Re: Content-Type (charset=EUC-KR) [was: rational exponentialisation problem of BABY RUDIN]
- From: Angus Rodgers <twirlip@xxxxxxxxxxx>
- Date: Thu, 10 Jul 2008 18:31:22 +0100
On Thu, 10 Jul 2008 11:10:48 -0500, galathaea
<galathaea@xxxxxxxxxx> wrote:
one that every newsreader has to handle in some form or another
is the Content-Type header
under which the charset variable may be set
the original article had charset set to EUC-KR
which correctly encodes all symbols in the message
and which all newsreaders i have read it on properly display
(mt-newswatcher, thunderbird, google groups)
so it is my belief that your newsreader is what is broken here
not the original post
[...]
the only section of the nntp protocol which i am aware of specifying character sets
is section 2.2 where it states the protocol directives themselves
(the commands sent over tcp/ip to control fetching of articles, search, etc.)
are in 7-bit ascii
if it were to be mandated that communications be in ascii
it would be the height of racism
that would mean no french or spanish or northern european or eastern european or any asian whatsoever
group could exist for communication in the language of those people
pretty much
it forces english and even english is limited by common symbols
i really think that it is time to start pushing the other way
maybe it was polite to try to stay on common ground way back in the beginning
where newsreaders were just being built
and no one wanted to assume large investments for entry into usenet
but now many free readers properly handle character sets
and the unicode consortium has generously given the world utf-8 as a great all-encompassing default
and i think the push to ascii has become regressive
all such comments only work to make communication on these groups
less natural and fluid
math has a language with many specialised symbols
the unicode consortium has added whole sections just for this purpose
i think it is time to jettison the broken and racist technology of the past
and use usenet the way it was designed to be used
[...]
It's not just Usenet and newsreaders, though. I may be a fool, but
I have got almost my entire life organised in terms of completely
standard and easily handled ASCII text files - a huge investment of
time! And even if (as might be sensible) I moved with the times, and
learned to use Unicode instead, isn't there still a loss of the simple
and direct tie with the keyboard (however "broken" and "racist" the
keyboard may be)? If we aren't going to dispense with keyboards (what
would you suggest to replace them?), isn't there still a place for a
character set which (pretty closely) corresponds to raw keystrokes?
(I'm not quibbling for the sake of it! I really am behind the times
in many ways. I haven't really got to grips with DVDs yet either, and
am worried about the future of my precious collection of videotapes.
Nor have I even converted my LP collection to MP3. So little time,
so little energy, must prioritise!)
--
Angus Rodgers
(twirlip@ eats spam; reply to angusrod@)
Contains mild peril
.
- Follow-Ups:
- References:
- rational exponentialisation problem of BABY RUDIN
- From: GabeCAI
- Re: rational exponentialisation problem of BABY RUDIN
- From: Angus Rodgers
- Content-Type (charset=EUC-KR) [was: rational exponentialisation problem of BABY RUDIN]
- From: galathaea
- rational exponentialisation problem of BABY RUDIN
- Prev by Date: Re: borel subgroups
- Next by Date: Re: proof for "symmetry"
- Previous by thread: Re: Content-Type (charset=EUC-KR) [was: rational exponentialisation problem of BABY RUDIN]
- Next by thread: Re: Content-Type (charset=EUC-KR) [was: rational exponentialisation problem of BABY RUDIN]
- Index(es):
Relevant Pages
|
