05/12/2021

062 – Software patents – 1981

Forward to PC Moments 063
Back to PC Moments 061 – Back to PC Moments Index
Back to Anecdotes

Inventions reached their limit long ago and I see no hope for further development.
Julius Sextus Frontinus, Roman engineer, lived from c 40-103 CE

When we consider how software like spreadsheets proliferated it is natural to ask why it wasn’t patented by the VisiCalc designers.

However in the late 1970s in the USA it was felt that software consisted merely of symbols that were matters of fact or natural occurrences that were not patentable.

The concern was that a patent would constrain mathematical formulae and algorithms and thus software was not appropriate material.

The US Patent Office advised the VisiCalc team that they had a less than 10% chance of getting a patent awarded for software.  Given the high costs involved and the poor chances of a positive outcome they decided to rely instead on copyright and trademarking.

In the UK a different view was taken and the first computer software patent was granted in August 1966 for a product described as

‘A Computer Arranged for the Automatic Solution of Linear Programming Problems’. 

It was software that managed memory efficiently in relation to the ‘simplex’ algorithm applied in linear programming.

By the late 1970s Europe had introduced its European Patent Convention which enabled software inventions to be patented; but it took until 1981 for first USA software patent to be granted. 

This was the culmination of a personal mission on the part of Satya Pal Asija, an Indian immigrant to the USA who had developed a data retrieval program named SwiftAnswer, an acronym for ‘special word-indexed full-text alpha-numeric storage with easy retrieval’.

The software was designed to allow a user without programming skills to ask in plain language for something to be retrieved.  The software could overcome problems with the user’s spelling, punctuation, grammar or syntax.

Satya Pal Asija wrote the software in 1969 and was not content to rely upon copyright alone.  He thought he deserved to be granted the same strengths as a hardware inventor.  He took the time to study law, particularly patent law, eventually taking and passing his bar exams. 

He lodged his application in 1974 and it took a seven-year process for his patent to be approved.  In 1983 Satya Pal Asija used his experiences to write a book entitled ‘How to Protect Computer Programs: A Case History of the First Pure Software Patent’.

In the preface he says,

‘Copyrights, design patents and utility patents all have their place in protecting software.  As a rule of thumb copyright protection is sufficient to protect a specific piece of code and a patent is needed to protect the underlying algorithm.  Design patents are needed to protect programs embedded in integrated circuit chips because utility negates copyright protection for chips [notwithstanding mask works].  This is so because computer programs are authored, chips are designed and algorithms are invented.’

In fact SwiftAnswer never amounted to much, but the successful patent meant that VisiCalc had missed its patent opportunity by just two years.  The USA Patent & Trademark Office was soon issuing many software patents every year.

Forward to PC Moments 063
Back to PC Moments 061 – Back to PC Moments Index
Back to Anecdotes

Leave a Reply

Your email address will not be published. Required fields are marked *