W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > January to March 2006

Re: Process on open RFC2518bis issues

From: Julian Reschke <julian.reschke@gmx.de>
Date: Wed, 11 Jan 2006 10:17:11 +0100
Message-ID: <43C4CD17.7080104@gmx.de>
To: Cullen Jennings <fluffy@cisco.com>
CC: WebDav <w3c-dist-auth@w3.org>

Cullen Jennings wrote:
> ...
>> OK, let me point to two specific ones:
>> Bug 188 (<http://ietf.cse.ucsc.edu:8080/bugzilla/show_bug.cgi?id=188>),
>> agreed upon on 2005-12-07
>> (<http://ietf.cse.ucsc.edu:8080/bugzilla/show_bug.cgi?id=188#c2>),
>> proposed text is in
>> http://greenbytes.de/tech/webdav/draft-reschke-webdav-rfc2518bis-latest.html#r
>> fc.issue.bz188>).
> This was one of the ones I was thinking about that seemed to require changes
> in many sections of the draft.

Three sections, basically. And the complete diff is available, it just 
needs to be applied.

> ...
>> OK, again two examples:
>> - The question of putting new requirements on ETags never was mentioned
>> in that original WGLC.
> It's very clear that the meaning of strong and weak ETags and how to
> implement them and what clients can rely on has been unclear to many, if not
> all, people for a long time. I'm not trying to put new requirements on
> ETags, I am trying to get ETags to a point where the WG agrees that
> 1) they are useful for clients
> 2) it is clear how clients can use them
> 3) and it is clear enough what they need to do that servers will implement
> them correctly

I think *that* summary makes a lot of sense. The problem is that this 
falls into the generic HTTP area. So adding non-normative text 
(appendix) explaining authoring scenarios may be a good idea, but 
forcing (!) servers to do things they potentially can't do won't work. 
In particular, requiring servers to do something where RFC2616 is silent 
about what it means is problematic, unless we can gather consensus for 
an RFC2616 erratum on the HTTP mailing list.

>> - The proposal to change the semantics for PROPPATCH (new issue 210,
>> <http://ietf.cse.ucsc.edu:8080/bugzilla/show_bug.cgi?id=210>) as far as
>> I can tell, even never have been raised *anywhere* until it suddenly
>> showed up in a recent draft -- and that's exactly the kind of change
>> discussion I'd like to avoid at this stage.
> Oddly, I view this as an example where I pretty happy with how things went.
> The editor of the document is writing down the details and considers the
> corner case where a client sent two contradictory statements in the same
> request (set X=1, set X=3), the editor decides to clarify this corner case

They are not contradictory. RFC2518 is clear about what it means 

"Instruction processing MUST occur in the order instructions are 
received (i.e., from top to bottom). Instructions MUST either all be 
executed or none executed."

> and puts in text suggesting that a client should not sent that. The draft
> gets published, some people read it and realize there may be reasons that a
> client does want to do that. Some quick discussion ensues and we decide how
> to resolve it. Presumably the editor will fix to reflect consensus on next
> draft.

I sure hope so, but the whole roundtrip could have been avoided.

> ...
>> Yes. But if A and B are acceptable, and A was in RFC2518 and this is
>> what people have implemented, why move to B? I'd really like to
>> understand that.
> I think you are missing my point on this one - If A and B are acceptable,
> and A is what people have implemented, I'm sure the WG would come to
> consensus on A. However, if neither A or B have consensus, then we don't
> have consensus. And if we don't have consensus, then we don't have consensus
> regardless of if the previous RFC did A,B, something else, or nothing at
> all.

Well, I *urge* people who want to make normative changes to RFC2518 and 
who can't get consensus for these changes to reconsider that change 
requirement, and to consider to stick with the original protocol. 
Otherwise it will indeed be hard to agree on a common document.

> ...
>> I think I clearly disagree here. If half of the people thinks the
>> original requirements (A) are just fine, and the other half wants to add
>> new requirements (B), going for (B) creates a spec that half of the
>> people aren't happy with and may not implement. Sticking with (A) on the
>> other hand means that we can go on as before, and those in favor of (B)
>> still have the freedom to come up with a way to describe (B), and let
>> clients discover that. Yes, that's a profile. Yes, I'm aware that
>> profiles should be avoided. But if the cost of not having the profíle is
>> that half of the implementors will either not implement the spec, and
>> implement it incorrectly (claiming compliance when they don't), that's
>> not better.
> Yes, and in that case you need to get consensus that going with A is the
> best path. I'm not disagree with you about why you might use the above
> argument in favor of A - that is fine, it is appropriate to make, and the WG
> might agree or disagree. But in the end I have to try and figure out if we
> have consensus for A, B, or no consensus. Lack of consensus for B will not
> imply a consensus for A. Only consensus for A will imply a consensus for A.

 From a practical point of view, I think we need to ask implementors of 
RFC2518 if they're willing/able/planning to implement RFC2518bis. If we 
can't get positive feedback from those implementors who *are* still 
active (so I'll explicitly exclude the Microsoft IIS developers here 
:-), then I think we have a problem.

>>>> Unless somebody can explain why it's likely that we can resolve these issues
>>>> between now and the end of the month, I propose to stop the discussion for
>>>> now
>>>> (as far as it affects RFC2518bis),
>>> Back in December I asked the people on the conference calls if we were going
>>> to be able to come to consensus on these issues. No one could guarantee
>>> anything but they all felt we could. Based on that, Ted decided to keep the
>>> working group open beyond the 2005. I sincerely hope we can come to
>>> consensus on these hard issues. However, I encourage people to come to
>>> agreement on something we can all live with - if that happens to be the same
>>> as what is in 2518, fine, but the idea that "if we can't agree, we use what
>>> is in 2518" is not how consensus works. Like all WG drafts,  we can't come
>>> to some form of consensus, it won't move forward.
>> Understood. So in this case, it would be nice if those who push for new
>> requirements for which there currently is no consensus is to re-consider
>> this.
> Yes, I agree. Perhaps I should say this in all caps or something but I
> really do agree - we need to be trying to shrink no grow the scope of the
> work. 

And that's why I was basically asking for a Feature Freeze.

Best regards, Julian
Received on Wednesday, 11 January 2006 09:45:46 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:01:35 UTC