7-Marcel-offense - Northwestern Networks Group
Download
Report
Transcript 7-Marcel-offense - Northwestern Networks Group
Offense: Planetary-Scale Views
on a Large Instant Messaging
Network
J. Leskovec, et al.
As a whole…
Provides a strong analysis
Considers a variety of approaches
However, it falls prey to a number of
critical pitfalls
The Data Set
Only includes data from MSN Messenger
Other networks may have different usage
patterns
Paper suggests MSN accounts for ~50% at the time
(seems to be correct)
Each network could draw an entirely different
type of user (CompuServe vs. AOL)
What about spam/bots?
Are these a relevant portion?
Communication Demographics
Only look at ages 10-60, to correct for any
misrepresented ages
Any theoretical backing?
Conversations depend only on number of
messages
What about non-symmetric conversations? (one
user sending messages, another not replying)
World Geography and
Communication
Show us distribution of MSN users by
World population
What does this really tell us?
Better to instead look per capita to computer
users?
Perhaps internet use?
Would take into account areas where no one is
on the internet and therefore not on MSN
The Communication Network
Only consider those who communicated
during time period
A limitation of the data set
Doesn’t acknowledge the possibility of large
amounts of potential links that were just not
active during the month
Really need a larger data set to deal with this
Milgram
Data is not actually comparable to
Milgram’s!
Milgram focused on drawing connections
with a letter
These connections are more incidental (based
around some sort of social network, but are not
“focused” in the same way)
Many people may be “on a first name basis”
with someone, but just not normally talk to
them
Milgram
Temporally incomparability
This data relies only on 1 month of
communication
In Milgram’s study, people could use anyone
that they knew on a first name basis
A whole lifetime!
Implication: MSN may be much better
connected than these results show!
A Purpose?
All this information is good to know
But what can we really do with it?
Generally speaking, seems people use it as
an extension of their real world networks
Experiment like this unable to describe any
motivation, which would be helpful