From: Asmus Freytag (email@example.com)
Date: Thu Sep 27 2007 - 12:17:25 CDT
Replying to John's message but also answering Dimitry.
On 9/27/2007 7:49 AM, John Hudson wrote:
> Dmitry Turin wrote:
>> JH> http://www.unicode.org/reports/tr25/
>> JH> http://www.unicode.org/notes/tn28/
>> But these are the markup language once again.
> Yes. Mathematics layout is not plain text.
Actually, the main point of these is to recommend which strings of
symbols to use to express a particular mathematical semantics. With
Unicode, there's no longer necessarily a need to express the
mathematical operators as markup, since there are now character codes
for practically all of them. That makes it feasible to render (simple)
formulae directly from plain text.
The linear format introduces several conventions that handle things like
scoping, super and subscripting in a way that keeps the raw text very
close to the plain text, making it possible to display the linear format
as fallback. Several clever inventions, such as using redundant ( ) to
mark scoping, utilize the inherent notational conventions of mathematics
towards this goal.
Similar conventions are used in other plain text context, from using
punctuation for emoticons to the use of paired * _ or / to indicate
>> Plain text is used in _all_ type of documents
>> (biologistic one, economistic one, texts of schoolchildren),
>> and manufactorers of software must be prevented from embeding of
>> _particular_ markup language into all system.
> So you want to force software developers to use a plain text encoding
> standard for *layout* to prevent them from using markup standards?
> This doesn't seem to me to be a good idea.
Actually, there's nothing inherently wrong if certain software
interprets such near-plain text conventions. Many e-mail clients
interpret emoticon strings or boldface text between *, etc. If text
display software wants to autorecognize math expressions and display
them according to the linear format, that would be just fine - as long
as it's optional, as is usually the case for substituting pictures for
The one thing you don't want to do is make that mandatory as part of
conformance to the standard. Which is something most of us agree on, anyway.
> John Hudson
This archive was generated by hypermail 2.1.5 : Thu Sep 27 2007 - 12:19:47 CDT