Re: Issue 6403: Enumertion Policy - some things for us to consider

Sounds like a good approach to me.
Katy



From:
Doug Davis <dug@us.ibm.com>
To:
Katy Warr/UK/IBM@IBMGB
Cc:
Christopher B Ferris <chrisfer@us.ibm.com>, Gilbert Pilz 
<gilbert.pilz@oracle.com>, public-ws-resource-access@w3.org, 
public-ws-resource-access-request@w3.org
Date:
22/04/2009 13:23
Subject:
Re: Issue 6403: Enumertion Policy - some things for us to consider




I agree that it would be odd to say "Optional=false" but I'm not 
comfortable with mandating that it always be "true".  If someone really 
did want to only expose the Enum operations then "false" would make sense 
to them.  Perhaps the easiest thing would be to simply show it as an 
optional attribute (as the proposal shows) but that we RECOMMEND it be 
"false" as to allow for other, non-Enum, interactions. 

thanks
-Doug
______________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com
The more I'm around some people, the more I like my dog. 


Katy Warr <katy_warr@uk.ibm.com> 
Sent by: public-ws-resource-access-request@w3.org 
04/22/2009 05:26 AM 


To
Christopher B Ferris/Waltham/IBM@IBMUS 
cc
Gilbert Pilz <gilbert.pilz@oracle.com>, public-ws-resource-access@w3.org, 
public-ws-resource-access-request@w3.org 
Subject
Re: Issue 6403: Enumertion Policy - some things for us to consider









Gil 

<gp>Maybe I'm missing the context, but (1) seems wrong to me (BTW - I'm 
assuming you meant "wsp:Optional=true"). Do we really want to constrain an 
endpoint to the operations defined in WS-Enum and no others? Why would we 
want to do that?</gp> 

In the example, I meant  "wsp:Optional=false", hence there would be only 
one policy alternative indicating that all communications with this 
endpoint MUST be enumeration. The point is exactly what you say: why would 
anyone use wsenp:Enumeration without the optional flag as it would be too 
restrictive?  I think that you would always want to add 
"wsp:Optional=true" to this assertion to provide the 2 alternatives as 
Chris says. 

We discussed this on the yesterday's call (I think you weren't on the call 
at the time).  Doug pointed out that the key point of this issue was to 
provide a way of specifying filter capabilities for enumeration and a 
starting point for a pattern to express capabilities.  The wider, more 
general discussions regarding indicating support for implicit operations 
and policy are covered in a couple of other issues [1] and (in order to 
move forward) we agreed to discuss these aspects in the context of these 
other issues. 

Thanks 
Katy 

[1] http://www.w3.org/Bugs/Public/show_bug.cgi?id=6694, 
http://www.w3.org/Bugs/Public/show_bug.cgi?id=6721 


From: 
Christopher B Ferris <chrisfer@us.ibm.com> 
To: 
Gilbert Pilz <gilbert.pilz@oracle.com> 
Cc: 
Katy Warr/UK/IBM@IBMGB, public-ws-resource-access@w3.org, 
public-ws-resource-access-request@w3.org 
Date: 
22/04/2009 04:13 
Subject: 
Re: Issue 6403: Enumertion Policy - some things for us to consider





wsp:Optional doesn't mean, "optional" it means that there are two 
alternatives, one with and one without the 
assertion containing the wsp:Optional attribute. I guess I am missing the 
point of what wsp:Optional has to 
do with there being some operations with and some without 
wsenp:Enumeration. 

Cheers,

Christopher Ferris
IBM Distinguished Engineer, CTO Industry Standards
IBM Software Group, Standards Strategy
email: chrisfer@us.ibm.com
blog: http://www.ibm.com/developerworks/blogs/page/chrisferris
phone: +1 508 234 2986 


From: 
Gilbert Pilz <gilbert.pilz@oracle.com> 
To: 
Katy Warr <katy_warr@uk.ibm.com> 
Cc: 
public-ws-resource-access@w3.org 
Date: 
04/21/2009 07:57 PM 
Subject: 
Re: Issue 6403: Enumertion Policy - some things for us to consider 
Sent by: 
public-ws-resource-access-request@w3.org






Maybe I'm missing the context, but (1) seems wrong to me (BTW - I'm 
assuming you meant "wsp:Optional=true"). Do we really want to constrain an 
endpoint to the operations defined in WS-Enum and no others? Why would we 
want to do that?

- gp

Katy Warr wrote: 

Just wondering whether we need to consider the following in the context of 
http://www.w3.org/Bugs/Public/show_bug.cgi?id=6403 proposal?: 

1) The wsenp:WSEnumeration policy indicates that WS-Enumeration MUST be 
used when communicating with this endpoint. I think this is shorthand for 
defining the data source enumeration operations at the endpoint (i.e. 
enumerate, pull, renew, getstatus and release).   
Because of the 'MUST', would it therefore be an error to attach 
wsenp:Ennumeration policy wsp:Optional=FALSE to an endpoint if that 
endpoint also supported other (non-enum) operations? 

2) We should consider whether an endpoint must advertise all the filter 
dialects that it supports using: 
<x:FilterDialect xmlns:x="xs:anyURI" wsp:Optional="true"/> 
If we don't mandate that all supported filter dialects are advertised in 
the policy, do we need a mechanism for an enumeration endpoint to indicate 
that it does not support filtering at all (i.e. including a filter on the 
enumerate verb will return wesn:FilteringNotSupported)? 

Thanks, 
Katy



Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU 








Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU 












Unless stated otherwise above:
IBM United Kingdom Limited - Registered in England and Wales with number 
741598. 
Registered office: PO Box 41, North Harbour, Portsmouth, Hampshire PO6 3AU