[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [xmca] internet and surveillance



I don't know the specifics of the snooper.  If every node in the internet
were equally ranked, and if there were multiple ways to get to each, then
packet switching would be fragmented in the ways that you suggest Michael.
 The thing is that this is not true, even in the U.S. and other similarly
highly connected countries.  There are hierarchies (for
technical/political/economic/security reasons) where some nodes see much
more traffic than others, and depending on where you are, a lot or all of
your traffic will flow through one of these "more important" nodes before
heading out into the wider info world.  To the extent that ISPs collaborate
with snooping initiatives, and even if they don't, to the extent that
"back-bone" carriers do, this traffic can be captured and duplicated (and
even stored for later analysis).  Hackerdom legend has it that the NSA does
this routinely today...

On a separate issue though, there is the actual format of the packets
themselves.  Anything coming out of my computer right now will have a header
that does not change, followed by data intended for some other computer.

HEADER-IVAN-MY-ADDRESS-00000000000DATA00000000000

Even if all of my packets took different routes to get to gmail, they are
not likely to take routes that are completely "path unique", in the sense of
not having at least one leg in their individual paths overlap.  So, more
than likely, if some entity with enough interest and money wanted all of my
packets (and they knew MY-ADDRESS), then they can capture everything that
comes out of my computer as long as they sit on one of the commonly used
legs, or simply have enough money to cover all the last legs that lead to
Google.

and so on...  But this is all still separate from what is in
the 00000000000DATA00000000000  part.  That part can (and is routinely)
encrypted nowadays.  That is the part that should be relatively safe, and
which would literally cost millions of dollars to decrypt given the fact
that there are still no known decryption algorithms that can do this more
efficiently than simply trying all the combinations (and each such encrypted
package wold require at least 2 to the 64th power combinations, or more in
the case of higher security encryption --- gmail uses 2^128).  So whoever
wants to snoop needs a lot of money and computing power.... doable, but
unlikely in the case of the millions of users that use Facebook and the
like...

Just found this

http://www.youtube.com/watch?v=9BGWrLiT9qs

which has only a few views (and is 12 minutes long), but covers the basics.
 It can be much better, but it does give you an idea of the "legs" of a path
and a way to imagine what you would have to do to successfully snoop.

Ivan

On Sun, Feb 6, 2011 at 11:04 AM, Michael Glassman <MGlassman@ehe.osu.edu>wrote:

> Hi Ivan,
>
> I may be wrong about this (and hope someone can correct me if I am) but I
> was under the impression that the Internet was very much a distributed
> information system.  That is the whole point of packet switching is that as
> soon as I hit the send button the information I am sending gets broken up
> into small packages with the same destination address and then moves in
> various directions through the big information universe out there until it
> reaches its end server where it is reassembled in transit.  This was a
> really important part of the origination of the Internet which was funded by
> ARPA.  It was important that those who intercepted the packet didn't have
> enough information to put together the entire mosaic.  It also helped keep
> the information from being corrupted (short packets of information have
> greater stability than long packets of information).  And of course for
> those who actually put together the Internet system it was important that
> nobody could shut down the internet simply by shutting down a transit
> server.  The packet would be re-routed, but even if it disappeared there
> would still be enough information from the other packets so that the overall
> information system could be reassembled.  I worry that fears of stealing
> information in transit, which if this is the system seems kind of
> impossible, is meant to make individuals fearful of using the Internet.
>  Most information that is appropriated is end product.  For instance
> Wikileaks, which has some of the best hackers in the world participating in
> it, is completely dependent on end product.  As a matter of fact, I have
> heard many urban legends about information being stolen in transit but have
> not actually heard of a single actual case.
>
> Then I wonder how much of this snooping story is true and how much of it is
> actually to keep people from using the Internet for fear of having their
> identities found out by dark powers.  I mean there are crawler programs that
> can let you know information that is in a given message (again only as
> document though, but not in transit) which results in this weird thing of
> writing to somebody about Japan on gmail and seeing an advertisements for
> flights to Japan appearing in your sidebar, but I don't think they are
> coordinated.  Meaning a crawler can't put together two completely unrelated
> pieces of information such as the word revolution and the ISP of the sender.
>  At least I hope it can't (Google claims it can't!).  So it led me to wonder
> if this snooping system can actually explore packets of facebook and twitter
> information transfers if this is because the individual packets actually
> contain more information because there is so little information?  Honestly,
> I don't know.
>
> Michael
>
> ________________________________
>
> From: xmca-bounces@weber.ucsd.edu on behalf of Ivan Rosero
> Sent: Sun 2/6/2011 12:59 PM
> To: eXtended Mind, Culture, Activity
> Subject: Re: [xmca] internet and surveillance
>
>
>
> I'm not a frequent contributor though, like many, I hang around.  Thank you
> all for so many great things to think about.  This issue of surveillance
> and
> "packets" is interesting to me.  This comment, from Michael, caught my
> attention:
>
> It seems like social media has a much more basic packet switching system
> > than say e-mail or longer postings.
>
>
> The way I understand packet switching technology, everything on the
> internet
> works on top of the same underlying system.  This means that whether it is
> the back and forth between your web browser and your bank, or Facebook, or
> Youtube, the same kinds of packets encode the information flowing in both
> directions.  I understood the article to be saying that the company in
> question has developed technology to snoop into the flow of information at
> this level, the bottom most.  This means that it is not social media
> specific, though of course the packets in which that information is encoded
> might be of special interest.
>
> There is a question the article left unaddressed.  If it is possible for
> this company to snoop on *any* packet whatsoever, this means that they have
> broken the various encryption protocols used on a daily basis to log in
> securely to your bank, or even simply to send email from Gmail (the latter
> encrypts email communication by default).  If this were true, the headline
> "Egypt Shuts Down Internet" would look like a mild headache in comparison.
>
> I assume plenty of information floats around that is unencrypted, and
> snoopable, and this may lead worried power-centers to take a hammer to the
> whole machine rather than let out even a tiny trickle of potentially
> dangerous information.  But the snooping and the shutting down of the
> packet-switching technologies are not directly related (the one cannot
> "cause" the other), you need a higher level of order (in this case the
> Egyptian status-quo) to connect the two.  Like Michael wrote, they "can
> only
> shut down the end servers that most Egyptians use".
>
> Ivan
>
> On Sat, Feb 5, 2011 at 8:56 PM, Michael Glassman <MGlassman@ehe.osu.edu
> >wrote:
>
> > This is interesting.  The spying really isn't done through the Internet
> but
> > through closed information social media systems such as Facebook and
> > Twitter.   It raises some really interesting issues about the difference
> > between open information systems and closed information systems - really
> a
> > contrast that has always been with us I think, but the Internet as an
> > information system is really changing all that.  There has been a great
> deal
> > of discussion of course of Facebook selling information, but I really
> hadn't
> > given much thought to differences in how the information itself is
> > translated.  It seems like social media has a much more basic packet
> > switching system than say e-mail or longer postings.  This would make
> sense
> > for Twitter which is limited I believe to 140 characters (don't use
> twitter
> > so I'm not sure).  It makes one wonder about the degree to which
> traditional
> > media is pushing social media as a phenomenon, but not so much other
> forms
> > of Internet communication.  Maybe because it is easier to control.
> >
> > Relatedly, Egypt can't such down the Internet, they can only shut down
> the
> > end servers that most Egyptians use, which is basically the last routing
> > point for Egyptians.  One of the big issues in Net Neutrality is the idea
> > that companies which own specific netowrks, such as Verizon, have
> complete
> > control over the end point routers for cell phones and their progeny.
>  That
> > means they can control information in much the same way that Egypt is now
> > controlling information, but instead of letting no information through,
> they
> > would only let the information they wanted thorugh.  I think for more
> > complex information where the end user has a choice of networks, this
> isn't
> > anywhere near as much of a problem.  And as Andy suggests, hackers can
> find
> > other end point servers (I am sure by this point Egyptians have hacked
> into
> > servers from nearby countries.  Hackers in China have become amazingly
> good
> > at this and it is one of the reasons I think that the government has gone
> > from trying to control the Internet itself to trying to control search
> > engines (I'm not sure how they do this, perhaps by having the search
> engines
> > not automatically translate to and from Chinese.  Maybe somebody else
> > knows).
> >
> > This raises so many questions about information and how we treat
> > information.
> >
> > Michael
> >
> > ________________________________
> >
> > From: xmca-bounces@weber.ucsd.edu on behalf of Andy Blunden
> > Sent: Sat 2/5/2011 10:59 PM
> > To: eXtended Mind, Culture, Activity
> > Subject: Re: [xmca] internet and surveillance
> >
> >
> >
> > I think there must be some law, a corollary of the law about offensive
> > and defensive military technology (that the technology of bullets is
> > always one step ahead of the advance of the technology of armour) to the
> > effect that the hacker is always one step ahead of the internet security
> > expert.
> >
> > Andy
> >
> > mike cole wrote:
> > > This story might be worth thinking about.
> > > mike
> > >
> > >
> >
> http://internetsgovernance.blogspot.com/2011/02/egypt-crisis-egypt-is-burning-and.html
> > > __________________________________________
> > > _____
> > > xmca mailing list
> > > xmca@weber.ucsd.edu
> > > http://dss.ucsd.edu/mailman/listinfo/xmca
> > >
> > >
> > >
> >
> > --
> > ------------------------------------------------------------------------
> > Hegel Summer School: The New Atheism: Just Another Dogma?
> > <http://ethicalpolitics.org/seminars/hss2011.htm>
> >
> > __________________________________________
> > _____
> > xmca mailing list
> > xmca@weber.ucsd.edu
> > http://dss.ucsd.edu/mailman/listinfo/xmca
> >
> >
> >
> > __________________________________________
> > _____
> > xmca mailing list
> > xmca@weber.ucsd.edu
> > http://dss.ucsd.edu/mailman/listinfo/xmca
> >
> >
> __________________________________________
> _____
> xmca mailing list
> xmca@weber.ucsd.edu
> http://dss.ucsd.edu/mailman/listinfo/xmca
>
>
>
> __________________________________________
> _____
> xmca mailing list
> xmca@weber.ucsd.edu
> http://dss.ucsd.edu/mailman/listinfo/xmca
>
>
__________________________________________
_____
xmca mailing list
xmca@weber.ucsd.edu
http://dss.ucsd.edu/mailman/listinfo/xmca