🚀 go-pugleaf

In article <a2cv5r$10lj4c$1@ID-106624.news.dfncis.de>, michael lefevre 
<michael.news@michaellefevre.com> writes
>i don't see
>that would have a huge impact on routing, name servers and all the rest
>of that stuff...

Let me try to explain it a slightly different way...

Giving you some example figures,  the  cost of all that investment (or 
at least servicing the capital required) plus the ongoing running cost 
is (say) �10,000 per month (note we're not talking about disk costs).

If an ISP's infrastructure supports 1000 subscribers with 10M disk 
space, each paying �10 per month then the ISP will break even 
financially. Increase each subscriber's webspace allocation to 20M then 
there will be a requirement for more bandwidth (there's a direct 
correlation between diskspace allocation and required bandwidth). More 
bandwidth (twice as much) will require more bandwidth (twice as much) 
into the Internet. There will be impact on the performance of the 
firewalls, routers and so on (in fact all parts of the infrastructure), 
so more investment to support the increased load will have to be 
financed - more investment - an increase in the cost of servicing the 
capital investment and the cost of maintenance. The revenue hasn't 
increased - so the ISP will lose money. Ergo, the ISP would be 
business-foolish to give away more bandwidth willy-nilly.

I would have thought that you could have worked this out for yourself. I 
would also have though that it is obvious that equating the cost of 
(say) doubling the amount of webspace entitlement with the cost of a 
disk at your local PC World lacks understanding of the issues - issues 
that I have tried to illustrate. If you can't understand simple 
economics then I'm sorry. Its called business. But please don't bitch 
when someone tries to explain it.
-- 
Tony Morgan