Today we have a visit from EPSRC folks, so we're giving them an update on what we do - viz this intro:
anyhow, thought for the day: router vendors used to howl negatively about per-flow state back in the day. Now we have "carrier grade NATs" and people doing more and more weird things to TCP flows in flight (see
good imc paper on how much the middle boxes do
so if there's all this end system state being shared in the middle, why can't it be done for GOOD (e.g. multicast:) as well as bad (e.g. early acking TCP segments:)???
We are all social scientists now (but weren’t we born that way, anyway?).
The last two days, a number of us were at Hamed Haddadi's excellently run workshop on social networks - see the web for some more info (and slides/talks and
http://www.commnet.ac.uk/node/42
challenge discussions to appear later) - One we had involved the idea of scaling up social science research so it can meet the dynamic/complex systems/graph theory people half way (right now most anthropology studies take 3 years to study 8 people and involve video diaries and interviews etc etc, whereas most the graph analaysis work is glibly working away on all of 500M facebook users ,or 10M phone call records, or 5M foursquare users locations etc etc) - so we suggesetedi) social scientists study social scientists studying something, and we then automate them. ii) we figure out better sampling metholdoogies for taking groups of 8 out of 5M iii) we scale up NLP and vision processing to cope with automatic social science/ethnology on 5M users and iv) we figure out how to train up and  get all the users to study temselves and incentise them to join in (free phones, all you can eat data plans, or even just plain old fashioned money).
meanwhile, the ccle route from King's Cross to QMUL/Mile ENd road is a history lesson in itself (go down Gray's Inn Road til you hit High Holborn, and turn east. You're then on the A11, effectively pointed towards Ilford. The name of the road reflects the history of the medieval city of london, as you go through westgate, newgate (yes, the prison), ten Cheapside, Poultry, Cornhill, Leadenhall (road off to onesise is Threadneedle St) and Aldgate (was Aldersgate)....and on to Whitechapel - awesome - a much less dangerous and fun route back is the Grand Union Canal towpath, along the side of Victoria Park and up past Haggleston to Islington...(5.6 miles by runkeeper:)
I was wondering about synaesthesia and data representation - you could touch, smell, hear, feel, taste, and even predict 6 out of 8 bits - the other two would be parity, to be fair and not too odd:)
roundtable followup, 19.1.2012
we had a roundtable followup to the social nets discussion day -my 4 cents were
1. why don't we build more things to test our theorie?
2 . these things might be new socieities...e.g. fit-for-purpose replacements for socialism and free market capitalism
(cyber-anarcho-syndicalism)?
3. how about we have a new way to do inter-disciplinary research in this area, using social networks to do the research???
People need to read more - for example, when I look at The Rational OPtimist, by Matt Ridley, I discover that he seems completely unaware of the work by Marcel Mauss nearly 100 years ago on The Gift, which covers all the ideas Ridley thinks he's made up about exchange! I discover that people don't know the work by Mancur Olsen on the Logic of Collective Action from 50 years ago. Even people who've heard of Dunbar's number, don't read Gossip, Grooming and the Evolution of Language.
More essential readings
Michael Kearns work on Graphs&Games
http://www.cis.upenn.edu/~mkearns/
Edward Cherry, On Human Communication (probably out of print....)
Lots of John Doyle's work on Highly Optimised Tolerant systems....
etc etc - maybe I should write a book about it:)
some people asked about our (CL work in social nets in RL and online - there are various papers from the project (some by Dunbar et al linked at
http://www.social-nets.eu/publications.html
quick summary of afternoon sessions
1. social science is nearly as dismal as economics
a) they couldn't predict the arab spring
b) even given all the twitter data, they can't even _explain_ the past of the arab spring, let alone build a model that predicts the future
c) they don't bother looking at places where there WASNT an arab spring (e.g. EU countries like spain, greece, ireland, where occupy/indigandos movements were MUCH stronger and online, but have not led to any political change, in fact so far, probably the opposite:(
2. no-one has thought that the online social nets technology (and web in general) might actually make completely new forms of government possible other than socialist or capitalist democracy, religious states, or dictatorships (do I hear you say cyber-syndical anarchism?)
Doesn;t any one in this community think big (e.g. Kropotkin, Marx, etc) even terry eagleton who is about a million years old is smarter:)
3. I find it depressing that social scientists think academics do research for fame (i.e. impact) rather than an altruistic model, or even just plain curiosity...are they so reductionist that the world splits into commercial motives or personal ego-soothing ? maybe it does in their disciple - maybe that's why it  isn't really a discipline:)
proof of deletion
in between reading SOSP liveblogging notes, I'm still trying to think up how one might implement a "proof of deletion" service for cloud storage - here's the latest
a user stores data in the cloud - the data is encrypted so cloud provder cannot simply read it, but is amenable to privacy preserving queries on some keys.
the user wants to delete a record, contacts a third party (the grim reaper?), and gives then the keys of records. the third party tells the cloud service to delete the data. and then, using an anonymous service (via TOR etc) queries the record - they should get a 404 response.
of course, the cloud provider can squirrel data away but not in any useful way, as the TTP can do the query at any time
why ot just let the user run the query? well they might want to go away, and rely on the TTP who might also be persistent and might have bigger TOR guns....