On 7 Jul 2001, at 18:40, Mike Andrew boldly uttered: 

> On Saturday 07 July 2001 15:28, Philip J. Koenig wrote:
> 
> > The problem is, regardless whether you change the User-Agent string,
> > there are fundamental differences in HTML parsing with particular
> > browsers so it still won't work right in many cases - ie because they
> > are putting code in the page that is proprietary to IE/Netscape, or
> > specific workarounds to things like CSS rendering bugs etc.
> 
> This holds true for the data being sent back to the browser, Ie, it's all 
> very well to 'pretend ' to be Netscape, but if you can't handle 
> whatever-it-is that Netscape does uniquely, then bang. Buit is it also fair 
> to say that the above is ALL that is required to phool a server into 
> believing it is a Netscape client, or does the server sniff elsewhere as well?


AFAIK, if you tell Opera to masquerade as say Netscape,
there is a portion of that string that still identifies
it as Opera. (same goes for IE's original aping of 
Netscape)

There are some various things that can be used to ascertain
the capabilities of a browser, many of them being javascript.
I know for example that you can determine the current screen 
resolution of modern browsers with javascript.  

MS has added all sorts of insidious things, active-x, cross-
frame scripting, and also including the ability of a website 
to GRAB THE CURRENT CONTENTS OF YOUR CLIPBOARD.  Be advised 
not to copy anything too personal to your clipboard if you 
use IE. (or at least turn it off, which is a security option)



--
Philip J. Koenig                                       [EMAIL PROTECTED]
Electric Kahuna Systems -- Computers & Communications for the New Millenium

_______________________________________________
http://linux.nf -- [EMAIL PROTECTED]
Archives, Subscribe, Unsubscribe, Digest, Etc 
->http://linux.nf/mailman/listinfo/linux-users

Reply via email to