ThinkPad T400s: Key to a Better Experience

  escdel_keyssm

I thought it would be useful to follow up on the sneak peek blog I published earlier in the week  with more information . Now that the product has officially announced, I can share a lot more of the research methodology and creative thinking that drove the changes to the keyboard. The video is  just too short and a bit vague, by design, to  share all the details. After all it was a bit of a teaser. Much of  the footage ended up on the cutting room floor due to pre-announce concerns from legal and others. My team felt there was a great opportunity to take the ThinkPad keyboard design and user experience to the next level with the introduction of the T400s. It only made sense, this was to be the best T series ever made. Achieving such a goal was not easy, however,  it required a lot of hard work, creative thinking , and most importantly information about how people used their exisiting keyboards. We thought that taking an updated look at key frequency of use would be a good place to start in order to uncover innovation opportunity. Things do change over the years. Who would ever have predicted the increased use of the @ and tab keys prior to the internet. To gather this kind of critical information we solicited agreement from several dozen internal users to install a special keyboard tracker on their ThinkPad.  The request to install a keyboard tracker on peoples'  laptops sounded a bit odd at first,  but eventually volunteers lined up once they understood exactly what we were trying to accomplish. We really didn't want to peer into their lives, we just wanted frequency of use data. After an extended period of time the data was translated into what we call a "heat map". The more frequently used the key, the more red we used to color it. This visual mapping technique quickly revealed patterns that suggested design changes.

editing-cluster_heat-map

Portion of the heat map based on collected data

The QWERTY section of the keyboard is highly controlled for obvious touch typing reasons. Outside of the QWERTY section,  some frequently used keys have larger than normal key tops. Enter, backspace, tab, caps lock, shift, control, and of course the space bar fall into this category. Surprisingly enough this idea has never made it past the 5th row. Frequency of use is not the only driver of increased key size. The style of gesture or stroke used for the function also comes into play. Enter falls into the category of a clear and deliberate action, a sort of salute to completion. There is a certain fervor of emotion connected to it's use. Sounds very similar to the delete key. Who doesn't enjoy hammering that one home when the boss sends you some ridiculous note that requires no action. When looking at frequency of use and gesture the delete key became an obvious candidate for increased size.  The upward trajectory of  movement towards the delete and escape keys also suggested making them taller rather than wider.  For example, the enter key is wider for a similar reason,  but with a more lateral trajectory. We didn't just rely on the frequency of use data collection technique. We also cross checked the data regarding  frequency of key use against the occurance of letters in language. For instance e is the most frequently occuring letter in language and it is the  most commonly used alphanumeric key. For those of you who are curious, space bar is the overall frequency of use winner by a wide margin. We also did visual inspections of numerous used ThinkPad keyboards to determine wear patterns. If you want  to check your own system the glossier the key the more frequently it has been used. The tilde key on my system has no visible wear.  Additionally we analyzed responses to a survey  that I posted on this topic that allowed us to get even richer data, including respondant comments. Thanks to all who participated. In the end we decided to change the keyboard for what believe is the better. We increased the size of the delete and escape keys to make them more comfortable to use and mapped the shape to the vertical trajectory of the stroke. Okay we had to move the insert key to make way for a larger delete key, but our data indicated it was far less frequently used. We tested early hardware with real people  and have used prototype machines with the keyboard modifications for over a year. As a side benefit we also think that we improved  accidental striking of the F1 key by moving it to the right. Nice fix to a known reported issue.  This has been quite a journey for my team, but I think it was worth it. I hope you've enjoyed reading  about the detailed thinking that went into these changes, and more importantly,  like the resulting experience we created.

David Hill