A usability story

Almost 10 years ago, I was talking to a large media company about how to authenticate subscribers who had paid for digital content. Being focused on cryptography at the time, my idea was to use a USB token that held a digital certificate and the corresponding private key. The user experience was to be like this: when the token was present, a user could get the paid content but when the token wasn’t present they couldn’t. Behind the scenes, the token was going to authenticate to a content server using the its certificate and private key, but all of the cryptographic details were going to be hidden from the user.

The people from the media company thought that this sounded interesting and thought that this could even be the basis for a trendy and cool product. Teenagers might wear USB tokens as a fashion accessory to show that they were cool enough to be subscribers, for example. They would make lots of money. I would make lots of money. Everyone would be happy.

The media company then proceeded to do a usability study. They were planning on charging roughly $20 a month for the digital content, and by their estimation, a support call from a customer once every three to four months would make that particular customer unprofitable for them. Because of this, anything that they added to their system had to be extremely simple and easy to use.

The USB tokens failed miserably. Depending on how they asked them, between 25 and 50 percent of users could actually find the USB port on their computer. This made the solution that I had proposed very impractical, and the huge multi-million-dollar deal that I was dreaming of never materialized. My kids never got the Sony Aibo robotic dog I was planning on getting for them and I never got the the beach-front condo that I was hoping for, but I learned a bit about usability in the process.

Users have become more familiar with their USB ports in the past 10 years, and if you did a similar usability study today, the results would almost certainly be very different, but at the time, what I proposed turned out to be impractical. What new technologies are this way today – seemingly easy, but not quite easy enough for the average user? I would probably be as surprised at the answer today as I was 10 years ago.