Michael Markov P.O. Box 17 Lockwood, New York 14859 U.S.A. Dear Friend, The demands on my time, both at home and at work, are increasing. This leaves me with less time for other activities. I therefore have been looking for ways to increase the efficiency of my correspondance handling -- the single most time intensive one. It is also the area most likely to yield results, because everybody in the handheld user community shares the same interests. Writing a a single reply to several correspondants will, I hope, allow me to take care of my mail without undue delays. There will be disadvantages, especially when I repeat old information. I also dislike the loss of the "personal touch". I hope the gentle readers of this epistle will bear with me, and keep in mind that I am trying to respond to other friends, as well as you. I should mention that it looks like that the union employees that work for our company could go on strike. If this happens, we expect the strike to last at least a month. During that period, I expect to be out of communication. Therefore, I am sending this letter out as is, somewhat unfinished. I will do my best to catch-up on any unfinished business on the next go-around. Private correspondance: to the extent that we need to discuss "private" information, I will enclose separate letters. Confidentiality will be kept. I am taking the opportunity to answer your last several letters, all at once. I appologize for taking so long to answer - I have been incredibly busy, both at work and at home in my so-called spare time... At work, we are preparing our budget for the coming year, which means we have to justify every project we are working on. This is very much like taking examinations at the end of the school year! At home - well, I am evaluating the CMT RAMDISK, managing the swap program as best I can, and trying to finish a dozen programming projects. This list does not include other jobs like helping dig-up a vegetable garden for my mother, going to the annual Chicago Conference... So, your requests for RIOWIO, SCOPE75, BASICLEX: yes, I can supply you with copies. See the enclosed disk(s). Please keep in mind that RIOWIO is an HP LEX file, and as such should not be distributed as part of the official swapdisk program. HEX listings were provided a while back in the CHHU Chronicle... SCOPE75 should be on the swapdisk CHHU03 or CHHU04. The X prefix is not likely to mean that this is a new version. Rather, it indicates that this is a duplicate file name being modified by BCOPY. I vaguely recall the problem as being associated with the CHHU05 disk. Anyways, it is the result of having too little time available for veryfying the contents of other folks contributions! This brings us to BASICLEX, which is a version of SPLEX by Chris Bunsen (HP). This LEX is part of the SPREAD71 program, a third-party software item, V7524-71-9. BASICLEX is available from Joe Horn's library of LEX files (see enclosed copy). However, BASICLEX is a flawed version, as it has the bugs reported by J.H., while SPLEX works well and seems bug-free. I strongly suspect that somebody disassembled SPLEX incorrectly... A LEX for the Quietjet??? Sorry, no such thing exists at this time. However, I could very easily modify John Baker's ESCAPELX to output any escape sequences to the current PRINTER IS device. I am willing to do it, provided you supply me with a list of the control / escape sequences. I suspect that ESCAPELX already implements the most common sequences, and that all we need to do is to add a few more. By the way - early versions of ESCAPELX were specifically designed to look for a device with the AID of the HP-IL thinkjet. Later versions, including the printer commands incorporated into PILEXTL4 were modified to send the desired escape sequences to any PRINTER IS device.. have you tried them? Now, SWAPGB : I have received the new, complete copy, thank you! SWAP program : Thank you for volunteering to help with distribution in Europe. I will add your name to a list of people who have volunteered to help. At this time, the list consists of: USA - Michael Markov P.O. Box 17 Lockwood, New York 14859 Mitch Hoselton P.O. Box 36047 Dallas, Texas 75235 ===>> Mitch has a Steinmetz & Brown dual drive. I recently talked to Bill Rudesdorf, of Houston, Texas, who has agreed to copy the material from 3.5" disks to 5.25" disks, so this would make it possible to keep Mitch up to date. Bill is also talking about making copies of swapdisks available in the format used by Southern Link. This would certainly help! Bill's address is : Bill Rudersdorf Southern Software 215 Hawthorne Houston, Texas 77006 USA ENGLAND : Mark Cracknell (HPCC) 6 Stratton Mews Off South Street Leighton Buzzard, Bedfordshire LU7 8NT ENGLAND Netherlands : Jan G. Buitenhuis Lage Duin 44 NL-2121 CH Bennebroek The Netherlands Several other people have more or less complete collections of swapdisks, and I am sure they would be happy to help make copies at local Chapter Meetings, provided that prior arrangments are made to bring the needed equipment : Stefano Tendon, Cantone Delle Asse 5, 29100 Piacenza, Italy Stefano Piccardi, via Antonio Panizzi 13, 20146 Milano, Italy (Stefano Piccardi is now doing his military service, and is out of touch) Janick Tallandier, 335 Rue Lecourbe, 75015 Paris, France Andrew Wallace, 10/247 Oberon St., Coogee, N.S.W. 2034 Australia (SydCHHU) Graham Fraser, 25 Octantis Street, Doncaster East, Victoria 3109, Australia Nick Reid, new address not available, (SydCHHU) Al Duba, 9925 S. Industrial Drive, Bridgeview, Illinois 60455 USA (Chicago UG) Sam Chau, 1200 Murchinson Dr., Millbrae, CA 94030 USA (BayCHHU) Tapani Tarvainen, Katsastajakatu 17 A 10, SF-15800 Lathi, Finland I should be very pleased if you gave the swap project some publicity in DATAFILE, and made people aware of the details of what is available, whom to contact, ect. See the enclosed material, files SWAPSYS & SWAPHOW2. A number of people have asked what to do when a swapdisk program is telling you that a ROM or LEX is missing: First, look for comments in the program that tell you what files are needed, and copy them to RAM. This should solve the problem, but not always. If you still have problems, look at the files LEXLOGSWAP and SWPLEXLOG on the CHHU06 swapdisk. These files should allow you to correlate the XFN numbers with the name of the LEX files that use the resources in question. Then, find the file on the swapdisk (assuming it is not a commercial product), or go find it on Joe Horn's LEX collection. This last disk is NOT an official swapdisk, but copies can usually be obtained from the local chapter custodian of swapdisks and other goodies. I would expect that an updated LEXLOG will soon be available. Finally - if you are still stuck, you can always contact me, preferably by mail -- looking stuff up in my databases takes time. CHHU ROM project: I have been urging people to support the Paris Chapter's JPCLEX as the most viable alternative to CHHUROM. Already, several excellent code writers such as John Baker, Jan G. Buitenhuis & Tapani Tarvainen agreed to let the Paris Chapter use their code. Group contributions (See SYDMELEX, from the Sydney, Australia Chapter) could also make a significant impact... This should help make the following versions of JPCLEX even more interesting! Yes, I do mean that the work on JPCLEX will continue, and that a new version will be coming out in another year or so, depending on the amount of new material contributed. As of May, 1987, I have in my hands JPCLEX version :B00, at just about 16K bytes. Version :C00 is in the final stages, with all resources allocated & approved by HP. This version will be released almost immediatelly. The French version of the manual (over 220 pages) is ready, and an English version is well on the way. This will provide some 100 new keywords and numerous handy poll handlers. See the enclosed order form. Typically, for 500.00 FF, they will burn JPCLEX onto your EPROM, and provide you with a copy of the manual. For me, this is by far the best option, because I already own 64K EPROMs.. and, of course, there are other options. This item also deserves good publicity, I would appreciate anything you can do... Incidentally -- in a recent phone conversation with Janick Tallandier, I was told that the Paris Chapter is eager to publish an English version of their journal. This version would include translations of the material found in the French version, to the extent that they manage to find time for such translations... (Anybody care to help? ). In addition, the English version of JPC would include any material submited to them in English. Production facilities are all in place, it is just a question of getting enough material in English. In a related area -- Graham Fraser, who is the editor of Technical Notes (Sydney & Melbourne Chapters in Australia) recently wrote to me and asked me to let Brian Walsh (now the editor of HPX) that he is free to use any of the material that appears in Technical Notes. At more or less the same time, Nick Reid (Sydney Chapter) sent me material to be forwarded to the Paris Chapter for inclusion into JPCLEX -- some 60K of source code for the HP-71! This spirit of cooperation is a very pleasant development. (In truth, I am thrilled!). It sure would be wonderful if all the surviving PPC/CHHU chapters would extend similar courtesy arrangments all around. In order to further this spirit of cooperation, I hereby authorize all active user groups to use any of the material I, Michael Markov, have authored that may be found in past issues of PPCJ, the CHHU Chronicle, or in any related publications, or on existing swapdisks. It should however be understood that I retain copyrights, and all commercial rights. I hope this will help these groups survive the rather lean times I see ahead of us. Insofar as the HP71 is concerned, I am told that version 2CCCC will be soon coming out, as soon as the existing supply of version 1BBBB is exhausted. This system is different enough to warrant a new set of IDS documments (now available) :0071-90104 for the new volume II, and 00071-90105 for volume III, at $50 and $200, respectively. All the old entry points are still supported, to insure upwards compatibility. Also, I am told that there have been a number of new supported entry points. There is also talk of two new CPU opcodes. This version is fully debugged, although you will find that the PRINT statment works as described in the Series 80 manual, rather than as described in the HP71 manual.. I am not sure what this means. You may be interested to hear that, now that the 2CCCC version has been completed, the HP71 design team has been reassigned to other projects, except for Don Ouchida, who will continue to handle HP-71 resource managment. Apparently, HP has produced all the HP-71 ROMs it intends to produce. I am afraid this probably means that we can no longer expect any significant HP support for the HP-71. So, in 2 to 3 years, we will probably be told that the HP-71 has been discontinued... Will the HP-71 be replaced by a new machine? Well, there are rumors, some of which involve third party sources. To make a long story short, there always is something being worked on. However, the market for such machines is being taken over by IBM clones, so that there is considerable doubt they will ever go into production. So, the next machines? Let's consider a variety of "wants" that have been expressed in DATAFILE & other sources. All I'll say is that design teams are painfully aware of the competition and of economic factors. Therefore, wishes WILL be fullfilled -- if it can be done at a profit for the manufacturer. I agree, any new HP (or other) handheld will have to provide at least a megabyte of RAM (expandable to 10 megabytes) in order to compete with the IBM clones that are flooding the market -- and that, at a competitive price. Also, the new machines will have to provide faster I/O, perhaps using optical coupling to implement HP-IB or MS-DOS for a 16 bit processor. This has become an industry standard, and I would expect that manufacturers will have to accept the fact that people need to transfer data between their personal equipment and the office IBM compatible system... I would expect that someone will sooner or later come out with a machine that meets the above specifications. Interesting variations would include DOS systems that, like the HP110, would be able to use HP-IL peripherals, assuming that they would also provide RS-232 & HP-IB or equivalent. You ask why MCOPY71 is so much faster than the MCODE versions for the HP75... First, let me point out that the HP75 MCOPY goes through the directory to make sure that there are no illegal (i.e., PRIVATE) files on the disk you want duplicated. Thereafter, it copies one sector of the source disk into RAM, and then it sends it out to be stored onto the destination disks, with some overhead for checking the status of each drive involved in the copying process. MCOPY71 operates on a totally different basis. It does not check status, except at the very start... Effectively, I have implemented a direct data transfer from the source drive to all the slaves at once. The HP71 could in fact be removed from the loop, since the only thing it does is retransmit the frames it receives. You can use it to do arithmetic or to run programs, provided that nothing is done to make the HP-71 take control of the loop. Thus, MCODE is totally irrelevant: no proccessing = no advantage in speeding up proccessing. In fact, my status checking loop could be slowing things down.. I will try replacing it with a WAIT 10 @ -- we might be pleasantly surprised. This brings us the Mass Memory software I wrote a while back. I stopped selling the product over a year ago, so that I am inclined to put the package in the public domain, for the reasons you mention (EXTIL ROM, ect.). So, yes, feel free to publish the stuff. I am adding the programs and the documentation files to SWAP07, enclosed. Enjoy & feel free to publish this stuff in DATAFILE. Educalc: A new catalog is now in the final stages of preparation... You will find quite a lot of new material for HP-71 and HP-41 owners. The new products that caught my eye included A) a barcode ROM for the HP-41, by Sqwid. It is supposed to be very fast, very accurate -- and already obsolete. A new LaserJet version is supposed to be able to print up to a page a minute by using a special set of character fonts... This is the way to go, assuming you have a LaserJet and if you are in a hurry. As far as I am concerned, the $200.00 price tag is ridiculous, considering the availability of barcode programs for the HP-41, HP-75 & HP-71 machines. B) GRAPH71 @$59.95 -- up to 19 simultaneous curves can be plotted at once using an HP71 /ThinkJet combination. This package sound VERY nice for typical engineering / business applications. It will very nicely complement another product I have heard of that will implement HP Graphic Language for a similar environment.. C) The PacScreen II, to be finally made available in the US, at about $6??.?? and, finally, D) Structured Programming Utilities LEX file for the HP-71, $44.95, by an HP employee. This file is suspiciously similar to work that has been carried out in France by Janick Tallandier & Pierre David. However, I have learned that this is a case of parallel evolution. The version by our French friends is somewhat slower, while the other increases the size of BASIC programs -- a matter of how return addresses are managed, ie, on the GOSUB stack vs. compiled in the BASIC code itself. Oh, well, I will have to try both and find out the (dis)advantages of each, and comaptibility problems of having both in my HP-71. Finally, the EQ solver 1 (#71-821, $39.95) by C. Bunsen also looks very interesting. See the catalog for further details. Now, for some HP28C news: Memory expansion will be provided by Educalc / SOS (David White). See the enclosed flyer. The price is rather steep, but working with hybrid/surface-mounted components is a job for the specialists. If you are experienced, "how-to" instructions are available via Richard Nelson... Your HP28C could have up to 64K of RAM. However, you should be aware that the HP28C editor is very slow when editing large objects, and that the presence of many objects all at once will slow you down further. That much investment into a machine without I/O does not seem justified! Perhaps we could copy the 28C operating system & modify it to work on an HP-71 2CCCC machine? (remember the two extra opcodes. The 2CCCC CPU is the one used by the 28C). For people who would like to learn more about the HP28C, Jake Schwartz is selling a videotape of the talk given by W.C. Wickes at Philadelphia. This tape gives you about 4 hours of very interesting "how-to" information by the master. If you want a copy, write to Jake at: 135 Saxby Terrace, Cherry Hill, New Jersey 08003 (USA). Now, some more about Richard Nelson and the end of CHHU. Most of this material was part of a talk give by Richard himself at the annual Chicago User Group meeting, the 29th of May, 1987. This entire talk was recorded on video cassette by Paul Hubbert, 2401 W Fairview Ave, Mc Henry, IL 60050. You can get copies from him. To summarize, Richard was faced with quite a few problems, to include declining membership, insufficient income, and lack of help. On top of that, he had to move to a new house with all the club property in his possesion -- and then the IRS (income tax) advised him that an error of $20000.00 had been made in the previous year and that he owed an additional $6000.00 to the government. This was the proverbial "straw" that broke the camel's back. There was no other choice, he had to get a job with regular pay that would allow him to clear his debts. So, Richard is now working for Educalc and you can reach him there during the day. Richard continues to be the Coordinator of the Orange County Chapter. He will continue to participate in Club affairs as best he can. However, we must now accept the fact that the time remaining available for such activities is very limited. One of the services he plans to provide is the telephone bulletin board, tel. #714-472-9580. Oh, before I forget, his new home phone is 714-472-9475. Incompability of HP/IBM 3.5" disks: If you try to exchange file between IBM machines and HP DOS machines such as the Portables via the HP9114, you will find that you get an error message "Non-DOS disk..". John Baker has identified the problem, and provides a temporary solution: " ... Basically, the problem is that HP logically uses zero-base numbering to describe the structure of the disk, to specify the #surfaces, #tracks/surface, #sectors/track & #bytes/sector. IBM (Immense Bowel Movement), instead of following the precedent set by HP, elected to use one-base numbering. Thus, when you try to read a disk formatted with HP equipment using IBM equipment, your double sided disk is believed to be one sided, and IBM formatted disks appear to HP machines to have 3 surfaces... To correct this problem, we need to implement an improved device driver that will read the boot sector and look for either "HP110" or "IBM" and respond accordingly.... For the time being, if you use Retry 4 times, you will override the error, and should be able to copy files back and forth.... Also, if the IBM formatted disk is full, you may not be able to read some of the files with the HP-9114, because the 9114 operating system assumes that the three innermost tracks are system tracks (2 spares and the actual system use track where HP systems keep track of media wear, ect.). This last problem arises because IBM allows you to access all 80 tracks per surface instead of only 77... In order to gain access to these tracks with an HP-9114, we will have to modify the operating system EPROM...". Needless to say, part of the problem is that you will almost certainly update the media-wear count, and what this does to any file stored on such a track could be hazardous to the health. A new approach to Memory Lost? One last note: the Portable plus reads IBM formatted disks, if somewhat slower than HP format disks... ============================================================================== Well, here I must quit for now, else this will leave here in a few months after the strike is settled. So, with my best wishes to you all, Mike Markov