I just sat in on another session covering research on the use of internet technologies and usability. Interesting little details:
- Error reporting systems (internet based or not) generally get reports of “wrong result” errors but rarely have any information on other types of errors: wrong behavior, wrong intention, wrong cognition, or wrong action.
- A study of blogs looking at size of posts and pages (short posts on short pages, short posts on long pages, long posts on short pages, and long posts on long pages) came to the conclusion that short pages don’t help; long posts help exploratory research; short posts help task-based research; short posts (summaries) are as good for recall as long posts.
- WikiTable is a collaborate table editing tool where many people can be making changes to the same table, only locking a cell at a time. (Done by Siemens Corporate Research).
- Spacial Navigation Ability (SVA) helps performance in hiearchical navigation regardless of strategy (the two focused on were relevance based and distance based), but SVA is not predicted by performance in these tasks.
- Consumer rating scales used on consumer websites that aggragate review scores by calculating the mean actually approach the “ideal” social choice rule of “majority”, and can even exceed it at small sample sizes, even though the mean is very easy to calculate, and majority rule is cumbersome and requires pairwise comparison of all possible choices, so is often impossible to implement anyway.
