- graap-wg - lists.ogf.org

telecon on 4/19
by Jim Pruyne 19 Apr '06

19 Apr '06

We will have a telecon. on Wed. morning/evening. Dial-in numbers will be the same: 9:00AM Central Daylight Time US (UTC-5) which should be: 10:00AM Eastern Daylight Time US 1500 UK 1600 Germany 2300 Japan 2100 Thailand Phone Number: 866-673-8466 in the US. 702-477-6031 for those outside the US. Conference code #8578310.

1 0

minutes from 4/14 telecon
by Jim Pruyne 14 Apr '06

14 Apr '06

They are attached. --- Jim Minutes from the April 7, 2006 GRAAP Telecon Attendees --------- Toshi Nakata Jim Pruyne Heiko Ludwig Wolfgang Ziegler Discussion ---------- - Spreadsheet row 23: Heiko will look into adding some text to clarify that when there are no service terms, there can also be no guarantee terms since the guarantee term must refer to a service term. - We determined that we do not need the pseudo-schema in 7.2 and 7.3 as it is not consistent with the real schema, and is covered in the 9.5.2 section on the service state information. - Spreadsheet row 25: Consensus is that would we don't feel like handling the complication of both parties providing a service in the first spec. - Spreadsheet row 26: We'll use Karl's proposed text in the intro. to the appendix as it does seem to explain our intent. - Spreadsheet row 27: Our intent will be to use string for Name type fields, and NCName for Id type fields. We realize this may cause some difficulty with current, though early, implementations. Heiko to make this change after verifying that it doesn't cause significant difficulty in the case of their implementation. - Spreadsheet row 28: Agreed, we should, and we'll watch for those recommendations. - Spreadsheet row 29: Agreed, changes have been made. - Spreadsheet row 30: We've made both Name and ServiceName indicated as MANDATORY. For Name, we did this because they service description terms are referred to by name in other places. We don't feel like tackling the issue of cross-agreement consistency w.r.t. the service names. - Spreadsheet row 31: We agree that it is redundant, but we like having the clarifying text here rather than requiring the reader to refer back to the partent type. - Spreadsheet row 32: We'll use the Name and Id types we agreed upon in row 27. - Spreadsheet row 33: Agreed, and done. - Spreadsheet row 34: Agreed, and done. - Spreadsheet row 35 & 36: Have already been addressed.

3 3

extra telecon on 4/14
by Jim Pruyne 14 Apr '06

14 Apr '06

We will have a telecon. on FRI. morning/evening. Dial-in numbers will be the same. The time in the US will remain the same, but the US has changed to daylight savings time over the weekend. I believe this gives us the times: 9:00AM Central Daylight Time US (UTC-5) which should be: 10:00AM Eastern Daylight Time US 1500 UK 1600 Germany 2300 Japan 2100 Thailand Phone Number: 866-673-8466 in the US. 702-477-6031 for those outside the US. Conference code #8578310. --- Jim

1 0

minutes from 4/12 telecon
by Jim Pruyne 12 Apr '06

12 Apr '06

Are attached. We'll meet again on this Friday for an extra session trying to complete the spec before the next GGF. --- Jim Minutes from the April 7, 2006 GRAAP Telecon Attendees --------- Toshi Nakata Jim Pruyne Asit Dan Wolfgang Ziegler Chris Dabrowski Discussion ---------- - GGF17 sessions are all on May 10th starting in the afternoon (1:45PM), and running into the evening. - Spreadsheet row 17: We do not see the need for the symmetrical discovery of guarantee terms by service as it can be derived. Can you provide us a use case to help us see the need? As for the structure, comment, we agree, though we do provide some structure for alternatives. We do assume that all terms are non-optional, though in the case of guarantee terms business value can make them optional to a certain extent. We do support some higher level language to be used in terms. - Spreadsheet row 18: We consider the agreement state and the service state to be independent. For example, agreement creation does not necessarily mean that the service is created. Thus, it does not seem possible to have a state transition diagram that joins the service state with the agreement state. To do this, we would also need to link the agreement to the service when the service is being accessed, and we have not specified that at this time either. We have added text to make it clear that the service states are only applicable when the agreement is observed. - Spreadsheet row 19: Customer obligations can come from the template, or can be added by the initiator in to the offer, but all of the initiator obligations are in the offer. As for the qualifying conditions, the customer obligations (e.g. request rate) or the non-controllable ones (e.g. time) specify when the guarantee is applicable. ** Meta-comment, there are excellent questions from Shamima, and we should capture these answers in a FAQ sort of document. It can make the work we're doing here to answer these in detail more useful and more visible. - Spreadsheet rows 20 & 21: We need further clarification (a use case?) on this comment as we would hope that this is already addressed in the spec. - Spreadsheet rows 22 part 3, and row 24, and row 35: Should we reference XQueryX at all? The consensus is no. Jim to remove all references to XQuery.

2 1

telecon on 4/12
by Jim Pruyne 12 Apr '06

12 Apr '06

We will have a telecon. on Wed. morning/evening. Dial-in numbers will be the same: 9:00AM Central Daylight Time US (UTC-5) which should be: 10:00AM Eastern Daylight Time US 1500 UK 1600 Germany 2300 Japan 2100 Thailand Phone Number: 866-673-8466 in the US. 702-477-6031 for those outside the US. Conference code #8578310.

2 1

minutes from 4/5 telecon
by Jim Pruyne 10 Apr '06

10 Apr '06

Are attached. Note that the next call will be on Friday, 4/7, at the usual time of day. Reminder announcement to follow. --- Jim Minutes from the April 5, 2006 GRAAP Telecon Attendees --------- Toshi Nakata Jim Pruyne Chris Dabrowski Asit Dan Wolfgang Ziegler Heiko Ludwig Discussion ---------- - On updated termination diagram: * Asit to send a note on the mailing list to verify that we really have a transition from PendingAndTerminating to ObservedAndTerminating. None of us seem to be able to come up with a message sequence that would cause that to happen. * Asit will do an additional diagram to show the initial states that may be seen. - Spreadsheet row 10: Heiko to validate the schema as it relates to ServiceTermState and GuaranteeTermState. Heiko to post those changes back to the mailing list, and we'll determine follow-up steps on the next call (which is on Friday). - The next call will be on Friday, 4/7, same time of day!

6 13

Minutes from 4/7/06 telecon
by Jim Pruyne 07 Apr '06

07 Apr '06

Minutes are attached. We will have another call at the usual Wed. time next week, and continue the extra Fri. calls at the same time until the next GGF. --- Jim Minutes from the April 7, 2006 GRAAP Telecon Attendees --------- Toshi Nakata Jim Pruyne Asit Dan Wolfgang Ziegler Discussion ---------- - Termination and state diagram have been agreed upon. Toshi to add some minor wording to introduce the bullets describing the states. - Spreadsheet row 12: Negotiation is outside the scope of the current document, this has been put into the gridforge tracker. - Spreadsheet row 13: We would like to have some more structure here, but that's been considered outside the scope to this point. Asit will look in to providing a simple structure which we can consider including, but it will not change the idea that this is extensible as it is currently. - Spreadsheet row 14: We need to add a couple of lines making the relationship of the template to the final agreement clear. Asit will add those lines, and send to me. Jim will update comment tracker as it relates to the advantageous of the provider stuff. - Spreadsheet row 15: We believe that the reply to row 14 will also clarify this. - Spreadsheet row 16: Jim to write response in the comment tracker, Asit to provide a sentence clarifying the one time vs. on-going fulfillment.

1 0

extra telecon on 4/7
by Jim Pruyne 07 Apr '06

07 Apr '06

We will have a telecon. on FRI. morning/evening. Dial-in numbers will be the same. The time in the US will remain the same, but the US has changed to daylight savings time over the weekend. I believe this gives us the times:: 9:00AM Central Daylight Time US (UTC-5) which should be: 10:00AM Eastern Time US 1500 UK 1600 Germany (I believe this is the time now that US & EU are both on daylight) 2300 Japan 2100 Thailand Phone Number: 866-673-8466 in the US. 702-477-6031 for those outside the US. Conference code #8578310. --- Jim

1 0

telecon on 4/5
by Jim Pruyne 05 Apr '06

05 Apr '06

We will have a telecon. on Wed. morning/evening. Dial-in numbers will be the same. The time in the US will remain the same, but the US has changed to daylight savings time over the weekend. I believe this gives us the times:: 9:00AM Central Daylight Time US (UTC-5) which should be: 10:00AM Eastern Time US 1500 UK 1600 Germany (I believe this is the time now that US & EU are both on daylight) 2300 Japan 2100 Thailand Phone Number: 866-673-8466 in the US. 702-477-6031 for those outside the US. Conference code #8578310. We'll attempt to complete the discussion of the terminate based on Toshi's drafts, and continue with other open issues. We'll also discuss whether people will be available at the same time of day on Fri. to continue discussions. --- Jim

2 1

minutes from 3/29 telecon
by Jim Pruyne 29 Mar '06

29 Mar '06

They are attached. We'll meet at the "usual" times next week. Usual is defined w.r.t. the US. However, the US is changing to daylight savings time over the weekend, this should make it one hour earlier in most other parts of the world. Further, there is some interest in holding additional phone conferences during April to try to make sure we're completed by GGF in Tokyo. If we do this, I would propose the same time of day as usual (i.e. 9AM Central time in the US), but on Friday's. I realize this makes it late night in Asia on a Friday night which might be considered unattractive. Please give feedback on the need for additional calls, and on preferred days/times for them. --- Jim Minutes from the March 29, 2006 GRAAP Telecon Attendees --------- Toshi Nakata Jim Pruyne Asit Dan Heiko Ludwig Wolfgang Ziegler Discussion ---------- - Next week's call will be at the same time in the US although daylight savings time will have started. Jim will try to update the reminder properly. - On terminate: We discussed the state transitions associated with the terminate operation thanks to Toshi's contribution. We concluded that the current model is somewhat mixed in that the "terminating" states are considered sub-states, and "terminated" is a base-state. The two approaches seem to be to either make everything related to termination a sub-state, or to make everything related to termination a base state. * The sub-state approach simplifies clients that are not concerned with termination-related states, and allows some state transitions, e.g. Oberserved::Terminating -> Completed which we may actually want to permit. It also allows some transitions, e.g. Pending::Terminating -> Observed that we do not want. * The additional base state approach means clients must explictly manage the termination-related states, but it makes various transitions more explict, and helps avoid undesired transitions. It complicates the case where an agreement completes normally while in a terminating state. We'll address this by allowing a domain to sub-state Terminated to indicate a normal completion prior to termination completion. Sub-stating Terminated may be something many domains want to do anyway to indicate additional information about the agreement's state post-termination (e.g. rolled back, etc.). The groupd decided upon introducing new base states. Toshi will write this up, and Jim will review, particularly w.r.t. the descriptive text related to Terminated sub-states.

2 1