waving android

I am currently a software engineer at Google, where as a member of the Android platform team I build frameworks and user interfaces.

The blog here at is mostly historical; you can find more recent posts on .

OSS usability, one more time

July 18th, 2004


Slashdot has yet another thread today on open-source
software usability
, prompted specifically by a NewsForge article on usability as a “technical problem”.

I had a much longer critique written, but it was boring and
self-indulgent. Instead I’ll give you the highlights:

  1. Claim: OSS usability is in bad shape. This is
    something of a consensus in the OSS community already. No argument
    here.
  2. Claim: The OSS community needs to develop
    some of its own HCI expertise.
    This is an important insight and is the
    chief contribution of the article to this open problem.
  3. Claim: Most UI problems are easy to identify and fix.
    (See the “smbUmount” example in the article.) Demonstrably false.
    Most HCI issues are more subtle than “one single thought to usability”
    can uncover. See also The Inmates Are Running The Asylum
    for why even that one single thought is likely not to be
    trusted.
  4. Claim: User interface design can, with sufficient documentation,
    be reduced to a mechanical process of right and wrong.
    Also false.
    Here’s a secret: UI designers disagree with one another about usability
    as often as OSS developers. Usable software can no more be created
    by referring to some (fictitious) “Usability HOWTO” than good software
    design
    can be achieved by reading a C++ syntax manual.

Good user experience is hard to achieve, and requires an
investment of study and practice on the part of each individual who
seeks it. The members of the OSS community can (and should)
take this problem into their own hands, but not by trying to
reduce the problem to a flowchart. (If such a thing were possible, we
could replace our entire community of HCI experts by computer programs.
A chilling thought!)

newer: older: