This is a busy time of year for those involved in preparing for the annual SIGGRAPH conference. One casualty has been the promptness of this issue of Computer Graphics. The combination of a larger than usual issue and a heavier than usual workload contributed to the delay. In producing the newsletter, we do share your concern for timeliness, especially publishing time-sensitive material such as calls for participation and deadlines. However, you will find several items of interest that I believe are worth the wait.

This issue introduces this ``From the Editor'' page with comments from the SIGGRAPH editor-in-chief and letters to the editor. Two inaugural letters appear here, one congratulating SIGGRAPH and the other raising some controversial concerns about SIGGRAPH and the art community. I urge you to read them both.

The major content of this issue is the report on the SIGGRAPH workshop on software tools for user interfaces. This workshop, sponsored by SIGGRAPH, was held in November 1986 to synthesize new ideas and directions for future research in user interfaces through discussion and interchange. The invited workshop produced four summary reports and several position papers. Dan Olsen chaired the workshop and collected the material published here. The manuscripts were delivered electronically and were copy-edited and reformatted for publication in Computer Graphics.

My significant concern for the quality of bibliographic citation of computer graphics research resulted in the article ``In-Citing Computer Graphics'' that appears in the SIGGRAPH Activities section. Coincidentally, Kelly Booth supplied an article about citations for many of the ``classics'' of computer graphics that are almost never cited in their original form; although these papers were presented at SIGGRAPH conferences they were published elsewhere in CACM or TOG.

The education committee continues to be active and has submitted a substantial education column on careers in computer graphics, complete with a questionnaire and a request for role model essays.

Continuing columns appear from Joe O'Rourke on computational geometry and from Jon Meads on standards activities.

If you want more technical substance to this newsletter, then Jim Kajiya has introduced a new column for you titled ``Graphics Goodies'' to publish small technical notes submitted in electronic form.

Other new material introduced in this issue includes tables of contents for issues 24 and 25 of the SIGGRAPH video review. We will index future issues of the SIGGRAPH video review in Computer Graphics on a regular basis, as we have done for the slide sets in the past few years. You should also know that new labels for your old issues of the SIGGRAPH video review are in preparation and will be available shortly. If you wish to get more information about these labels, contact Deborah Cotton, Single Copy Administrator, ACM, 11 West 42nd Street, New York, NY 10030, (212) 869-7440.

I welcome your thoughts and comments on SIGGRAPH activities, especially its publications. You can reach me via the electronic mail address below or via the information in the masthead. SIGGRAPH is obviously a busy and vital organization. Thanks for making it so.

Richard J. Beach
SIGGRAPH Editor-in-Chief
Beach.pa@Xerox.com

It has recently been brought to my attention that the SIGGRAPH'87 conference committee has decided to offer a set of 8" by 10" photographic enlargements of 4 computer graphics images for sale at the upcoming conference. These photographs are in lieu of a conference poster. As a long standing SIGGRAPH member who has been active for many years in the art show and related conference activities, I feel it is important to share with you my thoughts and concerns.

The issues of technical excellence, financial responsibility and esthetic judgment are the primary concerns of each conference committee. The individuals who comprise these committees have varying degrees of expertise in these areas. Success in each of these areas determines the success or failure of each conference. The individuals selected for the conference committees are all prominent professionals motivated to contribute their time and expertise for the good of SIGGRAPH.

It is my belief that the judgment of the conference committees is strongest in the technical areas and weakest in the esthetic sensibilities. The conference committees assign the selection and evaluation of technical papers to people with extensive knowledge of the areas covered by the papers. The judgments of this review committee are accepted without hesitation because they are educated professionals in the field. No one would ever say that their evaluations of the quality of either the writing or the research are ``just unsubstantiated matters of opinion.'' It is unthinkable that a conference committee would say that anyone, no matter what their background, is entitled to an equally weighted opinion in these matters, particularly in areas where those offering opinions have neither experience nor knowledge.

Nonetheless, these same arbiters of quality, with such high academic standards, repeatedly insult the art community by refusing to accept knowledgeable, educated opinions and suggestions from either these arts professionals, who are SIGGRAPH members, or reputable members of the traditional art world, even when these individuals have demonstrated by their hard work for SIGGRAPH their interests both in SIGGRAPH and in sharing their knowledge about art with computer graphics professionals.

In the interest of making money for their conference, the SIGGRAPH'87 conference committee has made certain decisions about matters concerning art and the exploitation of images created by artists and scientists. No one objects to SIGGRAPH conferences making money, quite the contrary. However, there are acceptable and unacceptable ways to exploit art. When the SIGGRAPH conference committee considers ways to produce revenues from art they should research in what ways reputable art institutions have done this in the past. Slides, posters and postcards are among the acceptable choices. It is my opinion that is is unacceptable to make photographs of art for sale. SIGGRAPH does not want to be in the business of making art for sale. Nor is it SIGGRAPH's intention to say to the world that these 8" by 10" photographs are what ``computer art'' is. Producing photographs of computer images results in a confusion of the promotional product with the artwork itself. The slide sets, which are an annual conference success, have clear educational value and are labeled with all the necessary information. Postcards are an excellent way of promoting all of the SIGGRAPH activities and were used successfully in the Japan SIGGRAPH art show of 1983. The decision not to have a poster this year is based on poor poster sales in the past. People love posters, so there must have been something wrong with some of the past SIGGRAPH posters if they did not sell well. Posters of quality and interest to conference attendees would certainly have sold.

The issues involved in creating a successful SIGGRAPH poster include the following:

In my opinion the most successful SIGGRAPH poster was the one in which an entire slide set was reproduced. My reasons are in part based on the fact that I have, over the years, seen this poster on so many office walls and in so many computer graphics laboratories. This poster has nostalgia value. It remains educationally valuable today. It shows how far the field has developed. Instead of making the mistake of using one or a few images to represent the whole field, everything is represented. The poster reads like a film strip. The poster's design makes the content more available to the viewer in contrast to posters where the design obscures the images. In the past SIGGRAPH has tried producing a variety of posters for each conference. I think that one really great poster is the best approach.

The issue of quality in promotional materials is an important one as it determines the financial success or failure of an item. When a SIGGRAPH conference committee looks at the previous years' failures they might realize that the cause of this is the obvious lack of quality and/or appropriateness for the audience of a specific item. This simple realization might lead this committee to find someone who has expertise in merchandising, computer graphics and design to supervise such items for sale. Expert opinion should be sought and heeded in these matters. It is common knowledge, particularly among academics and scientists, that expert opinions should be sought and new areas of knowledge researched before decisions are made. The SIGGRAPH conference committees are not going to reinvent the history of art. They might, however, learn the do's and don'ts in this field from members of the art community and from established art institutions and organizations.

While editing the workshop proceedings that are included in this issue and while preparing authors' instructions for the SIGGRAPH'87 conference proceedings, I was struck by the chaos in citations to published computer graphics research. Clearly, the advances in automated bibliography generation tools, which I saw being used by knowledgeable authors, have not improved the situation. I also believe that newcomers to our field will have some difficulty in obtaining the referenced material due to the sloppiness of the bibliographic citations encountered. As editor-in-chief, I decided to improve the situation and therefore spent some effort researching proper citations and subsequently editing the references in this issue of Computer Graphics.

I acknowledge a prompt from Chip Hatfield who forwarded the imprint page of the CHI'86 conference proceedings. That imprint page contains a model citation to a paper in those proceedings. He suggests that SIGGRAPH do the same and we will. However, SIGGRAPH has a more complex situation than simply referencing a paper in our conference proceedings.

SIGGRAPH publishes computer graphics material in several forms: our Computer Graphics newsletter, the annual conference proceedings, slide sets and video reviews. ACM SIGGRAPH is considered [by me] to be the publisher and the materials are all available from ACM in New York. Therefore this material is considered published and ``in print.'' The conference proceedings are published as a regular issue of Computer Graphics. Slide sets are published and available for sale from ACM, as well as being indexed in Computer Graphics shortly after the annual conference. The SIGGRAPH video review is an irregularly published video serial which is an optional supplement of Computer Graphics. The video review will also be indexed here.

SIGGRAPH also frequently produces several forms of unpublished material. The most frequently referenced material, notes for SIGGRAPH courses, is not considered ``published'' because they are available only at the conference and unavailable from a ``publisher.'' In fact, there are significant copyright implications if such notes contain original or reprinted material and were to be considered ``published.'' Nonetheless, material occasionally appears solely in these course notes and is later referenced in research papers. Catalogs for art shows and electronic theater presentations are also unpublished SIGGRAPH material. And images incorporated into SIGGRAPH posters and promotional material may be worthy of citation but are contained in unpublished material.

The following brief tutorial explains how to cite a variety of interesting computer graphics material. The references all use the style found in ACM journals. For more details, consult a friendly librarian or refer to the American National Standard for bibliographic references (ANSI Z39.29-1977), American National Standards Inc., 1430 Broadway, New York, NY 10018.

There was a recent article in comp.mail, the network news group for computer graphics, requesting information on a paper presented at SIGGRAPH'80 by Loren Carpenter. Only the abstract appears in the proceedings for that year. The actual paper appeared later in CACM 25, 6 (June 1982) 371384 as part of a joint article with Alain Fournier. This is one of a number of papers that were presented first at SIGGRAPH conferences but later appeared as journal articles.

The SIGGRAPH'83 conference proceedings have a complete ten-year cumulative index for the SIGGRAPH conference papers. Forward references are given for all papers that did not appear in the original proceedings. Because SIGGRAPH'82 was the last conference for which papers did not appear in the proceedings, the cumulative index is an exhaustive list of such papers. The cumulative index includes papers from the first SIGGRAPH'74 conference, although there was no proceedings distributed at the conference. The papers appeared subsequently in four issues of XXXX, a journal which no longer exists.

The reason that some papers did not appear in the original proceedings is related to ACM's rules on publication in ACM journals of papers that have appeared in ``widely disseminated conference proceedings.'' In the past this was interpreted to mean that papers appearing in the SIGGRAPH proceedings could not be published in JACM, CACM or TOG. In 19761982 the ``best'' papers were culled from the conference for CACM (TOG in 1982) and did not appear in the conference proceedings (although they were distributed in a separately bound supplement only to those who attended the conference).

In 1983 the decision was made to refuse to permit publication of conference papers in TOG unless the paper also appeared in the proceedings. A compromise was reached whereby the proceedings ``reprinted'' the papers as they appeared in a simultaneously published issued of TOG. That same year the SIGGRAPH executive committee passed a resolution that all future SIGGRAPH conferences would be required to publish all presented papers in the proceedings. This effectively ended the practice of any papers from SIGGRAPH appearing in ACM journals.

Recently the ACM publications board has informed us that the policy that caused all of this had been changed a number of years ago and that there is no a priori reason that a SIGGRAPH conference paper could not be accepted by an ACM journal. When all of this is sorted out it may be the case that some papers from the SIGGRAPPH conference will later appear (often revised somewhat) as ACM journal articles.

The one real casualty in all of this was a paper by Jim Clark ``A Fast Scan-Line Algorithm for Rendering Parametric Surfaces'' that appeared only as an abstract in the SIGGRAPH'79 proceedings. It had been selected to appear in CACM as one of ``the best papers'' from the conference. For reasons that have never been clear to me, the paper died in the refereeing system of CACM and was never published. The version distributed to conference attendees has subsequently appeared in SIGGRAPH course notes, but never in a form that is generally available.

The SIGGRAPH executive committee adopted the policy of requiring all conference papers to appear in the published proceedings precisely to eliminate the problem of people not being able to locate some of the best material presented at the conference.

Papers that originally appeared at SIGGRAPH conferences but were only published as journal articles include the following ``classics'' that are almost never cited in their original form. In some cases the later versions were the result of merging two or more papers, thus the originals (distributed only to conference attendees) often contained more detail than the final versions which appeared later.

Starting with this issue, I will be editing a column called ``Graphics Goodies.'' This column is intended to be a forum for the lively interchange of graphics ideas. The idea behind graphics goodies is to provide a quick turnaround publication vehicle, much like the forum in our sister publication SIGPLAN Notices.

I'm looking for timely submissions in short form (less than 20K bytes), of medium to high quality. The submissions should be well thought-out — more so than, say, the quality of typical computer mail — and should concern a topic of interest to a relatively wide audience of graphics people.

The subject areas are wide open. Here are a few suggestions: rendering tricks and speedups, clever applications, small algorithms, proposals of and discussions on interchange formats, position papers on research trends, interesting ideas, suggestions for research, open problems in graphics, etc. Sales or promotional literature is not wanted and will be culled.

Although some culling of unsuitable material by the editor may occur, no reviewing of submissions is planned. Thus it will be the responsibility of the author to ensure accuracy and of the reader to accept ideas critically.

Subscribers to the SIGGRAPH video review will receive a letter offering them new printed labels for their videotapes. Issues 24 & 25 were held up in distribution waiting for labels and eventually were shipped without them. The problem resulted in a committment to design a new label for the video review and to supply existing subscribers with sufficient labels to relabel all of their back issues.

These new labels are now designed and printed for issues 24 & 25. Labels for all of the back issues will also be printed and made available. If you wish to get more information about these SIGGRAPH video review labels, please contact Deborah Cotton, Single Copy Administrator, ACM, 11 West 42nd Street, New York, NY 10030, (212) 869-7440.

The sample below is the VHS cassette label, which shows the logo treatment and identification of the issue. Note that each label states that copyright is held by the contributors. The VHS box label and both the 3/4-inch cassette and box labels have sufficient room to print the issue table of contents.

The SIGGRAPH executive board at their February 1987 meeting declined as too costly a proposal that would enable SIGGRAPH members to receive automatic subscriptions to the magazine IEEE Computer Graphics & Applications as part of their member benefits.

Chip Hatfield, out-going editor-in-chief of CG&A, and Rick Beach, SIGGRAPH editor-in-chief, presented the proposal. The current circulation of CG&A is about 11,847 subscribers, of whom approximately 2,209 are SIGGRAPH members. The present SIGGRAPH membership is 11,327. The combined CG&A circulation would then be approximately 21,000. SIGGRAPH membership dues are $18 (increasing to $20 in 1987) of which about $4 goes to ACM and $14 pays for the SIGGRAPH newsletter, Computer Graphics. All four regular issues of the newsletter cost about $4 per member per year, while the conference proceedings cost about $10. In the CG&A proposal, SIGGRAPH would retain the conference proceedings but would eliminate the regular issues of the newsletter, providing $4 to pay for CG&A.

At its previous meeting in November 1986, the SIGGRAPH executive board was sympathetic to supporting CG&A. The magazine serves a useful role in the community, especially with its timely publication in color, and support was strong among the board members. It was anticipated that a dues increase of $5 to add CG&A would be acceptable while a $10 increase might be a tolerable upper bound. SIGGRAPH surplus funds were suggested as an additional source of interim funding to initiate the proposal.

Three options were presented by CG&A at the February meeting: 1) $26 for 12 issues @ 80 pages per issue = 960 pages (the current magazine size); 2) $14 for 6 issues @ 96 pages per issue = 576 (larger issues less frequently); 3) $19 for 12 issues @ 64 pages per issue = 768 pages (smaller issues). Thus the minimum cost to SIGGRAPH member would be $10 for the smallest magazine and $22 for the current one.

Additional discussion at the executive board meeting raised, without resolution, the issues of SIGGRAPH identity, coordinating policies with the ACM and IEEE publications board, and prepaid SIGGRAPH dues as a result of the recent tax law changes.

The bottom line was that the CG&A proposal was too expensive for SIGGRAPH and no action would be taken on the proposal.

ACM Transactions on Graphics is soliciting papers for a special issue on realistic image synthesis. Submissions are due by June 15, 1987. This is a 3-month extension from the previous deadline. Topics include, but are not limited to:

Four copies of the manuscript should be submitted to either of the special issue editors:

The January 1986 issue of TOG is presently in the mail. Indeed, those of you who subscribe should have received your copy by the time you read this. We are therefore approximately one year behind schedule. However, there are now nine papers at ACM headquarters being typeset and another five papers which have been accepted and should arrive at headquarters shortly. These will fill four issues and part of a fifth. It should be possible to complete this last issue soon (April 1987) from among the three manuscripts now in the ``conditional accept'' category or from those now in review or revision.

ACM is working hard to accomplish the typesetting as fast as possible. It seems probable to me that, by the middle of this year, we will be no more than one issue behind schedule — a vast improvement due chiefly to the efforts of Dan Bergeron (former editor-in-chief), Jim Foley (editor for the Special Issue on Interaction Techniques), our reviewers and the publications staff at headquarters. In order to eliminate the gap completely, and to stay on schedule, we will be pushing papers through the pipeline as rapidly as possible, consistent with careful refereeing. As a result the backlog of accepted papers is and will be nil in the near future — by implication it's a good time to submit to TOG, and I hope you will continue to do so.

Please note that we have moved the submission deadline for the Special Issue on Realistic Image Synthesis to June 15, 1987, as shown in the call for papers elsewhere in this issue of Computer Graphics. This change will allow us more time to advertise the special issue. Please contact Don Greenberg (607-256-7444) or Turner Whitted (919-962-0195) if you have any questions.

SIGGRAPH is publishing the proceedings of the 1986 Workshop on Interactive 3D Graphics held on October 2324, 1986 at Chapel Hill, NC. Abstracts for the papers presented at this workshop appeared in Computer Graphics 21, 1 (January 1987) 37. Participants at the workshop will receive their proceedings in due course. For nonparticipants, rather than publish the workshop proceedings in Computer Graphics, SIGGRAPH will distribute the first 1,000 copies free using the coupon offer below. Additional copies will be for sale by ACM; ask for order number 429861.