Rhys, I'm not sure on what basis you are making this assertion. Is it
because HTML is a less complicated markup language, hence more likely to be
successfully interpreted by a human reader in the absence of a mechanical
one, or is there some other measure that you're applying?
I don't think the complexity of the language has much of anything to do with
it. The real issue is how the language is *used*, which is a result of the how
the community uses the language rather than of the language itself. (I suppose
the language is indirectly involved, in that the nature of the language
tends to impose some selection criteria on the community that uses it, but
this is a second order effect at best.)
The legibility of messages presented without formatting capabilities is
entirely a function of how much care is given in producing the formatted
material in the first place. I've seen highly readable richtext, HTML, and for
that matter RTF -- so readable that nobody would have any reason for objecting
to it.
I have also seen completely unreadable richtext, HTML, and RTF. For that matter
I've seen a fair amount of plain text that is totally unusable in many
environments -- as a matter of fact Macintosh systems are one of the more
common sources of such material. I see ten times as many complaints about
unreadable plain text than I do about unreadable formatted material. I'm sure
this is largely a result of the overwhelming use of plain text in email, but it
does illustrate that you can turn out unreadable material in any format,
including no format at all. (And remember that in plain text there's no viewer
you can turn to that will fix it up for you.)
HTML is a language that it is often written manually, rather than indirectly by
a composition agent. It is also frequently read without a viewer. This has
compelled people using HTML to give some consideration to using it in a legible
way.
RTF, on the other hand, has always been envisioned as an interchange format
between applications. (It is also used in inconsistent ways by different
applications.) It is for all intents and purposes binary material, and in
practice often contains a bizarre extremely long and very short lines as well
as many other sins that make it impractical to read without a formatter.
Richtext, much as I hate to have to say it, despite being envisioned as a
readable format, has in practice fallen victim to poor composition agents that
produce fairly problematic output. Much of the richtext I receive I cannot read
without a viewer, and I know how richtext works!
Simpletext is as close as you can get to text and still have some markup
capabilities. Since I've never seen it in an operational context I cannot
say how readable it is in practice.
Ned