Using computers: your way is the slow way

By Lynley Hargreaves 03/12/2014 5


Professor Andy Cockburn
Professor Andy Cockburn

One of the mysteries of the modern age is why, when there are lots of ways to improve efficiency and productivity with computer-based systems, people don’t use the tools available. Sophisticated file search utilities, shortcut keys, and powerful commands are seldom used. Canterbury University’s Professor Andy Cockburn explains why most people trap themselves in ‘beginner mode’, using tried, trusted, and slow methods instead of the faster alternatives available.

Is it true that people don’t use keyboard shortcuts?

Most users know and use a very small set of control keys, such as copy (CTRL-C) and paste (CTRL-V). When considering that many people use the same tools (e.g., Microsoft Word) for several hours on every working day for years or decades, it’s surprising that they never get around to learning and using more extensive shortcut vocabularies. We’ve built tools that silently observe users working with their systems over weeks and months of use, allowing us to confirm that shortcuts are seldom used.

Why don’t people use shortcuts?

One of the key barriers to learning a faster method for doing something with a user interface is the temporary “performance dip” that comes with learning the new method – in order to become more efficient in the long term, most shortcut interface techniques will make the user slower in the short term, acting as a disincentive. For example, if the user wants to learn the “strikethrough” shortcut key they need to move their hand away from the keyboard, move the mouse cursor away from the worksurface, point to the strikethrough button, wait for a pop-up to show the shortcut, try to memorise it, move the cursor back to the worksurface, and finally return their hand to the keyboard. This is slower, once the pop-up menu is showing, than simply clicking the button. In our Marsden project we built systems that allowed users to browse shortcut commands on request without having to move their hands away from the keyboard, minimising the performance dip, and resulting in much higher use of shortcuts.

What sort of tasks do you focus on?

We focus on the most frequent actions during interaction, particularly those that form efficiency bottlenecks. This includes actions like making selections from hierarchical menus and navigating to files to open them.

What’s wrong with how people open files?

Although there are sophisticated and efficient search tools for file retrieval, the vast majority of users rely on file browsers such as the Mac’s Finder or Explorer on Windows. Retrieving files in this way takes about 15 seconds on average – working through successive layers of a hierarchy. We’ve built systems that reliably reduce this time by developing algorithms that predict which files the user will want to retrieve and by highlighting likely folders and files to help users quickly identify and select them.

Will this algorithm and system make your fortune?

We’re most interested in fundamental science – understanding barriers to efficient interaction and in finding ways to overcome them. We’ve often considering patenting our inventions, but haven’t done so yet. Instead, we rely on companies such as Microsoft, Apple, and Google reading our work and introducing our features into future versions of their systems. Industry leading companies are strongly represented at the top few conferences in the field, so technology transfer is fairly prolific.

Also, my group’s reputation is fairly strong, so big companies come to us when they need interns or recruits. For example, the three PhD students working on my Marsden project all served research internships with either Google or Microsoft, and both companies now own user interface patents invented by my students.

These interviews showcase researchers supported by the Marsden Fund which, since 1994, has been supporting fundamental, investigator-led research in New Zealand.


5 Responses to “Using computers: your way is the slow way”

  • I fully endorse Andy Cockburn’s research on this – I think its of great scientific and even greater practical value.

    Having started using PCs in the days of DOS 3.1 and CPM using MS Word, WordPerfect, Locoscript and assorted spreadsheets, I had to use keyboard as mouse input was very limited. This has given me a huge advantage now as I hardly ever use the mouse for standard Windows, Word, Excel and similar programs which I consider has given me a huge speed advantage over my mouse centric colleagues.

    At the same time, I feel that the mouse centric interface of HTML and especially the ribbon on post 2003 versions of office have achieved the exact opposite of what Andy is promoting, i.e. they have made the programs easier for beginners to pickup but make them much less effective for experienced users. I’d therefore love to see Andy have a look at these issues.

    Merf

  • Funnily enough Charles, when I was studying at UC our HCI course (Led by Andy) had a main assignment where we had to write a few thousand word report on everything wrong with the Ribbon. I’m sure he’s quite aware on that one 🙂

  • Ha – I used CP/M+, CharlesM..good ol’ amstrad days.

    re ‘Why don’t people use shortcuts?’..as a uni tech I recall showing a lecturer how to use keyboard short cuts with a program that lacked any obvious ‘cut n paste’ menu option.

    His response – ‘Im not going to use something from the dinosaur days.!!’

    One lecturer with a very complex folder filing system poking fun at another for storing all his data in one folder called ‘documents’. Later it occurred to me while waiting for him to locate a file among several hundred folders that the ‘one folder has all’ system would be quicker…and so now tend to do that still today…depending on finder or explorer to find stuff…and (usually) so much faster than I.

  • with any program that is to be ‘main use’ it would certainly pay to put time into learning those shortcuts. A user of ‘Blender’ (3D graphics) described how much faster he was being able to knock up an object seemingly extra fast with such familiarity.