Dave Crocker <dcrocker(_at_)brandenburg(_dot_)com> writes:
When people first get access to mechanisms for formatting, fonts, etc.,
they tend to overuse it. Over time, they learn to be more judicious.
Yeah, that's what I said when HTML first gained widespread use and first
gained those capabilities.
I'm still waiting for the learning to be more judicious part. :)
Microsoft's generated HTML sets the font for practically every word. Some
of the people who write these HTML generators should be forced to write
several physics lab manuals entirely in WordPerfect's view codes mode. It
certainly taught me an attention to tag cleanliness that I've never lost.
*grin*
The idea that email content should be required to have less visual
control than a web page or a document simply doesn't make any sense.
It doesn't make sense to require it.
It does make sense to recognize that most e-mail is written quickly
without a lot of attention paid to design, and that therefore complex
visual layout languages are frequently inappropriate for e-mail in the
same way that one doesn't generally use FrameMaker to maintain one's
shopping list. Trying to shoehorn normal e-mail into a page design
paradigm seems very wrong to me, and the cause of a lot of the worst of
the brokenness of multipart/alternative as it's seen in the wild.
There's also the entirely practical and non-standards-based, but
incredibly persuasive, argument that something like 98% of pure HTML
e-mail is spam. Regular e-mail senders are, by and large, not using it.
This is creating a fast-growing barrier to *ever* using HTML as the
preferred markup language for e-mail; it's easy to get things into
people's spam filters, and then *extremely* hard to get them back out
again. Try to send HTML mail to many people these days and they may not
even know that you sent it, because the last 500 messages of that type
that they've gotten were unwanted and offensive and made naked women pop
up all over their screen until they had to reboot their computer, and they
got tired of checking for the to-that-point completely nonexistent false
positives.
The well is getting pretty thoroughly poisoned.
--
Russ Allbery (rra(_at_)stanford(_dot_)edu)
<http://www.eyrie.org/~eagle/>