FIA Principal Investigators' Meeting

October, 2012




Day 1 - October 1, 2012

9:00 Project overviews, with a twist: XIA, MobilityFirst, NDN

One member of each deep dive team is elected to give a 15-20 minute introductory presentation on the project, plus Q&A.  The idea is rather than just recapitulate the project slides, help craft a clearer message: how would you explain the project's goals and ideas to someone coming in fresh?  What did you learn/find surprising from the dive?

10:30 Break

11 - 12 Project overviews, with a twist: Nebula, ChoiceNet

Lunch: 12 - 1:30

1:30 Case study: web browsing

Dave Clark has prepared an overview and initial comparison of the technical solutions for the different projects, using web browsing as the case study.

3:00 Break

3:30 Scenarios 1+2

Each session has two topics, to be discussed in turn.  For each scenario, we will start with the five projects each presenting how their architecture handles the scenario in 2-3 slides (5 minutes each); we will then open up for Q&A to the presenters as a panel to discuss common themes and differences.   Our intent is to allow the time split to be flexible -- we do the presentations and discuss the first topic until we've reached the point of diminishing returns, and then move on to the presentations and discussion for the second topic.

Scenario 1: Mobility: a smartphone user walks out of Starbucks and starts driving.  What changes from the earlier picture?

Scenario 2: Phishing: what assurance does the user have that the data they receive when surfing is not spoofed?

5:30 Break

6:30 - 8:30 Dinner and Next Steps

A group discussion of next steps for the FIA program and the individual projects.  What would it take for your project to be successful, and what should NSF do to help?  An important question for NSF is how much time the next phase of work will take, how much funding is needed over that time, and what visible outcome can be expected.

Day 2 - OCTOBER 2

9:00 Scenarios 3+4

Scenario 3: Operations: How does the job of a network operator differ in your architecture from how it is today?

Scenario 4: Transparency:  With a search warrant for a router, what can you discover about an ISP's users?

10:30 Break

11:00 Scenarios 5+6

Scenario 5: DDoS: suppose you had 1M zombies, what is the easiest way for to disrupt communications in your system?

Scenario 6: Data: a million people with cellphone cameras arrive in Times Square for New Year's Eve.   They are all both content producers and consumers.  How does your architecture support massive and spontaneous traffic production and consumption, as well as the potential for added value network services in this setting?

12:30 Lunch

2:00 Scenario 7

Scenario 7: Byzantine resilience: if network state becomes corrupted, what is the worst case outcome in your system?

2:45 Comments and advice from external visitors

4:00 Wrapup



Hosted by the Computer Science and Artificial Intelligence Laboratory at MIT under agreement with NSF.
Contact David Clark <ddc@csail.mit.edu> with questions or comments.
Content of this page extracted from NSF FIA solicitation and press release. See links above for full text.