isp in usa....

[Heiko Recktenwald <UZS106@ibm.rhrz.uni-bonn.de>]

In USA laeuft gerade eine nette Diskussion ueber quality of service,
zukunftschancen kleinerer ISPs, qualitaetslabels etc. Z.B. so:

> It is true that the need for additional computing resources and technical
> staffing does not significantly increase as subscribers are added.

This is not clear to me.  Some, perhaps even most, aspects of the
ISP business are fairly support intensive.  Each new customer requires
some number of hours of support to get started and some amount of
support on a continuing basis.  True, there is a base level of staffing
required to run an ISP, but there is also an increment of staff time
required for each additional customer.

> What
> *does* increase is purchase of  communications hardware (modems, terminal
> servers, routers) and the monthly costs of dial-in lines and additional
> bandwidth to the Internet.

> Larger ISP's do have an overall cost advantage
> and I agree that consolidations will increase in the near future.

While I agree with you in theory, I haven't seen this in practice; I
haven't seen large ISPs underpricing smaller ISPs.  Rather, it seems
that larger ISPs provide a higher level of service and/or have higher
cost structures, while smaller ISPs tend to provide a lower level of
service (e.g., are less likely to have 24x7 staffing).

I keep expecting this consolidation to occur, but, for the moment, smaller
ISPs appear to be holding their own.

If, as I assert, the ISP business is fairly support or personnel
intensive, it would be interesting to examine whether the availability
of good personnel is limiting the growth (or perhaps the quality) of ISPs.
It would be also interesting to compare the effectiveness of small ISPs
and large ISPs in competing for the finite pool of expertise.