Informal IETF Report

[[e-mail]]

 Apologies if you receive more than one copy of this. Jill


 IETF - Stockholm: July 17th - 21st, 1995
 ========================================    

 Trip Report:
 Jill Foster - Newcastle University, UK
 UK Mailbase Director
 Netskills Director

 Note
 ====

 Unfortunately I didn't quite get this finished before I went on
 holiday.  There may therefore have been further developments on the
 various WG mailing lists since the IETF meeting, which are not covered
 in this report.


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

 The IETF (Internet Engineering Task Force) met in Stockholm, Sweden.
 The final number of attendees was just over 600 which, considering
 this was only the second time outside of North America, was not that
 far below the 800 that attended in Danvers in March.  The local hosts
 were KTH with Bernard Stockman heading the local host team.

 A 34Mbps link to the US was installed just days before the start of
 the IETF.  This was a major achievement in international cooperation
 of the service providers.  (Paul Mockapetris, chair of the IETF,
 called this 'plug and play networking at the global level'.) The
 terminal room was up and ready a good day ahead of schedule and the
 tourist office of the City of Stockholm and staff from KTH had worked
 together to provide very comprehensive tourist information on a WWW
 server.

 The IETF sessions were held in the Grand Hotel, a truly splendid
 location.  The plenary sessions and some WG sessions were broadcast
 over the MBONE from the "Winter Garden".  This was an indoor hall with
 windows opening on to it, with chandeliers, balconies, trees and
 plants.  The multicasting set up was excellent; it was managed by the
 MICE team from Europe.  They used several video cameras (one for each
 mike around the auditorium) and provided a white-board function for
 the slides, as well as audio and video of the sessions.  The MBONE
 sessions were also put out on the hotel's cable TV network.

 The IETF started as normal with the new attendees session at 4pm on
 the Sunday, followed at 6pm by registration and the reception.  Some
 of the European Working Groups took the opportunity of meeting on the
 Sunday.  These were the Terena ISUS WG (Information Services and User
 Support) and the Terena MSG WG (Message Handling).

 My main reasons for attending this IETF were to:

 o    provide an ongoing informal liaison on behalf of the TERENA
      Information Services and User Support Working Group [Terena is
      the Trans European Research and Education Networking Association]

 o    join in the User Services and associated WG sessions.

 o    chair the 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.  Whilst
 this report 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 available via anonymous ftp from
 cnri.reston.va.us

 for example as
      html/html-minutes-95jul.txt

 or from the IETF Web server.  Note that the charters of most working
 groups and other WG and IETF related information are also now on the
 web.

      URL: http://www.ietf.cnri.reston.va.us

 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.


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


 Internet Schools Networking
 ===========================

 Chair: Jodi Ito - Uni. Hawaii

      (Jennifer Sellars - NASA)

 Jodi chaired this session as Jennifer was unable to attend this IETF.
 This group is chartered to "address issues related to the connection
 of primary and secondary schools worldwide to the network".

 Jodi reported that for the first time there had been an Educational
 colloquium for the K-12 (schools) community before the INET
 conference.  There had also been a separate "Education" track at INET
 this year which had been very successful.

 As part of the normal "sharing of experiences", Jodi reported that the
 University of Hawaii had an NSF supported state-wide project to put
 internet servers into all schools.  They had just finished a 2 week
 workshop for 250 educators as part of this.

 Recent ISN work includes the design of a template for the description
 of school projects.  These will be put in an on-line directory at the
 InterNIC.  It is hoped that it will be possible in the future to input
 the information online.

 RFC 1578/FYI22 needs updating.  This is the Frequently asked questions
 related to primary and secondary school internet users.  There was
 some discussion as to whether this was a US-centric document or should
 be more global in scope.  It *is* meant to be global, so input is
 encouraged from non-us educators too!  The value of a section on the
 relative costs of the different types of internet connection was
 queried.  It was generally felt to be too variable - and that a more
 beneficial approach might be to list the questions that should be
 asked of an internet service provider.  (The term "internet service
 provider" would also need clarifying.) Several new ISN documents have
 been written since FYI 22 was first written.  These should be
 referenced.

          Mailing list:           [e-mail]
          To join, mail to:      [e-mail]

          text:

          subscribe isn-wg <optional email address>

 Another mailing list of interest is:

 Mailing list: 
          [e-mail]
 To join, mail to:      [e-mail]
  text:

          subscribe isoc-learn <firstname> <lastname>


 User Services Working Group
 ===========================

 Chair: Joyce Reynolds - ISI

 This is the umbrella group for the working groups within the User
 Services area.  These are:

 IDS: Integrated Directory Services (joint with Applications area)

 URI: Uniform Resource Locator (joint with Applications area)

 ISN: Internet Schools Networking

 NISI: Network Information Services Infrastructure

 Trainmat: Network Training Materials

 RUN: Responsible Use of the Network

 SSH: Site Security Handbook (joint with Security area)

 HARTS: Arts and Humanities WG

 and

 USWG: User Services Working Group (the umbrella group)

 At this IETF, the User Glossary WG started up again to update the FYI
 document "User Glossary".

 Several "Liaison" reports were given.  Joyce mentioned the European
 RIPE meeting, the European joint networking conference (JENC) and the
 User Track at INET.

 TERENA ISUS WG:

 Dave Hartland, the chair of the TERENA Information Services and User
 Support WG, reported on the meeting that had taken place on the Sunday
 of the IETF.  (TERENA is the Trans European Research and Education
 Networking Association.  )

 Following the formation of TERENA in October by the merger of the two
 European networking associations: RARE and EARN, there had been an
 unsettled period as a result of 'political problems'.  These seem now
 to be resolved and TERENA looks set to move into a positive and active
 phase under its new President, Stefano Trumpy.  With a new VP for the
 technical programme, the recruitment of more Project Development
 Officers, the possibility of EC funding and the support for several
 TERENA projects, the ISUS meeting had a concrete basis for positive
 discussions.

 Members of ISUS had submitted a proposal under the European
 Commission's funding programme, Fourth Framework under the Telematics
 for Research section.  Other TERENA WGs and TERENA itself had also
 submitted bids.  The TERENA bid was to provide a supporting role to
 the Telematics for Research part of the programme (project SCIMITAR)
 to ensure that results of the programme were fed into the standards
 process by bringing them to the IETF.  The ISUS proposal, DESIRE, is
 to build an information infrastructure (based on much of the work on
 whois++, URIs etc).  This will develop tools to provide better support
 for multi-media information, to provide better indexing services and
 to provide better information management tools.  The projects SCIMITAR
 and DESIRE have been shortlisted - but with a very much reduced
 budget.

 The TERENA ISUS WG is similar to the IETF USV area.  It has a range of
 Task Forces (like IETF WGs).  Joyce and I have worked together over
 the past few years to ensure liaison and to minimise overlap.  Each
 ISUS TF has a home page with information about the charter, current
 documents, mailing list archives etc etc.  (USV WGs are encouraged to
 do the same!)

 See: http://www.terena.nl/


 USV-Web:

 As Susan Calcari had been unavoidably out of action, the InterNIC
 staff had worked hard to get the User Services Web pages in place.
 These contain information on all the USV WGs and their charters.

      http://rs.internic.net/usv-index.html

 Comments to Joyce Reynolds: [e-mail]

 The InterNIC staff had also put up their "NIC Locator Data".

      http://ds0.internic.net/ds/niclocate.html

 Web versions of FYI RFCs:

 Janet Marcisak (FTP Inc) announced at the last IETF that she had just
 completed marking up ALL of the FYI RFCs.  These have been moved to
 ISI now.


      http://www.isi.edu/in-notes/fyi/fyi-index.html

 Janet would welcome comments.  [e-mail]

 The WG asked Joyce to send an "official" thank you to Janet.

 USV-TV:

 USV and Trainmat WG folk wanted to start to use the MBONE both to gain
 experience and to input user oriented comments on the technology.
 Joyce reported that there has been no real progress in getting
 permission to use the MBONE (even though it had been used for a
 Rolling Stones concert).  She had been hoping to work through local
 ISI contacts.  It was suggested that perhaps a different approach via
 the IESG might be appropriate.

 Several members of the WG agreed to put in some effort on trying
 things out.


         Mailing list:           [e-mail]
         To join, mail to:       [e-mail]


 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.  April reported on
 progress in the US-WG session, rather than hold a separate NISI
 session.

 NIC Guidelines doc.  This is a replacement for FYI 12.  The current
 draft has expired.  It is still at:

 ftp://naic.nasa.gov/files/april-drafts/nic-guidelines.txt

 April would welcome comments either to herself
 ([e-mail]) or to the mailing list.

       NISI:
            mailing list:           [e-mail]
            to join, mail to:       [e-mail]


  HTML
  ====

 Chair: Eric Sink - Spyglass                                  

 It was reported that Dave Raggett (WorldWide Web Consortium (W3C)) had
 just released the "tables" document (part of HTML 3.0).  It was agreed
 that milestones were needed for the separate documents which would
 make up the HTML 3.0 spec.  - rather than try to set joint milestones
 for all the 3.0 features.  A lot of CALS compatibility had been put
 into the new tables doc.  (Some felt not enough.) Discussion would
 take place on the list.

 Stu Weibel (OCLC) proposed the use of a Meta tag to include Metadata
 in HTML documents.  This Metadata would be hierarchical resource
 description material (from one of a number of different schemes) which
 would not be displayed to the user.  The Metadata group (which had met
 in Dublin, Ohio) had already developed a meta-data element set.  They
 could proceed with their work outside of the HTML WG - but would
 obviously prefer to work with a new HTML tag for Metadata.

 There was a certain amount of opposition to the suggestion from
 certain members present.

 Stu argued that current use of Robots to provide the free text indexes
 is falling apart.  It only works with a constrained domain and
 vocabulary.  At the other end of the Internet resource continuum is
 the full cataloguing of resources (expensive).  In between is the
 "Forest Gump" approach; not very smart, not complete, but it gets it
 done and makes resources more visible than they currently are.  His
 proposal was for such an interim solution based on a new HTML tag.
 Stu said he would write an internet-draft for discussion on the
 mailing list.

 I18N: There was some discussion on support for multiple character
 sets.  Someone reported on a UNICODE implementation which used style
 sheets to determine the character set to use.  Harald Alvestrand
 (Applications Area Director) said that both languages and character
 sets were areas where the number of proposals that worked were very
 small, and the amount of noise generated was very large.  With the
 IETF being held in Europe, there were several new attendees who had
 experience of trying to support multiple character sets.  Informal
 discussions took place after the meeting.

 Mailing list:             [e-mail]
     to join, mail to:  [e-mail]
  the text: subscribe html-wg <your first name> <your last name>


 Uniform Resource Identifier WG:
 ==============================

 Chair: Larry Masinter (Xerox)

 The URI is 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
         { N,L,C, .... } = { Name, Locator, Characteristic, ... }

 There is now a URA: "Uniform Resource Agent" to add to this list -
 following the last IETF.

 Leslie Daigle from BUNYIP talked about the progress on their work on
 URAs and the URA architecture.  They have the notion of a UR Agency
 which will manage URAs and queries about URAs, requests to invoke
 URAs, the communication of the results etc.

 See http://services.bunyip.com:8000/products/silk/silk.html

 for more details.

 Karen Sollins (MIT) presented on the need for URN resolution
 standards.  Name assignment can be separated from name resolution.  It
 is important to consider longevity.

 She proposed that the IETF standardise on the client-server protocols
 and that as part of this standardisation, the form of URNs should be
 standardised.  She said that we should not standardise on a single
 server interface.

 Keith Moore talked about his LIFN scheme.  (Location Independent File
 Names).  He has a demo client (based on Mosaic) available from:

      http://mobile.netlib.org/

 This client is aware of various replication and caching services used
 for bulk file distribution.  It uses a lightweight UDP protocol, is
 low-tech, cheap and scalable.

 European Earth Observatory have developed a system based on harvester
 and Aliweb.  It uses Harvest on information from specialist sites
 which have assumed the naming and resolution responsibility - plus
 specialist AliWeb records of related sites.

 See: http://www.ceo.org/
 and
 http://elect6.jrc.it/~dirkx/alibrookek.html


 Documents:

 Ron Daniel: a draft document on a generic URN scheme.  A companion
 document describes the "DNS" scheme as a particular instance of the
 generic scheme.

 Keith Moore and Shirley Brown: draft document outlining the issues
 involved with URNs to help categorise the issues to help focus
 discussion.

      draft-ietf-uri-urn-issues-00.txt

 Ron Daniel: URC requirements draft nearly ready for last call but was
 held up while security section was reviewed.

 Ron Daniel: Draft on an SGML-based URC service (He has a demo)

 Nathaniel Borenstein's "Kidcode" draft had caused much heated debate
 on the mailing list.  He suggested URLs should be modified to give a
 minimum reader age.  The discussion on this was held over until the
 "Read the Label" BOF.  (See later notes.) It was suggested that rather
 than censoring information, another approach would be to use a SOAP
 (Seal of Approval) as part of the URC.

 John Klensin (Applications Area Director) said he was unhappy with the
 progress of URI and he felt the WG was not productive.  It had only
 produced 4 RFCs in as many years.  It was still holding discussions
 about what it was meant to do; it was behind schedule; its new charter
 was too far in the future.  He suggested splitting the WG into a
 number of WGs with more focused charters.  There was much disagreement
 with this viewpoint.  The parallel of the various "Directories"
 related WGs having been forced to combine into one WG to work on
 common problems was cited.  Keeping one URI WG to ensure
 interoperation between the various proposed schemes was felt to be
 essential.

 Ron Daniel and Leslie Daigle were asked to revise the charter.

 Chris Weider (BUNYIP) and Leslie volunteered to produce a draft on
 Uniform Resource Relationships by Dec.  95.

 There was some discussion as to whether the IETF was the appropriate
 forum for Metadata discussions.  Stu Weibel (OCLC) felt that at least
 part of it should be discussed along with URCs.  The Metadata group
 had been refused permission to hold a BOF - so had met at a different
 hotel that evening.  (I had a conflicting meeting - so can't report on
 this BOF.)

 Discussion on the finger and mailserver URL drafts etc was not
 possible because the time had been taken up by the Area Director.

 Note that Michael Mealling has a useful collection of pointers to
 related work at:

         http://www.gatech.edu/iiir/

      Mailing list:          [e-mail]
      To join, mail to:      [e-mail]
      Archive:               archives.cc.mcgill.ca
         Directory:          /pub/uri

 I don't have time to write up my IDS (Internet Directory Services WG)
 notes, but URI people might be interested in the work on WHOIS++
 directories (implementations etc):

 http://services.bunyip.com:8000/products/digger/


 BOF on Registration of MIME Types
 =================================

 Chair: John Klensin

 The process of registering MIME types is broken.  This BOF was held to
 hear the issues and to suggest a way forward.  Having been involved in
 a one-year attempt to get Powerpoint registered, I attended the BOF.

 MIME content types are used by applications other than MIME, such as
 gopher and WWW, in order to help maintain standards and
 interoperability - rather than inventing a new set of types.  Over a
 year ago it was agreed that the process of registering a new sub-type
 should not require an RFC, this process being too lengthy to enable
 rapid adoption of new content types.  A peer review mailing list
 (ietf-types) was set up and it was suggested that proposals for new
 content types should be sent to this.

 My experiences with trying to register Powerpoint are fairly typical.
 I'm a trainer and not a technical person (anymore).  Summer 94 I sent
 in the proposal for application/powerpoint to the ietf-types mailing
 list.  I followed the model for a type already registered.  It was
 deemed that the security portion was not sufficiently clear.  I
 gathered some technical advice over the next month or so (not full
 time of course!) and resubmitted the proposal in November.  I had been
 bounced off the mailing list for some reason - so did not see the
 responses.  Having discovered this fact, I then had 2Mb of old mail to
 wade through.  (The mail archive was one monolithic file - unlike our
 Mailbase hypermailed monthly archives.) Further discussion resolved
 the security problem.  I resubmitted the proposal.  The peer reviewers
 now said it should be mspowerpoint.  Fine.  New proposal submitted.
 Meanwhile I needed to distinguish between Powerpoint versions 3 and 4
 so submitted mspowerpoint3 and mspowerpoint4.  There was then some
 discussion about version parameters etc.  (Something MIME
 implementations can handle but most WWW browsers can't, as I
 understand it.) There were other discussions about definition of the
 data type.  (Microsoft apparently say this is 'secret' - so it is
 difficult to define this except by example.) It was June by this
 time....

 The basic problem for someone who is simply trying to register a type
 is that (apart from not being able to spend all day reading the
 mailing list) the goal posts move continually.  New "rules" are
 applied that have not been applied to earlier registrations.  Also the
 fact that it is by peer review on an open mailing list means that it
 is difficult for a type-proposer to know whether or not the objector
 is an expert or just someone stating an ill-informed opinion.

 Henry Rzepa had had similar problems.  He wanted to register the type
 chemical as (preferably) a top-level MIME type.  The chemical
 community want to go forward with an IETF "standard" to support their
 existing chemical modelling applications.  Their area is well bounded
 and well understood; the proposal for this type has been the subject
 of careful discussion and reasoned papers within the chemical
 community over the last 18 months.  Their proposals to the ietf-types
 list had been met with a series of objections.  They (the chemical
 community) were prepared to take responsibility for this new top level
 "type domain".  The MIME people were unhappy as a new top level type
 might break existing applications.

 Someone from the Aerospace industry stated that they had 50 new media
 types they would like to register.

 The method of adapting the existing MIME content type registration
 procedure with peer review was obviously not working.  This procedure
 is documented in RFC1590, but does not contain sufficient guidelines.
 The IANA had delegated mediator authority to John Klensin - who was
 stepping down from this role.

 It was agreed that the guidelines needed tightening up.  An updated
 version of the registration procedures would be published as an
 internet-draft for comment.  The question as to whether to allow more
 top level MIME types needs discussion.  A new WG will be set up to
 discuss the above with a view to completing the work by Dec.  95.

 John Klensin reported that he was trying to get Microsoft involved in
 the registration of their applications.


 Network Training Materials:
 =========================

 Chairs: Jill Foster - University of Newcastle
         Margaret Isaacs - University of Newcastle/Glasgow
         Mark Pior - University of Adelaide

 The following section is based on notes taken by Margaret Isaacs.

 Topics covered by the meeting included a round-up of training
 activities, the Catalogue of Network Training Materials, using the
 network to deliver training, registration of Powerpoint as a MIME
 type, and liaison with other groups.

 The meeting began with a round-up of training activities with which
 members were involved:

 -    Jill Foster reported on the recent Networking Workshop for
      Technologically Developing Countries in Honolulu in June 1995.
      The workshop was sponsored by the Internet Society, and other
      bodies, and comprised 180 students from 60 countries, divided
      into four tracks, of which 52 were included in the Network
      Navigation and Services track.  The course covered many aspects
      of networking including information services and services via
      e-mail, authoring and information provision, setting up clients
      and servers, etc.

 -    Jodi Ito from Hawaii described the Hawaiian training program
      using interactive video

 -    Ingrid Melva of Nordunet reported on a Norwegian distance
      learning course in using Internet utilising e-mail, WWW, and
      paper.  
      URL: http://www.uio.no/wwwfik/inter.html (Norwegian)

 -    Pien Voortman from Surfnet indicated the many varieties of
      user-tailored training and support provided by Surfnet, including
      the Surfnet Guide:     
      URL: http:///www.nic.surfnet.nl/surfnet/user-support

 -    David Hartland from the UK's Mailbase described how they support
      and train mailing list users and subject-based groups.  He also
      talked about a new large-scale training project called
      'Netskills' at Newcastle, funded under the UK Electronic
      Libraries Program.  Netskills will train librarians, information
      providers, and users.  At Newcastle, they are looking at WWW and
      training over the network.

 -    Nicky Ferguson of the University of Bristol described the current
      subject-based UK training and support program for social science
      researchers and users of social science research.

 -    April Marine, NASA contractor at Ames, described an interesting
      multi- agency program called Globe which seeks to involve K12
      students in taking environmental measurements, inputting them via
      Internet and eventually seeing the results of the analysis via
      the Internet.  Every teacher who is involved in the program is
      trained.  The program is exported to countries outside the US.
      In the US, the Project aims to get 2000 schools connected to the
      Internet, at no cost to the individual schools.

      http://globe.fsl.noaa.gov/

 There were some further reports of training activities from the
 Netherlands, the UK, Australia, Croatia, Sweden, and Japan.


 Catalogue of Network Training Materials:

 The fields in the template used to collect the information about
 network training materials was finalised in the light of experience.

 A new category of 'On-line Collection' was agreed to.  This category
 should encompass hyper-linked lists of on-line training materials,
 servers which offer organised collections of network training
 materials or links to them, and so on.  It would effectively broaden
 the scope of the Catalogue so that many more training materials could
 be accessed through it, without the necessity for describing each one
 in detail.

 It was agreed that the following flat range of categories should be
 used for the training materials included in the catalogue:
        User guides
        Resource guides
        Trainer Guides
        Presentation materials
        Workshop exercises
        Self-paced materials
        On-line collections

 The aim of the Catalogue of Training Materials is to provide a select
 list of quality items.  At the last IETF, volunteers from the Working
 Group had agreed to take ten each of the existing 100 items and cull
 them.  The catalogue is now in a reasonably good state, with much out
 of date and inappropriate material deleted or suggested for deletion.

 There was a question about applying some sort of quality control to
 input of new material to the Catalogue, as the Web interface
 effectively makes it possible for anyone to add any entry to the
 Catalogue.  It was suggested that the small review group be expanded
 slightly and volunteers to vet entries were called for.

 Once the amendments suggested have been made and the new categories
 retrofitted, the catalogue will be submitted for final review as an
 internet-draft.

 Using the network to deliver training: The Working Group has been
 interested in the use of the MBONE to deliver training.  Joyce
 Reynolds has tried to get support from technical people for this but
 was discouraged.  She will try a route via the IESG to the relevant
 MBONE Working Groups.

 Ingrid Melve mentioned a report on using the network to deliver
 training in Norway and volunteered to give a summary in English of the
 report to the network-training-tf list.

 Registration of Powerpoint as MIME type:
 Jill had been trying since last July to get Powerpoint registered as a
 MIME type.  There were initial concerns about the section on "security
 implications".  These had been addressed, but there were subsequent
 quibbles about the exact name of the proposed type, version numbers
 etc.  There was to be a BOF at the IETF about the problems with the
 registration procedures.  (See earlier notes.)


 Mailing list:             [e-mail]
     to join, mail to:  [e-mail]
  the text: join network-training-tf <your full name>

 Note this is a *working* list. Minutes etc are sent to the us-wg and
 wg-isus mailing lists.


 "Read the Label" BOF
 ====================

 Chair: Vint Cerf

 Objective: to determine whether the IETF wishes to provide mechanisms
 for access controls.

 Vint put forward some possible principles for discussion:

 o    if markings are used - they should be used voluntarily and
      accurately

 o    all forms of content should be dealt with (Web pages, files,
      telnet servers etc.)

 o    Browsers should be able to accept guidance from multiple sources

 o    no one body should be responsible for categorising or marking all
      content

 o    if marking is used - every effort should be made to make
      descriptive markings objective and not judgmental

 o    multiple sources of marking or categorisation for the same
      materials should be possible

 Tim Berners Lee than presented different scenarios for content rating:
 a 2 party and a 3 party system.

 A discussion ensued.  Here are some of the points made:

 Marking material as unsuitable for a section of the community may in
 fact make it easier for them to find it.

 Marking information may enable censorship at a national level.

 Publishers are worried.  The film rating system is not yet applied to
 books.  Why apply it to online information?  Can we learn something
 from the book industry?

 We are in danger of shutting off parts of the Internet to protect
 certain sub-sections.  What we don't want is to stop everyone getting
 to info that is deemed to be unsuitable for, say, 6 year-olds.

 Even within the USA there is a range of cultures.  Any marking system
 needs to be global in scope.  But that may be impossible.  Almost
 anything will offend someone.  It was pointed out that images from
 NASA might offend members of the flat-earth society!

 Can't rely on all information being marked.

 Vint: most people seem to want the IETF to do something about it.

 Need to provide users and providers with ways to be responsible.  A
 voluntary self-rating system might be a first step.

 If the Internet doesn't come up with its own solutions, others will be
 imposed on it.

 The Internet should not be restricted to a box, but should be an open
 sea with restricted islands.

 Stu Weibel (OCLC) said that this was part of a broader issue that
 should be addressed within the IETF - ie that of resource description.
 (cf the Metadata work and the URC work discussed in the URI WG).

 Vint summarised by saying that a WG was needed to look at these
 issues.


 Open IAB Meeting
 ================

 A show of hands showed that a much larger number of attendees were not
 using unprotected passwords to log-in back home over the Internet.
 The IPSEC specification is now out; key management is still being
 worked on.  Manual keying works bi-laterally.

 The challenge is to have:

 o    a measurable change by the Dallas meeting (Dec.)

 o    IPSEC with manual keying implemented for:
      -    the terminals in the terminal room
      -    laptops
      -    in terminal room fire-walls (some discussion re this one)
      -    available in commercial firewalls.


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

 A good IETF.  This was only the second time it has been held outside
 of North America.  Some WG session were bogged down by "procedures",
 some benefited from the presence of the European members who don't
 normally manage to attend the IETF, whilst a few were affected by the
 absence of key members unable to attend a non-US meeting.  The next
 IETF is scheduled for December 4-8 1995 in Dallas, Texas.

 Finally, a reminder that these notes are my view of the IETF and that
 I have had even less time than usual to "polish" them.  They may not
 be an accurate view, and certainly do not cover the wide range of
 topics discussed at the workshop, nor even all the sessions I
 attended.  This IETF was just before I was away on holiday - hence the
 delay.  Hopefully the information is not too out of date and it has
 given you enough of a flavour of the meeting.  If you are interested
 in subsequent developments or in more details on the above, please
 look at their full minutes, join the relevant mailing lists and look
 at their message archives.



 Jill Foster ([e-mail]) 

 21.08.95