Re: tlug: EUC & SJIS (correction)

["Stephen J. Turnbull" <turnbull@example.com>]

>>>>> "Klaus" == Klaus Kudielka <kudielka@example.com> writes:

    Klaus> Stephen J. Turnbull wrote:
    >> Fine if it's there.  If it isn't?  The point is that the
    >> administrator in question does not control the presence of the
    >> appropriate HTTP header.
    Klaus> [...]
    >> And the Content-Type charset parameter may be missing or wrong
    >> in the HTTP headers.  Your point is?

    Klaus> Again, the point is that a mirror should just re-transmit
    Klaus> the same Content-Type line (thus including the charset
    Klaus> parameter, if it exists). That way the user of the mirrored
    Klaus> page gets _exactly_ the same behaviour of the original
    Klaus> page, regardless of correct/missing/wrong
    Klaus> header/meta-tag. Whether the original behaviour is correct
    Klaus> or not, is not the business of the mirror.

I didn't say it was.  In context, I was explaining why it might not be 
possible to enforce directory organization according to content-type,
and thus that it might be necessary to have per-file information.  I
prefer that information to be included in the file itself where it is
less likely to get lost.

Ie, although using (and caching, if appropriate) the HTTP header is
the preferred mechanism, the META HTTP-EQUIV tag is a useful
alternative.  And a correctly specified META HTTP-EQUIV tag can't
hurt.

-- 
University of Tsukuba                Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
Institute of Policy and Planning Sciences        Tel/fax: +1 (298) 53-5091
--------------------------------------------------------------
Next Nomikai: 18 September, 19:30 Tengu TokyoEkiMae 03-3275-3691
Next Meeting: 10 October, Tokyo Station Yaesu central gate 12:30
--------------------------------------------------------------
Sponsor: PHT, makers of TurboLinux http://www.pht.co.jp