This wiki has undergone a migration to Confluence found Here
<meta name="googlebot" content="noindex">

Meeting Minutes-DSTU-Testing Team 4-23-2010

From HL7Wiki
Jump to navigation Jump to search


Attendees

Lenore Palma, Datafarm; Ragunathan Pathmanathan, Abbott; Scott Becker, Merck; Jill Geist, Abbott; Bill Friggle, Sanofi-Aventis; Harv Martens, Extedo; Rahul Mankad, Octagon; Jennifer Denomme, i4i; Kavitha Sivasubramanian, Independent

Discussion

Data Model Presentation

The meeting is scheduled for 11AM EDT on 4/28. Jason Rock of GlobalSubmit will present. There were two requests: a. Please show the relationship between RPS structures and those in the eCTD b. Please cover life cycle activities, particularly things like replace and append

Vendor-Sponsor Partnerships

No specific partnerships have been announced. Harv Martens of Extedo said that he has two clients, but did not have clearance to announce them.

It is recommended that sponsors contact the companies they wish to work with, or that vendors contact sponsors. Either the members of the Technical Team, or account managers, should be able to help get the relationships set up.

OIDs vs. GUIDs

Joel Finkle repeated the summary of the recent email conversation: HL7 is agnostic as to whether OIDs or GUIDs are used. The RPS team will likely make recommendations in the implementation guide similar to those made by SPL: Fixed items, such as lists of controlled vocabulary, are best served by OIDs – they’re well managed, and don’t require constant updating. Meanwhile, “unrelated” items such as files and context of use are best served by GUIDs. However, this will not be enforced: For instance, a regulatory agency may create their controlled vocabularies using GUIDs, and a sponsor with an infrastructure to support OIDs may wish to use them for identifying their objects. In either case, once an object is tagged with an ID, that ID should always be the only ID to refer to that object.