Hi Fred, My comments inline <HK>. ---- Hiro Kishimoto Fred Maciel wrote:

Hi all,

...

A very good question! ("can you define the goals of the BOF more precisely other than "not WSRF"?) I sent this email in part to hear from the community how "broad" they might like it to be. That is, we're being flexible. If someone wants to talk about one of these specific topics, we will certainly try our best to accommodate.

I think that different people will see this from different angles, and if we don't understand the different "problems" involved the BOF can get messy. So let me try to decompose the problem, hoping that it will help the steering of the BOF.

(1) Are multiple base profiles OK to begin with?

...

Let me answer the question in a different way. GGF Chair Mark Linesch said [...]

Yes, but some of us (I included) think that multiple base profiles is a bad idea, no matter what Mark Linesch said. Suffices to say that this is a religious discussion, in which all sides have strong opinions beforehand, spend a long time in a heated discussion, but nobody is be able to change anybody else's opinion. The ones we had in the OGSA-WG weren't fun.

<HK> I agree multiple basic profiles are bad idea wrt interoperability and architecture perspective. However, OGSA-WG or GGF at large has agreed to keep our door open. This policy is written in "OGSA Evolution Policy Statement" which was reviewed in April. http://www-unix.gridforum.org/mail_archive/ogsa-wg/2005/04/doc00004.doc </HK>

Try to steer the discusion past that one (or around it), so that we can get to the next problem:

(2) How do the protocols sets in this area (the WSRF+WSN set, and the WS-Transfer etc. set) compare to each other?

That one should be an interesting one, especially to end FUD. I'm looking forward to what you have to say. That should give the background to:

<HK> If WS-Transfer, WS-Eventing and WS-Enumeration is mature and some SDO already starts (or plan to start) standardization process, it is the time to start anther basic profile discussion using these specs. </HK>

(3) What are the plans for the "WS-I-only" profile?

...

Here is the information that Steven Newhouse provided for the GGF organizers:

"A BOF meeting to discuss the creation of a WG to define an OGSA Basic Profile that builds upon a minimal set of simple web services. The output

of

This seems to be the main subject of the BOF, right? As you said, you are being flexible, and you have an interest in (2), but if we don't take care we won't even get to (3) in 1 1/2 hour.

<HK> You can create "WS-I-only" profile but it does not provide standard state handling mechanism thus it cannot establish interoperability. We've discussed this topic in the mail thread titled "Modeling State: Technical Questions" started by Ian Foster, didn't we? </HK>

Anything else?

Regards,

Fred Maciel Hitachi America R&D

Hi Fred, Yes, these are all very good points. I agree that SOME time should be spent on your question [1] (Are multiple base profiles OK to begin with?), but that if we are not careful, we will spend all of our time there, with perhaps little progress. Your question [2] (How do the protocols sets in this area (the WSRF+WSN set, and the WS-Transfer etc. set) compare to each other?) and question [3] (What are the plans for the "WS-I-only" profile?) are clearly the most important. I know that I will speak from first-hand experience and attempt to be as objective and realistic as possible with regard to the two stacks that we have used/implemented (WSRF "vs" WS-Transfer et. al). We will also discuss our initial explorations at interop between the two stacks. With regard to a "WS-I-only" profile, well, this is perhaps one topic in which differences of opinion might exist within the BOF organizers (again, this is fine -- this is exactly the type of thing that should be discussed at a BOF, right?). While I, personally, certainly respect the goals/intentions of a "WS-I-only stack" even "WS-I+WS-Addressing-only stack", I see no reason to preclude WS-Transfer/WS-Enum/WS-Eventing, if they are deemed to have important functionality for OGSA. I have no doubt that these will appear in standards bodies relatively soon, although I really have no idea when they might appear in some profile in WS-I (this is again a possible topic of discussion for the GGF BOF -- a perceived need for such a profile). We all want OGSA to succeed. It is important that everyone needs to keep this overall goal in mind when either thinking about the BOF beforehand or participating in the BOF. The overall goal of the BOF is, of course, to determine if the pursuit of one/more "non-WSRF" profiles will help OGSA succeed. -- Marty Marty Humphrey Assistant Professor Department of Computer Science University of Virginia

-----Original Message----- From: Fred Maciel [mailto:Fred.Maciel@hds.com] Sent: Tuesday, June 21, 2005 11:12 PM To: Marty Humphrey; 'Ogsa-Wg' Subject: RE: [ogsa-wg] OGSA-MWS-BOF at GGF14 on Tues June 28, noon-1:30

Hi all,

...

A very good question! ("can you define the goals of the BOF more precisely other than "not WSRF"?) I sent this email in part to hear from the community how "broad" they might like it to be. That is, we're being flexible. If someone wants to talk about one of these specific topics, we will certainly try our best to accommodate.

I think that different people will see this from different angles, and if we don't understand the different "problems" involved the BOF can get messy. So let me try to decompose the problem, hoping that it will help the steering of the BOF.

(1) Are multiple base profiles OK to begin with?

...

Let me answer the question in a different way. GGF Chair Mark Linesch said [...]

Yes, but some of us (I included) think that multiple base profiles is a bad idea, no matter what Mark Linesch said. Suffices to say that this is a religious discussion, in which all sides have strong opinions beforehand, spend a long time in a heated discussion, but nobody is be able to change anybody else's opinion. The ones we had in the OGSA-WG weren't fun.

Try to steer the discusion past that one (or around it), so that we can get to the next problem:

(2) How do the protocols sets in this area (the WSRF+WSN set, and the WS-Transfer etc. set) compare to each other?

That one should be an interesting one, especially to end FUD. I'm looking forward to what you have to say. That should give the background to:

(3) What are the plans for the "WS-I-only" profile?

...

Here is the information that Steven Newhouse provided for the GGF organizers:

"A BOF meeting to discuss the creation of a WG to define an OGSA Basic Profile that builds upon a minimal set of simple web services. The output of

This seems to be the main subject of the BOF, right? As you said, you are being flexible, and you have an interest in (2), but if we don't take care we won't even get to (3) in 1 1/2 hour.

Anything else?

Regards,

Fred Maciel Hitachi America R&D