Net2 Meeting Notes from 01/30/2009
* Projects for 2009
* Razvan, JHU * Om, USC * Kaisen, UCSD
Om: Getting LEEP TEP comments soon. Zigbee and 6LP is mostly done. Steve is working on 6LP, Zigbee got transferred. Plans for Net2 projects.
OM: CTP works for LPL, but needs to work efficiently. More tutorial and debugging/instrumentation scripts that we want to do, but haven't had time to do. Plans for DIP and DRIP?
Kaisen: DIP and DRIP are in maintenance mode. No plans to work on it.
Om: No plans for MultihopLQI. One plan for 2008 was better documentation.
Razvan: Cleaned up instructions for Deluge. Missing ACKs slowing down the upload for Deluge. Look into that. ACK-related performance issues. Considering port of Deluge to IMote2. One problem is the USB bootloader.
Om: Nilesh is a student at USC. He is actively working with IMote2.
Om: About TYMO, nothing from Romain. My sense is we're not close to getting it released, but it would be nice to do it. It would be great to release IPv6 stack in 2009. Is there a name yet?
Razvan: The committed name is BLIP.
Om: Any other projects?
Razvan: Low-power probing.
Om: Is this Net2 item or Core? Related to LPL. Maybe agreeing on an interface. It will be good for Net2 to have two different link layers. People seem interested in traditional protocol like TYMO and TinyAODV, we should get one released.
Om: Maybe rearchitect BLIP code such that you can use any collection protocol.
Om: How about contributing Koala as an application? Koala as an example application for writing low-power applications and fast data transfer. Two categories of applications, regular transfer vs batch transfer.
Razvan: How do you discover network topology when it wakes up? Koala has extra support that CTP does not do.
Om: Is there any interest for point-to-point routing without necessarily using IP? As a general service, as opposed to being in Koala or Tenet.
Om: Most people interested in collection/dissemination instead of point-to-point routing. Let's think about it and make a decision. TYMO and TinyAODV seem to address the same problem space.
Kaisen: I think point-to-point protocols would be useful. Seems like people would use it for certain applications as opposed to general purpose dissemination/collection.