RE: Interesting (gross?) use of CHANNEL

"Eric Burger" <[email protected]> Sun, 17 Mar 2002 23:24:09 -0500
Newsgroups gmane.ietf.imap-voice
Message-ID <[email protected]>
I have to hop in here:  Is CHANNEL sensible if the end-system does not get
"transformed" data?

Transcoding *HAS* to be a legitimate use of CHANNEL.

Filtering *HAS* to be a legitimate use of CHANNEL.

Why?  If the bytes/packets are not modified, why use anything other than
IMAP?  The WHOLE POINT of CHANNEL is to get the data in a "more useful"
form, possibly from an "easier" place.

I do agree that the semantic content of the data may not be altered.
However, almost by definition the client is requesting alternate format,
possibly done on the fly.  CHANNEL transformations are (1) by client request
and (2) do not change the semantic value of the data.

What is the basis for the objection?

--
- Eric

--------------------------------------------

Lyndon Said, and Pete Replied:
[snip]
>Yuk. No. Again, CHANNEL is NOT alowwed to modify the data content in ANY
way.

What data content is being modified here? You are simply sending the
data (unchanged) to another process which is going to do something
with it. In some cases, that will be "Play the sounds to the user
using a speaker". In my case, it will be "Send this data via SMTP".
What's the problem?

pr
--
Pete Resnick <mailto:[email protected]>
QUALCOMM Incorporated - Direct phone: (858)651-4478, Fax: (858)651-1102