Square brackets & common inlines (was: [this] as a synonymfor[this][])

[Jelks Cabaniss]

Chris Purcell wrote:
>> 	[-#foo ."c1 c2" ="some title" some span of text-]
>> 
>> (or some such).
>> 
>> OTOH, versus the equivalent embedded HTML ...
>> 
>> 	<span id="foo" class="c1 c2" title="some title">some
>> 	span of text</span>
> 
> I'd have to say the latter is close to infinitely preferable. If
> nothing else, it's apparent at a glance what it means to anyone who
> knows HTML.

But if that's the criteria, we wouldn't even have

	![Alt text](/path/to/someimg.jpg "some title")

or

	[link text](http://example.com/foo.html 
	"some title") 

Note I stretched my example (which you quoted at the top) to the point of
extremes by including *three* attributes simultaneously, with one even more
uncommonly containing space-separated multiple class names!  *Normally*,
you'd have just a simple class or an ID:

	[-.foo some span of text-]

And you're right in that people who don't know Markdown might not instantly
get that.  But by that criteria, we wouldn't have inline image or linking
syntax (without escaping to raw HTML) either.  

You still have to *learn* Markdown.  The advantage (or rather one of the
many advantages) of Markdown is that you can use a text syntax (and a very
simple one at that) *or* escape to raw HTML.  Your choice, always.

The hard part is

A) designing a text syntax to handle the more common use cases, and that's
not *too* bizarre,

or

B) deciding "no more -- feature freeze!", and everything not currently in
Markdown from this point on is "You want it?  Use raw HTML."

I can live with the latter.  I'd prefer the former, specifically

1) common inlines (span, cite, sup, sub, abbr jump to mind)
2) class/id/title on those (and the currently supported inlines)
3) tables

Everything else of note IMO is already *in* Markdown text syntax.


/Jelks