Re: distinguishing browser types

Brian Gaines (
Sat, 22 Apr 1995 14:26:52 +0500

> Bill Perry writes:
> The main problem is not that they set the accept headers wrong (I don't
> think netscape does - are you sure its broken?).

One has to continually check such things version by version!

I just sent the server a request from Netscape with several additional
MIME types and helpers registered. The request came through as:-

GET /RealEstate/JohnMary.html HTTP/1.0
User-Agent: Mozilla/1.1b3 (Macintosh; I; PPC)
Accept: */*
Accept: image/gif
Accept: image/x-xbitmap
Accept: image/jpeg

That is, none of the helper MIME types are listed.

The logic being used seems to be to list only the types that Netscape
itself can display.

It would seem more appropriate to regard helpers as extensions to a
browser and list helper MIME types also.

This will certainly be necessary once we have Open Doc or OLE browsers
where the helper presentations are embedded and indistinguishable from
directly supported types.

It is particularly frustrating with Netwscape currently because its open
architecture enables the helper application to monitor and control
multiple windows and to integrate hypermedia links between them. It is
then a challenge to send clickable maps instead and emulate as much of
the more complex behavior as possible, rather as Lynx does an incredibly
good job of coping with forms.

It would be good to get all the mechanisms for content negotiation
cleanly specified and supported, and helper applications should be a
significant consideration in doing this.


Brian Gaines Knowledge Science Institute, University of Calgary Calgary, Alberta, Canada T2N 1N4
tel: 403-220-5901 fax: 403-284-4707