- From: Lisa Dusseault <lisa@osafoundation.org>
- Date: Tue, 28 Dec 2004 09:57:56 -0800
- To: Stefan Eissing <stefan.eissing@greenbytes.de>
- Cc: Elias Sinderson <elias@cse.ucsc.edu>, WebDAV <w3c-dist-auth@w3.org>
No, I certainly don't think that changing 3253 is necessary or even
desirable. Nor do I think it's necessary or desirable to block quota
on this issue, no matter which way the draft has it.
(and to be pedantic, it's not a consistency of terminology, it's a
consistency of naming style)
Lisa
On Dec 28, 2004, at 1:54 AM, Stefan Eissing wrote:
> Are you proposing to change RFC 3253 to get a consistent terminology?
>
> Am 27.12.2004 um 22:50 schrieb Lisa Dusseault:
>
>>
>> I generally agree that specification writers should re-use
>> terminology consistently -- great sentiment. However that isn't even
>> what we're talking about here.
>> - This isn't a case of re-using any terminology -- the issue is a
>> set of new error codes with a different naming style.
>> - We could go for a consistent style with RFC3253, however I found
>> the RFC3253 style confusing and prefer the quota style as-is
>> - The style used in the quota draft is consistent with RFC2616 style
>> of describing errors in text. E.g. HTTP generally describes the
>> error ("NOT FOUND") rather than the precondition ("RESOURCE MUST
>> EXIST").
>>
>> So it's RFC3253 that made the departure in style.
>>
>> Lisa
>>
>> On Dec 24, 2004, at 9:06 AM, Elias Sinderson wrote:
>>
>>>
>>> Julian Reschke wrote:
>>>
>>>> I'm convinced that re-using RFC3253 terminology in an inconsistent
>>>> way is a Very Bad Idea [...]
>>>
>>> Reusing /any/ terminology in an inconsistent way is a bad idea -- it
>>> leads to confusion among implementors, makes the ensuing discussions
>>> overly cumbersome, etc.
>>>
>>>
>>> Elias
>>>
>>
>>
>
>
Received on Tuesday, 28 December 2004 17:58:14 UTC