"hard interfaces" and HCI in the long haul

A very interesting conversation occurred in the pub this evening after work, which has re-sparked my interest in blogging about HCI.

The topic of conversation focused around the image of HCI and the abandonment within its research of expert users of difficult to use interfaces. The person to whom I was chatting (an avid user of Vim and emacs) made the point that specific systems such as emacs and VIM are the antithesis of the striving for an easy to use interaction. Yet although they have steep learning curves and fiddly interfaces, a huge user community exists around these editors and some bring enthusiasm in their users bordering on evangelism.

This got me thinking about whether as an area of research we ignore these types of systems in investigating HCI in the striving for easy to use, intuitive interfaces? Why is it for instance that these types of interactions bring such positive feelings, and a positive user experience to those who take the time to learn how to interact with these interfaces? I think it is true to say that HCI limits itself unnecessarily to the exploration of new types of interaction and to the novice user of systems without exploring in depth the development of usability over time with interfaces and how the user experience grows (a point expertly made by Hornbaek in his 2006 paper on the measurement of usability).

It is unlikely to be a linear growth and not one that is the same for all types of interfaces. Indeed the emotions users experience over time, as they learn a system, become proficient and then move to expert users are likely to be varied. Anxiety, frustration, elation and satisfaction are all likely to occur dynamically throughout this interaction development and growing literacy with an interface is likely to affect the frequency of these. Further to this a lot of learning initially may reap a high amount of benefits in terms of efficiency and effectiveness. These "extreme" users of interfaces that HCI would deem low in usability are certainly satisfied with their interaction too, so these interactions meet the criteria for efficiency, effectiveness and satisfaction for their users. Is it likely to be a specific type of user, personality, interest group etc that is attracted to these interfaces? Probably but these are interesting research topics to explore.


Even though Hornbaek highlighted this problem in usability measurement, not a great deal has been done about this. This is likely because of the necessary extreme longitudinal nature of such research and need for HCI research to explore "new" interaction methods and designs rather than dedicate a huge chunk of research time to one interface. For instance it would be unappealing to research the relationship users have with an interface over 5-10 years and explore the lifecycle of interaction, but really how can this be feasible/accepted in an area that it intrinsically dynamic in its focus as new technologies and ways to interact develop? Indeed how would this be feasible with the current research climate with the desire for real research impact in the short term?

These questions were running through my head after the discussion. Why do people start using these difficult to use interfaces? Why do they still attract young users rather than only having the original users enthusing about the system? Does a reliance on an easy to use interface such as a Rich text Editor in wiki editing leave the user with a lack of knowledge to deal with more complex tasks not possible using the Rich Text Editor, therefore leading to poor usability at this tipping point?

The exploration of these difficult interfaces and the reasons why these users are so evangelical about their use is an interesting research proposition. One which I will hopefully be exploring with colleagues at Birmingham as research work, as well as in the pub again, soon.