They believe that Barreau and Nardi draw the wrong conclusion from their study.
First because the number of users were exiguous.
Second because the preferences expressed by the users are contingent to the tools in use. For example they may prefer location based search instead of text search because the search tool is not adequate, not because they really prefer location. The common desktop metaphor favors certain types of interaction v.s. others.
The analyze all three claims with which they disagree:
Preference for Location-based Search
It is possible that subjects preferred location based search because it was the lesser than two evils. Better searches exist that may sway the results.
They claim that location search is a hit-or-miss as a logical search mechanism. It has many problems. Scalability, change over time, not applicable to archived things. So if archived is not pursued, then search will be less asked for.
They claim that since location based browse has many faults users should not use it. Do the users always use the best way?
Reminding is an important capability that software system should support. But the authors find that the use of location based storage as a way to remind is an unsatisfying , easily undermined method of creating reminders. They see the use of this strategies as a strategy used for lack f anything better.
Again, that's what the users do.
The few users studied did not archive: this may be a characteristic of those subjects.
Moreover since the software today makes it difficult to archive, this may be the reason why users don't do it.
The user then propose three different new metaphors:
Since Lifestream is their own work, they claim that that is a metaphor for the way people work. Lifestream allows to locate information in several ways, to remind and to automatically archive info.