The United States Patent Office maintains an archive of most all patents it has ever granted. The full text is made available for patents issued since 1975, and sophisticated search facilities are provided. For patents issued prior to 1975, only page scans are stored and searching is limited to patent numbers and the USPTO classification references; this does make historical research somewhat challenging.

 

Despite these limitations the archive is a valuable resource and the USPTO is to be congratulated oninvesting the means to create and maintain it. It is to be hoped that other patent offices will have the foresight to provide the same sort of resources.

 

Last year, after many hours searching the USPTO site, I opened a web site which lists over 600 patents related to the development of punched cards. Starting with the patents submitted by Herman Hollerith in 1894 and extending out into the 1930's. In addition to the effort involved in generating the index, I also developed a technology which provides for selecting and sorting the patents and for displaying statistics, all implemented in pure JavaScript, and thus not dependent on a web server. A word of caution regarding the statistics : it is certain that there are other related patents which have not been indexed and so any conclusions drawn from the statistics need to be treated with caution.

 

Examining the index raises some interesting questions which so far have not been researched. Looking at the statistics it is clear that IBM and its predecessor, Hollerith's Tabulating Machine Company (TMC), dominated the industry from the start. Well over half of the indexed patents are assigned to them or their associated companies in Europe. Here is an interesting point though, IBM was officially created in 1924 by the renaming of CTR (Computing Tabulating and Recording Co) itself created in 1911 by the merging of three companies including Herman Holleriths TMC. But the index shows no patent assigned to CTR. After the 1911 merger, patents continue to be assigned to TMC. Does this imply that TMC continued to be a legal entity even after the creation of CTR? Was this some part of the agreement with Hollerith? The mystery deepens in 1924 with the creation of IBM, the first patent assigned to IBM is granted in 1926, but other patents, even by the same inventor, continue to be assigned to TMC until as late as 1933, well after the death of Hollerith himself in 1929. What criteria were used to decide whether a patent would be assigned to IBM or to TMC?

 

Are these "anomalies" just administrative trivia or is there some significance to the fact that TMC apparently continued to exist until into the 1930's?.

 

Hans Pufal
22 March 2005