Telecon on Monday July 30

[Andrew Buttner]

It looks like Monday July 30 is the best day for everyone for a telecon
regarding the language portion of the proposed 2.0 CPE Spec.  I have
attached the latest revision that merged in the comments that were
received.  Obviously there are still some open issues so additional
comments are of course welcome.

Call details are below:

Date/Time:  July 30, 2007 at 11:00 AM America/New_York
Length:     60  (minutes)
Meeting ID: 1462
Phone Number: 703-983-6338 in Washington, DC
              781-271-6338 in Bedford, MA
              866-648-7367 Toll Free

** Toll Free number is for North America callers only.  Other countries
must dial either the Washington or Bedford numbers to attend. **


Agenda:

* need for two different representations?
* richness of string representation



---------

Andrew Buttner
The MITRE Corporation
[hidden email]
781-271-3515

[Andrew Buttner]

Thank you to those that were able to attend the CPE telecon yesterday.
Here is a quick recap of what was discussed.


Attendees
--------------

- Andrew Buttner (MITRE - CPE)
- Jon Baker (MITRE - OVAL)
- Matt Burton (MITRE - CCE)
- Kent Landfield (McAfee)
- Dick Whitehurst (McAfee)
- TK (nCircle)
- Jim Ronayn (NSA)
- Dave Waltermire (Booz Allen Hamilton)
- Doug Noakes (Booz Allen Hamilton)


Notes
---------------

** The first item to discuss on the call was the need for two different
representations.  The current 2.0 draft has both a string
representation and an XML representation.

 - It was agreed on the call that there should only be one
representation since it would be difficult to normalized both.  Also,
having two representations will add work for developers as they might
need to implement both.

 - The feeling was to drop the string representation and keep the XML
representation.

 - MITRE will start a thread on the discussion list related to this
topic and make sure the rest of the community agrees.


** The next topic was in regards to the name attribute of the
<platform> element in the XML representation of the CPE Language.  This
attribute currently has no format defined.  It needs to act like an id
field to distinguish between the multiple <platform> elements in an
instance document.

 - The discussion came to the conclusion that the name attribute should
be a local id, and hence it just has to be locally unique and doesn't
need a robust format.  If a global id is desired in the future, we can
learn from our past use of the language and make a change in version
3.0 of the spec.


** The last topic that was discussed was the richness of the CPE
Language.  This was actually brought up in regards to the string
representation so with the removal of this, the topic was dropped.


** It was mentioned at the end of the call that the spec should do a
better job of capitalizing key requirement words like SHALL, MUST,
SHOULD, and MAY.



---------

Andrew Buttner
The MITRE Corporation
[hidden email]
781-271-3515