Agenda for CCIT Meeting, 20100216

Invitees

CCIT, developers, Data Conservancy

John Cobb <cobbjw ornl.gov>
Roger Dahl <dahl unm.edu>
Michael Daigle <daigle nceas.ucsb.edu> (local to Santa Barbara)
Hilmar Lapp <hlapp duke.edu>
Jeff Horsburgh <jeff.horsburgh usu.edu>
Jim Green <jgreen iiaweb.com> (will be calling in. Marratech? EVO?)
Matthew Jones <jones nceas.ucsb.edu>
Giri Palanisamy <palanisamyg ornl.gov>
Paul Allen <pea1 cornell.edu>
Robert Waltz <rwaltz utk.edu>
Ryan Scherle <rscherle nescent.org>
Robert Sandusky <sandusky uic.edu>
Mark Servilla <servilla lternet.edu>
Todd Vision <tjv bio.unc.edu>
Dave Vieglais <vieglais ku.edu>
Bruce Wilson <wilsonbe ornl.gov>
Carl Lagoze <clagoze gmail.com>
Nick Brand <brand nceas.ucsb.edu> (local to Santa Barbara)
Tim DeLauro (funded by Data Conservancy)

Topics to be Addressed

The following topics needs to be addressed during the meeting.

Review of implementation progress. The overall state of progress for implementation of the CN, MN, and ITK client libraries. Comparison with timeline, overview of next implementation steps.
Review of activities and interactions with EOT.
Review of activities, progress and feedback from the EVA working group. This is directly relevant to the Workflow working group.
Mechanisms for incorporating feedback from the EVA and other working groups. CCIT is focussed on development and implementation activities. What is the mechanism, process for incorporating suggestions from that group into the design and implementation of the cyber-infrastructure.
Review of process for communications, tracking, and assignment
Clarification of the functionality to be implemented between now and the end of June
Roles and responsibility for the semantic groups for for the 0.x, 1.x, and 2.x implementation series
Resolution of any technical or other impediments to progress on implementation of CNs, MNs or other pieces of infrastructure. Some sub-topics:
- System metadata object class
- Versioning for schemas
- URIs for identifying object classes used in DataONE
- Mechanisms for node discovery. For example, a client should be able to use a single well known address to discover the locations of coordinating nodes and member nodes. Something like "cn.dataone.org"
- Mechanisms for introspection. A client (and other nodes) should be able to discover interfaces and capabilities available at a particular node.
- DNS operations. Currently single point of failure, no support for dynamic DNS or update by node administrators.
Query types supported by the CN.search operation
Design and implementation strategy for federated security
Specifications for implementation of three additional types of MN
Implementing the necessary logging and reporting capabilities to ensure that the operation metrics are available
Determine testing strategies for loading, failure, deployment, data consistency, querying, replication
Identify functionality, components, tools, packaging and distribution for the Investigator Toolkit
Design web user interface for system
Administrative procedures for CNs and MNs. Who has root access, point of contact at each institution, responsibilities for hardware maintenance, operating software and CN stacks
Assignment of responsibility for oversight of major components
Operations of the supporting infrastructure: dataone.org, trac.dataone.org, repository.dataone.org
Student project ideas (#216)
Identify additional and update existing risks
Selection of replacement Federated Security WG lead

Schedule Summary

The schedule is generally flexible except for four events requiring external coordination. These are in bold in the daily schedule, and include the overview of EOT meeting outcomes, the discussion with the EVA working group and discussion with the workflow group during Tuesday morning, and the discussion with the federated Security working group leads on Wednesday after lunch.

Note that Wednesday lunch should start early to ensure everyone is back in time for the call which starts at 1pm sharp.

Tuesday

(AM) Roadmap, progress, general state of things
(AM) EVA working group activities, Workflow working group activities
(PM) The investigator toolkit. Identify technologies to support EVA + workflow group, software that needs to be implemented
(PM) Web user interface design for system (CN)

Wednesday

(AM) Introduction to Data Conservancy project
(AM) DataONE - Data Conservancy interoperability discussion
(AM) Coordinating Node implementation issues and next steps
(PM) Discussion with Federated Identity WG leads
(PM) Planning and design for implementation of authentication and federated identity with emphasis on short term (year 1-2) goals
(PM) Metrics, logging, monitoring, privacy.

Thursday

(AM) MN implementation, with emphasis on integrating the various target member node types
(PM) Administration policies and procedures for CNs, MNs, software deployment
(PM) Other work that needs to be completed.
(PM) Finalizing actions, assignments etc for next couple of months.

Day 1 - Tuesday 16 Feb

08:00:	Introductions, meeting overview, agenda edits. Project overview. Quick introduction to DataONE CI with focus on year one activities and goals. General overview of CI development progress (Dave)
08:15:	Status updates for MN service (Roger), CN service (Rob), CN stack + deployment (Matt)
09:00:	Update of EOT activities, outcomes from Feb 2-4 meeting. (Steph)
10:00:	Break
10:15:	EVA update (Steve and Bob, via Marratech). Background information.
11:00:	Workflow group update (Bertram, Marratech). Draft workflow charter.
12:00:	Lunch
13:00:	Investigator Toolkit. Identify software components that need to be added to the investigator toolkit to support the science use case(s) of the EVA group and the Workflow group, plus any other initial use cases that need to be supported.
15:00:	Break
15:30:	System user interface. Discussion about the UI to be presented as part of the DataONE infrastructure. Focus on the 0.x release, ideally progressing through 1.x. Wireframes. Information flow.
17:00:	Review of day, actions, assignment
17:30:	Close

Day 2 - Wednesday 17 Feb

08:00:	Overview of Data Conservancy goals, approaches and architecture (Carl)
08:35:	Interoperability between DataONE and Data Conservancy (Matt moderates)
10:00:	Break
10:15:	Coordinating Node implementation issues encountered and plans for next stages of development. Plans for progression through version 0.6 - 0.7.
11:45:	Lunch
12:55:	Federated Security discussion with WG Leads (Von and Randy, remote). Draft federated security charter (1, 2),.
14:00:	Identity, authentication and authorization plan for prototype and progression to version 1.x
15:30:	Break Student discussions.
16:00:	Logging, monitoring requirements to support metrics needed for reporting to sponsor, the system administrators and the system users. This may also touch on privacy concerns.
17:00:	Review of day, actions, assignment
17:30:	Close

Day 3 - Thursday 18 Feb

08:00:	Member Node deployment to support the four types of MN (Metacat, Dryad, ORNL/DAAC, CDL Water Resources / Herbarium data + images)
10:00:	Break
10:15:	Member Node deployment to support the four types of MN (Metacat, Dryad, ORNL/DAAC, CDL Water Resources / Herbarium data + images).
12:00:	Lunch
13:00:	Administration policy and procedures for CN and MN operations. Mechanisms for node discovery. For example, a client should be able to use a single well known address to discover the locations of coordinating nodes and member nodes. Something like "cn.dataone.org" Mechanisms for introspection. A client (and other nodes) should be able to discover interfaces and capabilities available at a particular node. DNS operations. Currently single point of failure, no support for dynamic DNS or update by node administrators. Assignment of responsibility for oversight of major components Operations of the supporting infrastructure: dataone.org, trac.dataone.org, repository.dataone.org Student project ideas (#216)
15:00:	Break
15:30:	Additional discussion on implementation aspects for CNs, MNs, UI design, ITK or other topics that need further discussion.
17:00:	Actions, assignment, scheduling for remainder of year.
17:30:	Close