This is only a rough draft - Megan 04/15/92

The Services Location Working Group met and the following was discussed.

Scott Kaplan of FTP Software made a presentation of an architecture
proposal he has been working on.  Some of the principals that have been
made in the design of this protocol have been that the end nodes (these are
the nodes searching for services) would have no user interface
configuration.  This means that the information that is reported in the
user interface of the end node is information that is contained in the
network, in the services access points or in surrogates for these access
points.  Other protocol design principals are that a data representation
language can be borrowed or stolen, that some RPC mechanism can also be
borrowed.  Some issues that were brought up in the discussion of the
protocol is that this protocol is not SNMP even though there are
similarities between the two protocols (both protocols can transmit a list
of attributes that are supported by the service provider).  This protocol
should deal with both the dentist office case (plug and play IP in a small
unconnected network) as well in a well connected global internet.  The
protocol design document will be refined and posted to the mailing list.

Issues of multilingual support will be worked on by John Veizades and will
be presented at the next working group meeting in Boston.  It is assumed
that there is some character encoding standard that can be used for this
issue.

There will be a BOF session on this topic Thursday night of InterOp
Washington.

Scott and John met with the IRTF working group on resource location.  There
are many similar issues in the work that this group is trying to solve.

Submitted by
John Veizades
working group chair