Re: [cmis] Response to public comment on Paths

[Scott Malabarba <scott.malabarba@us.ibm.com>]

+1. Non-material.

A special case for the root path works.

We might also make a more general fix
(in italics):

A path for an object may be calculated
by taking the item's parent folder cmis:path property and append-
ing the '/' character, if the parent's
path does not have a trailing '/', and the object's pathSegment.

I don't know if the original intent
was to allow trailing /'s or not, but someone intent on defending them
could do so with a little semantic hair-splitting.
It wouldn't hurt to cover both cases.





From:      
 "Mueller, Florian"
<florian.mueller02@sap.com>
To:      
 David Choy <david.choy500@gmail.com>,
"cmis@lists.oasis-open.org" <cmis@lists.oasis-open.org>,

Date:      
 10/10/2012 03:55 AM
Subject:    
   Re: [cmis] Response
to public comment on Paths
Sent by:    
   <cmis@lists.oasis-open.org>

---

+1 for continuing the process.

Florian


From: Jay Brown <jay.brown@us.ibm.com>
Date: Dienstag, 9. Oktober 2012 22:42
To: David Choy <david.choy500@gmail.com>
Cc: "cmis@lists.oasis-open.org"
<cmis@lists.oasis-open.org>
Subject: Re: [cmis] Response to public comment on Paths

+1 for your suggestion:

  "... fix this bug; have the TC confirm the change being "non-material"; and then continue the process. "


Jay Brown
Senior Engineer, ECM Development
IBM Software Group
jay.brown@us.ibm.com

David Choy ---10/09/2012 12:47:25 PM---To all, We received the following public comment and we need to decide our action

From:	David Choy <david.choy500@gmail.com>
To:	cmis@lists.oasis-open.org,
Date:	10/09/2012 12:47 PM
Subject:	[cmis] Response to public comment on Paths
Sent by:	<cmis@lists.oasis-open.org>

---

To all,

We received the following public comment and we need to decide our action
quickly:

"2.1.5.3 Paths

The algorithm to calculate the path needs a special case for root (to be
in line with the example).
Otherwise the path of folder A would be //A.
An alternative more regular solution would be for paths of folders to always
end with a slash.
For example /, /A/, /A/B/
Although it is more difficult to spec to be compatible with 1.0."

Section 2.1.5.3 is meant to define Paths similar to conventional file system
paths. The apparent deviation reported in this comment indicated we have
a bug. My suggestion is to fix this bug; have the TC confirm the change
being "non-material"; and then continue the process. The recently
revised approval process allowed "non-material change", so our
delay would be just the editing time plus 1 week of electronic ballot.
However, if the TC failed to confirm that the change is "non-material",
then we have to re-start the entire approval process: first approve the
revised draft as a CSD, then approve it for public review and conduct a
new 15-day public review for the change. This would add at least another
month to the process besides editing.

An alternative is to keep the current description as-is. Then revise the
spec through an Errata after v1.1 is approved as an OASIS Standard. Time-wise,
an Errata for v1.1 won't come out until the end of 2013 or later. This
approach would reflect to the public a choice for expedience over quality
for v1.1.

Please reply to this email with your thoughts before Friday so we can decide
if we need next Monday's TC meeting to discuss this. If the responses seem
to agree, one way or another, we will proceed accordingly without further
delay. Please indicate if you consider such change being "material"
or "non-material", and which of the two ways suggested in the
public comment that we should use to fix this section (if you care).

Thanks.

David

---------------------------------------------------------------------
To unsubscribe, e-mail: cmis-unsubscribe@lists.oasis-open.org
For additional commands, e-mail: cmis-help@lists.oasis-open.org
[attachment "graycol.gif"
deleted by Scott Malabarba/Costa Mesa/IBM] [attachment "ecblank.gif"
deleted by Scott Malabarba/Costa Mesa/IBM]