Hello All, Thank you Chris for your comments, encouragement, and willingness to assist in moving forward. I do appreciate your contribution and assistance, and would like to also apologize for any perceived lack of response recently. I have been very busy and involved in a number of activities here at our lab, thereby finding it difficult to take the time to comment and respond to the many messages circulating. I would like to take this opportunity to clarify a couple of comments previously made:
I will refer to the portion of the re-factored RNS that provides the POSIX directory-like functionality RNS-Directory, for lack of a better term at this point. In fact, the sense I got was that not only is it generally agreed that this type of thing is needed, but that it is needed as soon as possible; I certainly feel this way, since much of the work of the GFS-WG should ideally consist of the definition of a "profile" for GFS usage of RNS- Directory.
This is a fair statement, however it may be misleading in this way: RNS currently covers all POSIX like hierarchical directory management of references to endpoints as well as nested "directories". To say that "this type of things is needed,...as soon as possible" may be interpreted to mean "we don't have this and we need it as soon as possible." This is not correct, RNS has for nearly 2 years proposed a solution to this obvious need and has been soliciting assistance (regardless of working group affiliation) to shape and mold it into the specification that would most practically meet the essential needs of hierarchical resource endpoint naming and location registration within the context of the Grid.
document definition includes an "{xsd:any} *" component. I, for one, feel that WS-Directory will be a a valuable reference point in the development of the directory service component of RNS 2.0, or whatever it ends up being called.
I don't quite understand this comment. First, what aspect of WS-Directory will be a valuable reference? What does it have that RNS doesn't? Second, I do not consider it the "development" of "the directory service component" of RNS, its fundamental function *is* a directory service.
to come out of RNS and how it might need to be changed in order to
This I agree, the effort to "refactor" RNS includes taking certian features out and changing other features to accommodate the refactored structure.
get to this magical future invention I call RNS-Directory. I've
Again, this is an interesting statement that implies the "development" of something new. By definition this is not exactly consistent with the terminology implying refinement or refactoring.
I've offered to help in whatever capacity I can, including arranging for mailing lists and conference calls if necessary, in addition to offering my opinions, which by this point is probably assumed ;-).
Thank you again!
My personal feeling was that the most difficult issue to resolve was the question of where the re-factoring effort should live and how it should proceed; , but it seemed to be generally agreed that GFS-WG is
I'm not sure I perceive the same degree of concern. For this effort to be successful and satisfy the requirements presented by the OGSA-WG, OGSA-Data, OGSA-Naming, GFS-WG, etc. this effort must proceed with participation by *all* interested parties (individuals); this is not an effort of "a" working group, but rather a collaboration of interested parties. The "official" charter held responsible for the current phase of this activity is GFS-WG, but this *should not* impose any lines of distinction between those who are interested, willing, and able to contribute and participate. It is not an issue of "what group" is responsible, if this is to be a specification that will strive to serve the hierarchical naming needs of the diverse Grid community, then we need participation from this same diverse community (represented by various individuals who have voiced opinions and convictions in the matter). Therefore, I suggest that the "forum" of discussion appeal to a higher, wider, and larger audience than GFS-WG, whilst the process responsibility remain with GFS-WG until it becomes a GFD at which time the refactored generalization core of the spec is adopted by OGSA-Naming. Discussion and ownership of the "factored-out" pieces and features will need a forum for discussion, it might be most appropriate to begin in GFS-WG with this as well. Does this make sense? Best regards, Manuel Pereira ------------------------------------------------------ IBM Almaden Research Center 1-408-927-1935 [Tieline: 457] mpereira@us.ibm.com ------------------------------------------------------