Re: [NTISP Digest]

David Payer ( (no email) )
Sat, 29 Aug 1998 08:40:21 -0500

I have abundant respect for your ability and of your understanding of
Radius. After all you are the one (or one of the originals) who pioneered
Radius on the NT platform. Your knowledge of the scope of Radius exceeds

Having said that... I think you yourself show (below) how it could be
utilized as a means for a unified database for an ISP. Just as the first
version of HTML did not have methods of dealing with graphics, Radius may
not be specific enough now for use as a Authorization server for multiple
projects but the Vircom people are indeed moving to use it in such manner to
its full capacity. It doesn't seem that far off to add attributes to a
Radius auththentication that will allow access to all types of systems and

HTML was never meant to display graphics or java or launch ActiveX but it
has been readily adjusted to do so. I think the growth of Radius as an
authentication method is less of a giant leap forward. As a matter of fact,
I encourage you to find ways of making Radius a more versatile
authentication service for multiple activities. There will be a huge market
for such authentication off a centralized database. (especially with the
inclusion of attributes that will allow access to some services but not
others, etc.).

Thanks for your contribution to this forum and to the enlargement of
knowledge of the NT base for many of us.

David Payer

>> Dale, that is like saying HTML was never meant for graphics.
>As a continuing member of the RADIUS charter and development group for
>over four years, I believe I DO know what I am talking about. What part
>of RADIUS development have you been that qualifies your statement?
>RADIUS stands for Remote Authentication Dial In User Service.
>So I stand by my statement that using RADIUS for authenticating mail
>users is using RADIUS for something is was not designed to do.
>HTML was designed to show graphics. Its likely that the first
>versions did not, but there were updates to the protocol to allow for
>it. Another protocol based on RADIUS may provide for additional
>authentication types (like mail, ftp, etc) in the future (check
>out DIAMETER) but the current RADIUS definition has no hint or
>intention of it.
>Dale E. Reed Jr. (
> IEA Software, Inc. | RadiusNT, Emerald, and NT FAQs
> Internet Solutions for Today |