While at MIT, Hollerith made what he would later call his “first crude experiments” on the census machine. Like the player-piano roll, his first approach involved punching holes in a long strip of paper, in this case with one row for each person.
Two incidents contributed to Hollerith's solution: conversations with Census Bureau colleague, Dr. John Shaw Billings, about count mechanization and the Jacquard loom card system, and observations of a railroad conductor punching riders' tickets for identification purposes.
It seems likely that he was also influenced by an automated
loom invented by the Frenchman, Joseph Marie Jacquard, about
eighty years earlier. Jacquard looms in Hollerith's time routinely
produced very intricate patterns guided by a sequence of thousands of punched cards, each with holes punched so as to specify
one step of the process. If not from other sources, he would surely
have learned about the Jacquard loom from a brother-in-law who
was involved in the silk-weaving business
He graduated from the Mowry and Goff School in Providence in 1891 and from the Massachusetts Institute of Technology in 1895. After graduation, he was employed by the Textile-Finishing Machinery Company of Providence
In 1882, Hollerith joined the Massachusetts Institute of Technology faculty, where he taught mechanical engineering for only one year; apparently, the thought of having to repeat teaching the same material the following year did not appeal to him. He was continuing to think about automating the census and first experimented with punched paper tape. The impossibility of reordering the information on the tape turned his thoughts to punched cards and solidified his ideas. He returned to Washington, DC and worked for a year as a patent examiner, gaining invaluable knowledge of the intricacies of that arcane art. He quickly applied this knowledge and filed a patent in 1884 for his method of automating the processing of census data.
When asked how he first had the idea of a census machine he would reply "chicken
salad" and then explain. When he first came to Washington he joined a boat club and
often rowed on the Potomac River. The club once had an evening entertainment at
which one of Dr. Billings' daughters, seeing father enjoying the chicken salad, invited
him to come to supper with them to have some of her mother's salad. It was at this
supper that Dr. Billings suggested the need for a machine to do the purely mechanical
work of tabulating population and similar statistics. Father talked the matter over
with Dr. Billings, who suggested using cards with the description of the individual
shown by notches in the edge of the card and a device something like a type-distributing
machine. Father went to Mr. Leland, who was in charge of the Population Division of
the Census Office, and asked to be taken on as a clerk to learn the nature of the job.
After studying the problem, father told Dr. Billings "to go in with him," but he was not
interested; his only wish was to see the problem solved