(Yaay! You're no longer using koi8-r!)
On Mon, 2 Jul 2018 17:06:08 +0000, Seymour J Metz wrote:
>Implementing any new RFC is a big order. If the mail clients on both ends
>implement RFC 3676, then cut-and-paste works. If a brain dead application
>simply slaps in format=flowed without following the specifications in the RFC,
>then blame that application. Again, do you know of any problem when both ends
>actually adhere to the specifications?
>
o By "the mail clients on both ends" do you mean only the software, or do
you mean to include the human beings? It is not practical to expect human
beings to conform all requirements of RFC 3676.
o Is there an available validator and data suite for format-flowed for both
sending and receiving ends? Human eyeball validation is impractical and
error-prone. (And quis custodiet ipsos custodes?)
o If such a validator were to exist, to whom would one report deviations?
What's the realistic likelihood of resolutions other than "WAD"?
o Does RFC 3676 guarantee round-trip integrity? Ed should be able to
copy-and-past his Rexx snippet and Don to extract it, byte-for-byte
identical to the original. Otherwise, the specification is a broken design.
Format=flowed had two (more?) design objectives:
o Support for comfortable viewing on hand-held fondleslabs.
o Support for word processors which reserve CRLF for paragraph separation.
These are pretty irrelevant to computer languages. Esmie's JCL was illegible
to Lizette; Ed's Rexx was illegible to Don.
If format=flowed imposes no requirement on the sending human being, it
SHOULD be possible to move all the flow-function from the sending MUA
to the receiving MUA and the receiving human being should be given a
{FLOWED|FIXED} button for viewing and extracting code respectively.
"Save as File" would suffice if that file could be guaranteed identical to
what the originator attached.
Many user agents do not allow disabling of "format=flowed".
-- gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
