Bram Dov Abramson on 25 Jul 2000 22:13:50 -0000


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

Re: <nettime> Terror in Tune Town


mwark@laurel.ocs.mq.edu.au:
>My concern is that if you can't 'propertise' the information, then all
>power resides with whoever owns the vector. The pipe -- the part that
>is still physical and material property -- will be where the power lies.
>The pipe guys will be king. 

tbyfield@panix.com:
>a PSTN (public switched telephone network)
>regime, because they were 'smart' networks, in the sense that the
>devices at the core determined what went where, when, and how.

>but the net is a 'dumb' network: its primary design goal is to en-
>sure that packets are delivered point to point, so 'intelligence'
>is pushed to devices on the periphery. 

Doubt it.  The Internet works great as a fairly dumb network when the
Internet's sole goal is to push modest amounts of ascii &
download-now-use-later data files to various people.  Which is fine by me,
but many billions of dollars seem to have been mobilized to turn the
Internet into something else.

Which means that the PSTN / IP difference is not so much the architecture
(romanticized old world / brave new world narratives aside).  It's what's
being pushed down the pipes, on the one hand, and the degree of vertical
consolidation among the different slices of architecture which do the
pushing, otoh.  And those feed into one another: the more people try and
push audiovisual stuff down pipes (aka "rich content", "broadband media",
"bitcasting", even ... "broadcasting", once), the more IP and SS7 (the
latest version of PSTN; yes, it's packet switched) look strangely similar,
and the more vertical integration takes place (in lieu of "slower"
decision_by_committee) as a way of making that architectural "innovation"
happen:

>thus there are the backbone providers
IP: backbone providers (and iisps = international isps)
PSTN: international carriers

IP:
>the routing registries, the naming registries,
PSTN/SS7: Service Control Point, Signal Transfer Point...

IP:
>the caching providers
PSTN: no equivalent, cause audiotex never caught on big enough (with an
important exxxception).  On the other hand calling caching an "edge" or a
"core" function is largely semantic within this kind of discussion.  The
cache and multicast boxes don't sit at the user's end, they sit inside the
network.  Content the "end user" wants to push onto the Internet is subject
to the routing policies of those boxes.  If looking for a political economy
spin or choke point, cf the various agreements between Akamai and various
backbone providers, for example; then note the revenue streams flowing from
content providers to caching providers.

>ISPs of various shapes and sizes
This is more and more true of the PSTN too.  Telephony's institutional
alignment is an artefact bound up far more in market and regulatory
decisions than the network architecture used to build it.

>these forces don't see eye to eye *at all*.

Here's where we agree a bit.  There is a lot more churn in the IP (here,
that's Internet Protocol) world than the PSTN has had (see above), and
that's a good thing.  For example the way the naming/routing registries on
the IP side are *comparatively* divorced from the backbone providers.

Regulators looking for anti-competitive and anti-trust flags -- and, say,
sideline observers worried about monopolies of knowledge infrastructure (to
gloss Innis) -- would probably do well to look for a layered approach that
makes sure things stay this away, ie that vertical consolidation doesn't
allow IP players to yoke together vertical layers in much the way that the
PSTN world does.

So to bring this back to the other kind of IP: sounds like Ken Wark is
arguing that if content producers don't "own" the content, then the
distributors will; that Ted Byfield is saying that that's a good thing,
because the means of distribution are so fragmented that nobody can own
that content at all (maybe the Publius project should be thought of as a
by-design version of that, http://www.cs.nyu.edu/~waldman/publius/).  I
don't know enough to say either way, only that the economic+technical
conditions Ted is describing won't hang around forever -- at least not on
their own -- and that that might be a source of concern & intervention.

cheers
Bram

#  distributed via <nettime>: no commercial use without permission
#  <nettime> is a moderated mailing list for net criticism,
#  collaborative text filtering and cultural politics of the nets
#  more info: majordomo@bbs.thing.net and "info nettime-l" in the msg body
#  archive: http://www.nettime.org contact: nettime@bbs.thing.net