LISTSERV mailing list manager LISTSERV 15.5

Help for NORDNOG Archives

NORDNOG Archives

NORDNOG Archives


Next Message | Previous Message
Next in Topic | Previous in Topic
Next by Same Author | Previous by Same Author
Chronologically | Most Recent First
Proportional Font | Monospaced Font


Join or Leave NORDNOG
Reply | Post New Message
Search Archives

Subject: Economic issues for ISPs (was: RE: IP limits from RIPE?)
From: Lars Michael Jogback <[log in to unmask]>
Reply-To:Network management discussion for Nordic region <[log in to unmask]>
Date:Sat, 14 Dec 2002 13:00:21 +0100

text/plain (79 lines)

> To be frank. I have never seen any figures about this. What are the
> actual cost for doing IP on ADSL to the home? I'm sure there are a
> lot of things to consider here. Hardware, software, amount of people
> working with it and so forth. Anybody here that are familiar with
> this?

I feel that this topic is a little off-topic for this list (it's not
very operational) but never the less it is an interresting question.
If we should move this discussion off-list please tell.

I think that you should divide the cost of producing the
"IP-packet moving" service in five different parts.
These parts should all include HW, SW and HR for their part.

The first cost element is the costs involved with getting some sort
of connection from the network out to the home. This could be pure
opto, barbed wire or radio. The cost is their regardless if you dig a
new fiber or rent a copper from someone that put it their several years
The structure we have in Sweden right now is a little bit wrong, because

the price that you need to pay to lease an already existing copper pair
almost as high as getting men and backhoes out digging new fiber, and
this is, IMHO one of the biggest problems.

The second cost element is the equipment for generating some kind of
service on the connection. The price is different very different
dependening the technology used. Different technology has their own
pros/cons, regarding price/bandwidth, scaling, initial cost, lifespan
and of course religion. The only problem here is to find the best mix
of pros/cons and to make the right descission. Easy. (Yeah, right :-))
Sometimes you don't have a choice since you stuck with the selection
done by the provider that make element #1. This is even worse.

The third cost element is the cost for building and running a backbone
network that transports the IP-packets around in Sweden and out over
the world. Again you have several competing technologies. Here you also
have to battle the "food chain" of Internet, with the different issues
for peering/transit.

The fourth cost element is the pre-sales costs. This includes marketing,
sales cost etc. I don't have a good knowledge of what the cost drivers
is for this element, but I don't think this is the problems.

The fifth cost element is the post-sales costs. I'm not counting the
for running the network (element #1, #2 and #3) here, because this
should be covered there. This more costs for billing, credit issues,
end-user technical support, customer-care etc. I don't have good
knowledge of these either, but I don't think the problem is here.

To summarize this I think the problems is in cost elements #1 and #3.
(Or #1, #2 and #3 if you only can get some bundeled L1/L2 access)

In Sweden the biggest problem is the last mile. The copper in Sweden
is very expensive to rent from Skanova, and Skanova has built their
own ADSL-network based on first/second generation ADSL-equipment with
no competition. There is very little options available for the operators
that would like to compete, especially if they'd like to make their own
descission on element #2.

> PS. First time I heard about the price for broadband to the home and
> comparing that with the cost of a monthly bus ticket it was Peter
> Löthberg in 1997. I'm not sure he knows more about bus tickets than
> networks.  :-) DS.

The problem with comparing residential broadband with bus-tickets is
that bus-ticket pricing is highly subsidized from the government.

I think that element #2-#5 should be market-driven, but element #1
is something that cost very much initially and the payback time is
quite long so it's hard to do this market-driven (especially when
there is a de-facto monopoly on the copper network, which where
not built market-driven)

Best Regards,

Back to: Top of Message | Previous Page | Main NORDNOG Page



CataList Email List Search Powered by the LISTSERV Email List Manager