Re: Thoughts on CLASS and ROLE attributes in HTML 3 draft

Dave Raggett (
Thu, 16 Mar 95 16:13:01 GMT

KEY: > is Peter Flyn >> is Bill Perry

>> Definitely would be useful - anyone? I'm going to implement the <lang>
>> tag that is in there sometime soon, and a list would be most useful.

The language values are composed from the two letter language code
as defined by ISO 639 followed by a period, and then a two letter
country code from ISO 3166.

I could include these lists as an appendix in the Internet Draft
for HTML 3.0 - its already 177 pages long so a few more won't harm! :-)

>> This is a one-way hashing algorithm. Will always be 32 base-64 encoded
>> chars long. ie:
>> MD5("foobar") = 3858f62230ac3c915f300c664312c63f

> Where does a user get a standalone base64 binary that will hash a string
> argument so the result can be included?

The C code for the MD5 algorith is published as an RFC, I can't remember
which one off hand though.

>>> l. Does nowrap in <p> effectively imply auto-<br>?
>> God I hope not. I read it to mean that <p nowrap> is basically a <pre>
>> segment not in a monospaced font.

No! It simply turns off automatic word wrap for the current element, in
this case a paragraph. This is useful when you only want the browser
to insert line breaks at particular places designated by <BR>. The SGML
entity &nbsp; can be used to include non-breaking spaces, when word wrap
is enabled.

I am still uncertain as to whether the Netscape NOBR and WBR tags should
be added to HTML 3.0 or not. These provide an alternative approach to
use of &nbsp; and the nowrap attribute. What do people think?

re the RANGE widget for fill out forms:
> I would like to see that broken out into <input type=range min=1 max=10>

I decided against this to keep a reasonable bound on the number of
attributes for the INPUT tag. Perhaps we ought to consider breaking
INPUT into a number of different tags?

>> What happened to the 'named' base tags from the HTML 3.0 spec that was
>> distributed at the IETF meetings in San Jose? I thought it was very
>> useful, and it was trivial to implement.

The way I heard it, was that most people didn't understand the concept
or couldn't see the value it adds. I therefore decided to move it to
the other side of the chop line in the interest of getting a 3.0 spec
out in a reasonable time.

-- Dave Raggett <> tel: +44 117 922 8046 fax: +44 117 922 8924
Hewlett Packard Laboratories, Filton Road, Bristol BS12 6QZ, United Kingdom