How much specification do we need in the data table?

[adammontville]

On page 5 of the draft some examples of endpoint data are provided, and readers are directed to see "Data Attribute Tables and Definitions". There are many data attributes listed in that table, but some attributes seem too abstract. Specifically:

• Operating system attributes (e.g., version, service pack level, edition, etc.)
• Installed software attributes (e.g., version, patch level, install path, etc.)
• Other software configuration information

Are we expecting that the software identification draft will provide the necessary level of detail?

[djhaynes]

I believe that the more specific details for this would be captured in the IM.

[jimsch]

I would want this document to be sketchy about the details. I do have some potential problems with the question of should the details be in the IM documentor the SWID document, but given that we have currently stated that the IM is going to a single IM and it will be in a single document then it needs to be in the IM document. (Not a favorite decision of mine.)

I think that the SWID document is going to be slightly vague as well given that the final details are going to be in the IM document and that the publishing order may not be totally synchronized (especially if there are updates later). The IM document is going to have to be the final authority.