North American Network Operators Group

Date Prev | Date Next | Date Index | Thread Index | Author Index | Historical

"Virtual" web servers (was Re: IP Allocation)

  • From: Martin Cooper
  • Date: Wed Nov 20 10:17:15 1996

Lyndon Levesley wrote:

> Michael Dillon wrote :
> ->   On Tue, 19 Nov 1996, Pete Davis wrote:
> ->   
> ->   > With all this talk of IP Allocation, does anybody know of a time frame
> ->   > for Prodigy/AOL/Compuserve to become HTML 1.1 compliant?
> ->   > 
> ->   > We have been trying to conserve IP space wherever possible, but the inab
> ->   ility
> ->   > for 6+ million people to see "software virtuals" based on HTML 1.1 has p
> ->   revented
> ->   > us from transitioning from /32's for each site to one single /32 for tho
> ->   usands.
> ->   
> ->   Selling a virtual website without allocating a unique IP address is fraud
> ->   and will continue to be fraud for the next few years.
> ->   
> 
>  Surely that's only the case if you misrepresent the service you're selling
> when you market/sell it ?
> 
>  It would be nice to see some stats about the percentage of 1.1 compliant
> browsers that people are using, such as what percentage of web hits to
> a reasonable sample of sites are made from "antiquated" browsers ? I imagine
> that as soon as that figure fell below 1% then the product wouldn't be
> entirely unmarketable.

[ ... ]

Am I getting confused here myself, or are we talking about HTTP/1.1
rather than HTML 1.1 ?

One good reason at the moment for not moving to only providing support
for HTTP/1.1 is the lack of support for it in lynx, which many blind
people use as a browser, and lack of support for which by ISPs would
probably be fairly politically unpopular.

I guess in terms of misrepresentation we're talking about the
fairly established term "virtual web server" which I would say
has been fairly well established in common parlance as being
indistinguishable from a real web server, so an HTTP/1.1
only server at the moment could probably be said to not always
meet that definition given the above.

M
-- 
Martin Cooper
Work <[email protected]> | Personal <[email protected]>


- - - - - - - - - - - - - - - - -