ietf-822
[Top] [All Lists]

Re: SWEDISH CHARACTERS IN EMAIL: THE SUNET INITIATIVE

1994-11-17 02:04:18
Gentlemen!
  Could we 
        1. cool down the debqte
        2. most important, put all your highly interesting comments
                in their appropriate context

The discussion is wrongly comparing discussing 3 completely separate
issues (at least). That is worse that "apples and oranges", worse
even than "chevres et choux" (ie. goats and cabbages).

We have:
        - discussion of composite document structure and format
        - discussion of character sets for text (usability, eficiency...)
        - discussion of capabilities of MUA/UI and installed base
        (- not including  the real goal:  user desire/requirements :-(

MIME is a document structuring model and encoding.
  - per se it has nothing to do with charsets
  - it just enables to label text parts with appropriate alphabet
        indicators and to use a transfer encoding suited to the transport
        mechanism underneath
        SO it is just plain silly to
        - propose to avoid MIME by using a given ISO-xxxxx charset
                how do I send external body-parts, sound ... with ISO-xxxx
  - even if/when there is one single charset we will NEED MIME to
        separate body-parts, to know that these octets are text or
        images, that they are encoded or not

Charset repertoire issue
  [[ I have no special skills in this domain, just plain user
        requirements]]
  - there exist SEVERAL charsets today and people are using them
        -- thus a need to "label" them within documents
  - thr worst situation is to use something without informing
        -- this is true for the language, for the charset and for
        the encoding/representation/compression
  - We users need somthings which works, we don't really understand
        what are closed community (how do I know the recipeint
        belongs to my community sharing unwriten assumptions/choices?)
        WE know that you experts don't know what are these colsed communities
        and often don't know well what are our (end-user) needs.
  - We don't want any solution that is limited/closed, we don't want any
        solution that aks us to directly understand/Handle these issues.
        Please provide us with UA/UI (MUA/MUI when mail) able to do
        all this for us transparently
         --> this clearly goes for MIME, needed to enable the software
            to process automaticaly on the user behalf representation and
          encoding
  - we don't know which choice is more efficient easy to implement
        and we don't care, it is up to the implementors to decide
        BUT   we want a workable solution/product
        -- this leads to the next point capabilities/availability of
        the UA/UI installed based

UA/UI capabilities/availability
  - even if it is proven that ISO-yyyy is far superior to ISO-xxxx, as a user
        I will not want any MUA/UI which encodes all my outgoing mail
        within ISO-yyy is I know that 82.7 % of the recipient will
        be unable to proces/read this message
  - even if ISO-xxxx is proven and widely deployed, I will not want any user 
        that will not allow me to view the 13.6 % of message that are still
        using 1984, 1989, and 1994 best solutions
  - We (USERS) do need MUA/UI able to handle several char-set and encodings
        ie. all the one with significant use
  - the choice of the charsets used/recognized by the MUA/UI is a
        software supplier problem, they do have to study the effective market
        and make their proposals
  - the choice of the MUA/UI is OUR (user) choice and we inevitably will
        not buy the products from the suppliers that
          - pretend (true or not is not relevant) that they have
                the best (theoretical) solution
          -prevent us from communicating easily with all the other people
        This is not theory, the market punishes very hard all the
        ones that tend to ignore its pragmatic laws.

---------

This being said, I believe the discussion as a lot of interest,
if it is focused
  1.  MIME
        alternatives (to MIME)?  
        improvements?
        gatewaying  (eg. X.400 and/or proprietary things)?

  2. charsets
        which choices?
          existing practice must be covered
          but what is recommended? what is likeky to happen?
          will ISO-xxx Unicode... really deploy...

  3. today operational issues
        we have several charsets, several requirements
        do we need recommended guidelines?  (IMHO yes)
        which guidelines?
          how to facilitate/speed up migration toward better
          interworking practice (eg EMIL approach)?


---

  these were my 2 cents of "plain stupidity", "boring whatever",
        "brain-damaged stuff"....
        -- I take these too lines to self qualify my message, in order
        to save bandwidth as it seems a new etiquette of this list
        to honour contributors with "birds names".
        -- if some of you however insists in sending me that kind
        of comments, I am ready and prepared to receive and ignore them
        but PLEASE, do send these privately and not the 1000 of recipts

regards

-- PAP



<Prev in Thread] Current Thread [Next in Thread>