Re: Slug header encoding

[ASAKURA Hiroshi <asakura@xxxxxxxx>]

Thank for your reply.

Bjo:rn Ho:hrmann wrote:
>HTTP uses mime word encoding in headers, you cannot correctly implement
>many core parts of RFC 2616 like the Content-Type header without
>support for mime word encoding.

What is "mime word encoding"? Does it mean a string such as 
"=?ISO-2022-JP?B?GyRCJDMkb..." ?

#If it is so, sorry, I can't understand above sentence. I think that
#Content-Type header doesn't use "mime word encoding". Could you tell
#me why?

Bjo:rn Ho:hrmann wrote:
>The content of the Slug header is not a URI, it would be confusing to
>treat it as such. As such I find your argument not very convincing.

Of course, the Slug header is not URI. But I'm looking for other 
encodings than RFC2047. Because as a reality there are many 
incompletely Mailer implementations using RFC2047 which influence
interoperability.

Bjo:rn Ho:hrmann wrote:
> This is a somewhat artificial distinction and argument. Just because
> many character encodings could be used, does not mean you have to
> support them all.

Sorry, my point and intention was not cleared. The 3rd reason is
unrelated to percent-encodings as said by you.
This reason relates only character set.

>Further, percent-encoding neither in IRIs nor in URIs requires the
>use of the UTF-8 encoding, it is very well possible that your 
>percent-encoding library uses ISO-8859-1 resulting in Slug headers
>that cannot be decoded. 

Yes. So, I think that we should decide encodings and character set
that can surely be recognized by all APP servers.
Sorry, I don't know the situation in Germany, do you have two major
encodings, iso-8859-1 and utf-8?
If the encoding that should be recognized by all APP servers is
assumed to be utf-8 like XML, I think it is good for interoperability.

Anyway, this issue is unrelated to percent-encodings.

---
ASAKURA Hiroshi / NTT Communications Corp.
<asakura@xxxxxxxx>, <h.asakura@xxxxxxx>
# Sorry, I don't know how to write "o:" character to this mail.