nettime's_timekeeper on Wed, 19 Jun 2002 14:34:06 +0200 (CEST)

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

<nettime> the download times they are a changin' digest [plasmastudii, Geer, Beaubien]

   Re: <nettime> where has all the bandwith gone?                                  
     uospnJ <>                                                

   Re: <nettime> the download times they are a changin'                            
     Benjamin Geer <>                                      

   Bandwidth redux, was EboneKPNQwest going down                                   
     Beatrice Beaubien <>                                               


Date: Wed, 19 Jun 2002 00:11:44 -0400
From: uospnJ <>
Subject: Re: <nettime> where has all the bandwith gone?


What you describe is a really cool use for the web.  But on a day to 
day basis, the average user just is not as resourceful.  We look up 
the medical sites and long-winded (basically ads for the medical 
world).  We aren't always culling useful information from the sea of 

In your case, you found some helpful stuff.  I have never found 
useful health-medical type info.  But I have no idea where to look 
and very little interest in giving it much of my time (as the great 
majority of surfers who aren't particularly enthralled by the net). 
Not that someone hasn't posted it, but that after wading through few 
hundred MBs worth of junk, I have to get up

The net COULD be a beautiful garden of uses but first we have to clip 
off the 3000 layers of weeds.  Bandwidth is only going to make it 
worse for now.  If it helps for one movie, it'll facilitate 20 
innocuous ones.  What's the hurry?  The web will eventually grow into 
some purpose, whereupon we can decide then if it's worth more 
band-width.  Not give more bw and hope a use fills it.



223 E 10th Street
PMB 130
New York, NY  10003


Date: Tue, 18 Jun 2002 17:56:41 +0100
From: Benjamin Geer <>
Subject: Re: <nettime> the download times they are a changin'

Florian Cramer wrote:
> - - Free Software (and software downloads in general).  Much of the
>   server/downlaod bandwidth for Free Software is provided by company
>   servers like; many of them originate in the dotcom
>   area and, producing no revenue for their operators, have a doubtful
>   future.

There is a similar problem with the hosting of web sites for activist groups. 
Some Indymedia sites, for example, are hosted on PCs with DSL connections in 
people's homes.  Commercial web hosting is cheap if you're happy with static 
web pages or a limited variety of basic, pre-configured software, but if you 
want to run an application server or a custom content management system, or 
if you need to do any system administration yourself, you need a dedicated 
server, which is very expensive, considering the budgets of most volunteer 
groups.  A lot of the best Free Software for running dynamic web sites is 
currently supported by only a few, very expensive commercial hosting 
packages.  Therefore, many activist groups run their own servers on home DSL 
connections.  Naturally, this is completely inadequate if your site gets a 
lot of visitors.  Also, DSL in Britain is extremely unreliable; it's quite 
common for a connection to go down for several days.

If bandwidth were a lot cheaper, it would dramatically increase the ability 
of small volunteer groups to run dynamic web sites built on the best 
available technology.  There would be a lot more things like Indymedia, and 
they'd be a lot more reliable.


This email has been scanned for all viruses by the MessageLabs SkyScan
service. For more information on a proactive anti-virus service working
around the clock, around the globe, visit


Date: Tue, 18 Jun 2002 13:56:13 -0400
From: Beatrice Beaubien <>
Subject: Bandwidth redux, was EboneKPNQwest going down

Dear Nettimers,

The beat goes on.

Thanks to Pit Schultz, enbyoire e, and Morlock Elloi for their further 

Firstly, I solicited the insight of someone in operations at a major 
European ISP. Here are his comments, lightly edited:

> The majority of the KQ Network is maintained and built by EBONE which
> was aquired last year..... The EBONE people got in a very weird 
> situation
> because EBONE BV isnt bankrupt all, but KQ is.
> Thus they didnt get a cent of this weird construction, but they kept on 
> maintaining the network for a few weeks.
> They were fed up and said if they didn't get payed they would shutdown 
> (sounds normal doesn't it? They are proud of the things they have built 
> over the past few years).

> The people who look after the bankruptcy agreed with payment, no 
> problem there.
> No shutdown, no inet falling apart, nothing.
> And believe me.... if KQ should go down it will matter and you will 
> feel it, even
> if europe is a long way from your home.

> There IS a shortage of bandwidth, we have got to the level that if we 
> miss one carrier
> we experience slow and unreachable sites (route is a world of its own), 
> and btw next
> up is Qwest holding.
> (ever did a traceroute that didnt say qwest in eg canada?) so bummer... 
> we will
> all loose like 55% (yes 55%) of our so-valued but empty virtual world :)
> About sitting on an empty network, are you nuts? Ebone/KQ did 25% of 
> european
> network..... thats not empty.
> Seems it isnt as easy as you thought to pinpoint european internet isnt 
> it? Or heck seems
> you dont understand internet at all :)  <-- WATCH THAT SMILEY


Although I find your language a bit hard to follow, I agree with some of 
your points.

> the broadband issue is of great interest for nettimers
> for different reasons. the materiality
> of the internet besides the invested labour of users,
> is represented by its lowest layers, the physical one
> and the the one of switched packets. controling those
> layers means controling the 'means of production'.

And any constraints on it, like service interruptions due to providers' 
financial problems, impacts us all.

> non-mainstream content shouldn't be hirarchized by
> data-types. a usual text needs 40 kb, a usual audio track 4 MB,
> and a usual movie takes 800 MB (divx) and in DVD quality 4,5 GB.
> it would be a bit absurd, to claim that trough 'bandwidth
> scarcity' textual production would play a higher role in
> cultural production than other formats.

The converse is also true. Bandwidth is sucked up by divx files clogging 
the ether every time a blockbuster film is released, and this does 
effect the average 80 kb/day user.

> the cultural politics of the net are very much co-determined
> by it's economical basic conditions. the implications might
> be different for producers, distributors, consumers and their
> various mixes, but it certainly matters if i have to pay
> 6 euros per gigabyte or just 10 cents. the latter would be
> possible if kpnquest,  firstmark, globallcrossing and however
> the fibreoptics backbones are called, would have made their
> full capacity available to resellers.

I can't comment on this. The suggestion seems to be that it is in the 
best interests of providers to hold back bandwidth and I haven't heard 
of this practice. Also, in the current climate it doesn't seem fiscally 
viable, i.e. their shareholders might not approve.

> interestingly there is not much investigative online journalism
> concerning the developments of the bandwidth market.

There has been a river of ink spilled on this, but it is in the 
financial press which apparently some nettimers don't read.

> a long time ago, i tried to remember nettimers to george gilder's
> dream of the bandwidth glut. like other messianic out-of-control
> consultants he influenced the dot-com mania a lot, in its
> underlaying ideology of hypergrowth. the negativity of this
> absolute optimism turns out today as a phase of restauration,
> a rethorics of dull praxis, a culture of looking backwards and
> historify the last glorious years. that not all nice ideas
> about the future (like free bandwidth) get real as fast as promised
> by the cyber-prophets mean that they are completly wrong.

Yeah, and glorifying the past results in a present day nihilism; trading 
one extreme position for another.

> first it seems that the users themselves stand in the way of
> an unlimited bandwidth future. enjoying the new
> possiblities of digital fluidity (p2p etc.) they activily
> supported the crash of almost every 'soft' content driven
> audio, video, streaming or otherwise broadband driven
> dot-com project, which started with 'free services' in
> expectation to "buy out" the "free floating" productivity
> of its users. this 'fall of the profit rate' resembles the story of
> immaterial labour, the promises of communism... the question
> is what went wrong in building up a consciousness of the users?

Lesson learned: whether or not you embrace the realities of global 
marketplace you will be subjected to its fluxes and (mis)fortunes.

> clear is, that there are no larger numbers of customers which
> pay for the infrastructure layed out arround these glorious
> info-autobahn-plans, might they be called broadband, next generation 
> mobile
> phones, video-telephony or flying cars. it is the tragedy
> of the wired economy, that the leading american technological dreams,
> were inspired by a immature or at least incomplete intellectual culture
> and not by social needs. the broadband future was mostly described
> in scenarios resembling the techno-futurism beginning in the
> 50ies, with fully automatic homes, and a user experience directly
> inspired by the narratives arround extraterrestial space 
> colonialisation.

and the banks got sucked into this futurism to the tune of 4 trillion 
dollars in a span of months. Hence the situation we find ourselves in 

> let's take another look at the laws of internet traffic.
> today just the costs of streaming down a 90 minute divx
> video with 1000kbps at the side of the host (website)
> is at least as high than what one has to pay to rent out
> two VHS tapes at the local video shop. (6 euro) any royality
> payments have to be added to it. if the video content industry
> would be willing to compete with p2p services they would have
> to first find a way to pay prices for bandwidth which are
> not 'free market prices', it possibly means they would simply
> own those backbone infrastructure, which is available for
> a low price at the moment. once described heroically by
> neil stevenson, the big fibreoptic grid is "for sale" like
> a ghost town after the gold rush.

Hmmm, that is chillingly accurate.
> another technology, called multicasting which would make internet
> broadcast economical, is not happening yet, because some smart
> companies are customizing old switching protocols to the max,
> developing 'smart routers' and therefore adding complexity on
> the backbone level, instead of upgrading to the next stage.
> legislation of standards is not taking place, because the
> 'out of control' ideology still supports in the market forces.
> (see above)

One consequence of the meltdown in the past year is that ISP's (and most 
other tech businesses) are oriented to make do with less. There is a 
global gun-shyness of investment in radically new technology, and this 
is going to delay implementation of broadband, wireless and other 

> it is obvious that the future of broadband is delayed
> as long digital rights management, supported by copyright
> laws which are written for and by the industry, plus all
> kinds of dirty tweaking of routing protocols, and caching
> traffic is basically *closing* the internet modeled after what
> is known from private sattelite tv. in the moment payment
> schemes will be in place and alternatives can be shut down
> technically and legaly, very likely broadband will be
> there immediatly.
> strange is that in germany, the royality collectors are
> claiming to get payed back for every piece of hardware,
> cd-burners, harddrives. the other option, that internet
> traffic itself, through some kind of copy-tax, a per
> gigabyte fee for using the internet like radio stations
> or public tv is not considered yet.  it would  possibly
> enable the smaller content  holders and providers to run
> viable business models. it is again a mix of cultural
> backwardness and "mafia lobbyism" which supports models
> which are not innovative and do not support small business.
> the time in which filesharing is sponsored through
> DSL flatrates can be also over soon.

Bell Canada is trying exactly such a byte-levy in Canada for their DSL 
service. We will see how far they get.
> there are plenty of projects which really need "public
> bandwidth" to contine to provide their cultural content
> for free, and i speak about many terabytes a month.
> there is plenty of interest on the side of the users.
> i'd be interested if nettimers are interested to discuss
> these practical issues on this list.

Hear Hear. But come to think of it, what cultural content would suck up 

> it is one thing to make a "we want bandwidth" campaign
> sucessful, it is another to find 'open' ways to
> distribute and allocate available public bandwidth
> usefully once you got a lot of it from your local
> cultural authority... so far institutions of
> media culture do not see their role in the tradition
> of a project like  maybe that can
> be changed?

I see the lack of familiarity with issues facing the 
telecom/cable/backbone providers as one of the major impediments in 
getting people to realise bandwidth is a resource to be appreciated. 
After that would come the challenge of how to distribute "free" access.
Morlock Elloi responded:

> Aside, there is no such thing as "social needs" - that is the phrase 
> used
> to justify whatever needs to be justified at the moment (to paraphrase
> famous propagandist, "when I hear 'social needs' I go for my gun").


> The issue with bandwidth is really simple. There is no content (outside
> movie industry) to justify it. Average user has nothing to offer to
> average user. Zilch. Zero. Average user is a dumb empty nitwit that may 
> be
> able to create 0.5-1 kilobytes of original material per day. And outside
> his own house he can't really force his family videos onto anyone. The
> only other possible use would be videoconferencing, and guess what -
> people don't really like to videoconference.
> So almost all P2P bandwidth today is used for re-distribution of
> commercial stuff or material made high-value by government intervention
> ("illegal"). Assuming that someone will find interest in making this
> cheaper in today's global juristiction is plain silly.


(I'm reminded of people that have downloaded recent blockbuster films to 
discover they were shite when they viewed them.)

>> in scenarios resembling the techno-futurism beginning in the
>> 50ies, with fully automatic homes, and a user experience directly
>> inspired by the narratives arround extraterrestial space 
>> colonialisation.
> As opposed to some "natural" societal evolution ? Bullshit. Masses were
> always and always will be buying dreams.

What I find particularly fascinating is that the people who were sucked 
into this la la land were the banks (Canadian banks were remarkably 
vulnerable to this hype). Call it greed. The average consumer was and is 
pretty sanguine about utopian dreams, but will respond to marketing 
efforts eventually.


Enb's comments were well-spotted:

> Well- it does matter! Even if there have been overcapacities, there was
> "some!" routing trough the cables of KPNQWest/Ebone. I've heard of a
> capacity of 80+ Terabits of possible transfers trough their net- if 
> only 5
> Terabits were used(for example), it is sure an overcapacity- but to stop
> that net now, it would mean that 5 Terabits must get routed trough the 
> nets
> of other companies. The question for reprogramming all those routers is:
> Where does KPNQWests/Ebones "net" begin, where does it end?

> And a second thought:
> They've served not only thousands of B2B-customers like small/medium 
> ISP's
> and Webhosting/Housing-Providers, but also ten-thousands of 
> B2Endcustomer
> like companies and SOHO. Those big enough to have lines from 2 different
> ISP's are the lucky ones now(if at least one of them doesn't rely on
> KPNQWests backbone), the others rush to find an alternative from
> KPNQWest/Ebone (or hope that a solution will be found).


> Somewhere I've read
> the following from an ISP "usually, we've had 5-10 requests for a big 
> new
> line per month- now this number increased to 100+". Sure these dealings,
> coordination with local telco and installations all will take their 
> time,
> 2-3 weeks for a complete migration in normal times are ok, at the 
> moment,
> 4-6 weeks are fast...

Not sure where you dig up your sources, but they are spot-on. And if 
this happens again in, say 6 months time with another backbone provider 
(a situation which is developing) the fall-out will be even more severe.

> But also: All these assumptions of "what will happen" are useless as 
> long as
> they find new investors every other week ;-)

Which they can't, hence the seriousness of the situation.

Thanks for your indulgence.

Be well,



#  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: and "info nettime-l" in the msg body
#  archive: contact: