Friday, March 28, 2008

things any good (CS) lab should do

1. encourage almost ALL phd students to go on internships to industry labs at least once and possibly twice

2. have a really wide net for visiting talks which is broad church (include industry and social aspects)

3. involve PhDs in teaching and paper reviewing

4. take students and post docs out for dinner from time to time, and to the pub a lot

5. teach phds about processes (papers, PCs, conference management systems, grant applications) viz
http://www.cl.cam.ac.uk/~jac22/talks/jon-cfip.ppt

Friday, March 21, 2008

piety - irrational numbers and social network rganisation

pi is an interesting number - so is that fact that social groups are structured around tiers of trust, centered on each person with nearest trust group being kith and kin, and then
close friends etc...the degree of this graph is something like 3.x (e.g.
3.1415957...)
somewhere at 5 hops (3.x^5) you reach the limit of human's ability to infer intentionality

this tells us that a social networking tool (TM) should be built around making levels
and group at each power visible and should change how it supports people outside of 5 hops and groups larger than 150 or so...

i think such a tool could be called piety since that has PI in it and should be easy peasy to use

ageing (churn in friendship groups) could determine when fols are evicted from the close circle of friends into the outer darkness of cyberspace (or facebook:) and
frequency and duration of contacts (including email/IM, co-location) can be used to include people in...

coud make a nice startup:)

Thursday, March 13, 2008

references for money, and conference reviews for fun

1. so here's my idea of reference writing

I only every write 3 references - the person asking me for one as to pick one of my pre-canned references, and I will only sign if it what they pick matches my belief about them - oh, and i charge 1 euro (stable currency of choice)

2. here's my idea for conference reviewing and PC management

we have PC meetings after all the reviews are written - how about we haver the meeting just after all the papers are submitted and (modulo conflicts) the pC meet to choose who reviews what - this
a) calibrates them
b) socialises them
c) load balances (from each according to her needs, to each according to his means)

then we dont need another meeting - everything else can be done online

3. choosing PCs - easy - ask potential PC members (mostly from pool of PC members from this or related conferences in previous years) to furnish th PC chairs (and conference organising ctttee if there is one) with 2 or 3 example reviews of papers accepted in previous years - these are used to judge if the potential PC members understand the job (fairness, accuracy, etc).

4. alternative, use my reference scheme above - there are only really 9 different kind of reviews (7 if you are in CS theory) so we publish those, and authors submitting papers pick which reviews they think it, and the PC reviewer signs it if she agrees, and doesn't if he disagrees.

Tuesday, March 11, 2008

More on Mobile Social Networks

so lots of people think social networks are worth lots of money

and lots of people think mobile phones are worth lots of money

what can you do with a mobile phone in terms of social networks

1. obvious (I wrote this ages ago) is run a browser (flock) on yr smart phone and go to a social net site
2. obvious (everyone wrote this a zillion years ago) - run a location service and map people's geo-relationships to each other or to locations or both (see MIT reality mining, Intel's project doing location service using just about anything a few years back etc etc)
3. do pocket switch net and decentralised link discovery (non obvious, but we did it in pre-haggle project work about 4-5 years back) and build temporal graph in a distributed way, then use it for
i) establishing relationships (co-lo duration and frenquency above some threshold)
ii) to actualyl forward info (person to person or pub/sub, via percolation, gossip, epidemic, but filterered by relevance (e.g. interest/tags), and by trust/tribal/dunbar/closeness etc) to get efficient delivery and focus...

This is gonna be the topic for the Million People project, and is discovery (not invention) and therefore not patentable:)

clever algorithms to make it work relaibly are more interesting though...

Thursday, February 28, 2008

patention deficit disorder

Just been reading (prompted by last months CACM article) about peer to patent:

http://www.eupaco.org/
and
http://dotank.nyls.edu/communitypatent/

seems a fairly sensible movement, but really lacks teeth ...there's some
big company guns behind more large scale US federal changes

Microsoft:
http://www.microsoft.com/presspass/features/2005/mar05/03-10PatentReform.mspx

and Cisco:
http://www.cisco.com/web/about/gov/markets/patent_reform.html


viz...costs keep mounting just to fight the mountains of nonsense...
and there are some recidivists
http://www.computerworld.com/action/article.do?command=viewArticleBasic&articleId=9025841

Kind of old news...but I am catching up....

for me, the key thing is that
a) software patents are silly - we have software copyright, and algorithms are discoveries and should be in the public domain as they are knowledge that underpins many things as in mathematical theorems, whereas the effort lies in turning them into useful computing products and that is supported adequately by copyright for software and patent for hardware
b) all patents should have
i) use it or lose it clauses based on reality check - does someone intend to
do something other than hold other people to ransom with their invention?
ii) lifetimes that reflect the domain of activities typical time to product, service, and profit - for example, in Internet time, this might be about 5 years - in Big Pharma, maybe 10. In vacuum cleaners, perhaps 7.

c) Clearly patent applications need testing properly by motivated experts - one trick would be to tax patent lawsuits (say 10%) and use the tax to allow the patent office to hire patent reviewers (see peer to patent above).

d) intellectual property is not a good - it has some rather different flavour; as with spectrum, and other new aspects of 21st century life, we need the law to move on from simple notions of property and commons to some wider range of notions....

to be continued... ... ...

Sunday, February 24, 2008

squeezy throbbing phone skins - new idea for staying in touch

So phones vibrate - so do game console controllers - but what if they provided a "remote touch" mechanism? how hard could this be?

one end simply senses a few "pressure points" - the other end turns this into
the opposite force - a first cut at such a system would be a phone that you could "squeeze" and the phone at the other end of a call just expands a bit - so if someone has the phone in their hand, it "feels" like the remote person is squeezing their hand.
A more complex system would put the phone in a glove and have full on remote robot glove controller sensor and motors (actually you just need motors - in reverse, they work as sensors so it is really really simple to build) - there's a little bit of concern about feedback and the amount of force so the system would need to be sloppy
but that isnt too hard to design.

I wonder if there's anything out there one could adapt to this right away?

Monday, February 11, 2008

Small Worlds and Flat Earths

Reading Flat Earth News which is much better than yesterday's guardian review would have it, but not perfect - one interesting item was the use of plagiarism detection software to track where news items came from and how many of them were basically direct or close transcriptions of wire feed (AP, Reuters) stories - would be Interesting if Google were to build on this - some stories come from PR, some from real reporters, some from the public - a tool that provided a provenance tree for online information should be something easy to knock up - sort of a backwards in time, patient-zero finder, that allowed you to see if somethign was cooked up by a government Press Officer, a company
publicity dweeb, or had some basis in the real world (or even more than one).


This would also obsolete reportage:_

You could call it "newsrank", since basically it is just souped up pagerank....

feel free to pay me a small royalty for the bad idea

Monday, February 04, 2008

SybillRus - a new idea for anonimity in online social networks

oh, ok, so it aint that new an idea, but I think there is a business case for a company that provides ready-made identities (in large numbers) that look plausible, have existing interest, group membership, friendship groups etc, and slot in to facebook or myspace or any of the other social networking environments (cyberspace, we used to call it:) - recently I noticed from some friends that their facebook pages had inaccurate birthday dates -this turnbs out to be deliberate on their part (a mechanism to mitigate identity theft).

So sybilRus.com would provide lots of these - then the bad guys wouldn't know who was real and who was virtual

This idea, or something like it a lot, is also in Ben Elton's latest SciFi novel, Blind Fait, (in my opinion, a return to form by this amusing, contentious, but thoughtful satirist)...it also has some very funny reality-tv extremism...

Thursday, January 31, 2008

Tiling Docking (Real and Virtual) of iPhones/Touch screen gadgets

So here's an idea I don't want anyone to patent, so I am publishing it right here, right now.

If you have a household of n people all with mp3 players with displays and phones with displays, why not have a dock for them that tiles them?

so the idea is you have n areas on a panel, and people drop their devices in their area - when all the devices are there (esp. if they are say htc tuch or iphone) you have a larger continuous area of (touch) screen - actually quite high resoluton too - so it could be your tv too...or your computer display...

you could coordinaet this either via the mirousb, or wifi or bluetooth....

it could work approximately so crowds of people could hold up their iGadget at a sports event and do a mexican wave without moving...

Tuesday, January 29, 2008

google crime wave

an interersting feature of large systems run by large organisations is that, somewhere, sometime, someone is going to turn out to be bad - an argument used by the security research group in cambridge against many systems that share too much data (e.g. children's data base in UK, the HMRC fiasco, the NHS patient record spine).

So to date, we havnt seen any massive abuse of google (or yahoo or hotmail) huge repository of personal data evidenced by large scale misbehaviour - why is this? the penalties for someone accepting bribes, or being blackmailed would be no higher than those for someone working for a UK government agency that abuses their access to private data.

I don't think the procedures employed by the large scale search engine/mail/socal net systems are inherently immune from misuse of power by an insider especially more than the UK governemtn's "transformational" systems - by the way, what a great phrase
transformational government is ! given what most of the attempts to federate government databases have achieved, there has surely been a transformation
from Blair to Brown....but it has largely been one of rapidly increasing entropy, as the pathetic IT-consultant-ignorati that they contract to for these giant projects screw up again and again...oh well. maybe thats it - maybe most people in yahoo and google are paid well and enjoy doing a good job too much to do a bad thing:)

Saturday, January 19, 2008

virtual libraries and french literary criticism

Just finished the awesomely clever-clever, witty
How to Talk About Books You Haven't Read by Pierre Bayard ...

far from being a naff "self-help" "make-good" type book, this is a witty and ingineous dissection of the different ways society reads, avoids reading, and discourse about books - it could well apply to the way the scientific community works on reviewing papers too (indeed he talks about academics reviewing each others academic books). and to films, and music and so on.

each time you choose a book, you leave out 18M other books - in your life, you read a small fraction (infinitesimal ) of the available literature of the world. When you skim, you make rapid interpolations and extrapolations. once you finish reading a book, your memory plays tricks. You may remember reading books you have not read and don't even exist - indeed, I was disappointed that his discussion of imaginary books cited
I Am a Cat by Soseki Natsume, which actually does exist, rather than creating a fake book for the purpose - especially since he deliberately gets several details of other books (and films) such as The Name of the Rose and The Third Man subtly wrong to prove a point about imperfect recall, and "desired" or even "screen memory books".

everyone, especially geeky nerdy computer scientists with a literary bent, should read this...it is a hoot....ever so slightly cynical (all the examples are chosen to maximise relevance to a general argument, but then the argument is claimed to apply to everything when only extremum example are used - clearly the "norms" of reading and of virtual, screen, private, shared and other types of libraries are not the same as the extremes.

Also, annoyingly, he uses several arguments which are extremely clearly illustrated more amusingly in the Library of Bable by Jorge Luis Borges (indeed, whose is namechecked by Eco since the blind monk in the Name of the Rose - jorge of borgos- who is the crucial part of the argument about talking about books you havn't read (Aristotle's lost volume of poetics on comedy, of which Wiliam of Baskerville deduces the existence and content, and then debates with jorge, who is blind so hasnt read the book for decades and only, possibly falsely, recalls the contents).

Other Lacunae exist, where, if this was a paper submitted to a scolalry journal, I would have to ask for minor revisions...

I give this a SB++

Wednesday, January 16, 2008

peer assisted power over DSL

for today's lesson in green networking, I'd like to consider the problem of
heat dissipation in the PoP, and a radical new approach.

YOu;ve heard of power over ethernet? You've heard of broadband over powerline?
You've heard of Peer-assisted TV?

Well, now we propose: Peer-assisted power over DSL.

Here's how it works.

Take away power from the DSLAM from the exchange building, and provide it from the consumer side - 900 consumeers per DSLAM should be plenty to run the whole thing - as in the old days, when telephones were powered by the exchange, we can certainly run
20 milliamps at 25 volts over a few hundred meters, per consumer - easily enough.
We then put sails and solar panels on peoples houses, and voila,
entirely electrically self-sufficient broadband.

If there is a drop in wind, or a cloudy day, some users can get on an excercise bike.
others can walk up and downstairs. Others can just reduce their downloads.

By decentralising the power suply, we make it more robust, and we make it much greener as we dont have to dissipate all that heat in one place, but it is naturally dispersed by the wind (in the sails)

We can even load balance power and data - it might be possible with AC to network
code the different components in the power too, together with multipath routing of power

multihoming would have to make sure that they got the polarity right (perhaps a BGP
option could be added to convey this?)

Monday, December 31, 2007

mad world...

no, not tears for fears, or donny darko - no, the sci fi story i just remembered after someone commented that URI's have feelings too - fred pohl (I think) once described a future where science was based on electron psychology....mainly so he could get away with saying how impossible it would be for someone from the 20th century to understand. of course, we are from the 21st century, where nothing is incomprehensible, and everything is accelerando...

frankly, my pohl, after string theory, I am prepared to believe in electron fenh shui, if it passes occam's razor...

Thursday, December 20, 2007

combat reductionism! [1]

bl**dy g**gle are banin on about capturing all of human thought in things called "knowl"s - obviously meant to sound like atoms of irreducible knowl-edge - but, as we all know, much really useful knowledge is holistic, and therefore not reducable to some sort of lego ball and spoke trivial model

i dunno, really!

[1] after the Marx paper, Combat Liberalism, a rant about how all liberals turn into bourgeois irrelevances)

Friday, December 14, 2007

spam from pluto

I recently gathered a large body of spam recently and sent it into the SETI@home project, and it turns out, you know what, it is from an extra-terrestrial source. Yes, Asimov was right, but not about jokes. Spam is from approximately 100 light years away - it turns out friendly aliens heard Marconi's early broadcast SOS message across the Atlantic
and realized that earthlings needed help, and so they cooked up lots of useful things like viagra and low cost loans of 25M pounds of gold from Sierra Leone, and sent off messages immediately on how to get these

the only reason that people don't get the thing they want, when they reply to spam, is that we havn't connected the internet to the Intra-Galactic Internet yet....real soon now, when the Interplanetary Internet reaches Pluto, we will be able to respond to the right email addresses and web sites - then the health and welath of the planet will be assured for the unforeseeable future.

Then you will here "this is the voice of the mysterons" coming over your skype connection loud and clear.

Thursday, December 13, 2007

computer scientist + coffee => code - for example, carbon copy offsetting...

not sure that this is quite what they meant...

anyhow, too much coffee, and bjorn scheuermannn and I came up with the idea of a carbon copy tax - well he did with soem friends, but then I thought of a use for it...

I call it carbon copy offsetting(*)

the loss of privacy, just as with the (temporary) loss of identity is not
a simple black&white matter - for example, I have a bunch of cards/devices that let me pay
electronically for various things, but only work in limited domains and have limits
on the amount of credit they can be topped up with - so while someone masqerading as me
can do some damage to me with them, it is a (roughly) known quantity

so rather than saying to the government:
"do not hold this data because you don't know how to do it"
say to the government:
"here's some escrow rules for holding this data"

i.e. say I trust a bank some level X to not reveal my bak account information -
X is likely high because the bank has reputation to lose in its primary business area.
I am likely to trust other agencies much less with the same information,
as their incentives are not setup right to take due care with the information.

So instead of taking them to task legally after they screw up and
instead of preventing them from potentially usefully doing somethign with my data,
I ask them to offset the higher risk by putting some money in escrow somewhere, that,
in the event of a leak, accrues to me

so for example, if they screwup with my bank account, I can use money to get on with
life while the bank fixes things up. if they screwup with passport or other ID, I get
some compensation without having to do anything ...

if the escrow schemes are run by 3rd parties,
then we just let the market set the prices....

its a bit like the government wanted all my keys in escrow
because they didnt trust me
but the boot is on the other foot...

[(*) originally the idea was to do with spam, hence "carbon copy"]

Monday, December 10, 2007

five degrees of seperation - make it so

so someone on facebook is trying to reach everyone else to run a "complete"
six degrees of seperation experiment (given the original experiement only counted people that answered and used postage, both of these guys are severeely misled if they think they aer doing science - ever heard of random samples, or control groups, guys?)

anyhow, my goal is one better. social scientists tell us that since time immemorial, humans have gathered in grooming groups (tribes, clades, etc) of around 150 members. But now we have technology, we could make the _degree_ that a given person can handle much much higher - already, electronic address books let us handle, remember, and outcast far more people than in olden days with long tedious letter writing by candlelight, smoke signals, or even (god forbid) sunday afternoon tea.

So I want to start tht "five degrees" of seperation movement- we will bring the world one step closer together - this will, of course, bring about prosperity and world peave and banish hunger, ignorance, and english weather along with most illnesses.

cambridge high fives

s i got given two nice books for a recent birthday - the cambridge quintet, a sort of phantasy intellectual foorball match with Haldane, Wittgenstein, Turing, CP Snow, schroedinger et al, asking and trying to answer the big old "Hard AI" question...
the other was the re-issue of the Night Climbers of Cambridge, about the boys-own derring-dos of a bunch of mad students scaling the physical rather than intellectual heights of colleges and other buildings

I wont say which was more enjoyable....

Monday, December 03, 2007

social networks of social network researchers?

everyone's talking about social nets -but we've been here before several times - the
original fancy 6 degrees of separation of milgram and axelrod and social games, and even nash. then the out-of-work theoretical physicists bored with too many string theories, started to look at the degree distribution of the AS level of the internet and the linkages in the web, and even the encounters and popularity of nodes in cellular and Wi-Fi nets. now we are trying to put it all together in some sort of network science (which has obvious bio/medical applications as well as anti-terrorism, economics, and transport planning).....

so how long is it before someone decides to do a bibliometric study of social network
papers/authors? surely we can see where the gaps are and choose those as alliances to form to probvide for more robust research groupings (mixnets by design)...

post human auto-uploading

so reading more posthuman (upload mind to swarm computer embedded at qumatum level in matter throughout solar system) speculative literature, there are appear to be a few missing considerations
1. not only do you need enough MIPS and storage to run a virtual human (and their sensorum) you need logging and backup
2. latency is going to be a major question - maybe one would replicate the virtualised person across multiple sites and run them in paralelle, and then synch them up every now and then - interesting problems with consistency (and reconciling differences) would then occur
3. one would need a "virtual human hosting" economy

indeed, many of the ideas conceived in the Xen (oh, server) work would apply equally well in this world - perhaps the Sci Fi writers who write this kindah stuff are not quite as hip as they think, even though they do cite Scottish Samba Punk bands:)
(come to think of it, the Spider SYstems folks in Bloco Vomit is a very very old story too:)

alas, alackaday, that we now see "retrospective" speculative, extropian literature:)