(This article was first published in the ICPUG
Journal SEPT/OCT 1988 issue.
Permission from Joe Griffin to republish on the Internet has been
As promised last time, here is an outline of some of the packages available for the PET.
PET Program Packages
When the original 8K PET came out it did not take long for a number of entrepreneurs to realise that, while the machine made an excellent hobby computer, there was tremendous potential if it could be turned into a business tool. A set of programs was required which the user could operate without much knowledge of computers. (Commodore eventually got round to this way of thinking with the 8296 which was packaged with Calc-Result, Superscript-II and The Manager). Ideally, it should be possible to just switch the computer
on and start using it. In fact, this idea was not implemented until the advent of the cartridge ports on the Vic & C-64. However, it was at least possible to have a program which loaded from tape when the user held down the < SHIFT > key and pressed <RUN /STOP>. Once the program was up and running all dialogue could be in English, (or French, German etc.), rather than 'computerese' and no in-depth knowledge of programming was required.
With the arrival of the disk drive and printer the scope for packages widened. Previously, it had been desirable to hold the entire program in memory at one time, because of the speed of tape. With the disk drive, loading speed was substantially increased and it became possible to use program overlays. By this means a very large program could be used without sacrificing
data storage. As programs became available, the uptake by business users grew and the market expanded, providing room for even more applications. However, as the decline set in and Commodore was displaced in the business market, so the number of packages dwindled. Now there are few available new, though many expensive programs can be picked up quite cheap secondhand
The one type of package which probably did most to place Commodore and Apple machines into businesses was the electronic spreadsheet - specifically VISICALC. As Brian Grainger has started a series on spreadsheets, I shall not describe them here, but refer the reader to Page 266 of Volume 10 Issue 3.
Visicalc was originally made available for both 40 and 80 column machines. The program was 'protected' with an EPROM chip which plugged into one of the spare sockets in the PET. I recently saw an interesting tale in an old copy of 'The Transactor', concerning Visicalc.
Apparently, an enterprising hacker removed the need for the protection code within the program, (rumour has it that the 'code' was merely re-located ROM routines). At the same time he modified the program to allow use of all the memory in the 8096 machines. As Visicorp had not developed the program to include this capability, he offered it to them tree of charge! They declined and instead it spread by piracy. A nice story, but I guess it doesn't hold much water as in fact in 1982 we bought a copy of Visicalc
for work and it had both 8032 and 8096 versions, both of which were protected by the EPROM.
Another spreadsheet, still available, is Busicalc, marketed by Supersoft. At the last count this was available for the 8096 as well as other PETs and the 64.
One late arrival, no longer with us alas, was Handic's Calc Result. This was originally available on the 8032/96, using a security EPROM, though with the advent of the 8296 the protection changed to a key disk. For some reason this system only works on an 8250 drive. It fails to load on our 8050s at work but is perfectly happy with my own 8096/8250 setup. Calc-Result is far more 'User Friendly' than Visicalc and contains a number of rudimentary
Help Screens which are called from disk as required. Extensive use is made of program overlays. Calc-result was also available for both the 64 and the 700 series.
There has been very little Public Domain software in this area for the PET. I have seen only two programs, both rather limited in scope.
By contrast there have been many text editors and wordprocessors developed within the public area. PET library disk B2 contains a number of fledgling programs with varying features. Indeed, around the time I joined ICPUG, I wrote my own text processing system, comprising separate text editor and output programs, eventually made redundant by the purchase of a commercial package.
In the early days there were two 'heavyweight' contenders and a number of 'lightweights'. The latter included Papermate (or Simply Write) - a nice package, not as slick as some later offerings but quite fast, and written in BASIC, so it could be tailored to individual needs. My manual includes a list of subroutines and variables used by the program, with a brief description of their function.
The two heavyweights were WordPro and Wordcraft. Both evolved through a series of versions, adding features to keep pace with the machines. Both were very expensive. It always seems to be the case that a product marketed to 'business users' can carry a far higher price tag that the same product sold to hobbyists. In 1982 both WordPro and Wordcraft sold for about £400.
At that time Simon Tranmer produced his own wordprocessor which was Beta-tested by friends at ICPUG-SE. I believe it was Kevin Viney who coined the name 'Superscript'. The SE group began to sell the program for the unheard of price of £35. The heavyweights at WordPro took a bit of an exception to this as Superscript was superficially a similar program to theirs, (Simon had added compatability to allow users of WordPro to upgrade
easily), but sold for a fraction of the price. Legal action was threatened against a number of ICPUG-SE members and at the end of the day Precision Software took over distribution of Superscript, their legal counsel Alistair Kelman became a good friend of ICPUG, and WordPro retired hurt. Subsequently, Commodore adopted Superscript as the official wordprocessor for the 700 range and commissioned a version for the C64. The latter version was marketed
as 'EASYSCRIPT', and became a world leader. The original Superscript came on a protected disk, (WordPro used an EPROM, Wordcraft a dongle), which contained versions for 3032, 4032, Fat40 and 8032 all, with similar commands and features.
Superscript-ll was a later version for the 8096 and 700, followed by SS-3 for the C64 and C-128. Script Plus for the Plus4 was from the same stable.
The third major business use of micros is for data storage and retrieval. Here too there have been a number of runners over the years. One of the first commercial packages was Compsoft's DMS (Database Management System, I guess). The story goes that Nick Horgan wrote a database system, in BASIC, on a DEC computer as part of his final year project at university. When the PET became available, he typed the program in from the DEC listings. I am told that the program included statements not allowed in PET BASIC. However, it worked pretty well, and being in BASIC, the user could tweak it to his own needs. Mick Ryan, one of our former Chairmen, was a great fan of DMS. Around 1985, Compsoft released a compiled version called DMS Diamond. This was faster and more robust but could no longer be modified.
As I recall, the major competitor to DMS was Commodore's own The Manager, written in Canada, specific to the 8000 machines. The program suite contained around 20 separate programs, many in BASIC. It was possible to interface with the data, which was stored in relative files, using one's own routines. For a long time I used The Manager for a project control system at work, using the routines of The Manager for data management and my own programs
for retrieval and reporting. The biggest criticism I have of The Manager is the poor retrieval facilities. It is only possible to search for a field beginning with a particular string. Hence system design was constrained by the need to ensure that fields were set up to allow access for searches.
By comparison, the last arrival in this field was a joy to use for retrieval. I do, of course, refer to Superbase, again from the Precision stable (this one was only available on 8096 and 700). As with Superscript, it was written in machine code and supplied on a protected disk, so there was no way it could be modified by the user. However, Simon, together with Tom Cranstoun,
had given Superbase most of the facilities one could want in a database. I upgraded my project control system to Superbase and using the built-in programming language I was able to do the entire job within the data base, without external programs. In particular the search functions are delightful; in contrast to The Manager's 'find a record with a field beginning with...', Superbase allows you to find a record containing a text string (anywhere in a number of consecutive fields) or where a field does not contain a particular string and various combinations using 'and', 'or' and 'not'.
My own personal 'bitch' about Superbase concerns the programming language. I am essentially a lazy programmer, so I use keyword abbreviations (eg: lI for 'list', gO for 'goto', etc.). Now everyone knows that nE in BASIC is 'next', so when I was developing a very large program and wanted to clear several lines of the screen, I used a for...next loop, except I abbreviated
'next' to 'nE'. That's when I found out that in Superbase 'nE' is 'new' - I hadn't saved the program I had been working on for 2 hours! After switching the computer off and going for a long walk, I investigated the order of keywords and found one called 'OLD' - it's not implemented!
Next time I will be delving into the subject of toolkits.