W3C home > Mailing lists > Public > www-validator@w3.org > March 2003

Re: validation in Opera

From: David Bryant <davidbryant@att.net>
Date: Fri, 14 Mar 2003 17:42:09 -0700
Message-ID: <3E7276E1.8020305@att.net>
To: Håkon Wium Lie <howcome@opera.com>
CC: www-validator@w3.org

Håkon Wium Lie wrote:

 >[snip ...]
 >Opera always adds the string "Opera" at the end of the UA string.
 >

Well, that's good to know.

 >Users sometimes need to change the identification to get into their
 >favorite web sites. For example, www.xbox.com refuses access to Opera
 >when it identifies itself as "Opera".]
 >

You're right. I just tried it out, and I couldn't get inside xbox.com
when I set the browser ID to "Opera."

On the other hand, once I was inside xbox.com, and had set up a user
account and all the rest of it, the browser started to hang. It looked
like the site stopped responding at certain points in the dialogue,
possibly because the server still recognized my browser as Opera. So
changing the preferences for browser identification got me part of
the way into xbox.com, but it didn't get me as far as the public
forums, where I intended to post a message asking about this
apparent discrimination against the Opera browser.

I will have to try that out some more with other browsers, etc, when
I have a little more time.

 > > How can standards be standards if we lower them just to
 > > let some software authors put a "cool" feature in their
 > > programs?
 >
 >Which standard are you referring to in particular?
 >
 >

Well, perhaps I misunderstood your original post on this subject,
Håkon.

I understood you to be asking for a special access hook into the
W3C validator that would _assume_ the presence of a <!DOCTYPE>
statement even when no such statement was present.

Since the use of <!DOCTYPE> is part of the enunciated standards,
I couldn't see why such an assumption should ever be made.

I probably spoke too brusquely, for which I apologize. The
standards are not up to me, at all. But I do have a lot of
software coding experience (mostly in IBM 360/370 assembly
language), and I have some very definite opinions on coding
standards.

Standards don't do anybody any good unless they're followed
religiously. Methodically. Continuously.

Sometimes standards need to be changed. I understand that. But
it did not, and it does not, appear to me that a standard which
says "Identify what language you are speaking as soon as
possible" is the sort of rule that ought to be bent ... not
even a little bit.  dcb
Received on Friday, 14 March 2003 19:44:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 25 April 2012 12:14:06 GMT