Earlier this year, the University of Newcastle upon Tyne hosted Networkshop 18. The major themes of the formal presentations were:
Of these topics, three were chosen for discussion in greater depth by three parallel groups. The major points of discussion were summarised by a rapporteur for each group and presented to the full Networkshop on the last day. In order to bring the conclusion of the groups to as wide an audience as possible, the rapporteurs were asked to provide summaries for publication in the Networkshop proceedings. As in previous years, they are also being published in Network News. My thanks to all those who contributed.
Sue Weston, Joint Network Team
Chairman - Willie Black (Joint Network Team) Rapporteur - Tim Clark (University of Warwick)
The discussion started by considering the question, What should be done about RPC in the academic community. One simple option was to do nothing at all; suppliers were very interested in RPC and so we could perhaps leave them to do all the work. After further consideration it seemed reasonable to keep out of the area of developing toolkits and interfaces (the suppliers would certainly be doing that) and concentrate on the uses of RPC. There was, however, the opinion that perhaps we should not completely opt out of all RPC development work; we should at least keep track of what is going on and possibly use interim RPC's and track developments to gain experience.
After clarification of some of the technicalities of RPC, the ISO timescales were raised. It was believed that RPC had a very high chance of being adopted as a New Work Item in ISO shortly. One might then expect to see the work progress to DP stage late 1990 or early 1991, then to DIS stage middle to late 1991 with the possibility of an IS at the end of 1991 or start of 1992.
It was thought sensible that the academic community should get involved in the actual use of RPC. This would hopefully promote awareness of RPC and encourage software developers to make use of ii.
In discussing what use could be made of RPC, it became obvious that there are a number of areas where RPC might be used but should not, because there are other standards more appropriate. The following were thought unsuitable, with the named standards a better alternative:
It is obvious that RPC is a general purpose tool and that there is a potential for misuse. However, it was pointed out that just because a job can be done with different standards, it does not mean that it ought to be.
The major use foreseen for RFC in the academic community was in the export of processing to more powerful machines. Three subjects were identified as possibilities for work.
Of the delegates present there was only one who might be interested in doing work in the above areas; that was for the second, the split processing. Though it was pointed out that those present were unlikely to be the appropriate people to volunteer an interest in the other two areas.
Chairman - Les Clyne (Joint Network Team), Rapporteur - Denis Russell (University of Newcastle upon Tyne)
Les Clyne chaired the meeting. He opened the Security part of the discussion by emphasising that one main aim of the session was to obtain community input for the security project primarily in terms of whether the appropriate threats were being faced.
There was a wide-ranging discussion in which it emerged that the primary worry for most people was the stealing of passwords from Ethernet, and the onward problems that this could cause. In addition dial-in access was widely used by hackers presumably because of the difficulty this introduced for back-tracing. The primary protocol involved seemed to be X.29.
With mail, most worries concerned the authenticity of messages. There seemed to be little current threat, but a growing unease about the threat in the future.
The worries over FTP were comparatively small. The main threat here seems to be concerned with the vulnerability of interactive passwords used in associated X.29 access.
Several people expressed interest in the facilities offered by the Kerberos authentication system.
Several people said that confidentiality was important for some of their users for some of the time, but the extent of the need was not generally agreed.
It was suggested that a review of security products on the market would be useful - not least in reassuring administrative types. However, the scope of such a review was not obvious.
There seemed to be a general feeling that we needed to deploy counter-hacking measures urgently, and feed the results of that experience into the standards-making process, and yet at the same time there was clear reluctance to do much retrofitting on coloured-book implementations.
There was some discussion of the legal aspects of the matter, but about the only useful thing that emerged was that data on a network is in transit, and is thus apparently not covered by the act.
In the summary of the session the helpful (?) remark was made that we should scrap Ethernet and Unix!
In the management part of the session an attempt was made to evaluate where we should go next. It seems that no management systems are sufficient as they are.
Surprise was expressed that the JNT was not more active in fostering protocols in this area, Les responded that it was a matter of person-power and invited concerned users to volunteer effort.
There is a real problem with proprietary management protocols running in parallel with OSI protocols. There was some discussion of the problems of working to these protocols, and the consequences of protocol changes and disclosure. There was much discussion of a possible way forward that concerned licensing rather than disclosure of protocols. The participants seemed to think that the scheme had possibilities and so we must hope that the discussion is followed up.
Chair - John Dyer (Joint Network Team) Rapporteur - Mike Whitehead (University of Dundee)
This discussion of X-windows followed papers given earlier in the day by John Dyer and Brian Gilmore.
The session began with an attempt to assess the current use of X. No sites were making extensive use of X at present but many expected the situation to change quite soon, Cranfield, for example, reported that some of their major applications would soon have X as the only interface. This was just one of the reasons why X would become vital for service provision within the near future.
The networking community itself is encouraging the use of X by writing X User Agents for Mail and Directory Services. Examples of these were presented in demonstrations and sessions at this Networkshop.
It was agreed that the overwhelming use of X will be over LANs. Remote traffic will amount to less than 10% of the total for most sites. There is no serious cause for concern about the load this will place on the network because:
As the use of X over WANS grows the importance of having a suitable OSI standard supporting network stack increases. The work being done by the JNT and others was felt to be on target. Brian Gilmore thought that running X over ACSE would be unnecessarily slow and that using Transport Service Class 0 should result in response times equivalent to TCP/IP. Everyone was happy with the current JNT strategy.
There was a discussion about the relative merits of equipping student teaching areas with X terminals or with PCs running X software. The considerations included:
Several people made the comment that whatever the system you were using you should add in a few more MegaBytes of memory if you intended to run X.
The meeting was pleased that the JNT were trying to standardise the use of X over OSI stacks and hoped - as X was designed to run over any reliable byte stream - that this would be agreed by the standards bodies.
The JNT will try to make available the Edinburgh code that was used in the demonstrations at Networkshop. It supports a daemon running on a Sun which will run X over JANET using TCO as the carrier.
The proceedings of Networkshop 18 have been published. A copy has been sent to everyone who attended.
During the past few months there has been a lot of discussion about a Networking Association. Some Network News readers are involved in the discussions, others may be wondering what it is all about. This article attempts to provide a brief summary of what is happening and why.
Over the past decade the UK academic community's networking programme has developed to serve a large and increasingly diverse user community. The existing organisational framework is largely improvised having been created over ten years ago to support a relatively small collaborative networking venture between the Computer Board, SERC and the other research councils. The programme is managed by the Joint Network Team located at the SERC's Rutherford Appleton Laboratory. This group forms part of, and reports directly to, both the SERC and the Computer Board. In addition the Group reports to the Network Advisory Committee which advises the Computer Board and has no executive powers.
After much discussion, including widespread consultation in the academic community, the Computer Board has agreed that the present structure is inappropriate for the size and complexity of the current networking programme; it increasingly acts as an inhibitor to progress, which puts the whole programme at risk. Several other countries have gained a significant advantage by creating better adapted organisational structures, and the Board has decided that the time is ripe for a re-structuring for me benefit of the UK academic community's current and future needs.
The Computer Board has therefore agreed in principle to the formation of an autonomous Networking Association, and has set up a Steering Group under the chairmanship of Dr David Hartley to prepare detailed proposals. The terms of reference and membership of the Steering Group are given on the opposite page. The Board is concerned that the proposals should be developed with further consultation with appropriate representatives of academic community interests and that they have the full support of the academic community. A Consultative Group has been set up to provide this liaison, the membership of which consists of representatives from different interests in the academic community. A list of the organisations involved in the Consultative Group are shown opposite. The first meeting of the Consultative Group was held on 29 June.
In general terms the Networking Association will take responsibility for the UK academic community's networking programme. The intention to make it autonomous refers to its relationship with the funding bodies and not the institutions of the community itself. Indeed, it is desirable that the Association is owned and controlled ultimately by those in academic institutions which have a direct interest in its activities and services. The networking programme is now large enough to require an organisation and management structure which is sufficiently independent of the funding authorities and which can act in the best interests of the
community's networking requirements.
The major benefits resulting from the proposed changes are expected to be:
It is intended that the Association will be responsible for managing all the activities, projects and services which comprise the current networking programme, which will continue to be funded via grants or contracts from the appropriate funding bodies. The Association will employ staff to undertake this work and would replace the existing Joint Network Team and Network Executive. The Association will be responsible for promoting the networking programme and encouraging collaborative participation with appropriate external organisations.
The Steering Group intends to use the Consultative Group as a sounding board; draft proposals will be submitted for comment and discussion, and although the frequency of meetings and the time table are difficult to predict at this early stage, it is likely that more than one round of consultation will need to take place. The timescales are not yet well defined but the aim is to ensure that any course of action which the Computer Board and Research Councils decide to take can be put into effect by 1 April 1991.
Bob Cooper, Director of Networking
At the end of 1988 the JNT created the FDDI Advisory Group to provide a forum for technical advice and community liaison. The membership now covers most universities together with several Research Council sites and Poly technics. One function of the FDDI Advisory Group to date has been to educate the Group on FDDI, A number of suppliers have attended afternoon sessions of the group to give presentations on FDDI products, standards, etc.
In April 1989 the Computer Board approved a new local area networking initiative with the aim of supporting the installation of an initial high performance backbone LAN in each university over a four year period (starting in Financial year 1990-91). The Computer Board's contribution to this initiative is sufficient to provide starter kits or pump priming. It is intended that every university will receive a share of this funding.
The JNT in conjunction with the FDDI Advisory Group has identified the following as objectives of the High speed (or Fibre) LAN Initiative:
Having surveyed the current standards activity in this area, it is clear that the only contender at this point in lime is FDDI (Fibre Distributed Data Interface). Although the FDDI standards are mature enough to begin to make plans and to purchase products with a commitment to the published standards, the JNT felt that it would not be prudent to commit to a full programme of spending in the financial year 1990-91. The JNT then proposed a strategy to the Computer Board of supporting a limited number of FDDI pilot projects in financial year 1990-91 in order to assist in the preparation of the installation of a high speed LAN backbone on each campus which will commence in financial year 1991-92. This was accepted and Universities were invited to bid to be an FDDI pilot site.
The FDDI Advisory Group has defined a set of objectives for the pilot phase. The principal objective of the pilot studies is to make recommendations on how to deploy the available funding in order to meet the objectives 1-5 above.
The following sub-objectives were identified in order to meet this main objective:
A number of Universities submitted proposals for pilot projects which addressed one or more of the objectives and the proposals received were evaluated by a small group of experts drawn from members of the FDDI Advisory Group that had not submitted proposals. Five sites were selected as potential FDDI pilot sites: Birmingham, Edinburgh, Glasgow, Manchester and Queen Mary and Westfield College. The Computer Board approved funding for these five pilot sites.
The progress of the pilot sites will be monitored by the FDDI Advisory Group. If anyone from an Academic site would like to join the FDDI Advisory Group they should send an electronic mail mes$ageioFDDI-REQUEST@UKACJNT.
Sue Weston, Joint Network Team
For several years a JNT LAN Advisory Group has been in existence to advise the JNT on issues associated with layer 1 -3 protocols and services operating over ISO 8802 LANs in general and CSMA/CD LANs in particular. Many of the lower layer issues that arise on LANs are related to similar concerns on WANs. In addition the distinction between use of various technologies for LAN (on-site) and WAN (inter-site) communications is becoming blurred. For example, X.25 packet switching is used for both LAN and WAN communications, and the introduction of remote CSMA/CD bridges and FDDI backbones can extend the range of LANs greatly. There is also the need to take account of the range of protocols and interworking scenarios that are now commonly found on sites and between sites.
In response to these changes the LAN Advisory group has been reconstituted as the JNT Lower Layers Advisory group. Its detailed work plan is still being developed but in general its remit is:
New areas will include routeing protocols, non-OSI LAN protocols, CO/CL interworking, Transport layer protocols etc, subject to there being the relevant expertise and effort available within the group. One of the first activities of the group has been the consideration of Level 3 routeing versus MACbridging between LANs and a report has been produced outlining the factors to be taken into account by sites when connecting many LANs together.
Anyone in the academic community with an active technical interest or involvement in one or more of the areas described above is welcome to join the group. If you wish to join or want further details, please contact:
Les Clyne, Joint Network Team
The JNT is funding a study with the above title. The study is in two phases, with an initial report due fairly soon and a final report in the Autumn. The aim is to suggest ways in which the security of network access to services can be improved, and the final report will suggest practical ways of achieving this in a timely fashion.
In order that any suggested solutions are both relevant to the perceived needs of users, and capable of practical and timely implementation, it is essential that there is the fullest discussion within the community both of the project's assessment of the problem and of the suggested course(s) of action, A talk on the project was given at Networkshop, and there was valuable input both in the discussion session and in the corridors. To facilitate further discussion a mailing list is being set up at Newcastle. The list is expected to run until about the end of the year.
The mailing list will be run using the Mailbase facilities of the NISP project at Newcastle. This provides automatic list management facilities, including allowing users automatically to join and leave the list themselves. All attendees at the last Networkshop in Newcastle should have received details of how to use mailbase on a blue card together with their Networkshop material.
The list is called PORTCULLIS@UKACNEWCASTLE.
Denis Russell, Newcastle upon Tyne
In recent years RARE (Reseaux Associes pour la Recherche Europeenne) had organised an annual European Networkshop and EARN (European Academic and Research Network) had organised an annual conference. In 1990 RARE and EARN collaborated to organise a joint event which was held in Killarney, Ireland from the 15th to 17th May. The conference took place in an atmosphere of real cooperation despite the networkers of RARE and EARN being divided by their different views on networking.
RARE is OSI oriented and the sponsor of COSINE. EARN is derived from the BITNET model, and was initially funded by IBM. It is now independent and its network covers many smaller countries and there imminent connections into Eastern Europe.
The joint conference attracted an impressive attendance of over 340 delegates from all over Europe as well as other parts of the world. Representatives of the following countries attended the conference.
By merging the two events, the Joint Conference programme emphasis was on:
There were also a number of technical sessions, such as high performance networks, organising the access, standards, OSI transition, TCP/IP, Security and lower layer protocols. There were a number of panel discussions including a highly entertaining one on User Support.
Readers may find the enclosed cartoon by Francois Fluckiger from CERN entertaining. This was only one of a series of highly entertaining cartoons drawn by Francois and available to delegates.
The conference tried to cram too much into too short a time but the delegates seemed to take advantage of what spare time there was to talk to their counterparts in other parts of the world and the overall view of those attending was that it was a successful conference. All those that were unfortunate to miss this conference can read all about in the proceedings of the conference which will be published in Computer Networking and ISDN Systems. Don't forget to book early for next year's conference.
Shirley Wood, Joint Network Team
There is a report above on the Killarney Joint European Networking Conference organised by RARE and EARN in May. There have been many favourable comments about this Conference and we are receiving a growing number of enquiries about the possibility of a similar event next year.
It is planned to organise a Joint Conference in 1991. Planning has only just started so it is not possible to give any details but the dates and location are known. If you would like to attend the 1991 Conference please make a note of the following in your diary:
2nd Joint Networking Conference 13-17 May 1991, Blois, France
More information will be published in Network News and elsewhere later this year.
The proceedings of the Killarney Conference will be published in Computer Networks and ISDN Systems in the Autumn.
Bob Cooper, Director of Networking
JANET Starter Cards May 1990
The JANET Starter Card has been produced as part of the JANET Starter Pack project. This is a Joint Network Team project which is being carried out by Caroline Leary from the University of Sussex Computing Service.
As mentioned in Network News 30, the aim of the project is to produce introductory information about JANET that should be useful to new institutions connecting to JANET, and to new staff at institutions that are already connected.
The JANET Starter Card is a quick reference card to accompany the JANET Starter Pack, which will contain more detailed information. They are both being written for people who are familiar with computers but are not networking specialists, such as computing service advisory staff.
The information on the card is of a general nature and is intended to supplement local documentation. It can be copied as long as the Joint Network Team copyright is acknowledged.
Batches of cards are being distributed by the Joint Network Team to the secretaries of all the regional and special interest user groups, as well as to the Computing Service Director, Computing Service Information Officer and the Library JANET User Group contact at all institutions connected to JANET.
The card will be updated later in the year and distributed again. The Starter Pack is being prepared and will be distributed in the same way.
If you require any more cards then please contact the Joint Network Team.
Caroline Leary, University of Sussex
PADs (Packet Assembler/Disassemblers) are widely used to network terminals and PCs. Currently there are at least 3000 PADs in the academic community used to provide over 50,000 network access points for terminals. This represents a capital investment of over £5M, The PAD market is growing rapidly and new products and suppliers are emerging to meet the demand. It is expected that there will be a substantial PAD requirement in the community for the foreseeable future.
The JNT undertakes product evaluations to identify suitable networking products for use in the community and over the past year has been taking steps to improve the evaluation procedures, particularly in product areas where there is a large demand and growing supplier interest. In the past, the evaluation procedure has led to problems of consistency of approach and poor response when new suppliers bid to enter the market or existing suppliers introduce new products. In order to overcome these problems and to cater for requests from sites for information on PADs, a PAD Evaluation Centre has been established.
The Evaluation Centre is intended to be in a position to react quickly when new products become available. On arrangement with a manufacturer, PADs can be tested against an established test procedure. The results of the testing are fed back to the supplier and to the JNT, with the expectation that a recommendation can be made on the product's suitability for purchase by the Academic Community. A technical questionnaire provides an initial screening mechanism.
Suppliers wishing to avail themselves of this service are invited to contact Dr G Howat, Edinburgh University Computing Service,
The Evaluation Centre caters for both X.25 packet switching and Pink Book ether PADs.
Dick Gillman, Joint Network Team
The CHEST Directory will shortly be mounted on the NISSPAC service, providing the user with a much more sophisticated interface to information on CHEST deals than is currently available on the NISSBB. We are also planning to provide a version of the Directory on the NPDSA (National Public Domain Software Archive) Service at Lancaster which can be easily downloaded.
The current 1989/90 printed edition of the CHEST Directory will be superseded this Autumn by an updated volume, and the 1989/90 version is now available to members of the Higher Education Community at the cost of postage and packing only. A bargain not to be missed at £2.00 ......
contact Deborah Sawyer in the CHEST Office at Bath for copies.
NISSPAC evolved from the original NISS Software and Dataset Catalogue, and is a more advanced system providing a common, easy to use, interface to a range of related databases. We also plan to provide an additional facility which allows the user to interrogate all the databases simultaneously. NISSPAC currently includes 3 datasets :-
This remains at the heart of the service and continues to grow rapidly. In addition to standard bibliographic descriptions of software and datasets, each record provides details of technical requirements and a brief descriptive text. Details are also included about sites holding the material and, where appropriate, information on access and utilisation.
The Directory is a collection of records supplied by the VAX/VMS University and Polytechnics Application Committee, and provides details of the VAX applications software in use in the UK academic community. As with the NISS Catalogue, details of both the software item and the sites holding the item are provided.
The Database is supplied by IBM, and gives details of IBM supported research projects undertaken by the UK academic community in computer related areas. The dataset provides a description of each project, lists hardware and software used and the anticipated products or results. Contact details in both the academic community and IBM are also provided for each project.
If you have a connection to JANET and you wish to use NISSPAC, the command
CALLNISS.PAC (numeric address 000050300009)
will access the host machine. A single dot prompt indicates a successful connection, and typing
L NISSCAT
will enter the opening NISSPAC menu.
As a consequence of the continuing popularity and high levels of usage of the NISSBB, an upgrade to its underlying hardware has become necessary. Thanks to further support from DEC, we have been able to purchase a MicroVAX 3800 as a replacement for the existing MicroVAX 2. The service will be switched to the new machine during the summer months.
As part of a drive to make access to our online services as easy as possible, we plan to launch the NISS Gateway by midsummer.
Once the Gateway is fully implemented, users will need enter only a single command,
CALL NISS
at the PAD in order to gain access to all NISS's online services. The Gateway will provide a top-level menu from which a selection (eg NISSBB, or NISSPAC) can be made, with ensuing automatic connection to the preferred service.
The CALL NISS (numeric address 000062200000) command can continue to be used to access the NISSBB until the Gateway is implemented. Following that, the NISSBB will then have a new NRS name, UKAC.NISSMB, which can be used if direct access to the Bulletin Board is required.
Annette Lafford, NISS
Acorn have developed a Coloured Book package for the R140 which will be distributed to end-users by the Edinburgh University Computing Service (EUCS). The R140 runs the Berkeley variant of the Unix operating system into which the Edinburgh/York/Bristol UnixCommunicationsPackage has been integrated.
The product provides Pink Book (the CONS over CSMA/CD technology) and the associated ColouredBook software, providing:-
The software will be distributed to end-users over JANET. A single machine licence costs £200, whilst substantial discounts are available for bulk purchase. Interested parties should contact Joyce Anderson of EUCS on 031-667-1011 E\l,6n2oTQmai\:Joyce_Anderson@UKAC.EDINBURGH
Dave Mercer, University of Edinburgh
Earlier in the year the JNT became aware of the NETMON Network Monitor which enables a PC to be used as a network monitor. The JNT subsequently asked Dr Alan Buttle, Exeter to review this package. Copies of his report are available from the JNT. The JNT now recommends this package for use within the community.
A number of other Network Monitors are also recommended for use within the community. A report on Ethernet LAN Monitors was produced by Scott Currie in 1986. This report reviews seven Ethernet monitors. This report is also available from the JNT. If you would like either of these reports please send a mail message to
Sue Weston, Joint Network Team
Networkshop 19 will be held between the 25th and 28th March 1991 at the University of Aberdeen.
The format will be similar to Networkshop 18, however the programme has yet to be decided. If there is a networking issue that you would like to see discussed which would be of general interest to those working in the Academic Community, please submit it to the organisers for consideration.
The organisers are:
Wallace Anderson at Aberdeen, Shirley Wood for the JNT
