Produced by the Computer Board And Research Councils' Joint Network Team and Network Executive.
Sue Weston joined the JNT/NE in June. Sue is well known in UK and European networking circles for her work on VAX/VMS networking and the GIFT project and we are very pleased to welcome her into the Team. This was the only success in our most recent recruiting campaign and three vacancies remain unfilled.
The JNT has participated in the sandwich student scheme for many years and we have had the pleasure and benefit of working with a number of undergraduate students who are interested in data communications. Our student for the past year, Mike Wooldridge, has returned to college to complete his degree course. Mike worked on a project to develop converters between ODA and popular word processor mark up schemes and as a result acquired considerable expertise in ODA and ASN.l. We are pleased to welcome three new students, Collette Truman, Brian Toner and Andrew Blowers, and hope they will have a stimulating and productive year working with us.
Bob Cooper
The Computer Board recently agreed to fund a replacement for the NRS computer at UMRCC. The service has recently been running on a Prime 2655 loaned by Prime while an evaluation was carried out to determine the size of machine required to run the central service. This was recently completed and a Prime 4050 with 1 Gigabyte of disc storage has been installed at UMRCC.
The new machine which is about twice the power of the 2655 has been sized to cope with the increase in size of the NRS database expected over the next few years and will be able to distribute incremental updates to the expected 50 campus NRS servers during the early hours of each morning. This is required as the servers will themselves be expected to put out new versions of the database before work starts in the morning.
Now the Prime 4050 is in service the NRS distribution centres will be phased out. One of them UK.AC.NRS.DC-1 has already been withdrawn as it was believed to be interfering with the operation of the Janet News service.
John Carey
The new initiative to improve interworking between JANET and the US Internet has been reported in recent editions of Network News. The new gateway is now in service at ULCC and is currently connected to ARPAnet via the MoD link. There have been serious delays in installing the new link to NSFnet and this is now expected to be available early in 1989. The agreement on the interim use of the MoD link has therefore been extended to allow a continuation of service and I am grateful to the MoD for their continued cooperation and support in this area.
The current restrictions which prohibit terminal and file transfer traffic over the transatlantic link will remain until the new NSFnet link is available. However, the widespread support for this project from funding bodies in the UK academic community allows an improved mail authorisation scheme to be introduced for that community. It is planned to authorise all Grey Book mail systems that are NRS registered in the UK.AC domain to send and receive mail through the gateway and this will be implemented as soon as possible. This registration will happen automatically and will not require intervention by account managers as at present. Authorised mail systems which are not in the UK.AC domain will continue to be registered using the present scheme.
The new authorisation scheme implies that all registered mailers are trusted not to permit unauthorised access to the gateway. UK.AC mailers should not allow access to the gateway for users who are not part of the UK academic community and may lose the right to use the gateway if unauthorised access occurs. This scheme is in line with the general move to trusted host authorisation for all gateways. Following the introduction of the new scheme, the JANET/NSFnet gateway will become the recommended and supported service for access between the UK academic community and the US Internet community.
Bob Cooper
Following the evaluation conducted on behalf of the JNT by Leeds University, the High Level Hardware CR82 PAD is now recommended by the JNT for use in the academic community.
Sue Weston
At the JANET Networkshop in March 1988 there was a discussion session on the need for Networked Information Services. The report on this session appeared in the last edition of Network News. Following discussion at the JANET National User Group the JNT asked the University of Newcastle upon Tyne to submit a proposal for a project in this area, and the proposal submitted was accepted at the Computer Board meeting in September.
It is not the aim of the project to set up and run Information Services over JANET, but rather to develop the mechanisms to allow special interest groups to set up their own information servers, electronic journals, bulletin boards and of course electronic-mail-list servers as part of a Networked Information Services Infrastructure. These tools and services will be aimed at serving the basic needs of all JANET users, whether they be casual users of JANET, members of specialised research interest groups (for information sharing), local user support staff, or the JANET Network Executive and Joint Network Team (for disseminating information about the networking programme). The aim is to make the user interface to these facilities as natural as possible to encourage use by academics of all disciplines, and to enable efficient group communication and information dissemination, thereby improving the quality of national collaborative research.
The first stage of the programme will be a short initial study. This will consist of a survey of the tools and facilities currently available both in the UK and on other networks, and an assessment of the requirements and needs of the various users, groups and organisations that use JANET. The first stage will provide input for the second stage of the project.
During the second stage of the programme a functional specification for JANET Networked Information Services based on the results of the survey will be produced. A pilot system, based on the functional specification, will be developed and used to mount a prototype distributed National Information Service at a small number of sites within a fairly short time-scale. This will act as a demonstrator project to prove and refine the specification and model, and to establish routines for the collection, validation and distribution of the information. It will also allow users to gain experience of such a system and to provide useful feedback.
Consultation with the academic community will, of course, form an important part of both stages, and progress reports will be issued in Network News from time to time.
Jill Foster University of Newcastle upon Tyne
The English Law Commission has published a report on Computer misuse with a request for comments by February 1989. The JNT proposes to prepare a submission on this report from the UK Academic Community and would welcome comments and discussion. Discussion will be by electronic mail via the mailing list Lawcomm@UK.AC.Rutherford. If you wish to be included on this list please contact Lawcomm-request@ UK.A C.Rutherford.
The report is Law Commission Working Paper No 110 on Computer Misuse and copies can be obtained from HMSO at £4.50 net.
Please note that the JNT is not able to supply copies of the report.
Bob Cooper
Networkshop 17 will be held between 11 and 13 April 1989 at the University of Warwick. Application forms for registration will be sent out towards the end of the year. These should be completed and returned to Warwick before 5 February 1989. If you think you may not be on our distribution list for application forms or have not received an application form by mid December, please send a request to the electronic mail box : networkshop @ UK.AC.WARWICK
The program for the event has not yet been finalised. If you have an issue that you would like to see discussed, or an area on which you would like tutorial information, please send your ideas to the Networkshop organisers at the above mail address.
John Dyer
The JNT is setting up an advisory group and a mailing list (PC-COMMS@UK.AC.RUTHERFORD) for users of MS-DOS, PC-DOS and OS/2 machines. The group and the mailing list will cover open networking matters of interest related to these operating systems, and other closely related matters for which there is no suitable forum. The first aim of the group will be to formally define its terms of reference.
Sue Weston
JANET, in common with most other X.25 networks, currently supports the version of the X.25 standard agreed by the CCITT in 1980. A major revision of the standard took place in 1984 and a number of new facilities were included which are summarised later in this article. The most important changes are the new DTE facilities introduced to allow a network call to be made across a number of concatenated subnetworks, each with its own addressing scheme. The Yellow Book interim standard provides this capability within the Coloured Book architecture and the enhancements in the X.25 standard will allow use of the Yellow Book to be phased out.
The International Standards Organisation (ISO) has developed an addressing addendum for X.25 level 3 which introduces the concept of a Network Service Access Point (NSAP). An NSAP is a globally unique number for a given networked service which is, in principle, independent of the route taken to reach the service. The carriage of NSAPs together with the specification of quality of service parameters has led to the introduction of what are called CCITT-specified DTE facilities to support the OSI Network service. These facilities form an important enabling part of the UK Academic Community's transition to OSI Standards and must be implemented early to support other parts of the transition.
X.25(80) is a subset of X.25(84) and it is possible to support both styles of working over the same network with each DTE configured for either X.25(84), in which case it can support both styles of working, or for X.25(80) only. Clearly, to obtain full connectivity with a mix of DTEs on the same network some form of interworking unit is required.
The CCITT works in 4-year cycles, so a new revision of X.25 is being agreed this year. Indications are, however, that the changes this time are not so dramatic, although some additional new features have been added. It is too early yet to even speculate how long it will be before implementations appear, but if the 1980 to 1984 change is anything to go by it is likely to be several years yet before X.25(88) is in widespread use.
The 1984 version of the X.25 specification contains a small number of major changes and a much larger number of more trivial changes, most of which seek to clarify rather than modify the sense of the text. Within the scope of this article it is only possible to deal with the major differences, which for convenience are split into the following four areas:
In X.25(80) the link level frame sequencing has been modulo 8. To cater for fast links, in particular satellite links with comparatively long transit delays, a modulo 128 scheme has been introduced. This has an impact only on those who wish to implement the option.
Another new feature is the Multi-Link Procedure which allows the sharing of several links between the network and an attached DTE. Packets for a given virtual circuit can be routed down any of the available links. This provides increased resilience against link failure and can also be used to increase link bandwidth. This is an optional facility and may not be available in all implementations.
Finally, there have been a number of changes in the details of the actual procedures used across the interface between the network and a host. It is somewhat difficult to predict exactly what impact this will have, if a host running a 1980 implementation is connected to a network which supports 1984 facilities, or vice versa. Likely problem areas would be with link set-up/disconnect and error recovery during data transfer.
X.32, a new standard, allows the connection of a packet mode DTE via the telephone network or a circuit switched network. In this case, there has to be some form of dialogue to identify what is being connected and what facilities are required. To assist in these activities, two new packet types have been introduced into the X.25 specification:
A new facility in call request packets, the network user identifier, can be used as one way of identifying who is making the call.
Five new facilities for call negotiation have been introduced to cater for requirements associated with the OSI Network Service. These are
The address extensions are aimed primarily at the carrying of NSAP addresses across concatenated networks. These addresses are independent of the subnetwork addressing scheme, and will be used at the boundary between subnetworks to derive the SubNetwork Point of Attachment (SNPA) for the next stage of a call. For an X.25 network, the SNPAs are the calling and called DTE addresses carried at the start of a call request packet immediately after the packet type identifier. The CCITT originally assumed that the address extension field needed to cater for 32 decimal digits. ISO subsequently settled on 40 and this is expected to become the standard in the 1988 version of X.25. In the meantime we expect implementors to go beyond the letter of the 1984 standard and support the full 40 digits in their implementations. Work has still to be done on specifying how the conversion between NSAP and SNPA addresses will be achieved.
Minimum Throughput Class and End-to-End Transit Delay allow the user to specify the quality of service required over the whole network in terms of throughput (bits/sec) and transit delay (millisecs). Because concatenated subnetworks may be involved, the numbers represent the result of summing the effects of individual subnetworks. It should be noted that it is not the subnetworks themselves, but the relays at the boundaries between them, which alter the contents of these parameter blocks. It should also be noted that analogous facilities exist in the subnetwork-specific facilities, called throughput class negotiation and transit delay selection and indication, which apply to each hop of a connection rather than to the total connection.
Expedited Data Negotiation is intended to allow a user to specify a requirement to send expedited data, eg. to force through an interrupt regardless of the state of the connection. A capability to send a single byte interrupt already exists in X.25(80); the new facility allows up to 32 bytes of expedited data to be sent.
Datagrams, single packets sent through the network without acknowledgement or call establishment, have been deleted from the standard. This feature of the 1980 standard was not implemented.
A number of other new facilities have been introduced. These include additional types of Closed User Group (a mechanism for restricting who can call whom); facilities associated with charging; hunt groups (with a similar meaning to telephone hunt groups); and facilities for redirecting calls and telling the new called system and the original caller what has happened.
Other changes include extending the facilities field in call request packets to 109 bytes (to cater for the extra space likely to be needed to carry the OSI facilities), larger packet sizes for data (2048 and 4096 bytes), and some new clearing codes, including one which indicates 'ship absent' to be used in association with the provision of maritime X.25 services.
JANET uses Packet Switching Exchanges (PSEs) supplied by GPT. GPT has been contracted to provide an implementation of X.25(84) for these PSEs supporting the new features required for the OSI Network Service. It will be possible to mix X.25(80) and X.25(84) traffic and the PSEs will be able to identify which version of the standard each attached DTE is capable of supporting. Since the Transition is expected to take several years, it is clear that it could also be necessary for hosts to be able to select, on a per call basis, which version they use-Accordingly, a host connected via a link which supports X.25(84) can set up calls using either version, but those connected via an X.25(80) link can only set up calls using this version and any calls using X.25(84) facilities will be cleared.
GPT are now working on a full implementation of all the new X.25(84) features, which will be available in a later release of their software.
The GPT software has been delivered and tested, and installation on JANET service network has begun. We expect to complete installation on all the PSEs during 4Q88. This software is also available to site networks based on GPT PSEs. SEEL has an implementation of X.25(84) for the MULTIPAC PSE although some parts of the development have still to be completed. Accordingly the majority of sites in the Academic Community should soon be able to start providing an X.25(84) service.
The immediate requirement is for each site to upgrade at least part of its network to complement the JANET upgrade. Not all PSEs need to support X.25(84) calls immediately, but can be used to cater for hosts which still operate only to the 1980 specification. It may be that some hosts will never be used in 1984 mode, in which case it may be possible to retain 1980 mode only on some switches during their active lifetime. This is a matter for each site to decide, and due consideration needs to be given to various aspects, including projected lifetime of equipment.
Ian Smith
Whilst progress in the development of OSI protocol stacks covering layers 1-7 for applications such as FTAM and Electronic Mail has been impressive, until recently the development of associated OSI standards for management has lagged behind. However over the past couple of years substantial effort has been put into this work. An addendum to the OSI Reference Model entitled 'OSI Management Framework' has almost reached International Standard status. Standards are being developed in particular areas such as fault management, accounting, configuration management, security and performance, together with service and protocol definitions for the transfer of management information between systems.
It has been recognised that the management standards need to refer to so-called managed objects and the actions that can be carried out on them in a consistent manner, and appropriate documents are being produced. In order to relate what are often abstract definitions to real systems, the individual groups of experts associated with the various layers and applications of OSI need to define how the management standards apply to the requirements of their layer/application. In June a meeting took place between OSI management experts and experts from the Transport and Network groups. As a first step in developing layer-specific standards, lists of mandatory managed objects are being generated. Similar work on LAN-related managed objects is being done within IEEE for 802 LAN technologies. It is estimated that it will be approximately two years before an implementable set of management standards is available. In the meantime the JNT wishes to stimulate activity in the following areas
In the first instance anyone who might wish to participate in this work is asked to send an e-mail message to jnt-b@.uk.ac.rutherford,gec-b. Please indicate if you have an interest in a particular area of network management.
Les Clyne
The JNT are funding a project by the Edinburgh University Computing Service (EUCS) aimed at the provision of Coloured Book and X.25 functionality on the Meiko Computing Surface. The Meiko runs a version of the Unix operating system and the Edinburgh/York/Bristol Unix Communications Package will be mounted with the LAN and WAN technologies.
The Computing Surface is a transputer-based parallel processing engine based totally on British technology. The machine exploits the price/performance gains to be had from employing concurrency and can be assembled in modules to attain the power of a conventional supercomputer. It is a very powerful tool for users who have large scale computational applications.
The Meiko is built round the concept of an underlying spine of transputers (the Computing Surface Network) which run a proprietary datagram protocol to serve various transputer-based servers (which all run Unix). Access to the machine is currently through directly connected terminals by routing interactive traffic through a transputer-based asynchronous server and over the underlying network (CSN) onto a Unix environment situated on a transputer-based processor farm, each user being allocated a transputer from the farm. Each Unix environment (which can be thought of as discless Unix workstations) communicates over the CSN to other transputer-based servers (File and Compute), which also run Unix. EUCS are working with Meiko to produce a Communications Server based on the Meiko MK050 ether and MK064 synchronous boards.
The first stage of the project covers the provision of a Pink Book implementation (the CONS over CSMA/CD technology) and is due for beta testing by October 1988. The second stage of the project will incorporate a Yellow Book Transport Service (YBTS) running over X.25 rev. 1980 for the WAN and is due to be ready for beta test by November 1988. The associated Coloured Book software will be run over both technologies to provide:-
The availability of JANET connectivity for the Meiko is perceived as vital for the Academic Community due to the fact that both the Computer Board and SERC have invested heavily in this technology. Amongst the interested parties in the universities are:- Bristol/Bath, Kent and Edinburgh.
Dave Mercer Edinburgh University Computing Service
This article has two main parts. The first part is a survey of the current state of networking in the Australian Tertiary Education scene. The second part is a brief survey of developments now unfolding in the establishment of an Australia-wide backbone network.
The perspective is from within the academic community; there are wider issues of networking within the research and commercial communities that are not covered fully. The scope of the academic community may be of interest, partly illustrating our problems. Australia currently has about 20 Universities and over 50 Colleges of Advanced Education, such as the South Australian Institute of Technology. These institutions are spread over an area the size of Europe, from Madrid to Moscow.
The views expressed in this article are personal and do not necessarily reflect the views of my employer,
Australian Tertiary Education networking appears to be currently in a state of transition. We have at least three nationwide networks with inadequate links between them. We have only the very beginnings of a nationwide backbone network, although more low speed long distance links are being installed, and permanent local links are also increasing. Meanwhile, plans are being made for a large scale, high performance backbone network.
ACSnet (the Australian Computer Science Network) is by far the largest of the Australian networks. It was begun in 1979 and now connects hundreds of nodes across the country at well over a hundred sites. Many of these are research or commercial sites; ACSnet has certainly reached out well beyond the Computer Science community. This expansion has occurred with no central funding.
ACSnet is essentially a messaging system using store and forward mechanisms, with attendant variable delays. It primarily supports electronic mail and file transfer; a much narrower range of services than JANET.
ACSnet is primarily UNIX based (UNIX is a trademark of A T & T Laboratories). The software underlying ACSnet is called SUN III (Sydney Unix Network Version III). It is available at very low cost compared to the cost of colourbook software.
Inter-machine links range from local area networks to dial-up connections, leased lines and X.25 connections (pretty well any form of wet string). Electronic mail under SUN III is largely based on ARPAnet RFC 822 style headers. SUN IV, now under development, will extend this and also offer X.400 style headers, although it will probably continue to use ACSnet's current wide variety of communications. SUN III supports file transfer in a very interesting way; the sender does not have to know a remote password, and the received file is held in a spool directory while the recipient is notified by mail of its arrival. There is also a weak name service.
One of the nice features of ACSnet is that sites need take no action to keep their address records up to date. The state files that describe the network, all its nodes and connections, are automatically maintained by the network itself. Any significant change to the node configuration or connections causes a state message to flow through the network, triggering changes to state files on all relevant machines in the network.
Melbourne University maintains an international gateway at the node munnari.oz. They have connections to more than a dozen overseas networks including JANET,
In the last year or so, links to non-UNIX networks have become more widely available. Several VMS sites have connections from their DECNET networks into ACSnet. Its not particularly easy, unless you have both a VMS system and a UNIX system on your Ethernet, with TCP/IP available on both. There are still many sites in the College sector which do not have access to ACSnet and where access will prove difficult to achieve.
In the absence of an adequate backbone, performance can be a serious problem with ACSnet. Many users in the Sydney/Melbourne area claim the performance is adequate. However, from my site performance can be woeful. Round trip message times from this Institute to Melbourne and back currently average 25 hours! (We are generally regarded as existing in some sort of black hole here). Wide variations in performance occur because message delivery is multi-hop, often with intermittent links.
To cure some of these performance problems a backbone is gradually being installed. A permanent connection at 2400 bits per second already exists between Sydney and Melbourne, and another between Perth and Melbourne. Other links are now going in, and there is a major proposal, described below, to provide a high performance backbone.
A lot more work is needed before ACSnet can truly claim to be the national network. Its performance must be improved, probably through installation of a better backbone, and coverage of the interested community must be increased through much better links to non Unix systems. However, ACSnet generally works very well, and provides a high degree of communications functionality to end users.
The second largest network relevant to Australian Tertiary education is non-standard, manufacturer specific and almost accidental. It came about when a number of VMS sites wishing to use PSI (DEC's X.25 product) to communicate over Austpac agreed to share a common naming convention. The InfoPSI mailing list was established to distribute these names; the distribution is a command procedure to install the names in PSI. There are now over 45 sites participating in InfoPSI, spread over the academic, research and commercial communities.
However, not all the sites on InfoPSI communicate using PSImail. Many of these sites have installed PMDF (the Pascal Memo Distribution Facility) and some of these use the XSMTP protocol rather than PSImail. XSMTP has some claims to being standard, although of rather mixed parentage: it is SMTP over X.25, an Australian invention now distributed with PMDF XSMTP has advantages in that mail messages can be transferred across multiple hops (DECnet, X.25, then DECnet again, for example) and have their headers re-written in such a way that reply will still work at the final destination. About 20 sites use XSMTP
PMDF sites also have to establish a configuration file and there is currently discussion about how to transfer this file along similar lines to InfoPSI.
Another facility offered on InfoPSI is the PSI _ COPY command, which is equivalent to anonymous FTP.
It is also possible for sites without access to X.25 to use InfoPSI using the Phonenet protocol (used in CSnet in the US), which is supported by PMDF Phonenet works over a dial-up connection; sites without X.25 would dial-up a site with PMDF and X.25 access, willing to act as a gateway for them. As yet only a few sites are using this approach, which offers the possibility of connecting totally non-UNIX sites via InfoPSI into ACSnet.
Although announced with some fanfare more than 2 years ago, the Australian coloured book network (South Pacific Education and Research Network, or SPEARnet) has not been very successful. I believe about 14 academic sites currently have coloured book software plus 3 research sites and 1 commercial site. All but 2 of these sites can more conveniently be contacted by this Institute using InfoPSI. In New Zealand however, I believe all University sites have coloured book networking.
Why have coloured books been relatively unsuccessful? First of all, at the time of its introduction some sites felt that they should go straight to X.400. I believe almost as many sites have X.400 capability as have coloured book but there is not yet much widespread use of X.400.
One problem SPEARnet shares with InfoPSI is the cost of using Austpac. Communications costs with ACSnet are often hidden in telephone budgets, as most of the links are local dial-up. Often someone else pays for the long-distance links!
As people began to investigate coloured book networking further a series of problems became
apparent. First of all, was the quality and cost of the software. This Institute runs the UWIST VMS implementation. We have not been at all impressed with its quality to date, especially the difficulty of linking it to our DECNET network. We are hoping for improvements in the VMS 5 version. I believe that users of UNIX versions have also had problems.
As a result of some of the software problems, it is difficult to create a gateway between coloured book and other protocol networking, using the UWIST/VMS implementation. This effectively isolates coloured book sites from the ACSnet sites.
Another problem early on and still to some extent persisting is the difference in domain ordering. Because of the dominance of ACSnet which follows the ARPAnet domain ordering (high level domain last) we have decided to adopt that order for coloured book. However, the software expects the reverse order, as used in the UK. This results in problems in expanding partially qualified domain names, and also where messages cross the boundary between ARPAnet and UK domain ordering (particularly with REPLY addresses).
A serious problem with SPEARnet is that there seems to have been little attempt to keep users of coloured book software up to date with address changes. This process is what has made InfoPSI so comparatively successful. ACSnet handles the task automatically, In the UK the NRS performs this function, but there is no similar mechanism in Australian coloured book networking. In researching this article. I discovered several coloured book sites that were not in our configuration file.
Coloured book software can be very useful for contacting UK Tertiary Education Sites. Messages arrive there almost instantly, compared with delays of many hours using ACSnet. However, until recently most UK sites sent messages to Australian coloured book sites via ACSnet because most do not have IPSS accounts. As a result, we had to pay for un-solicited incoming mail! Happily, this difficulty has been overcome with the decision recently by the JNT to relay mail to Australia via the BAN-RELAY gateway.
All three of the major Australian networks use domain naming based on the ARPAnet model. ACSnet domain names currently have some restrictions which hinder full conversion to ARPAnet-compatible names. These problems should disappear with SUN IV, which includes improved domain support.
Australia's top level domain is AU and this is now divided into at least 5 sub-domains. The first is OZ for ACSnet; the remainder are EDU, GOV, COM and CSIRO. Each domain has a domain administrator whose job it is to resolve ambiguities and make sure names are clear.
The OZ domain administrator is the SUN III network software itself. This means that anyone can join ACSnet if they install the software, find a site to connect to, and select a name not already in use. Unfortunately this has resulted in many obscure names like goanna.oz.
Last year a Joint Working Party on Networking was established by the University Vice-Chancellors and the Directors and Principals of the other Australian tertiary education institutions. They appointed an Executive Officer and gave him and the Working Party six months to produce a report. This report has just been released and its major recommendations have been adopted in principle.
The report recommends the establishment of a small management team followed by the establishment of a nationwide high bandwidth, high reliability communications network. The goals for this network are quite ambitious: a full resource sharing network is envisaged, with remote login, remote job submission, file transfer, remote spooling, and remote database access as well as electronic mail. Voice, facsimile and video are considered as possible candidates as well. The implementation suggested is 2 Mbps circuit switched links forming the backbone. Dedicated international links are also proposed. It has since been suggested that these detailed parts of the proposal were for budgetary purposes only.
I recently visited a number of UK Universities. The Computer Centre staff I spoke to saw JANET as extremely successful. They put this success down to several factors:
The new Australian initiative should provide the first of these factors. The report currently suggests that network running costs should be recovered from member institutions, although it is hoped that establishment costs can be centrally funded. The report itself ignores address control, but this issue is certain to be dealt with in the technical analysis. Central control of purchasing as a means of enforcing standards is unlikely to be a factor in Australia; we will have to find a somewhat less blunt instrument to get acceptance of standards. I hope we can achieve the critical level of acceptance that will ensure success.
I have great hopes that this endeavour will in fact succeed in unifying Australian Tertiary Education Networking, as I believe JANET has done in the UK.
Chris Rusbridge Academic Computing Service Manager
For some time now each of the National Centres has had a degree of host support for SSMP, and more is on the way. This note summarises the position and prospects at each of the centres at mid-September 1988.
Examples of calls use the syntax of a JNT PAD with the X.29 and TS29 prefixes set to 'A' and TS' respectively. This is for illustrative purposes only. In a particular case the syntax will have to be adjusted to reflect the type and configuration of the physical or soft PAD being used.
RAL runs version 1.1.2 of the Edinburgh SSMP/3270 emulator. The emulator drives an SSMP terminal over the network, and converts to 3270 protocol inboard, emulating a 3277 model 2. Thus all VM/CMS screen-mode services are available at the SSMP terminal as at a 3270 terminal.
The service is available over X.29 and TS29:
CALL A000000000002 D = .XXXE CALL TS.000000000002.TS29E
It is NRS-registered, in TS29 context only, as UK.ACRL.IBSSMP.
For further information use the standard VM/CMS help facilities. FIND SSMP lists key mappings. Other queries and problems may be emailed to US@UK.AC.RL.IB.
RAL provides a limited VMS service for access to the CRAY X-MP/48. Curlew is available on that service, but is not a formally supported facility.
SSMP/3270 conversion is not supported under MVS at ULCC. SSMP is available only for specific full-screen applications. The only such application currently in service is the ZED editor.
Access is over X.29 to DTE address 000005001000. The service is NRS-registered in X.29 context as UK.AC.ULCC. TS29 access is not available.
For further information on full-screen ZED, use HELP ZED under Phoenix and consult MVS file SPEC.ZED.FULLSCRN. Other queries should be emailed to SUPPORT@UK.AC.ULCC
By the time of publication of Network News ULCC should be running a limited VMS service for access to the Cray supercomputers.
As for MVS, SSMP support is at the application level. Specifically, the Curlew editor is available.
The service is accessible over X.29 and TS29. in both cases via a Spider gateway and an Ethernet to the VAX. The service is NRS-registered in both contexts as UK.AC.ULCC.VMSFE.
For an X.29 call, it is necessary first to call the gateway, and then to specify the NRS title of the VMS system in response to the prompt from the gateway. Only the last element of the title need be given since the gateway knows its domain.
CALL A000005000900 VMSFE
A TS29 call is passed through the gateway without the need for a second interaction.
CALL TS.000005000900.ISO.3882611000005000005013
Curlew causes VMS to set the PAD into message mode. The service at ULCC should provide a PAD setting to suit all PADs. Slightly improved performance using Curlew may be obtained using the VMS command
SET TERMINAL/X29/HOLD_TIMER = 2
Queries and problems should be emailed to SUPPORT@UK.AC.ULCC.
ULCC plans to introduce a limited VM/CMS service for access to the Cray supercomputers, probably early in 1989. SSMP will be supported via the Edinburgh SSMP/3270 emulator, as at RAL. No further details are available at present.
UMRCC runs version 1.1.2 of the Edinburgh SSMP/3270 emulator. As at RAL, all VM/CMS screen-mode services are available at the SSMP terminal.
The service is available over X.29 and TS29:
CALL AOOOO1000102011 D = .XXXE CALL TS.00001000102011.TS29E
It is NRS-registered in both contexts as UK.AC.UMRCC.CMSS.
Information is available from the standard VM/CMS help facilities. HELP SSMP provides information on SSMP support. Queries and problems should be reported to 061-275 6010.
Roscoe and TSO may be accessed via the VM/CMS SSMP/3270 emulator. VM/CMS should be called as described above; DIAL VTAM then provides a route to MVS. Normal 3270 services are available in the session. Queries and problems should be reported to 061-275 6010.
John Powers on behalf of TRIKE
Network News 26 contained information on SSMP implementations.
The Edinburgh Rainbow products for PC/MS DOS provide a consistent user interface over various networks (ATS, X.25, via a PC Lan and ethernet CONS). The product support combined Blue book file transfer with terminal emulations for VT100, SSMP, Tektronix 4010, Network/3270. Available from Edinburgh University Computing Service, 59 George Square, Edinburgh, EH8 9JU. Contact Pat Moran or Chris Adie.
A terminal emulator supporting VT100 with SSMP for the Apple Macintosh is also under development and will be available towards the end of this year.
John Dyer
CBIGAL (Coloured Books Implementors Group on Asynchronous Lines) has changed its name to ANIG (Asynchronous Networking Implementors Group).
Sue Weston
City University, 12th and 13th December 1988
The subject of this event is Open Networking solutions to satisfy end-user requirements in their use of networking Unix systems. Talks will be given covering the topic with particular reference to the networking applications that users require, to generating and porting appropriate software, and to supporting it in service. There will also be coverage of some general OSI matters. The following items will be on the agenda:
It is hoped that both regular attendees at the annual JNT-NE workshop held every spring will find this event useful, as well as those who are usually unable to attend.
There will be ample time for questions and discussion, an open forum with all speakers invited to join a panel, and a parallel session to address special subjects.
The registration fee for the workshop is £70 for residents and £60 for non- residents. All those interested in attending the workshop should contact Peter Tinson at City University (Tel. 01-253 4399 ext. 3758 or via email: P.A.Tinson@ciiy } for a booking form.
Phil Jones
The COSINE (Cooperation for OSI Networking in Europe) Project was adopted as a part of the Eureka programme at the 2nd Eureka Ministerial Conference in Hanover in November 1985. The objective of the project is to provide the European academic and industrial research community with a computer based open communications infrastructure. All Western European countries, the European Commission and Yugoslavia are participating. COSINE is therefore the most European project in Eureka.
The first intention of COSINE is rapid establishment of an environment to make data communication services available to users from both academic and industrial research organisations. These services will:
The COSINE project has reached an important stage. Recently the first part of the project, the specification phase, was concluded. The specification phase was executed by RARE, the European association of research networks and their users. RARE set up a detailed plan for this work, the execution of which involved specialists from all over Europe, from consultancy bureaux, individual consultants, software houses, universities and, of course, the RARE Working Groups. Areas covered by the specification phase are: user requirements evaluation, selection of urgently needed and possible future services, evaluation of present and planned public data communication services, examination of operational requirements and migration strategies. The specification phase provides a blue print for the next phase of the COSINE project, the implementation phase. The results of the specification phase will be published this autumn. The first detailed technical reports are already available.
In January 1989 the three-year COSINE implementation phase should start. The main organisational structures have already been designed. A large part of the activities between 1989 and 1991 will take place at a national level. For activities at a European scale, a COSINE Project Management Unit will be established. Proposals for this CPMU are being prepared at this moment. The total costs of the international activities within the implementation phase will amount to some 35 million pounds. For developments at national and local level the expenditures of the various countries will probably exceed 300 million pounds, according to estimates from the COSINE Policy Group.
In June this year, the Eureka ministerial conference in Copenhagen already expressed its support for the further execution of the COSINE project. The Eureka ministers promised to promote the availability of the necessary finance in their own country. The final decisions on the content, organisation and financing of the implementation phase of COSINE will be taken towards the end of this year.
Ruud Olihoff
With the stabilisation of the majority of ISDN specifications in the CCITT 1988 Recommendations, it was decided to investigate the use of ISDN infrastructure for the support of OS1 data communications in the service environment over at least the next five years. This role would be in addition to other uses of ISDN for non-data communications such as voice, video etc.
In parallel with the CCITT activity, a program of work has been established in ISO on the use of ISDN for OSI data communications. The first detailed specification from this work is a document currently at the Draft International Standard level (DIS 9574) which specifies various ways in which the OSI connection-mode network service can be supported over ISDN. A technically similar document exists in CCITT and this work is being used as the basis for developing European Functional standards for OSI over ISDN. Work by PTTs on the provision of national ISDN infrastructure for OSI data communication services is also well advanced.
It is likely that national services will emerge before their international equivalents because of technical difficulties and variations associated with international connectivity. In addition suppliers of hosts, PABXs and LANs are developing ISDN equipment and interfaces for data communications.
A feasibility study has been initiated by the JNT to review at a detailed level all the above areas in order to consider the possible use of ISDN by the UK academic community for OSI communications over at least the next five years. This study is being done in the context of existing WAN and LAN infrastructure and will take into account the need to maintain continuity of services where appropriate and where technically possible. The study is being carried out by Professor Kirstein and his colleagues from the Department of Computer Science, University College, London who have long and wide experience of ISDN technology and services.
The results of the study are expected early in 1989 and will be used as input to discussions on future developments of open systems infrastructure for the academic community.
Les Clyne.
A survey conducted for the JNT by Daresbury Laboratory and Edinburgh University in 1987 identified four categories of LAN Bridge,
The immature state of the IEEE 802.1 Part D standard for MAC Bridges at the time was reflected in the state of the products then available, and it was possible to recommend products in the local bridge and remote kilostream bridge categories only. The recommendations, which were circulated to ether sites in November 1987 were
Since the survey was completed, a number of new products claiming substantial price and performance improvements have become available, and more are expected. To cope with this fluid market situation, the idea of an Evaluation Centre was conceived. The Evaluation Centre would have a two-year mandate to watch the market place, and test and make recommendations on the most promising products available, in other words, perform an ongoing survey over the two-year period.
Proposals were invited from the JNT LAN Advisory Group, and, after discussion, a contract has been placed with Rutherford Laboratory Informatics Department.
The immediate objectives of the Bridge Evaluation Centre are to make recommendations in the categories of bridge where it was felt not possible to recommend in 1987, and to advise on whether the 1987 recommendations should be updated. The initial market survey will be by technical questionnaire, with the most promising products being given hands-on testing. First results from Informatics are expected early next year.
Dick Gillman
Pink Book is the specification of OSI protocols for use over CSMA/CD (Ethernet) LANS and is compatible with UK GOSIP and European Functional Standard ENV41103.
The state of Pink Book products is evolving and hence it is only possible to give a snapshot as of 3Q88. Also since many developments are being done by manufacturers in-house or in some cases in collaboration with sites in the academic community, the information may be slightly out of date. The JNT's current understanding is as follows:
Dick Gillman
The NISS Bulletin Board (NISSBB) is an online information service which can be accessed from any terminal or workstation which has access to the Joint Academic Network (JANET). In most cases terminals are connected to the network via a PAD which is used to set up and manage the network connection to host systems including the NISSBB. If you are not familiar with what is involved your computer service will be able to advise you.
NISSBB is connected directly to JANET and its X.25 address is 000062200000. It is also registered in the NRS as NISS and PAD managers are encouraged to configure this name into PADs so that users are not required to remember the JANET X.25 address. Assuming your PAD is configured to recognise the NRS name you can call up NISSBB by simply typing:
CALL NISS
So, why should you be interested in NISSBB. No matter what your involvement in computing it is likely that you will face problems which others have already faced, or you may be working in a fairly specialised area in which there is little local expertise. Some people who are "in the know" already use JANET to participate in special interest groups, mailing lists. bulletin boards, etc. which already proliferate. However, many are not so well-informed and for some time there has been a need for a national online information service; a focal point for the numerous information sources which already exist. Needless to say this is an ambitious undertaking, and one which can only be realised if you participate and share the knowledge that you have.
NISSBB is so easy-to-use that I will not waste your reading time by describing it here. There are no registration formalities to complete, so having established a network connection using the CALL NISS command you are led directly into the top-level menu:
A Introduction to NISSBB B The NISS Catalogue C Academic Computer Services D National Computing Initiatives(CHEST, CTISS, NISS) E Library Services F Online Information Sources G Computer-Related Job Vacancies H Computing Groups and Committees I Machine-Range User Groups J Software User Groups K Conferences and Courses L Joint Academic NETwork M Main Menu N Sciences O Social Sciences P Medicine Q Engineering R Arts and Humanities Y Comments Z Table of Contents
Each of these sections lead onto other menus and other sections. You can display the contents of any one of these sections by simply typing in the appropriate key.
Because NISSBB is designed to be easy-to-use over a computer network, from any one of a wide range of terminals and workstations, and by a potential audience boasting varying degrees of computer-literacy, the facilities offered are few in number and deliberately simple. The most glaring limitation at present is the absence of a sophisticated and comprehensive index/search facility and this is something planned for NISSBB Mark II. There is however a useful, if rather simple, basic section search facility.
Section A (together with the HELP, HINTS and SEARCH commands) tells you all you need to know about how NISSBB is structured and what facilities it offers. For example Section A4 describes how to download information from NISSBB onto your local computer system, should you want to print it out.
Section D covers some national computing initiatives such as NISS, CHEST and CTISS. Under Section D3, on CHEST, you will find details of the current educational discounts available on all kinds of computer software, the CHEST Directory of Micro-Software Prices, and the current list of public-domain software held by the Lancaster Micro Service, with instructions on how to download the software to your preferred system.
Under Section D4, on CTISS, you will find details about the majority of the 139 projects in the Computers in Teaching Initiative. Many of the projects have come to fruition, having produced CBT software which may help you in your teaching. Why re-invent the wheel when a quick check will put you in touch with colleagues who may have something which you can use or easily adapt
Under Section D2, on NISS, you can read about the NISS Software and Datasets Catalogue which is the other main service under development within the NISS project. Work on the Software Catalogue is progressing well, and details are given on how to register to look at the current pilot service.
Section F holds details of other sources of information which are available to you either over JANET or further afield. There are hundreds (if not thousands) of special interest groups who use electronic mail or Bulletin Boards to correspond; we have details of a few.
Section G has been set aside to advertise computer-related job vacancies in the UK Academic Community. Advertisements can be placed free of charge for the remainder of 1988 - after that there will be a charge; mail NISS@SWURCC for more information. One of the current vacancies is for a Bulletin Board Administrator within the NISS Team!
Section H is given over to publicise the work of the many computer groups and committees which are run by the IUCC and PCCC. (Also of interest in some quarters will be Section C2 which has entries for each PCCC/IUCC computer service.)
Finally you are directed to Sections N-R, which are devoted to publicising information which is academic discipline-specific. Although there is not a lot of information here at the moment, these sections are set for healthy expansion. Geography, for instance, will have a substantial amount of information online within the next month.
You should be aware that the team looking after the Bulletin Board information consists of about half of one person (bits of three people in reality), so you can help us out by supplying information which is currently lacking, and please be patient. Our aim is to establish NISS as a worthwhile service by November 1990!
The Bulletin Board is gathering contributors as it rolls along, but I would particularly like to thank Peter Abbott (Aston University) for his help with sections C and D, Janet Wheeler (University of Newcastle) who is looking after section F, and Jim Morris (University of Leicester) who is the author of the underlying software.
Please try out NISSBB and pass on the word if you find it useful. If you feel that there is a section of information not represented and you think that you are well placed to manage such a section, please get in contact with us.
Rob Armstrong
