IRTF Open Meeting @ IETF-88 Vancouver, Canada 1420-1550 PST Tuesday Afternoon Session II State of the IRTF Lars Eggert 5+5 min Network coding proposed research group will be discussing chartering after this meeting. Mat Ford: Could we use a different acronym for the network coding resesarch group (NWCRG) to make it less easily confusable with the network complexity research group (NCRG)? Lars Eggert: suggest one! Mat Ford: I don't know - CODERG? Applied Networking Research Prize (ANRP) Award Talk *** IDILIO DRAGO *** for characterizing traffic and workloads of the Dropbox cloud storage system: Idilio Drago, Marco Mellia, Maurizio M. Munafo, Anna Sperotto, Ramin Sadre and Aiko Pras. Inside Dropbox: Understanding Personal Cloud Storage Services. Proc. ACM Internet Measurement Conference (IMC), November 2012, Boston, MA, USA. ?: clarifying question on slide 4 Lars Eggert: This mechanism seems quite ad-hoc and simplistic - did that surprise you? have you seen ongoing optimisations? ID: Yes, it's surprising. But they are managing to deliver the service regardless. It seems that they are learning how to improve as they go along - perhaps not surprising for an Internet start-up. Borje Ohlman: Can you say a few words about the business model? ID: I don't know about that. Jana Iyengar: Did you show how much of the traffic was peer-to-peer? ID: We don't observe that traffic, so we couldn't measure it. We tried to use folder names to infer that, and it seemed that the potential for using P2P is low, but I'm not very confident of that result. Jana: That's very interesting. So using a central repository seems to be adequate? ID: I think so. Measuring and Circumventing Internet Censorship and Control Nick Feamster Dave Myer: I understand you want reproducibility NF: Repeatability would be better DM: OK, repeatability. Is it isotropic - looks the same from everywhere? Do topology aspects have an effect? NF: Yes probably. That's why I said repeatable. We may see different things but we run the same measurement DM: So is a legitimate outcome that a censorship event appeared in one place and not in another, and then what does that say about the result. NF: It's posssible, but I don't have a good answer. THese results are a lot more fine-grained than you might be led to believe. LE: for some of these more researchy topics, the IRTF could be a home for some of these kind of responses to recent revelations of pervasive surveillance. It will need someone to push the creation of a research group, but it can be done if there's interest. Internet Research Grand Challenges Eliot Lear 10 min Edward Wallis: could be uses to match up research proposals and funders Stephen Farrell: don't believe researchers would post their proposals, but funders might support - just try it Matt Mathis: an excellent idea - long history of trying to make laundry lists of research problems - hard to answer when you don't know what the right question is - need a research problem canary EL: will you help me on this? MM: will add it to my long list Dave Taht: putting researchers and funders together it would help if the agreement was for the good of the Internet - resulting work should be openly available EL: purpose here is to provide a rendezvous point Robert ?: having sponsored research project, the hard part is that i don't have time to look for researchers or research projects, so idea of a kickstarter like project would probably help a lot christian huitema: i don't fund anything, you don't manage research projects, you manage researchers - you don't fund projects, you fund people because you believe they're good. that's a great idea, but it's a problem for people without a track record - so high value is to outline existing succcesful efforts that people don't know about. EL: i'm hoping i can also volunteer you to help me out on this Welcome by Lars. Invitation to join mailing lists call for 2014 research awards Review of the IRTF mtgs from this week. Concerns about IPR. A question about acronyms - network Complexity & Network Coding 9 chartered WG Closing SAMRG and RRG - would like to reboot RRG. One RFC published and one coming the ANRP award talk: 36 nominations, 4 awards questions: Lars: bytes being pushed by ad-hoc mthods. was this a surprise? ID: Yes, but the work is being done and people are tuning their protocols ??: Clouds/Data Centers for storage - what is their business model? ID: Don't knwo business models Jena: peer to peer traffic % ID: not measured due to instrumentation Jena: cloud may not need P2P Nick on perpass: 1st: measurement work that is WIP 2nd: Manipulation of data ?? : please define censorship? Nick: won't do it now, but will highlight characteristics Dave Myer: Want tests to be repeatable? Are the isotropic? Nick: Topology will have an effect, not be repeatable but reproducable Dave: Censorship may appear in one place/test and not another? Nick: No good answer ??: which destinations? Nick: a selection, alexia, external/internal, others Elliot & Challanges: Find me the Money! or... Find me the Researchers! ?? - a kickstarter for research projects? A list of interesting questions? Not so much. Add research proposals to answer the questions. Steve: not sure folks will post proposals matt Mathis: An excellent idea - a laundry list of problems - can't quite figure out the right questions - a research problem canary is needed Bernard?: if there was mutual ground/neutral it would be good Roberto: the hard part is time to do the matching, even if the Christian: can't underestand (outline existing successful efforts