I think that you might be interpreting the phrase "middleware- specific" incorrectly here. You are correct that use cases are generally provided by users, but seeing as how PGI doesn't have any users we could ask to provide use cases between now and tomorrow, we opted for the next best thing which is to ask the PGI group members (and OGF at large ofcourse) to provide some. The use cases are necessary here because without use cases driving requirements, it's hard to justify or even categorize whether or not requirements are actually required. Since the PGI group has already produced requirements, the belief is that the members of that group must have some use cases that they believe are driving the requirements that they came up with. In this context, the term "middleware-specific" use cases is misleading and incorrect. The point was rather that each middleware group (or pgi/ogf member) would supply use cases that s/he believes are important for PGI and that as a group we would collect these use cases together to form a use case document for the PGI group with which to drive the requirements process and to lend backing support to the requirements that we carry forward into the following standards phases. So, it's not that the use cases are specific to a piece of middleware, but rather that middleware folks represent a type of user (in addition to anyone else with uses cases) and in the absence of having real users to supply use cases, they are the next logical choice. -Mark On Jul 1, 2010, at 3:34 PM, Oxana Smirnova wrote:
Hi all,
I'd second this request - I learned that we are supposed to provide middleware-specific use cases (which sounds a bit odd for me), and I couldn't find out what is the status of requirements, and how do the use cases relate to them.
Regarding use cases: in my experience, they are provided by *users* (surprise, surprise ;-) ), and every middleware should be able to address the same use case(s). The proposed split of use cases per middleware suggests that middlewares are inherently non- interoperable because they address different use cases. Is this the new approach (no interoperability needed), or is the goal to find overlapping use cases?
Cheers, Oxana
01.07.2010 14:38, Balazs Konya пишет:
Hi PGI,
I heard from the nordugrid people that many things had happened during the productive OGF29 sessions. I was wondering if meeting notes containing decisions are available somewhere? Or the decisions are implicitly contained in the google document?
bye, Balazs _______________________________________________ Pgi-wg mailing list Pgi-wg@ogf.org http://www.ogf.org/mailman/listinfo/pgi-wg
<oxana_smirnova.vcf>_______________________________________________ Pgi-wg mailing list Pgi-wg@ogf.org http://www.ogf.org/mailman/listinfo/pgi-wg