[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: updated document



Dan, A profile is just a document - as such it carries a version number.
The schema carries a Version qualifier on every class.  Also, the DMTF
Interoperability WG is defining a machine-readable version of a profile
(basically for testing and compliance) - which would capture
Mandatory/Optional classes, properties, methods and basic operations.
(BTW, by "operations" I am referring to the CIM Operations that let you
create, delete, enumerate, modify, ... classes and instances.) 

So, right now, our document can (and should) have a Version, our schema
classes will also have version info, and ultimately, the
machine-readable form of the profile will carry a version.  The latter
isn't there quite yet.

Andrea

-----Original Message-----
From: owner-nm-wg@gridforum.org [mailto:owner-nm-wg@gridforum.org] On
Behalf Of Dan Gunter
Sent: Wednesday, April 30, 2003 3:52 PM
To: nm-wg
Subject: Re: updated document


Yes, that helps.

But from this I take it that there's no "CIM way" of versioning 
profiles. So, maybe instead of the "Grid Middleware" profile it should 
be the "Grid Middleware profile, version 1". If others have a strong 
preference for adding a "version" field instead, I'm all for it. As long

as we can distinguish it from the inevitable version 2.

- Dan

Andrea Westerinen wrote:
> The profile reflects the CIM Schema classes - those are versioned and 
> the schema is what is sent "on the wire".  The profile tells you what 
> should be sent.  I think that the profile needs a version also - but 
> it is the requirements/compliance document and not the 
> schema/classes/data.
> 
> 
> Did this help or make things worse?
> Andrea
> 
> -----Original Message-----
> From: Dan Gunter [mailto:dkgunter@lbl.gov]
> Sent: Wednesday, April 30, 2003 10:37 AM
> To: Andrea Westerinen
> Subject: Re: updated document
> 
> 
> Andrea,
> 
> I'm not sure what you mean by "the schema", here. Is that a synonym 
> for
> "profile"? If not, is there any pre-defined way to put a version # on 
> the profile itself -- or should this just become part of the profile
> name?
> 
> - Dan
> 
> Andrea Westerinen wrote:
> 
>>I think that this is indicated in the schema.  Each of the classes
>>carries a Version qualifier. Andrea
>>
>>-----Original Message-----
>>From: owner-nm-wg@gridforum.org [mailto:owner-nm-wg@gridforum.org] On
>>Behalf Of Dan Gunter
>>Sent: Friday, April 25, 2003 7:42 AM
>>To: nm-wg@ggf.org
>>Subject: Re: updated document
>>
>>
>>I have nitpicky versioning question for the group, and of course for 
>>poor beleaguered Andrea.
>>
>>Let's assume that this profile has a version number (how is that 
>>assigned? 1.0?). Then I think it's reasonable to expect that the
> 
> example
> 
>>delay.roundTrip that Brian shows in the document, when it actually is 
>>sent "on the wire", will carry this version number with it -- how this
> 
> 
>>happens is up to the implementation.
>>
>>Where do we indicate this version number? In the profile itself? In
>>the
>>example delay.roundTrip measurement, as well? Somewhere else..?
>>
>>- Dan
>>
>>Brian Tierney wrote:
>>
>>
>>>I made some updates to the document, based on the last phone call:
>>>
>>>http://www-didc.lbl.gov/NMWG/NMWG-profile.html
>>>
>>>Andrea: there are a couple CIM questions in there that hopefully you
>>>can answer (just search for "Andrea")
>>>
>>>Also, a reminder that we have another call next week, usual time.
>>>
>>>------
>>>
>>>Call: 510-665-5437
>>>
>>>When: April 30, 2003, 08:00 AM America/Los_Angeles
>>>
>>>Meeting ID: 5217
>>>
>>>-----------
>>>
>>> 
>>>---------------------------------------------------------------------
>>>-
>>>--
>>>-------------------
>>> Brian L. Tierney,   Lawrence Berkeley National Laboratory (LBNL)
>>> 1 Cyclotron Rd.  MS: 50B-2239,  Berkeley, CA  94720
>>> tel: 510-486-7381    fax: 510-495-2998   efax:  240-332-4065
>>> bltierney@lbl.gov   http://www-didc.lbl.gov/~tierney
>>>
>>
>>----------------------------------------------------------------------
>>--
>>
>>
>>
>>>------------------
>>
>>
>