Welcome to Graphics Newsletter No 2! If you have not seen this Newsletter before and wish to get on the mailing list. please send your name and address (preferably by email) to me. New readers are always welcome! New contributors are also always welcome!
The main activity since the meeting in September has been getting approval from the Computing Facilities Committee and the Computer Board for the recommendations from that meeting.
It has been agreed that the GKS-UK code will be completed to full ISO standard and validated by NCC. The GKS-3D developments at Manchester will be incorporated and the source of the combined system made available to the community. It is hoped to complete the GKS validation in April. This version of GKS is seen as an implementation for teaching, enhancement, modification and general academic use. It is not seen as the main supported commercial product for Computer Centre use. UNIGKS is the recommended supported commercial product for Computer Centres where appropriate.
In the interchange area, it has been confirmed that CGM is the main protocol for exchange of pictures and we will aim to get a number of tools available in the community to support it. It has been agreed that we will use SGML as the markup language for transfer of documents between sites. For example, the GKS Documentation for GKS-UK will be in SGML with drivers for standard text formatting systems such as troff.
The Computer Board has funded the production of an Introductory Guide to PHIGS with Viewgraphs for general use in the academic community. This will be distributed in SGML format also.
Most of the other decisions concerned directions rather than specifics. For example, we plan to assess PHIGS implementations and look at the colour printer market in the near future.
Bob has outlined most of what is going on at the moment. I would just make a plea for people to keep me informed as to what they are up to. Do you have any good ideas which can be shared with others? Let me know. I see one of my roles as spreading the good ideas that are around.
The UNIRAS training materials meeting showed the level of concern of sites and also a great willingness to work together to get something good off the ground quickly. Can you help? Let me know.
If you have any experience of GKS versions (PC or mainframe); PHIGS implementations; colour hardcopy, please send me some information. This will help us to assess these areas.
From some recent discussions, it has become clear that UNIRAS UK Ltd regard CHEST as one customer especially where contractual matters are concerned. There has also been a hint that this view of CHEST may not be restricted wholly to contractual matters and this has been of some concern to the UUUG Executive Committee since the current membership of of the UUUG is heavily CHEST -biased. UNIRAS have kindly helped us to contact non-CHEST sites by mailing our publicity material to all their Industrial/Commercial customers. (UNIRAS cannot give us the mailing list because of contractual problems). However, so far, there has not been a huge flood of new non-CHEST members. We suspect this may be due to the fact that the mailings may not be reaching the right people i.e. those who actually install, support and use UNIRAS.
I am writing this note therefore to encourage anyone who is, or knows of, an industrial/commercial user of UNIRAS, to let me know the contact's name, address etc. This may help us to twist a few arms and to balance the UUUG membership so that we can claim to be truly representative of the whole UNIRAS community in the UK.
It is getting very confusing in computer graphics in the Higher Education and Research communities. From the position of 2 or 3 years ago when effort, coordination, communication and quality software was sparse, to the position today with a multiplicity of committees and people actively working in this area, there is a stark contrast. This is, of course, a welcome and long overdue change in the graphics scene.
This appears no less confusing than to UNIRAS Ltd who receive all manner of communications by a variety of media, about many aspects of the CHEST deal for UNIRAS. It seems apposite to try and clarify the situation for UNIRAS and the rest of us in the community who are totally confused:
Cranfield put a paper forward sponsored by the IGWP for funding a member of the UNIRAS support staff at Cranfield part time, enabling JANET to be exploited in communicating with UNIRAS. Further news is anticipated about this.
In October, CHEST circulated details of the offer to education and research establishments about USEIT, UNIRAS UIMS. This equates to the OSF's MOTIF and UNIX International's Open Look X Windows software. SERC have a team based at UMIST who are currently reviewing USEIT. Watch this space for further news about this.
An open meeting was held at Loughborough University on 15th November to discuss training materials for UNIRAS. The result is a wish list which is to be costed and a submission made to the Computer Board for areas which need funding. The aim is to put together a coordinated package of materials which can be used in the community. The package will be for UNIRAS version 6 and will be available across the network where appropriate (e.g. documentation). The aim is to produce a range of training materials for the UNIRAS software and to produce these in a timely fashion. It is appreciated that there will be a time-out on the materials.
The following areas are being investigated:
I am also interested in knowing who would like to put together the various bits of the documentation. Let me know if you would like to get involved. We also need some data sets and examples. Does anyone have any good ones?
This is a report of the October 1989 meeting of ISO/IEC JTC1/SC24, the international standards committee responsible for computer graphics, held in Olinda, Brazil.
PHIGS and GKS-3D, the three-dimensional application programmer interfaces, are now complete and published. Language bindings for Fortran and Ada are also complete and the C ones (much changed from earlier drafts) are stable but drafts will not be available till mid 1990. Implementations of PHIGS are starting to become available, but be careful: none are complete and some bear only a passing resemblance to the standard. GKS-3D remains commercially neglected, though having GKS segments and the same viewing pipeline as PHIGS, it is a very sensible product. The PHIGS central structure store that replaces GKS segments is horrible to use and much less useful than it seems. I would advise programmers not to go too deeply into it, if they value the robustness of their code! Work has started on PHIGS PLUS (PHIGS Lumiere Und Shading). Here again products are becoming available that likely will bear even less resemblance to the final standard than some current PHIGS offerings!
There is now standard GKSM, audit trail metafile for GKS which uses CGM codings. Hopefully products will follow.
CGM itself is improved, with the soon to be published Addendum 1. CGM now supports GKS functionality, including segments and workstation transformations. Pictures are truly static, with no vestiges of audit trail dynamic effects like colour table changes within a picture, just like our Reference Model says it should be! CGM is a commercial success and is now THE standard non-proprietary way to transfer pictures between CAD, CAE, DTP and graphics arts systems and even to hardware. Many will have seen the excellent demonstrations at Eurographics UK, and at the Computer Graphics Show where it was a major attraction and space was donated by the organisers. CGM is being amended to handle curves, precise control over picture appearance, and true colour. Completion is scheduled for 1991.
There is to be a CGM-3D, matching the functionality of GKS-3D. It is now technically stable with the next draft to be available in early 1990. This will be a major benefit for users of three-dimensional graphics, where there is currently no standard picture transfer format.
CGI, the Computer Graphics Interface, is going to be completed at last. The stable texts will be available in early 1990. The market for CGI is probably going to be smaller than had earlier been hoped, but there still remains considerable supplier interest, especially among the specialist companies.
Registration of graphical items, additional lines styles for example, is now in operation with the first batch of items agreed.
The standard for conformance testing of graphics systems is now well advanced. What, you may well ask, is the need for a standard peculiar to computer graphics? There are particular problems in testing interactive systems and with computer graphics, a human observer has to agree that the picture looks CORRECT.
Indeed all the first generation graphics standards are complete or going well. First generation means that we know they do not fit together as well as they ought. The work on the Reference Model, the framework for the second generation of standards which will fit together properly, is progressing. However, the review of GKS, that the UK hoped would take GKS into the second generation, has suffered a major setback with SC24 being unwilling to accept the change in approach that is required.
One development for the future is the likely standardisation of the X11 protocol. ANSI are currently processing this, technically unchanged. The UK have said that they want the completed work brought into ISO. SC24 endorsed this. X is however incompatible with everything else, but there is little to be done about that with products starting to come onto the market. X needs to be frozen so we can buy it and use it. Work on a rational replacement is also necessary.
Image processing is another new area where work is starting.
The standardisation process becomes ever more complex, with relationships between graphics and previously separate areas springing up all over: with office systems over colour, raster, font and page description technologies, with communications, with product data and many others. We need more expertise. Contact Chris Cartledge, if you are interested in helping with the work.
