W3C home > Mailing lists > Public > ietf-dav-versioning@w3.org > April to June 2001

RE: label header (was: Re: Versioning TeleConf Agenda, 4/6/01 (Fr iday) 12-1pm EST)

From: Clemm, Geoff <gclemm@rational.com>
Date: Wed, 11 Apr 2001 08:11:28 -0400
Message-ID: <3906C56A7BD1F54593344C05BD1374B102A755D0@SUS-MA1IT01>
To: ietf-dav-versioning@w3.org
I agree that nobody is trying suggesting that we try and solve the
semantics of common version labeling requirements.

On the other hand, I believe it is a reasonable expectation on the
part of the user that if they declare a label in the English language
(and indicate as much with an attribute on that label), that this
definition be respected, and that a request for that label not come
back with some non-English byte string that happens to match.
Similarly, any other attribute declared on that label should be respected.

So unless we define an "XML attribute encoding" syntax for the
Label header (which I strongly advocate we not do), the Label
header will be missing key information needed for a correct
label match to be performed.

Cheers,
Geoff

-----Original Message-----
From: Tim_Ellison@uk.ibm.com [mailto:Tim_Ellison@uk.ibm.com]
Sent: Wednesday, April 11, 2001 4:21 AM
To: ietf-dav-versioning@w3.org
Subject: RE: label header (was: Re: Versioning TeleConf Agenda, 4/6/01
(Fr iday) 12-1pm EST)




I'll just restate my position for the record.

I want to keep the label header and don't think that labels should be
language encoded.  Using a well defined encoding we can match on labels
sufficiently well (canonicalizing if necessary) and that is all that we are
required to do.

The issue of two strings being interpreted differently in different
languages is bogus.  There are strings in a single language that are
ambiguous and I don't think anyone is suggesting that we try and solve the
semantics of common version labeling requirements!  Others have already
described situations of languages permitting variations on spelling etc. of
the same word.  We just don't need to go there -- a label is just a string.

The use of the Label: header is useful for operations on a single resource,
and invaluable for operations on groups of resources that are labeled with
a single label.

I'm in favour of keeping the label header as is.

Regards,
Tim
Received on Wednesday, 11 April 2001 08:10:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 13:57:41 GMT