W3C home > Mailing lists > Public > w3c-wai-er-ig@w3.org > June 1999

Re: An Open And Extensible Personalised Proxy Framework

From: Al Gilman <asgilman@iamdigex.net>
Date: Fri, 04 Jun 1999 09:58:42 -0400
Message-Id: <199906041353.JAA05258@relay.interim.iamworld.net>
To: WAI ER IG List <w3c-wai-er-ig@w3.org>
At 09:32 AM 6/4/99 +0200, Daniel Dardailler wrote:
>
>> In terms of client-side proxy filtering, I've been looking at the following
>> five-piece plan:
>
>My experience (working with other W3C groups, like Ecommerce) is that
>there isn't a portable solution for client-side proxy, and you have to
>adapt your setting and filter to IE, to Netscape, to Lynx, etc.
>
>For Lynx, for instance, there's a way to compile it (so not supported
>in every binaries) so that it recognizes some URI schema (say
>xhttp://) as special and runs a filter thru them. See lynx documentation.
>
>For Emacs/W3, you need to write you own piece of elisp.
>
>For Netscape, there's a panel (Automatic Proxy Configuration) that
>let's you play with client side script. (see
>http://www14.netscape.com/navigator/admin/v3.0/proxy.html)
>
>etc.
>
>It's too bad that there isn't a simple client side CGI interface,
>fired thru some .mailcap file, like for helpers, that would let anyone 
>install a filter on the client side.
>

Incidentally, this is the argument for relational programming or investing
in the overhead of a Problem Solving Environment framework.  In the
abstract you define once the outcomes in terms of the relationship between
UI behavior and the document content.  Then the PSE generates for you a
solution that delivers these outcomes built in the native resource
vocabularies of the different execution environments.

Al 
Received on Friday, 4 June 1999 09:53:31 GMT

This archive was generated by hypermail 2.2.0 + w3c-0.30 : Thursday, 9 June 2005 12:10:33 GMT