Re: Requesting a revision of RFC3023

John Cowan a écrit  :
> Rather than having thousands of ad hoc mechanisms for encoding declarations
> in each of the thousands of text formats now extant,

But Makoto is precisely suggesting that there should be a generic 
mechanism, usable with (almost) all formats.  Since most extant formats 
do not have a mechanism (viz. programming languages) they could all 
introduce the same.

> file systems should have
> a convenient mechanism for recording the encoding of each file,

Yes, file systems, and databases and all kinds of repositories, and all 
the connections between these (NFS, SMB, FTP, TFTP, DB connectivity 
protocols, etc.), all suitably and robustly configured and aware of each 
other, so that the charset parameter never gets dropped on the floor.

Would work in principle but it looks like Utopia.

-- 
François