Yesterday the Guardian had a special pull out section dedicated to the
LHC. If you browse through the articles you find lots of comments along
the lines of
"temple to mystery and imagination", "a journey to the edge of
understanding". "a modern cathedral to our relationship with the
universe", and so on. From the superlatives that are being written one
would think that the LHC is the best thing to happen to enlightenment
since some fat chinese guy sat beneath a tree, and that it is the summit
of human imagination, achievement and art. Well, I just don't buy all
of that crap.
Reading these articles got me thinking about what the LHC is, and
fundamentally it's just a larger detector than what we already had
before. As I see it, it's an inevitable extension of what you do if you
want to measure something that we already know how to measure (particle
tracks), with better precision over a higher energy range.
The bottom line is that we have been doing this since the 1920's.
If you look at it as just being an artifact then it is neat, but there
are many other piece of artistry that required as much imagination,
effort, skill and chutzpah to bring together. The moon landings are one,
the regularity of probes landing on mars another. The engineering
required to make a large city like New York work always blows my mind,
and that emerged from a bottom up self organization of 15 million souls
trying to find a way to survive in an area of land a little too small
for them all.
As we look around the world at the things we as a species have built
there are many such artifacts that can inspire our awe and wonder. I
don't think that the LHC can lay a claim to be at the pinnacle, though
no doubt it is a good example of a big complicated object that make
people look small when they stand beside it.
There is something to wonder at in all of this, and that is the idea
behind the inevitability of something like the LHC. That idea is the
atomic and quantum electrodynamical nature of the world. In that there
is something to be proud of as a species. I don't see the LHC as being a
radical departure from this idea, but rather an object whose existence
is quintessentially rooted in that idea.
One could almost argue that the LHC represents a failure of the
imagination. We are faced with limits to our ability to test the
mathematics that we have written down against the atoms that we write
with. We cannot tease apart the Fynemann diagrams to tell us more about
the world, and so we resort to a bigger hammer rather than a more subtle
approach that might look to other ways to coax the mysteries of the
universe out of their hiding places.
There have been some papers that have come out recently looking for
connections in the physics of super fluids with the imagined state of
the early universe, the idea being that looking at the behavior of
vortices in super cooled liquids could demonstrate identical physics to
the phases of matter at the point of various decouplingings in energy
scales. It's pretty clear that these models are yet toy models, but
perhaps they point out an orthogonal direction to building massive atom
smashers.
I want to be clear that I do applaud the work of the thousands of people
working at Cern, and I do think that the billions of euro that something
like this costs is more than worth the investment. I appreciate how hard
it is to deal with systematics on something of this scale, and it is a
minor miracle, but I just don't think that the artifact deserves unconstrained
adulation over the ideas that is reflects.
The second talk in the evening CREEN parallel session is from J. Holyst
about phase transitions in coupled complex networks where the network
properties are different in the two groups.
In one group there are a smaller number of members, but they are tightly
coupled. The second group is larger, but less tightly coupled. The model
uses an Ising model. The less coupled group has higher fluctuation in
opinion, and when it is brought into contact with the smaller group the
larger group undergoes a phase transition. To reverse the process if you
can find a hub in the tightly coupled system then when you convince them
their opinions gain traction.
This is a really nice paper and there are obvious tendancies to draw
parallels with real communities, such as for example two scientific
communities, or an immigrant community, but I am fairly certain that
this would at the moment be too simplistic.
He goes on to show a face off between ER graphs vs BA graphs, and this
shows that network structure has signifigant effect on the suseptability
of the community to opinion change.
One big application could be to look at these results in the context of
the new vote that Ireland will probably undergo early next year to
re-ratify the Lison treaty (I mean, I assume that is what is going to
happen in Ireland next year).
I'm at the netsci08 conference and there is a really delightful talk
about the network of papers published in the creationisim/evolution
debate.
The group looked at key people in the ID debate and people who acted as
strong defenders of evolution. One can then make a graph of the links
between the groups and intra-groups based on citation and co-citation.
Now one area of social science that is pretty interesting in this debate
is looking at triads, as there are clearly going to be antagonitic
relationships in this debate.
In the graph evolutionists are blue, creationists are yellow and Dawkins
is red on his own.
From the graph it is clear that there are some people who are opinion
leaders.
Ok, what is interesting is that there are far more mixed triples than
similar triples in the graph, meaning that people from both sides seem
to be spending more time slagging each other off than agreeing with
their friends.
This is the first time that I would be tempted to say that the study of
creationisim could be considered science.
I'm really torn by the number of great talks on today. There are three
parallel sessions, and for each time slot I want to be in at least two
places at once. I'm going to try to pick out talks that have some
relation to online social networks, community detection and scientific
networks, but some of the talks on the theory of clustering are
conflicting directly with some use cases of looking at some online
social networks. Ahh, what a dilemma.
The opening talk of this session was from Stwphen Uzzo talking about the
The next talk was by M.C Gonzales looking at the network of travel
patterns. This was the paper that made the cover of Nature.
The big question is trying to find out what the travel patterns of
people are. Thhe big problem is that getting data is apparently quite
hard.
The solution is to follow mobile phone signals, following 10^5 people over
10^6 locations over six months.
I'm looking at the movie of their data, and it is clear that many people
don't move very much, and other people move a lot. Of course one wants
to know some information about the people to see what effect like age,
wealth and occupation will have on these results. Again I'm looking for
something surprising.
There is a nice graph showing corellation over time, it is hugly spiked
on 24 hours. Not surprising, but a good reality check on the data.
I'm going to head to one of the other sessions after this talk.
This talk, though, is very nice. Once again there is evedence that our
behaviour is depressingly regular. Also the longer a journey the more likely
that a journey is going to be linear.
Mark Newman, Networks in Space,
This is about networks in geographic space.
Mark is looking at properties of networks that are tied to geography.
Transport networks are a good example, and we are looking at the
difference between road and air networks.
The road and air networks are very different, even though you use both
of them for getting from A to B.
There is different bahviour, could I say 'driving' the use of these
networks. For roads we want to minimze the length of our journey, but
that's not such an important factor in flight journeys. When we fly we
like to take direct flights, and minimize the number of flight hops that
we take.
If you model this behviour you get out networks that look a lot like
road and flight networks.
Their first model looked at connecting randomly distributed nodes.
In order not to get influenced by population density they made a map
that is rescaled by population density. This is called a cartogram.
You can see some really nice election cartograms that Newmann and
Gastern made here: http://www.cscs.umich.edu/~crshalizi/election/
There is a really nice historical example from Raisz from the
Geographical REview from 1943.
It looks like most recent attempts have been hand-drawn, but they look
pretty shit.
Newman and Gastner made a difffusion algorithm that allows you to do
this quickly.
OK, it looks like this started off as a network talk, but segwayed into
a demo of this mapping technique. Ahh, no, we are back to looking at
airports.
The interesting result from this talk is that the best covering fro
utilities such as airports or post offices does not grow lineraly with
popultaion, but to the power of 2/3.
This was also a pretty nice talk.
]]>Nicholas Christakis, Harvard, "eat drink and be merry, the spread of
health phenomena in social networks".
This talk is looking at the spread of desies throgh social interactions,
rather than other types of interactins. The main study was looking at
obesity using the Framingham Heart Study Social Network. This seems like
a very famouse social network health related study, so I'm not going to
go into detail about that, but the bottom line is that they were able to
construct the social interactions from this study by digging through the
huge paper archive. They were able to look at friend, relative and
co-worker ties.
The main study was looking at about 5k individuals out of 12k, taken
from 1973 onwards.
Nice, node ssize is related to a person's weight!
There is clear clustering of obese nodes in the network, now is this
clustering random or structured?
Well, it's more clustered than random.
There are a couple of reasons why this might be the case. It could be
that obese people like each other, people might be susceptible to local
factors, or there might be some kind of peer pressure.
By looking at time evolution the hope is that they might be able to find
'patient 0' for the obesity epedmic. OK, video is coming up now ...!
OK, looking at people getting fatter all over america from 1972 onwards,
I'm going for a run later!
The effect is not centered in one location, but it seems that it's an
epidemic that had multiple starting points in the network.
Looking at the directionality of ties of friendship helps you make
inferences about causes. Wow, if you are friends with someone who is
friends with you, and they get obese, you have 300% greater chance to
gain weight. Stay friends with thin people!
It looks like much of this is driven by social norms.
They also have gwo data from the network, that is really cool.
They can convert location to wealth, and can take this into account when
looking at the evolution of the network.
This data is really really cool.
No drop off in effect with distance, it is really the social tie that is
important.
They also looked at the effect of smoking, and were able to take this
into account.
So their working hyppothesis is that it might be the spread of behaviour
and habit, perticluarly shared behaviour, going runnning vs going for a
beer.
It might be the spread of an idea, the spread of what an acceptable body
size might be.
OK, that's pretty amazing, and you can tease a hugh amount of
information out of this study. Liklihood of quitting smoking, of how
that is effected by education, and friendship tie.
I have to say, there is not a lot of results that are amazingly
astonishing. They have food networks, like the bannana network and the
friend chicken network.
They are also looking at emotions. We know that emotions can spread
through groups, on diads. Could emotions spread hyper-diadically, and
over longer time frames?
There is strong clustering of happienss, and your happiness seems
coreelated with people who are outside of your direct social horizon.
Interestingly happiness does not spread in the workplace (I think that
was the point), but happy people have higher clustering and better
centrality in the network.
There seems to be a half life for catching happiness from your network,
this is about 6 months. There is also a strong local effect, you need
happy people to be within about two miles of you, and to be having happy
events happening to them every six months or so.
Ahh, you can look at smiling on facebook. Right, I gotta put up some
happy pictures on my profiles!
Ahh, thiness also spreads, but the reason they have been looking at
obesity is that this study is looking at the obesity epidemic. The
network shows you the magnification of the phenomena, not the cause or
origin of the phonomenon.
Interesting question, if you wanted to hire flight attendants who you
didn't want to gain weight, should you hire them based on the bmi of
their friends? Well, the answer is that in a workplace if a certain
behaviour begins to spread it is likely to have a network effect. The
flpiside is that you can use these network effects to more economic
effect by trying to promote certain behaviour through targeting core
groups in the workplace.
I'm in Norwich all this week attending Netsci08
http://www.ifr.ac.uk/netsci08/, the internatinal workshop and conference
on network science. It's a week long event, and broadly speaking it
looks like there are three types themes that are being discussed here:
biological networks, pure networks science and community detection in
networks, principlaly emergent networks of the kind we see in the
internet.
I'm twittering about the meeting using the tag #netsci08, but it seems
that I'm the only one out there in the twitterverse who is also at this
meeting. Not enough power in the lecture hall, and wifi is a little
ropey, but the conference is pretty good so far.
The talks on Monday were about some basics on network mathematics, and
on network science in the social sciences. I'll go back over my notes
and give a quick report on them when I get a chance to catchup, but the
discussion in the evening was pretty interesting, and the talk in the
morning touced on some very important topics.
The Tuesday morning tutorial is on economics and networks. The morning
model was very simple, and I think that's fair enough, but I got the
feeling that the level of the audience, at least on the side of the room
that I am sitting on, was high enough to have taken a bit more robust
model, so I got the feeling that there was some discomfot with the model
presented.
The after-coffee section is focussing on social influencers, now this is
interesting.
How is it that information flow is highly assymetric in the world?
The model is a mmulti-state model with differeing outcomes. Individuals
don't know the true state of the system, but they have beliefs about the
states. Sounds like a hidden markov model.
The model is stationary, and we want to see how the choices we make
change the beliefs that we have. Could be a bayseian network? Let's see.
What I am hoping to see from this model is how reccomendations can
travel throgh a network. There is a network of communication between the
network. The model can integrate dynamics, the dynamics of belief.
There is also feedback between actions and beliefs. The main result is
that as time goes by new information has less effect, and so beliefs
converge in the network. This is a consequence of Martingale's theorm.
The big question is whether we get optimal actions, and the big result
is that the ability to explore the action space and find the best action
is depenant upon the structure of the network. That is really
interesting.
Oh my God, someone has an OLPC machine in the audience, how cool is
that!!
Anyway, back to the talk. So this is indeed a Bayseian network. The
anti-intutive outcome from this model is that if you have to build a one
time only network that can't be changed later, then you have the best
chance of getting optimal behaviour if no one person has undue
influence, hoever I think that for online social networks there is a lot
of dunamics going on that can pull you out of local sub-optimal minima.
I've just signed up for an account with twidox, which is a start up
that is collecting shared documents of interest to scientists, I
believe. They are in the private beta stage, but had a link on their
homepage for requesting an account. I got the following message when I
hit the verification link:
Registration is taking place!
Many thanks for you interest in twidox. Your account has been activated.
We will send you your private-beta lock-in details very soon.
We thank you for your support.
Your twidox-team
]]>There is a tension between the providers of social software, and the
way we behave. When I move from one city to another my social network
changes as that's very location dependent, but when I do have that
network set up for the most part, I don't expect restrictions on where
I can go in that city with my friends. For sure, some friends of mine
might not be caught dead in the palace bar, they only drink in the
stag's head, but I could drop in with my palace friends for a quick
pint and catch up on news.
On the internet distance only affects us on the scale of timezones,
and even there our tail of interaction is much broader. Our changing
activities very much determine the networks we hold on to. I no longer
practice science, but I'm still in contact with my old climbing
buddies. However a big change at the moment is that the places we go
on the internet still don't play well with each other in the same way
that they do in real life.
I hope that truly mobile social networks will emerge, and I think they
will be driven my our address books on our phones. First we will have
real time tracking of the location of our contacts (to the point that
mutual permission is granted), and then this will start to seep into
awareness of location on the web. It's something that has been faces
before, with IM and VOIP walled gardens. So far only email and phone
numbers and physical mail addresses don't have this problem, and
perhaps for that reason those will be the media that crack the problem
first.
I'm reading a great article by Joe Esposito on Open Access. You can read
the article
here:http://quod.lib.umich.edu/cgi/t/text/text-idx?c=jep;cc=jep;rgn=main;view=text;idno=3336451.0011.203.
The article's main claim is that attention is the barrier to
disemination rather than access. Another claim of this article is that
paid for publishing platforms are good for readers in that they act as
an attention filter.
One of the main claims of the article is that Open Content in a sea of
other content will exist beyond the scope of peer review and editorial
review. That is certainly the case at the moment for the most part.
Beyond Google's search rank, there are few other effective ways of using
machines to rank the quality of content, and certainly none at the
moment that apply specifically to academic content.
I think that one of the issues is that the volume of signals about the
quality of data that can be data mined compared to the data that needs
to be mined is still tiny, and the tools to do this mining almost non
existant. Social signals from tools such as Connotea are only just
beginnig to create traces, but they are still small.
There is a hope that automatic data mining of sources will push out the
need for manual editorial control, and there are platforms out there
with enough computational power to do this, it's just that these
platforms at the moment have much more important jobs to do.
Even when hardware and storage really are commodotized (and the
existence of S3, EC2, AppEngine should leave no doubt in our minds that
this is the way things are going), who is going to write the algorithms
to review papers for us. Perhaps we only need a few people to do this,
and so I shouldn't worry too much about that, but I still don't see this
kind of thing happening for a few years yet.
There is another interesting aspect of Open Access that springs to
mind while reading this paper.
The article does not address the machine redability of published work,
and how restrictions on that type of access can hinder auto-mining of
facts and figures, the aggregate results of which may command
signifigantly more attention than the individual componenets.
Finally Joe has a nice little analysis of the kinds of costs associated
with building new publshing platforms to market.
Alltogether I really liked this paper.
]]>