Re: SRC attribute for HR/UL/DIR/MENU?

MegaZone (megazone@livingston.com)
Mon, 17 Jun 1996 13:15:09 -0700 (PDT)


Message-Id: <199606172015.NAA27770@server.livingston.com>
Subject: Re: SRC attribute for HR/UL/DIR/MENU?
To: Daniel.Glazman@der.edfgdf.fr (Daniel Glazman)
Date: Mon, 17 Jun 1996 13:15:09 -0700 (PDT)
Cc: www-html@w3.org
In-Reply-To: <9606170922.AA04854@cli51ak.der.edf.fr> from "Daniel Glazman" at Jun 17, 96 10:22:49 am
From: MegaZone <megazone@livingston.com>

Once upon a time Daniel Glazman shaped the electrons to say...
>May I ask then why HTML is standardized ? Usually, standardization (or at leas
>discussion) precedes implementation, right ? The current process seems to make

Not in the real world.  Most 'standards' are implemented first to see if
they really work, then someone decides it would be a cool idea to make it
a standard - possibly it gets revised:

V.FC -> V.34
V.34+ MAY -> V.34bis
RADIUS (Remote Authentication Dial-In User Servers) (a comm server security
server) was developed in house here and used first, now is an IETF draft 
and used by many other companies.
GIF - CompuServ first, then it became a deocumented 'standard' for the net.

There are some standards that are ideas first - PNG is basically something
people thought was a good idea and decided to spec it out from the ground
up - but it is based directly on experience with the current image formats.
HTML has *always*, since day one, had implementations ahead of the spec 
which were later 'standardized' and documented.

It is like mountain climbing.  Occasionally you drive an anchor into the
rock to give yourself a firm foundation to call back on.  Then you climb
some more, and when you find another good level, you drive the next anchor.

Wha we have is a number of forces going in different directions, all making
new tags - and IMHO some of them are damn nice and some are fluff - and
from time to time we skim off the cruft and document how things are, and
then the next wave of creativity builds on that.  Without the occasional
standard level to unite browsers, we'd soon have a complete Tower of Babel
with proprietary and incompatible systems.  As it stands now you can write
with the reasonable assumption that if your code is solid HTML 2.0, or 3.2,
with or without extra tags, any browser will be able to deliver a base
rendering of the document.

This same thread has come up before - 3.2 was designed to provide a solid
base of current implementation NOT introduce new tags.  And the fact of the
matter is, NS had done more to introduce new tags to wide use than other
browsers.  So most of the 'new' 3.2 tags is documentation of NS tags that
other vendors agreed on, and some are M$ tags, and others, like tables,
are distilled from the full W3C spec.

With 3.2 as a base to build on, we can start working on improvements - 
such as SRC for lists, PLAIN, STYLE, ID, CLASS, etc and put them in '3.3'
or whatever.

-MZ
--
Livingston Enterprises - Chair, Department of Interstitial Affairs
Phone: 800-458-9966 510-426-0770 FAX: 510-426-8951 megazone@livingston.com
For support requests: support@livingston.com  <http://www.livingston.com/> 
Snail mail: 6920 Koll Center Parkway  #220, Pleasanton, CA 94566