Re: Browser sniffing for customisation

> 'browser sniffing' - the use of HTTP information to determine what User
> Agent is accessing a resource, and customising it accordingly, is already
> common in some areas such as language negotiation.

One needs to separate the build-in formal negociation aspect present
in HTTP 1.1, such as Accept, Accept-Language, Accept-Charset, which
obey a particular syntax, from the guess-work one can do using random
fields (e.g. does the User-Agent string contain the substring "Mozilla
3.*" and the Authorization field longer than 12 characters, and what
that means in reality: it probably supports JavaScript 1.017)

'browser sniffing' is just the second aspect, the former is just
regular negociation of file format, language, etc.

Received on Wednesday, 30 December 1998 09:04:50 UTC