Re: SPEC: DNS conventions & "naming" of VRML sites

[email protected]
Thu, 15 Jun 1995 16:48:51 -0400 (EDT)

I don't abree with what Andrew has written below. A convention for server
names would be nice if it stayed static, because chanbing a directory
name is easier than chanbing a machine name/alias. Thi , combined with
the fact that vrml is not going to be wired forever, which if we wanted
only the most wired server name , would entail chanbing server name each
time a new standard type comes out. If we just had one convention for
server name www or whatever it may be and derived content info from file
ending or directory name, it might be more elegant a solution. The only
reason that www is a logical choice for the server "standard" is that the
World Wide Web doesn't imply an information type, all info can flow
through thi I don't truly believe that this standard should be enforced though. If
someone really wants to have a wired server name, then they should be
able to do it.

Owen


Owen D. Johnson -------------------------------- voice: (617) 577-9800
net.Genesis fax: (617) 577-9850
56 Rogers St [email protected]
Cambridge, MA 02142 --------------------------- http://www.netgen.com/

Once again, my opinions are my own and not the opinions of my company.
----------------------------------------------------------------------
On Thu, 15 Jun 1995, Andrew C. Esh wrote:

> Following your reasoning, we could apply the same naming ideas to www,
> since it is a superset of both FTP and Gopher. We should have not allowed
> any new servers to be named "www.", and they should instead be
> "gopher.host.com/www/..." or even "ftp.host.com/gopher/www/...". We don't
> see much of that out there.
>
> Thi
> I would expect it
> formats as well. Advertising "www." in a VRML-fixated market i
> leave you behind. "www." is tired, "vrml." is wired.
>