W3C home > Mailing lists > Public > xml-dist-app@w3.org > January 2004

Re: Propsed new issue: variability of encoding in Miffy

From: Mark Nottingham <mark.nottingham@bea.com>
Date: Wed, 7 Jan 2004 09:31:21 -0800
Message-Id: <4EEB8F3A-4137-11D8-A34B-00039396E15A@bea.com>
Cc: noah_mendelsohn@us.ibm.com, Martin Gudgin <mgudgin@microsoft.com>, Anish Karmarkar <Anish.Karmarkar@oracle.com>, "Xml-Dist-App@W3. Org" <xml-dist-app@w3.org>
To: Amelia A Lewis <alewis@tibco.com>

Yes, q-p, 7bit and 8bit aren't really relevant; however, if the group 
disallows base64 in MIFFY, it won't be possible to use it in a 
non-binary-clean transport. There was talk of doing so; I agree that 
the most sensible thing is to just let it be (at least in MIFFY).


On Jan 7, 2004, at 9:30 AM, Amelia A Lewis wrote:

> I hate to break into the discussion, but I think that you guys may be 
> over-engineering.
>
> 7bit and 8bit are *text* encodings.  As such, they permit padding of 
> lines with whitespace, stripping whitespace, and conversion of line 
> terminators to NVT ASCII normal form (CRLF).  I do not believe that 
> you can safely convert binary to either encoding and back again 
> without the possibility of lossage.  base64 was invented to permit 
> transport of a complete eight-bit octet over a seven-bit transport 
> (gatewaying to five-bit transports or similar crude tricks remains the 
> province of unstandardized gateway-specific algorithms).  In other 
> words, I think that this is out of scope.  Although it's possible that 
> I've converted common practice into a requirement.  I do believe 
> someone should swot up the specifics before going further, 'cause I 
> think it's irrelevant to transport of binary materials.
Received on Wednesday, 7 January 2004 12:37:30 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 22:28:13 UTC