Notes on HTML3 draft

Maurizio Codogno (
Tue, 18 Apr 95 07:41:03 EDT

I spent part of the Easter long weekend to take a look at the HTML3 draft
by Dave Raggett (draft-ietf-html-specv3-00.txt of March 28), and I have
some doubt. Maybe they are already resolved, anyway here there are.

page 9, Character Sets

(yes, again!)
Since I have plenty of accents in Italian text, I'd like to write
"perché" instead of "perché" for example. HTTP is a 8-bit protocol,
so it should work, and HTML specs say nothing against it, until I use
ISO8859-1. But the character "é" is the same as entity é , so we fall
in the previously discussed problem. BTW, can I add to my document a
META charset directive to be sure that even if someone copies my files they
are correctly displayed with ISO 8859-1?

Body Element

I think it could be clearer if all attributes which remain the same for
most elements (ID, LANG, CLASS...) had a section just for them before
elencating the elements. At the element level, there would just be a link
to that page, unless there is some difference (for example, ALIGN).
In this way, both HTML and text presentation should be lighter, and the
differences in the meaning of the attributes depending on context would be
highlighted (footnote: Dave, where can I find the robot which converts
from a HTML set of documents to the text form, like you did with HTML3 specs?)

page 39, tab alignment

Instead that align=left/right, maybe align=here/flush is politically
correct for languages which write right-to-left.

page 80, columns in table

Could it be useful in some case to have short columns (because of small
width) and text clipped? Just a thought.

page 83, COLSPEC units

I am not sure whether units "relative" are in range 0.-1. or just whatever
relative range you prefer. Could you amend the definition?

page 98, SUB/SUP

I undertand that it is not possible to write X^a^^i^, but maybe it can be
allowed X<sup>a^i^</sup>, since context is now clear. Is it feasible to
amend the DTD to allow it?

ciao, .mau.