Best of GEnie..... October 1990 News from the GEnie Forth RoundTable by Gary Smith Two expressions pop to mind as I realize how long it has been since I recapped GEnie Forth RoundTable Real Time guest conferences. The first is, "Time sure flies when you are having fun.". The second has something to do with posteriors, swamps, and alligators. These trips down memory lane are without doubt one of the more popular features of this column, and certainly they are among my favorite to present. I have said this before, but it bears repeating. We have really been graced with an array of guests with divergent views, but without exception all have been interesting and a delight to chat with. I would encourage you to participate in at least one of these. The guests generally appear one per month on the third Thursday except for the third quarter (October through December when they are moved to the second Thursday to avoid holiday conflicts. If you are interested in knowing who might be scheduled to next appear this is posted iin the GEnie Forth RoundTable Bulletin Board in Category 1, Topic 6 with the current invitee pre-announced in the door banner that greets users on entrance to the RoundTable each day. Enough of that. We have a lot of catching up to do. This issue we will focus on vists with Robert Smith, "Floored Division and Floating Point", Phil Koopman, "Stack Machines", Charles Curley, "A Minimalist View of Forth", and John D. Hall, "The Business of FIG". As has become a standard format for these guest sessions I will recall only the guests opening remarks. These can be used quite accurately as a pseudo abstract of the conference. If you determine you want to follow the discussion more closely, the complete transcripts are available for capture in Library 1 of the GEnie Forth RoundTable Software Libraries. Phil Koopman September 1989 Senior Scientist, Harris Semiconductor Some of the things I have found out about stack machines go against widely held (at least, outside the Forth community) ideas. For example stack machines: don't need stacks bigger than 16 to 32 elements, need not have a significant context switching time, and can cycle their clocks every bit as fast as (or perhaps faster than) RISC processors. One thing I run across continually is that folks confuse the requirements for real- time embedded control with those of workstation environments. One of my professional goals is to understand more about Forth-derived stack computers in order to help them gain acceptance in applications for which they are well suited. Stack machines seem to be superb at real-time embedded control (although I still want to do more research to quantify this notion). But, what about other application areas? If stack machines are the answer, what are the questions? .. Robert L. Smith October 1989 Research Specialist, Lockheed Palo Alto THANK YOU. For floored division, it helps to focus on the modulus or remainder rather than the quotient. Most users use only positive argu- ments, so floored or non-floored give the same results. For almost all cases that I know of, if you have at least a negative numerator, you propbably should use floord division. For Floating Point: (1) should Forth have it at all? and (2) If so, should it be in the Standard? And (3) IEEE floating point? Charles Curley November 1989 Neologist Grief, after that lead-in.. I guess I see forth as a tool, and I like my tools to be simple and easy to understand. A lathe is (conceptually) simple, and it does almost anything one needs to do. So are wrenches, hammers, etc. I don't think there is such a thing as a single tool do do everyting, so I like lot's of simple, semi- custom tools. Hence minimal Forth. That's all for a start. <[Gary] GARY-S> Charles - your original non-standard committee is something of a standard rejoiner do you still feel anti-standard? Well, since no-one ever joined it, it can't be a joiner but, yes, I do still think that standards get in the way more than they help. Note that standards are not the same as models, which can be very useful. I think Chuck said it best: standards are wonderful; everyine should have one. I think that coding standards would be more useful than a language standard, but try getting any two forth programmers to meet even miniimal standards in coding. ... I think that standards are too vague to be of any real use. They are too open to interpretation and fudges. If they are too tight, the resultant forth is useless for its natural applications, high speed & compact code stuff. If the're too lose, then you have no portability. John D. Hall December 1989 Programmer, Lockheed Palo Alto Forth Interest Group was organized in 1978 and one of the first things that was done was the formation of Forth Implementation Team lead by Bill Ragsdale to build figForth and put it in the public domain. Because fig- Forth was implemented on many microprocessors based on a single model and released with complete source listings, it became the defacto standard of Forth. The listing were made available for $15 and were extremely popular. To encourage extensions and modifications to Forth, the same people started Forth Dimensions, FORML and the Forth Standards Team. Many of these same people wrote or encouraged others to write articles for Byte and the famous Byte Forth issue was published in Oct of 1980. Money was coming in faster than new uses for it could be found and sales taxes were collected but no formal organization existed to pay the sales tax and avoid any income tax. At this time, FIG was only a loose confederation of interested Forth users that were spread around the world, so a California non-profit corporation, Forth Interest Group,Inc., was formed to centralize the contacts, establish a central distribution point for Forth information and establish the formal function of publishing Forth Dimensions and the FORML proceedings and the distribution of these and the figForth listings. No great thought was given to the other reasons for having a corporation. It was primarily to centralize the loose confederation of people and ideas and dis- tribute contributed literature. FIG was guided by a Board of Directors con- sisting of Bill Ragsdale, Kim Harris, John James, Dave Kilbridge and Dave Boulton. The corporation wasn't the answer to the problems of organization, it took people to make ideas happen. Organization ideas were and still are plentiful, people to implement the ideas were not. From the beginning business meetings were held one evening a month on the Tuesday before the 4th Saturday of the month. Before the corporation, FIG and the Silicon Valley chapter were the same and the planning for the Sat. meeting was finalized on the previous Tue. As the corporation was formed, the corporation business meeting began to take over the topics of the Tue. meeting until finally the Silicon Valley chapter planning was split off to another night. Since actions took more time than was available at the business meetings, committees were formed to plan and guide activities as existing or new functions were needed. FORML and Forth Dimensions needed guidance. FORML split into a conference and a convention. People around the world were forming group meetings and wanted to be kept informed, so a chapter committee was formed. Literature needed to be pub- lished and new books and publications needed to be reviewed. A publication committee was formed. Direct communication with the Forth community was needed, so a connection to GEnie was established and a committee of sysops was formed. All of the literature needed to be distributed and orders needed to be filled, so an outside organization was hired. Money was being received and distributed on a daily basis and this was more than the volunteer treasurer could handle. From the beginning, all of the functions of the FIG, Inc were handled by a group volunteers with the help a few paid people such as the editor of Forth Dimensions and Mountain View Press for distribution. When we saw our way clear enough, a management organization was hired to try to tie all the office functions together to establish a central office for mail distribution and phone under the direct control of FIG, Inc. We are now at a point now where FIG, Inc, includes 1) Editing, publication and Distribution of Forth Dimensions 2) Production and convening the Annual FORML Conference 3) Publication and distribution of the FORML Proceedings 4) Production and convening the Annual Forth Convention 5) Support and Sponsorship of the GEnie RoundTable 6) Support and organization of approx 50 Chapters 7) Distribution of over 90 publications 8) Production and distribution of a growing disk library 9) Membership enrollment of over 2000 members FIG, Inc is governed by a 7 member Board of Directors: Dennis Ruffer, John Hall, Terry Sutton, Mike Elola, Robert Smith, Jack Brown, Wil Baden Business is conducted by the Business Group including the above and: C.H. Ting, Jan Shepherd, Bob Barr, Marlin Ouverson, Bill Ragsdale, Robert Reiling, Tom Zimmer and others who are interested in the business of FIG The growth of FIG was probably not as chaotic as I am remembering it, but the organizing was driven by the needs that had to be filled rather than with a planned structure and the organizing was done by those of us who are technically competent in Forth but not necessarily in organizations. Was FIG the answer for Forth? Well, yes, if the question was how do we establish a central focal point for Forth. What we had done is created a "passive" organization, given it ability to reach out to the Forth community and made it financially self- supporting as long as there were enough people who were interested in Forth and willing to support FIG. What we had not done was create an "active" environment for "encouraging" those who were interested in Forth, "instructing" those who were new to Forth, "supporting the creativity" of people who were advancing Forth or even "establishing the solid technical foundation" and background for the growth of Forth use. There are probably other things the FIG is not and should be. We have the organization and it has the room for new directions. How do we now make it the organizational answer for Forth?