Very interesting, thanks for sharing. Jeff_Birt -----Original Message----- From: saipan59 <saipan59@q.com> Sent: Thursday, January 3, 2019 10:26 AM To: cbm-hackers@musoftware.de Subject: WordCheck app, circa 1980 Hi, I was looking through some old stuff, and it occurred to me that folks might possibly be interested in this. I am the author of the "WordCheck" app for PET/CBM, circa 1980. Here's the story: At the beginning of Junior year in college (1979-80) I bought my first computer, a Commodore PET, which ran at 1 Mhz, had 8KB of RAM, and stored software on cassette tapes. I bought it used for $400 from Tom Callen, who ran a shop called Micro Computer Industries in Ft. Collins, Colorado. I got to know him and his family well, and later bought other computer gear from him, including a KIM-1 board that I used as part of my Senior Design Project (I still have the KIM-1). Tom did not have a good technical background, rather he was a marketer/salesman, and he had a lot of ideas for creating and selling software apps to his customers. Since I had some ‘formal’ software expertise, I helped him with various projects. The biggest project for me was born when Tom and I talked about how he was such a lousy speller, and he used a certain word processing program called ‘WordPro’, and wouldn’t it be great if the computer could help check his spelling? So it became my project in 1980. I decided that the program would need a ‘dictionary’ to check words against, because there is no adequate algorithm that could be used (many exceptions and variations to spelling rules, etc.). The next problem was that a useful dictionary would need thousands of words, and there would not be enough memory in the computer. But I realized that most spellings can be resolved by looking at the ‘root word’, then checking for various suffixes and prefixes. For example, ‘spell’, ‘spelled’, ‘spelling’, and ‘misspell’ all share the same root. The dictionary needs to contain only ‘spell’, and the possible pre/postfixes of ‘mis’, ‘ed’ and ‘ing’ are assumed to be correct. Other known suffixes included ‘s’, ‘es’, ‘ly’, etc. It’s not a perfect system, but it works in a large number of cases. The program would highlight unverified words on-screen, and allow the user to add that unknown word to its dictionary. So, a business user would have their clients’ personal names in the dictionary, so they would not be called out as errors. In the end, I created a standard dictionary of 2100 roots. Tom marketed the program under the name of ‘WordCheck’, and sold a number of copies of it for $200 each. Tom was concerned about software piracy, so a common practice then was to include an EEPROM with the program, which must be installed in the computer to allow the program to work. So, I wrote some pieces in Assembly, and we burned it into 2716 EEPROMs. As far as we knew, WordCheck was the very first spell-check program ever sold for microcomputers! We probably should have patented it… Around the time I graduated, my total share of the WordCheck royalties was about $1400. That felt like quite a lot of money, and I was very proud of the accomplishment! Pete -- Sent from: http://cbm-hackers.2304266.n4.nabble.com/Received on 2019-01-04 01:00:04
Archive generated by hypermail 2.2.0.