> > First, not all browsers do this. Should people really have to
> > reconfigure their browser to visit your site? Second, this isn't going
> > to achieve the desired result. If an Opera user visits your site, do
> > you really want to send them a random choice between Netscape and IE
> > HTML, depending on what their User-Agent happens to be set to?
>
> Well, they ask for it, don't they?
> Come on, setting the user-agent to whatever and then complain that your
> browser doesn't support what's coming back?
> Like, asking for the French translation and then complain it's not
> English? ;-)
> Parlez vous pommes frites? ;-)
Most people do not make well-imformed choices on how to fake their
User-Agent field. They don't understand the subtle differences between
different browsers, and so they just try things until whatever web site
they are worried about works.
You suggest that browsers should emulate Netscape or IE depending on
what setting the user has chosen. While this would be nice, it doesn't
seem practical. These are *big* programs, and they aren't logical in
their responses--bug compatibility is a tall order. Furthermore, a
fundamental reason to use a different browser is because you didn't like
a behavior of one of the big two -- is it a good thing to emulate them
exactly, even if it is possible?
Really, I see the faked User-Agent headers as a feature to let people
deal with web sites that are too clever for their own good. When a web
site refuses to send you HTML because they "know" your browser can't
handle it, it can help to lie and to say you are a different browser.
This sounds more like an iffy last-ditch way to deal with mistakes, not
a mechanism that you want to rely on up front.
-Lex
