W3C home > Mailing lists > Public > www-html@w3.org > March 2001

Re: Make Microsoft follow the spec.

From: Scott E. Lee <sandman_001@netzero.net>
Date: Sun, 04 Mar 2001 04:53:51 -0600
Message-ID: <3AA21EBF.1A078477@netzero.net>
To: www-html@w3.org
Chris Wilson wrote:

> You hardly need to inform ME of that - since I've been participating in Web
> browser development for roughly eight years now, and have spent in ordinate
> and expensive amounts of time figuring out how Netscape implemented weird
> things in their browser in order to satisfy customers complaining that IE
> didn't work on their page (and then getting blamed, e.g., for font
> properties not inheriting through table cells, which was Netscape's design,
> not ours - which really chafes my shorts).  That is, however, a fact of
> life, and innovating features is a competitive sport on both sides as far as
> I'm concerned.  If you want Web pages that are accessible from a variety of
> devices, operating systems and browsers, then write to the standards; if a
> content developer is happy to limit their market to use more powerful
> features, then that's their business as far as I'm concerned.  Microsoft
> still participates in the DOM Working Group (among other groups), and tries
> to innovate in those bodies of work in what we feel are the right directions
> for our customers.

See, the problem I have with the last sentence of this paragraph is simple.  Why
"innovate" in the direction you feel is right for the customer and not in the
direction that is right for the internet?  While the direction of the internet
should be driven by the people that use it, some consideration should be given
to the people that actually write the documents that provide the information
that the users are accessing.  By "innovating" non-standards-compliant objects
in the most popular browser out there, MS is sending out a message that IE is
the only valid UA and documents containing proprietary code are "correct".  This
is a fallacy that the average user won't understand.  By "innovating"
non-standards-compliant objects, MS is again attempting to use its market share
to dictate to others.  "Innovate" as we do, or get squashed, or absorbed, or be
perceived as not being user-freindly.
</RANT>

> This particular discussion started with the fact that IE supports
> "document.all" - a collection of all the elements in a document exposed in
> the object model.  I would point out that IE shipped support for this
> collection in IE4 in September of 1997 - more than a year BEFORE DOM level 1
> became a W3C Recommendation (only a week or two after the _Requirements_ for
> DOM level 1 were published, in fact).  Indeed, the innovations in IE4 were
> certainly partly responsible - at Microsoft's urging, I would point out -
> for the formation of the DOM Working Group, and the eventual development of
> the DOM specification.
>
> If you think we are going to remove support of such widely-used object model
> from our implementation, then you are deluding yourself.  Microsoft argued
> long and hard in the DOM WG that the usefulness of document.all should not
> be ignored; regardless, it was not incorporated into the standard. Mind you,
> document.getElementById() does nearly the same thing - and we implemented
> that, along with a lot of other duplicate object model functionality, to
> provide standards support rather than just proprietary versions of the same
> thing (despite having implemented our own functionality long before).

And huzzah for IE.  But, by implementing document.all (even with the
document.getElementById()) IE allows hack document writers to exclude people not
using IE as their UA.  Is that good for your customer base?  Probably.  Is it
showing that MS is an inclusive company?  Hardly.

> -Chris Wilson
> PS - You should probably consider this diatribe to represent my personal
> views, not Microsoft's.

--Scott E. Lee
--Also my personal views.
PS-A bug is an unsupported "innovation" too.


Shop online without a credit card
http://www.rocketcash.com
RocketCash, a NetZero subsidiary
Received on Sunday, 4 March 2001 05:53:53 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 March 2012 18:15:45 GMT