/listinfo/piclist