IETF Amsterdam: Report

Jill.Foster@newcastle.ac.uk Fri, 23 July 1993 18:51 UTC

Received: from ietf.nri.reston.va.us by IETF.CNRI.Reston.VA.US id aa10917; 23 Jul 93 14:51 EDT
Received: from CNRI.RESTON.VA.US by IETF.CNRI.Reston.VA.US id aa10912; 23 Jul 93 14:51 EDT
Received: from sun2.nsfnet-relay.ac.uk by CNRI.Reston.VA.US id aa02074; 23 Jul 93 14:51 EDT
Via: uk.ac.mailbase; Fri, 23 Jul 1993 19:51:03 +0100
Received: from [+JANET.000001500100/FTP.MAIL] by uk.ac.mailbase; Fri, 23 Jul 1993 19:34:46 +0100
Received: from eata.ncl.ac.uk by cheviot.ncl.ac.uk id <AA20591@cheviot.ncl.ac.uk> (5.65cVUW/NCL-CMA.1.35 for <nir%mailbase.ac.uk@newcastle.ac.uk>) with SMTP; Fri, 23 Jul 1993 19:34:24 +0100
Date: Fri, 23 Jul 93 19:34:16 BST
Sender: ietf-archive-request@IETF.CNRI.Reston.VA.US
From: Jill.Foster@newcastle.ac.uk
Subject: IETF Amsterdam: Report
To: nir@mailbase.ac.uk
Message-Id: <emu-ct08.1993.0723.183416.njf@eata.ncl.ac.uk>
Reply-To: Jill.Foster@newcastle.ac.uk
X-Orig-Sender: nir-request@mailbase.ac.uk

 IETF - Amsterdam: July 12-16, 1993  
 ================================== 

 Trip Report:
 Jill Foster - Newcastle University, UK
 Chairman: RARE Information Services and User Support Working Group 


 Introduction
 ============

 This was the first time the IETF (Internet Engineering Task Force) had
 been held ouside of North America.  Phil Gross, the IETF chair, opened
 the meeting.  He pointed out that the European section of the Internet
 was the fastest growing segment.  In spite of the concern about the
 effect on attendance of holding the IETF ouside the USA there had been
 480 pre-registered attendees.  There was a roughly even split between
 US and non-US attendees, with 66 WG/BOFs meeting in 90 sessions.
 
 My main reasons for attending (thanks to funding from RARE) were to:
 
 o    represent RARE Information Services and User Support Working
      Group (which I chair)
      [RARE is the Association of European Research Networks]
 
 o    join in the User Services and associated WG sessions.
 
 o    co-chair a WG session on networked information retrieval tools.

 o    co-chair a WG session on network training materials.
 
 The following informal report is in note form and deals mainly with
 the areas of User Support and Networked Information Retrieval,
 although reports of some of the plenary sessions are also included.
 Whilst it is as accurate as I can make it, it is naturally a personal
 account and may be inaccurate due to lack of background information or
 misinterpretation of what I heard.  Corrections of fact are welcome,
 but any discussion of items contained here would be best directed to
 the appropriate mailing lists.  Minutes of individual sessions are
 also available via anonymous ftp from nri.reston.va.us
 
 This report will be stored on the UK Mailbase Server.  To retrieve a
 copy, email to Mailbase@mailbase.ac.uk with the following command in
 the body of the message:
 
      send wg-isus ietf.07.93

 Alternatively use anonymous ftp to:  mailbase.ac.uk

 file:  pub/wg-isus/ietf.07.93

 [Also available via gopher]

 Note: in general I have not expanded acronyms as those readers
 involved in a particular topic should know them whilst those who
 aren't familiar with the acronyms should still be able to get a
 reasonable overview of the topic.

 All addresses quoted in the report are in internet (rather than UK
 JANET) order.

 Each section has a double underlined heading - to enable you to skip
 sections not of interest.


 Plenary
 =======

 Erik Huizer (SURFnet, NL and IESG (Internet Engineering Steering
 Group)) gave a "European's perspective of the IETF".  Each of his
 overhead transparencies had a title in Dutch in order to give the
 native English speakers a flavour of what an IETF is like for a
 non-native English speaker.  The fact that the titles contained jokes
 only decipherable to those understanding Dutch added to the
 experience.  (In describing RARE and RIPE's relation, for example, the
 title was "Hatred and Love".)

 Erik (who is the IETF Applications Area Director) stressed that the
 current networking activities focussed too much on infrastructure and
 band width issues and not enough on applications, services and users.

 The European networking scene consists mainly of national research
 networks (such as SURFnet) with some multinational networks such as
 Nordunet and EARN.  Most of these are multiprotocol networks, many
 with up to 2Mbps links.

 There are now two international backbones in Europe.

        EBONE           IP only, 2Mbps
                        participants pay

        Europanet       Multiprotocol
                        2Mbps
                        part EC funded

 RARE and RIPE:

 RARE is concerned with network politics and the technical programme
 and provides a secretariat.  RIPE is concerned with coordination of
 the IP networks and runs the RIPE NCC (Network Coordination Centre).

 RARE Technical work structure:

        RTC: RARE Technical Committee

        8 Working Groups
          International Character Sets          WG-CHAR
          Interactive Multimedia                WG-IMM
          Information Services and User Support WG-ISUS
          Lower Layers Technology               WG-LLT
          Mail and Messaging                    WG-MSG
          Network Applications Support          WG-NAP
          Network Operations                    WG-NOP
          Security Technology                   WG-SEC

 In Europe the commercial networks are mostly X.25 networks with
 closed user group networks.  X.400 is important here.  However the
 PTOs are losing their monopoly.

 The commercial IP based networks are growing rapidly.  There are
 various national providers and Eunet.

 In Europe we have (at least) 2 protocol stacks:

 TCP/IP
        still mostly academic and research
        growing fast
        not subsidised
        not really accepted for doing business yet

 OSI
        CEC
        government
        business

 ISO liaison with IETF/Internet Society is politically very important
 for IP networking in Europe.

 Europe has a tradition and large investment in OSI networking.  In
 order to get widespread acceptance of Internet technology, we need to
 see the internationalisation of ISOC in order to give us a more stable
 organisation, with an open and accepted standards process.  The ISO
 liaison is valuable from this point of view.

 Involvement of all kinds and at all levels is needed from the academic
 and research community, the commercial sector and government agencies.
 A balance is needed however as we must take care not to destroy the
 IETF through bureaucracy.

 In preparing for the IETF, SURFnet had set up a gopher server which
 contained a great deal of useful tourist information about Amsterdam
 and the Netherlands (scanned in with the permission of the Tourist
 Board) as well as information about the IETF itself.  Local volunteers
 wore T-shirts bearing "volunteer@ietf.amsterdam.nl"m.nl".  These were in
 unbleached cotton, reflecting the Dutch concern for the environment.


 Uniform Resource Identifier WG: Alan Emtage (Bunyip), Jim Fullton (CNIDR)
 =========================================================================

 The URN is now the union of Uniform Resource Location and Uniform
 Resource Name.  The idea is to identify information resources uniquely
 and to allow the location of these by navigational tools.

        U = Uniform
        R = Resource
        { L,N,I, .... } = { Name, Locator, Citation, ... }

 In the interests of making real progress in this area, this group had
 three separate sessions scheduled during this IETF. 


 Session 1
 ---------

      There was some disagreement as to whether or not URLs (Uniform
      Resource Locators) should be human readable.

      Alam Emtage said they should not be human readable, but should be
      human transcribable.  Tim Berners Lee wanted to define the set of
      characters that could be used in a URL.  Peter Deutsch was
      against this, fearing it might restrict further use.  He also
      felt that there should be a separation between the human readable
      part and the access method (cf the gopher readable title and
      selector string).  Brewster Kahle commented that they had tried
      to use a general mechanism in WAIS and that it had gone badly
      wrong.  He felt it would be better to go for a reasonable
      compromise and that a URL should have a restricted character set.
      There had been much discussion on the mailing list and no
      concensus reached at the meeting.  It was decided that, having
      aired the issues to a wider audience, discussion should take
      place outside the meetings and the topic would be revisited in
      the last URI session.

      The three proposals were:

      -    no restriction on character set

      -    significant restriction (ie current proposal)

      -    restriction - but to include more characters (including
           space).

      Chris Weider gave a short tutorial on URNs (Uniform resource
      names).  This is designed to be a persistent reference to a
      resource on the network.  Some form of directory service is
      required to map a URN to zero, one or more URLs.  The URN is
      needed to detect duplicates.  The URN determines whether the
      intellectual content of objects are = (same URN).  (This
      equivalence is of course open to debate.  It is assumed the agent
      assigning the URN decides on equivalence.)

      The URN is deliberately simple.  The difficult bits had been
      punted to the URC (Uniform Resource Citation).

      URN Structure

      URN : naming authority :: opaque string :::

 Session 2
 ---------

      Karen Sollins gave a presentation on InfoMesh, a research project
      at MIT.  She stressed this concentrated on research into the
      issues rather than providing operational services.  The location
      and naming of information objects were separated.  The
      information and relationships between information objects had
      values.  Information objects and links could evolve and change.
      Therefore it might not be possible to guarantee the resolution of
      a link.  Users should not be shielded from these problems.

      Karen was of the same opinion as Peter Deutsch, that various
      aspects should be kep as general as possible.  She talked about a
      "bag of factoids" about an information object.  We should define
      how these are passed around but should not restrict what these
      could be.

      Brewster Kahle suggested that URN, URL pair should be the norm
      and that the URN->URL resolution via a directory service should
      only occur if the URL failed.  Chris Weider said this was exactly
      what was coming out of discussions on URCs.

      URN Questions

      Format : URN : NA :: OPSTR :::

        -  why naming authority?

        -  why can't we use something like an MDS check sum?
           (no restriction on a local URN?)

        -  what is the definition of "Readable"?

        -  where do naming authorities come from?

        -  why does the syntax of URL and URN differ?

      Peter Deutsch:

      We have: a taxonomy
          -  URCs
          -  URNs
          -  URLs

      Have specific URL paper
           generic URN paper

      Main concern: Don't have a generic URL paper and this is causing
      problems.

        Issues for URL paper
        1)   what we will be dealing with
        2)   how we will transport them


      How is a URL moved around the net?  We shouldn't put restrictions
      on it, but should allow arbitrary strings and define mappings.
      Need to say what we are trying to encode.

      What we will be dealing with:

      1)   opaque strings (no semantics)

      2)   we should allow arbitrary selectors without predetermining
           semantics allowable

      The Scheme should be extensible - but it may be necessary to make
      some compromises on this to get this going.

 Session 3
 ---------

      There was a discussion on which characters to allow in URLs.

      Compromise proposal

      Include: space : ; = " ' ( )

         national character sets

         problems: noting URLs on the back of an envelope
            passing around filenames with o" etc within country

      Exclude: > < | { } \ ~ [ ] \ ^ # %

      Peter Deutsch had proposed that the URL definition should *not*
      restrict the character set used.  These discussions had been
      going on for many months without a concensus emerging.  Some
      people said this was holding them back from implementing the
      scheme.

      URN: It was felt that the persistent naming feature was more
      important than the capability of testing for equivalence.


 Summary: (points 1-3 from Tim BL's message)

 1.  The afs: URL type was removed from the spec due to lack of interest
     and because it had not been really discassued or thought out.

 2.  The mailto:mail@add.ress url was put in.

 3.  The practice of wrapping URLs within plain ASCII documents
     within <> (as is commonly done with mail addresses and news
     message IDs) was put in.

 There will be a new URL paper and a paper on URCs (Uniform Resource
 Citation).

 URC: URN, URL + format specifier for citation information (attribute:
 value pairs).  There had been several proposals and these will be
 discussed on the list.

     Mailing list:          uri@bunyip.com
     To join, mail to:      uri-request@bunyip.com
     Archive:               archives.cc.mcgill.ca
        Directory:          /pub/uri


 User Services Working Group    Chair: Joyce Reynolds/ISI
 ===========================

 Joyce introduced this as being the umbrella group for the working
 groups within the User Services area.  It spawns other WGs and tries
 to avoid overlap with the activities of the RARE ISUS WG.

 The FYI (For your information) series of RFCs (Requests for comments)
 are a group of "living documents" that are periodically updated.

        FYI 2:  Network Management Tools Catalogue
                Updated June '93
                This is very long (192 pages) and should only be
                browsed on-line.

        FYI 19: A short bibliography of introductory Internetworking
                Reading

        FYI 20: What is the Internet    May '93

        Draft Revision of FYI 4: Questions and Answers for new Internet
                Users

                Currently at ftp: naic.nasa.gov//files/fyi4-june93.txt

                Gopher has been added and the suggested reading section
                updated.

 The mailing list has moved.  It is now at

        us-wg@nic.near.net      (see below for details on joining it)

 Joyce thanked BBN NSC for hosting the list for so long.

 The need for four or five well known repositories of introductory
 information was discussed.  Guidelines on publishing and on ethics and
 netiquette should be made available to new connecting sites.  Need
 guidelines for major ftp mirror sites.  Perhaps the first screen of
 information (ftp) or gopher front page could list the other major
 sites and suggest users use their closest site.  The California
 Department of Education has written something - which was actually
 intended for the K-12 (school) community - but which might prove
 useful here.  Details of this will be posted to us-wg.  We also need
 to remember that there are now very good books available on the
 Internet which cover these topics.

 Cindi Mills of BBN reported that the nnsc machine went on June 30th.
 Everything of importance should now be on the Internic machines:

        rs.internic.net         (registration services)
        is.internic.net         (information services)
        ds.internic.net         (database services)

 A new on-line form of BBN's resource guide is on the internic and
 available via gopher.

        gopher.internic.net

 The same gopher allows "seamless" access to the three parts of the
 distributed Internic.

 The Hypercard tour has no home at present.  The Internic will be asked
 to host it.

 David Sitman reported on the EARN "Guide to Network Resource Tools"
 (May '93).  He wanted input for this and would be submitting it as an
 FYI RFC.  He stated that it was aimed at end users who had email
 experience - but not much more.  The guide is intended to give them
 enough information to get them started using the tools (i.e.  how to
 use the network not to learn about the network).  Some EARN specific
 tools (such as Trickly and BITFTP) will be "demoted" from the main
 section for the FYI.

 The "Guide to Network Resource Tools" is available via Listserv in
 plain text and Postscript versions.

 Send mail to LISTSERV@EARNCC.BITNET with the line:

 GET NETTOOLS MEMO      (plain text version)
        or
 GET NETTOOLS PS        (Postscript version)

 The guide should now be available from ripe.net for access via FTP,
 gopher, WAIS and WWW (at ftp.ripe.net, gopher.ripe.net, wais.ripe.net,
 etc)

 Comments about the guide should be sent to:

        EARNDOC@EARNCC.BITNET

 I gave a report on the RARE Information Services and User Support
 Working Group and its various task forces:

        User Documentation
        Document Delivery
        Networked Information Retrieval
        Coordination of Networked Information Retrieval Services
        Training, Publicity and Awareness
        Support for Special Interest Communities
        Multi Media Information Services

 See the report ftp: mailbase.ac.uk//pub/wg-isus/isus-028
 for more details

 Joyce circulated a message from Naomi Courter who was interested in
 finding people who would be willing to work on defining a "clean user
 interface" for novice users (probably menu based).
 (naomi@boogie.concert.net)

        Mailing list:           us-wg@nic.near.net
        To join, mail to:       us-wg-request@nic.near.net


 WHOIS and Network Information Lookup Service Working Group: WNILS:
 =================================================================

 Joan Gargano (UC Davis)

 The WHOIS++ project aims to develop a lightweight useful Internet
 Directory Service using simple technology.

 Status of WHOIS++ Architecture: Peter Deutsch
 ---------------------------------------------

      Draft architecture document written.  Designed to be general as
      well as being used for URN-URL look up.  Extensible.  Yellow
      Pages type services.  There are hooks in for support for
      different character sets.  There is a mechanism in the protocol
      for querying the options supported.

      Quoting from the document:

           "This document describes WHOIS++, an extension to the
           trivial WHOIS service described in RFC 954 to permit
           WHOIS-like servers to make available more structured
           information to the Internet.  We describe an extension to
           the WHOIS data model and query protocol and a companion,
           extensible distributed indexing service.  A number of
           options have also been added that enable such feautures as
           the adding or updating of information records, the use of
           multiple languages and character sets, multiple views onto a
           single set of data and a number of other useful features.
           An optional authentication mechanism for protecting all or
           part of the associated WHOIS++ information database from
           unauthorised access is also described."

      The design philosophy is to keep it simple and to provide lots of
      options.  Ouput formats need to be defined (and put in the
      document) so that people can start to write clients.

 Status of the Distributed WHOIS++ Model - Centroids: Chris Weider
 -----------------------------------------------------------------

      This defines a mechanism for handing around indexing information
      in such a way as to allow "pruning" of the search tree.  (Only
      those parts likely to yield hits are searched.) There was some
      question as to whether this would scale.  It might be OK at
      present when there might not be a "Smith" (or whatever) on every
      server, but as the information becomes more dense, the result
      might be that every node (or a large percentage of nodes) were in
      fact searched.

      At present the aim is to put in place temporary solutions.  It
      was felt that some hefty testing by loading in existing data
      (from CSO directories for example) would be an appropriate first
      step, before persuading people to install and populate WHOIS++
      servers.

 Status of WHOIS++ Clients: Jim Fullton
 --------------------------------------

      Mark Prior had developed a WHOIS++ Server, Chris Weider was
      working on Centroids.  Simon Spero and Kevin Gamiel were each
      working on WHOIS++ clients.  It was hoped to have all components
      ready by the end of the month and working together by the end of
      September.


 User Documentation:  Ellen Hoffman (Merit)
 ================== 

 Bert Stals introduced the work of the User Documentation Task Force
 which is a sub-group of the RARE Information Services and User Support
 Working Group.

 The aim was to provide documentation for end users on an introductory
 and intermediate level.  The first phase had produced a set of
 pamphlets, each on a separate topic (gopher, e-mail, WAIS, WWW,
 archie, file transfer, Mailbase, etc).  The aim was to give the user
 answers to the questions what is X?  what do I need to use it?  where
 can I find more information?  in order for them to be able to decide
 whether it was worth finding out more.

 It is intended that user support staff could use the text in their own
 documentation series (using their own house-style).  Alternatively,
 simple formatted versions of the pamphlet are available for those who
 wish to use them directly.

        Mailing list:      rare-userdoc@mailbase.ac.uk
        to join, mail to:  mailbase@mailbase.ac.uk
          the text:        subscribe rare-userdoc firstname lastname

 Ellen Hoffman asked whether the IETF should get Introductory
 documentation translated into other languages.  Various points were
 made.  It was end-user information that was the problem, not technical
 information - as most technical people understand English.  Tools
 should have user interfaces in different languages.  David Sitman
 (EARNInfo Officer) said that in his experience people were unwilling
 to translate documents, as unlike writing their own documentation they
 had no feeling of "ownership".  He also made the point that
 documentation written in the States tended to be chatty and verbose
 which was not user-friendly for non-native English speakers.

 It was felt that Applications and Tools should have in-built help and
 that the user documentation group could usefully help with this.

 The FYI - bibilography of introductory texts - will be revised soon.
 Non-English introductory texts could also be included in the
 bibliography.  (Suggestions for additions welcome.)


 Networking Multimedia Applications BOF: 
 ======================================

 Chris Adie (Edinburgh University) 

 Chris Adie started with an overview of existing systems, many of which
 are described in this survey of Distributed Multimedia (RARE Technical
 Report 5).  These included: Gopher, WAIS, WWW, Hyper-G Microcosm,
 Athena Muse 2.  He also covered the standards in this area: MIME,
 HyTime, MHEG, SGML and the standards bodies: AVIs, SMSL, IMA, MMCF.

 The aims of the BOF were:

 o      Define Goal
 o      Prioritise the list of important issues
 o      Discuss how to make progress
 o      Discuss how to interact with existing groups

 The BOF was attended by, and received good input from, members of the
 remote conferencing WG.

 Issues identified were:

        Responsiveness (QOS)
        Reliability (QOS)
        Isochronism
        Handling out-of-sync
        Reverse flow control
        Hypermedia linking
        Presentation

 We need to look at providing a toolkit to take applications produced
 using authoring packages such as Authorware, and making them
 networkable.

 The respective rights of the author and user on defining how a
 MultiMedia production is presented were discussed.

 html+ is designed to handle on-line multi media hypermedia.  (Example:
 An active area within a figure could refer (point) to a video clip).
 Hytime is complex and not widely deployed at present.

 Some issues/questions:

        Conversion tools
        Feasibility
        Which authoring package
        Use of intermediate format
        Relevance of existing tools and standards

 Need a pilot to try to get as far as possible with existing tools and
 then come up with a list of requirements.

 We need to define the scope of the project.  For full blown Multi
 Media Information Services, users will wish to browse and search.
 This will need thumbprint repression of videos, etc.

    Mailing list:       mmis@mailbase.ac.uk
    To join, mail to:   mailbase@mailbase.ac.uk
    The text:           subscribe mmis <your firstname> <your lastname>
        Archives in Directory:  /pub/mmis


 Networked Information Retrieval WG: 
 ==================================

 Jill Foster (Newcastle University) 
 
 This section taken from the minutes written by  Kevin Gamiel

 Jim Fullton gave a general overview of the MCNC Clearinghouse for
 Networked Information Discovery and Retrieval (CNIDR).  CNIDR is
 funded by the National Science Foundation to promote the use and
 development of networked information tools and systems.

 Anders Gillner from Sweden gave a presentation on the Eurogopher.  The
 Eurogopher is an attempt to establish a geographical-based as well as
 subject-based gopher infrastructure throughout Europe.

 The NIR report was discussed in some detail and edited by the group.  

 -  It was decided that NCSA's Mosaic for X should not be a stand-alone
    category, but rather indicated as a client under the World Wide Web 
    category.  

 -  The WAIS and freeWAIS categories will be merged as WAIS/freeWAIS. 
    Brewster Kahle and Jim Fullton will write parallel descriptions of 
    each and will resubmit those writings to Jill for the next update.  

 -  The Grateful Med item was voted out of the report as it was 
    generally considered not to be an open network resource.  Also voted 
    to be removed from the report was the group Lister Hill National 
    Center for Biomedical Communications, National Library of Medicine. 

 -  CNIDR are considering publishing the report online.  Access could be 
    via ftp, gopher, wais, and web.  Any details will be posted to the 
    NIR list as they become available (after the meeting, a representative 
    from the InterNIC expressed interest in doing this as well).

 -  It was agreed that the full report should be updated once per year,
    although more regular updating of the individual sections held online
    should be possible. 

 Reinhard Doelz discussed his NIR tool checklist.  The checklist was
 very extensive and ambitious.  Several schemes for maintaining the
 list were discussed including online telnet "holes" allowing the user
 to complete the checklist.  As a first cut, however, April volunteered
 to prune the checklist into a fairly simple list.  Reinhard's work was
 deemed a useful resource for users deciding on NIR tools and will be
 posted to the mailing list.

 The group agreed to remove the sections concerning the provision of
 documentation and training materials for NIR tools from the charter.
 It was agreed that this is too ambitious and out of scope.

 Volunteers for the checklist were Luc Boulianne, Ton Verschuren,
 Brewster Kahle, Jodi Chu, Marc van der Noordaa.

    Mailing list:       nir@mailbase.ac.uk
    To join, mail to:   mailbase@mailbase.ac.uk
    The text:           subscribe nir <your firstname> <your lastname>
        Archives in Directory:  /pub/nir


 NISI - Networked Information Services Infrastructure WG
 =======================================================

 Chair: April Marine (NASA)

 This group is concerned with co-ordinating NICs (network information
 centres) and improving the service they provide.

 The procedures for interNIC interaction and the need for guidelines on
 acknowledging referrals was discussed.

 NIC-forum list needs kick-starting into action.

        To join, mail to:       nic-forum-request@merit.edu

 There will be an NIC-fest which is like a trade show for the NICs and
 Internic at the SIGUCCs meeting.

 Nethelp: Impractical to implement
 Susan Calcari (Internic) had said the WHOIS template could be modified to add 
 NIC contact information.

        mailing list:           nisi@merit.edu
        to join, mail to:       nisi-request@merit.edu


 Integration of Internet Information Resources Working Group (IIIR):
 ==================================================================

 Chris Weider (Merit)

 The purpose of this working group was to start to pull together some
 of the applications (WAIS, gopher, archie) and to work on
 interoperability issues, what new tools should do and to discuss
 gateway protocol design.

 The HTML Specification will go through as an infomration RFC.  The
 Internet draft, vision document: draft-ietf-iiir-vision-00.txt (Chris
 Weider, Peter Deutsch) was discussed.

 The main problem is that some information on the network is only
 accessible via one type of service.  There are gateways, but there is
 often loss of functionality.  The vision document provides a general
 architecture to allow integration.

         ------------------------------------------------
        |  gopher  |  WAIS  |  archie  |  X.500  |  WWW  |
         ------------------------------------------------
             |                             |
             |                             V
             |                  ---------------------------
             |                 | resource location service |
             |                  ---------------------------
             |                             |
             V                             V
             -------------------------------
            |     URN - > URL               |
             -------------------------------
                        |
                        V
                    Resources

 This allows the data to be seen by a number of tools and also allows
 inter-operability.  For example: if a gopher menu points to
 information in WAIS, the URL could be passed back and then passed to a
 different tool for retrieval.

 I suggested we needed a further layer above the tools to "unite" them.
 Ideally a user should not need to know whether to ask X.500 or WHOIS
 for some information.  (UNITE = User Network Interface To Everything.
 This is the name of the RARE ISUS task force looking at this
 requirement.)

 Peter Deutsch reported that he was building something like UNITE.
 Peter and Mitra discussed the need for a common library of routines.
 Mitra saw this as being part of the client, Peter as a separate
 element.  (There was real danger of adding a seventh layer:-))


 Resource Transponders:  Chris Weider
 ---------------------

      Internet draft: draft-ietf-iiir-transponders-00.txt

      Chris went through the basics of his paper.  The problem this
      addresses is that of being able to resolve URN -> URL when
      resources move.  The idea is that each resource has an associated
      "transponder" which is responsible for updating URN->URL resolver
      services when the resource is moved.

      Possible implementations were discussed.

      o      Separate file with meta information
      o      Change operating system commands to activate transponder if
             resource moved
      o      One transponder per resource or one per machine?
      o      (futuristic) resources on the net are active processes  

      Transponder could deal with security and charging mechanism.  No
      code at present.  Karen Sollins suggested Chris look at the Argos
      (?) project and their "guardians" which are active objects.  Luc
      Boulianne from McGill is working on an information broker which
      is an active agent.  Resources are checked in and out.
      Implementations are needed to provide *real* feedback on the
      transponder paper.


 Taxonomy Document
 -----------------

      Do we need this?  If so why and who will write it?  The idea
      would be to see what tools we have in the toolbox and what is
      missing.  Mike Schwartz, Alan Emtage, Cliff Neuman and Brewster
      Kahle are apparently looking at problems of resource discovery
      and categorising NIR tools to help us understand them better.  A
      paper on this is being submitted to a journal and there are other
      related activities.  Cliff will post details/references to the
      IIIR list.  A separate taxonomy document from IIIR is therefore
      not necessary.

 HTML
 ----

      Tim Berners Lee HTML Internet Draft: draft-ietf-iiir-html-00.ps
      was discussed.  (Sorry no further notes here.)

 Gopher 0
 --------

      The original gopher protocol was being put through the RFC track.
      It would go as an informational RFC at least initially.  There
      was some concern that it did not adequately relect current
      practice.  The Gopher selector string is defined as being opaque,
      but many applications take advantage of the structure that is
      currently in it.  This needs to be noted at least in the RFC.

 Prospero
 --------

      Cliff Neuman gave a presentation on Prospero and on integrating
      information resources and improving interoperability by building
      a common framework for all applications.  This framework should
      have five functions.

      o      presentation
                - the user interface - application specific

      o      storage - common convention
                - optimised for specific data - service specific

      o      search
                - automated scanning

      o      access (how to get data once identified what want)
                application specific (eg netfind)
                - data access protocols - common library - ideally
                  common protocol

      o      organisation (how construct links between documents)
                - export of meta-information - common protocol

      Cliff naturally believes that Prospero is the appropriate choice
      for building such a framework.


 IIIR Types
 ----------

      IIIR types and MIME types have different requirements.  With
      MIME, an object is sent without knowing whether the recipient can
      handle it.  In IIIR - the client and server can negotiate what
      can be supported.  There had been problems trying to register
      gopher types as MIME types (due to the number of them) and the
      need to issue a new RFC each time to cover them.  There is now no
      such restriction on registering more types.


 Data Elements
 -------------

      This was the topic on multiple bar BOFs in Ohio.  The Library
      Community has its own set of Data Elements; the IAFA Documents
      used a different set.  (Data Element names, e.g.  "Author:",
      "Title:", and possibly data element values in some cases, need to
      be standardised.  A small closed group composed of librarians and
      people from IIIR and others are working on this.  There was
      concern that this discussion was closed.  It was pointed out that
      it was being kept to a small closed group in order to get closure
      on the initial recommendations.  These would then be posted to
      the normal open WGs for discussion and comment.

 Quality: Mitra
 -------      

      The concepts of quality, responsibility and reliability of
      Information Services needs to be addressed somewhere.  We may
      need extensions to the protocols to allow a process to
      systematically test out important links (e.g.  to "ping" gopher
      menus to check they are still there).

 Z39.50: Jim Fullton
 ------

      As a first step towards bringing the Z39.50 community into the
      IETF, Clifford Lynch will put a paper to IIIR with a view to
      having it as an informational RFC.  The paper will describe a
      mechanism for providing Z39.50 over TCP/IP.  (Well known of
      course - but needs documenting.)

      Brewster will also document the WAIS suite as a draft
      informational RFC.

 Gopher+
 -------

      Some difference of opinions here.  Some felt that gopher+ and the
      discussions had not been as open as they'd have like.  Others
      felt that the gopher folk saw the IETF process as slow and
      potentially constricting.


 Network Training Materials: Jill Foster (University of Newcastle)
 ==========================

 Ellen Hoffman announced that, due to a change in her role with Merit,
 she could no longer act as co-chair.  At the end of the meeting two
 volunteers offered to consider becoming co-chair.

 The session started by a sharing of information on training activities
 around the world (Europe, Pacific Rim, North America and Australia).
 These included subject based training, involvement of librarians in
 training, and training educators (in NASA, K-12 and German schools).

 The group would like to see network training materials used in a pilot
 on networked multimedia delivery.

 I described the work of the Newcastle based NISP/ITTI network training
 materials project.  This aims to produce generic training materials to
 be used by trainers as is - or tailored to their particular
 circumstances and/or subject groups.  The first Unit of this pack
 which is an overview entitled "The World from your Desktop" was
 produced by Margaret.Isaacs@newcastle.ac.uk.  This is a presentation
 (Powerpoint), accompanying handout, speakers notes and suggestions for
 demonstrations.  There are worksheets for an accompanying hands-on
 workshop.  I ran quickly through the disk-based presentation and
 handed out examples of the accompanying material.  This is UK biased -
 but should be easily tailorable by trainers.  Margaret had mailed
 details of how to get the training pack plus the powerpoint viewer
 (for Mac or PC) via ftp to the wg-isus and us-wg mailing lists.

 Catalogue
 --------- 

      The catalogue of network training materials was discussed.  To a
      certain extent this is being held up by the Data Elements
      discussions.  Various volunteers will look at defining the
      template for collecting the information and will each take say 10
      items from the NISP/ITTI catalogue and update the information on
      these items as well as collecting information on further items.

 Subject Headings
 ----------------

      Volunteers will agree on a *small* set of agreed subject headings
      based on the actual material collected for the catalogue.

 Subject Resource Guides
 -----------------------

      It was agreed that subject resource guides were a useful aid in
      making training relevant.  Many exist and we should pool what we
      have and develop others.  These should be along the lines of "the
      top 20 most interesting resources in subject area X" rather than
      an exhaustive list.  Australian and UK Librarians are keen to
      help here.

 Whilst we have some volunteers for these tasks - others are welcome.

 The RARE WG and US WG mailing lists are currently being used for this
 Training WG.

    RARE ISUS:         wg-isus@rare.nl
    to join, mail to:  mailserver@rare.nl
    the text:          subscribe wg-isus <your firstname> <your lastname>

    US-WG:             us-wg@nic.near.net
    to join, mail to:  us-wg-request@nic.near.net



 ISN: Internet Schools Networking WG: Jennifer Sellers (NASA)
 =================================== 

 This started by introductions and attendees explaining their
 involvement.  Klaus Fueller from the German Schools Networking Project
 (ODS) gave a presentation and gave some comments on the Internet draft
 put out by the ISN WG.

 He said the main reason for using the network was to get pupils to
 read and write better, to follow a written discourse/discussion and to
 use a foreign language as a means of communication.  Getting teachers
 communicating was another good reason.

 Klaus said the FYI should concentrate first on cultural exchanges.
 Email may well be quite sufficient in the beginning.  He liked my idea
 of an email "conference" (emeeting) and would try it.  It was useable
 even by schools with poor network connections.

 There was a need for pre-packaged installable software for using the
 network.  The University of South Africa had produced a package for
 underdeveloped countries - which would be useful for schools too.


 Technical Presentation: An Experiment in Remote Printing:
 ========================================================

 Marshall Rose (DBC)

 Marshall Rose described a facility for "remote printing" using MIME to
 convey information across the network to a remote "print server" which
 would send the information to the nearest fax machine.

 This was very much an experiment and definitely NOT a bypass
 operation.  Remote print servers had been set up to serve particular
 areas.  Geoff Huston currently runs one for the whole of Australia.

        Mailing list for discussion:
           tpc-rp-request@aarnet.edu.au

        Prototype implementations from:
           ftp.ics.uci.edu
           area: mrose/tpc
           file: fp.tar.Z
           mode: binary

 Volunteers to serve areas are wanted.  This _is_ an experiment and
 servers will be withdrawn if abused.  A question from the floor asked
 about the use of MIME audioparts and a remote loudspeaker and
 microphone!

 Concluding remarks
 ==================

 An interesting IETF with (at last) good participation from the
 Europeans, and the North Americans getting their turn of walking
 around in a jet-lagged daze.  I hope there will be a much greater
 level of European participation in IETF WGs (both via meetings and
 email) as a result of this meeting and better understanding in the
 IETF of the European requirements.

 The 28th IETF is scheduled for November 1-5th 1993 in Houston, Texas.

 Finally, a reminder that these notes are my view of the IETF.  They
 may not be an accurate view, and certainly do not cover the wide range
 of topics discussed at the workshop.  The IETF was just last week.
 I'm off now for four weeks - so the report is a little rushed.  Please
 excuse any typos.
 
 Jill Foster         (Jill.Foster@newcastle.ac.uk)
 23.07.93