I think Ron's comments are outstanding. I pointed out many problems with usability in Trouble in Paradise: Problems Facing the Usability Community but I didn't give enough space to the issues Ron is talking about. In effect, I think he's politely saying that usability is broken. It is a form of black magic or voodoo that, while based on science, is not scientific. If it was, we'd be able to compare usability studies and we could expect to return reliable results.

Here's what I would like folks to consider...

1. I'd like people to consider computational usability. In effect, I wonder if we can create useful models of usability that are described by mathematical functions. I would like some usability functions where I plug in some numbers and get some results that tell me if my method is good. In other cases, I'd like math functions that allow me to actually analyze raw data, easily and effective and reliably. I'd like these functions to be relatively simple, so that they speak louder than any gurus. There is such a thing as Computational Psychology so this isn't a total fantasy.

2. I'd like to see something like a Consumer Reports of Usability. An actual print version of usability issues and resolutions, ranking of best and worst products by usability, usability versus cost analyses, and so forth. I'd love to start something like this if there was some *serious* interest in it. Further, a radio show about usability, with perhaps a humorous slant, could be fun to do once per week.

3. An organization such as the *cough* UPA *cough* ought to put together a repository of usability research, along with an analysis and synthesis of that data. We have the HCI-Bibliography but there should be an organizational effort to generate a high-level summary of useful tactics, methods, and research.

I'm kind of tired so these might be insane ideas. Still, I hope they give Ron, and others, some ideas about how to solve this problem. And yes, it is a problem and it deserves attention.

Posted by: John S. Rhodes on January 12, 2003 05:05 PM

---

 

Thanks for the feedback, John! You're jumping right into solutions, while I'm still focusing on describing the problem in a way that is well-understood...

In response to John's ideas:

1. Computational usability: This is a very long-term solution. While there has been some interesting research published, progress is very slow.

2. Consumer Reports of Usability: This requires a reliable and valid testing method. Same for the various existing initiatives to standardize and/or certify usability.

3. Repository of usability research: Usability.gov is just that. UPA has a working group that is creating a body of knowledge. It will be interesting to see what they come up with.

Posted by: Ron Zeno on January 12, 2003 06:50 PM

---

 

Interesing. See: http://www.stcsig.org/usability/resources/index.html I am currently using an internet connection in a hotel room in Shanghai, has anyone else used a connection in a hotel and what has the user experience been like?

Posted by: daniel szuc on January 12, 2003 07:04 PM

---

 

now this I like a lot.

Your comments ooze with thoughfulness and precision. This is exactly the work that need to be done so that I can repel the 'usability chameleons' with their 'finger in the air' methods, ever changing job titles and pocket full of small knives. I want to spend time working with people who actually know what they are doing, and care about doing something that they can be proud of, rather than something they can mis-represent in their next job application.

I think this clearly states that Usability Does Not Work. I would like to see Why Usability Does Not Work and Usability That Works. By next week, if you don't mind.

Posted by: Mac on January 17, 2003 02:30 PM

---