[Odrl-version2] re: ODRL V2.0 - Model Semantics

[Steven Rowat]

Greetings,

Here are some comments on "ODRL 2 Model Semantics"

Overall, very impressive. A substantial increase in scope.  After reading it, I have only two main concerns; (and only the second of these is substantive on the content):

[2.0]. I have serious trouble following the visual lines in the main model diagram. For instance, under "Offer" you state:

"The Offer entity must contain an Asset entity, one or both Permission and Prohibition entities, and a Party entity with Assigner (rights holder) role."

However, when I try to follow this on the diagram, I at first see only one line, with no arrowhead, going straight down from Offer into Permission, and one line, with arrowhead, going from Permission into Asset. I see nothing to indicate that Prohibition and Party are indicated. Then, on second look, later, I've realized that it might be interpreted that there is also a line going across from Offer to Prohibition, but there's no way to know since there are so many lines superimposed there.

I strongly suggest that the superimposed lines be separated so they run parallel instead, so that they can be followed. Especially on that first set under the Rights types, there are too many possibilities.

[2.1]. An overall conceptual concern I have is that you have attempted too many levels of abstraction at once in attempting to include Transfer Rghts, Next Rights, and Tickets. This concern is in part a selfish one, because I am in this working group as a representative of those who will wish to sell their own works directly to the public (and which, as I've argued in several essays, I believe an important new possibility for the Internet and for society as a whole). Yet several of the levels of abstraction you have laid out in this model are targeted at corporations who will broker the works of others. I agree that these are welcome additions to anyone who would use ODRL if they work well. However, my concern is that ODRL runs the risk of becoming too complex and of attempting to be too many things to too many people.

Of course, that is exactly what ODRL is attempting to do - be a very high level language for the management of rights. I just hope it does not, like many other things in our world at present, fail *because* of its success at doing what corporations want.

[2.4] Note typo error: "...indicates the *grated* operation on the Target Asset." should be 'granted'  (Or, possibly, 'gated')?


Again, much thanks for this work.


steven rowat

p.s. I hope to see a browser that can read ODRL in...what, a week or two?  ;-)