4
In 1969, Frank Heart and his team at BBN were awarded a contract to build the first Interface Message Processors (IMPs) for the ARPANET, a project funded by the Defense Advanced Research Projects Agency (ARPA) to create a packet-switching network. The team had already put in a lot of work to generate a proposal, but now they were faced with the challenge of actually building the IMPs.
There were many skeptics who didn’t believe that a packet-switching network would work. They argued that while building the hardware was not the hard part, making it all work together was the real challenge. They also questioned the profitability of such a network, as there was no mass market for computing like there was for television or telephones.
Before winning the bid, BBN had doubted whether ARPA would trust such a small company with the project. There was a lot riding on the success of the IMPs. If they didn’t work, the concept of networking and packet-switching would be deemed a failure. Some competitors expressed astonishment that BBN had won the contract, as they didn’t think the company was capable of handling such a large systems job.
Despite the skepticism, Frank Heart remained focused and ignored the detractors. However, he did occasionally worry about the technical challenges that lay ahead. One of the main challenges was ensuring that data packets were sent reliably, despite errors introduced during transmission over phone lines. Computers are much less tolerant of noise than human ears, so the IMPs would have to be able to compensate for these errors.
Another challenge was dealing with circuit outages, especially if the network expanded coast to coast as ARPA intended. Bad weather or other interruptions could knock out service on long-distance phone lines carrying network data traffic. Heart’s team had to anticipate and deal with these issues.
Additionally, there were complex routing problems to be resolved, to ensure that messages didn’t circulate endlessly in the network without reaching their intended destination. The team also had to consider the possibility of jam-ups in the memory buffers, as messages were broken into smaller packets for transmission.
The system had to deliver packets and messages within half a second, as specified by ARPA. This required processing data at speeds of around 100 messages per second, which was possible but challenging to synchronize.
In addition to the technical challenges, there was pressure to meet the project’s fast-track schedule. The schedule was tied to budget and political considerations, and BBN had only eight months to demonstrate that the network concept could work. Heart knew that compromises would be necessary to meet the deadlines, but there was always tension between his perfectionism and the need to deliver on time.
BBN faced numerous other issues that had caused other bidders to drop out of consideration. Now, all these problems were handed over to Frank Heart and his team, as they worked towards building the first IMPs and proving the feasibility of a computer network.
Words: 496