Mosaic, Netscape and Net Expenses--Was Re: Network Abuse by Netscape?

Thu, 27 Oct 1994 13:17:19 +0100

On Tue, 25 Oct 1994, Robert Raisch wrote:

> On Sun, 23 Oct 1994, Phil Trubey wrote:
> > If you're worried about how Netscape will impact web space providers,
> > well ths choice is simple - either keep up with the technology (ie.
> > get a Unix machine with better i/o performance), or don't operate
> > a web server. Nobody is forcing you to operate a web server.
> Ok Phil, forgive me if I interpret your words in slightly different way:
> If you're worried about how Netscape will impact your web site,
> --which you provide as a public service or as a labor of love--
> make sure you keep up with the technology because the tools
> are now going to assume that you are fully responsible for the
> performance bottlenecks.
> I can see this being a real damper on the provision of free information
> resources on the net. Mosaic and Netscape have significantly up-ed
> the ante when it comes to providing content....

> But what really worries me, I think, is that design decisions seem to
> be made with the idea that we can do anything we want because the
> network will react by becoming faster and more efficient. I do not
> believe this to be the case.
> And even if this appears to be the case now, there must be a point
> where this assumption will cause the providers of the network to
> radically change its economic model.
> Without understanding the implications of technology, we may end up
> inadvertantly creating the pressures that will drive the net to
> something we will all regret, like packet charges.

Has anyone been keeping track of what the economists have to say on this
topic ? According to Hal Varian (U. Michigan) packet - or similar -
use -based charges are more or less inevitable (though difficult to
implement right now because of the router problem). There has also been
discussion of this topic on list VPIEJ-L - primarily at the behest of
Stephen Harnad.

If I can be permitted 2x my $0.02 worth...

1. We have had Mosaic running at this site on an experimental basis since
May. Many academics - including myself - would be overjoyed to join the
Mosaic "club" - but the blunt truth is that we don't have the capacity to
handle the traffic that the widespread use of this browser is likely to
generate; for similar reasons the provision of a www server at our site
is out of the question. We are likely to upgrade our capacity within a
year (or two) but the rise in connection charges is likely to be steep,
and in any case improved Internet connectivity is going to take a back
seat to LAN connectivity - as, in my view, it should - and to supplying
computing resources to students.

2. I don't think that anyone should be under the illusion that improved
technology will leave the network unchanged. Two developments are likely
- increased pressure to upgrade will eventually drive limited resource
providers off the net altogether or into niches (backwaters ?), and, via
pressure on the capacity of the system, the increase in traffic will have
the same effect - by hastening the packet
charge scenario Robert Raisch mentions. Packet charges will have the same
effect as any charges for previously "free - to-end- user" goods -
consumption (and possibly packet charge levels) will gyrate around until
some equilibrium is reached - an equilibrium, I suspect, which many
individuals - and not a few institutions - will find unacceptable. Result ?
Welcome to MY world ... I happen to be an accountant.

I DO NOT find this latter scenario attractive, and my preference is for a
low flat rate fee structure for non-commercial and academic use, a
distinction between academic and commercial users and a "provider pays" fee
system for the latter linked to user size. However, I doubt that this is
realistic - the Net is too close to a free market in structure, and
although the Net community may protest - even ostracise - innovations
such as Netscape, the authoring company is no Cantor and Siegel. We shall
have to get used to the fact that the Net is going to change in reponse
to graphics - based browsers, and find some way of living with the

Roger Collins

PS Just a thought - is there any way for a site to restrict access to
browsers having attributes the site finds "unattractive" (my guess is
"not right now").

(Sorry I don't have any URLs handy as references for the above - this was
written "on the fly"...)