(I have directed replies to ietf-822(_at_)dimacs(_dot_)rutgers(_dot_)edu --
there is no
need to keep this blowing everywhere ..)
Excerpts from junk.info-mime: 16-Nov-94 Re: SWEDISH CHARACTERS IN E..
Peter Sylvester(_at_)inria(_dot_)fr (1556)
PS: There are people who think that 'backspaceA
where backspace is the backspace char gives you
a much better idea of an A with accent than =XY
True enough. There are also:
.... (fault-cases deleted for brevity)
Without a standard, you can't possibly make everyone happy. -- Nathaniel
Yes, and that's the problem of the unregulated combinations of
combining characters of ISO10646/UNICODE.
People who are satisfied with Latin-1 or other precombined
characters may not be able to notice the problem, though.
Masataka Ohta
And we whom fancy ourselves to be from the pragmatic
side of implementing things usually try approach:
One problem at the time
If the basic system is flexibly enough extensible, creating
such enchangements should not be very difficult, nor break
anything. (If not, the basic systems should be changed, which
often is not possible..)
If we want to make things new, how about creating a new set
of standards to compete with RFC-822 and X.400 email systems
about the popularity of the global email standard covering all
the possible aspects, and _Finally_Doing_Things_Right_ ?
Would it be necessary ? At my dark moments I feel like it.
Would such succeed to take over ? I doubt it...
Would it make sense ? For a purist yes, hardly for a pragmatist.
As I see it, what MIME gives us is a way to label things so
that we don't need to write extensive AI programs to GUESS
what some piece of email means (exotic charactersets, or
structures). It is entirely another matter HOW they are
encoded, largely these encodings are due to the need to be
compatible with existing old systems.
Things are well enough, when local users do not need to worry
about such trivialities, as encodings when exchanging email.
(Within one enclave, or a set of enclaves sharing similar
preferences -- national charactersets, for example.)
Things are fairly well, when people who can use (and understand)
the non-local stuff are able to handle such email -- I do know
Finns who read and write Japanese, though I am not able to do
it, still I have installed the software they needed.
A fact which should be obvious is that when an enclave wants
to represent something in their own characterset, and can do
it (like ISO-2022-INT allows), they are free to run a translator,
which does it. But please allow us narrow-minded latin-based
cultures to read our email in ISO-Latin series which have common
hardware supports in these parts of the world.
Things would be ideal when nobody needs to worry about encodings,
and could concentrate on the message contents, but the world
(and the systems in it) is not yet on that state, so shall we
quiet those useless flames, and be programmers/designers for
a change ?
/Matti Aarnio <mea(_at_)nic(_dot_)funet(_dot_)fi>