On Thu, Aug 14, 2008 at 4:22 PM, Erik Garrison <[EMAIL PROTECTED]> wrote: > On Thu, Aug 14, 2008 at 03:03:45PM -0400, Benjamin M. Schwartz wrote: >> Erik Garrison wrote: >> | Can you reasonably >> | expect to navigate more than 20 or 30 different entries without indexing >> | and search? >> >> Indexing and search are precisely what the Journal is supposed to provide. >> ~ In fact, there's a nice, fully functioning search box at the top. The >> search even includes the full text of your Write entries, etc. The >> Journal approach was sparked in part by the move, among all major desktop >> environments, toward search as the principle data retrieval mechanism >> (Google Desktop Search, WinFS, Spotlight, Beagle, etc.) >> >> Improvements to this system are needed, but all major design work on the >> Journal has been on hold for a long time, waiting for the new datastore. >> > > I understand. > > I think that well-functioning indexing and search are far better than > expecting users to manage hierarchies (a skill that is incredibly useful > but takes years of practice and observation to master). I just believe > we should provide as many avenues to data organization and access as > possible. Obsoleting the hierarchical filesystem crowds out a large > number.
Agreed. Our goal was never to obsolete it, but to provide a fresh and interesting alternative. It turned out that this led us to obsoleting it by obfuscation, due to lack of time and resources, but our long term goals, olpcfs among them, aim to rectify this. In any case, I think that our responsibility is to focus entirely on the Journal model, to be sure we get it right, rather than trying to provide all possible solutions in this space. I'm happy to see others attempt a File activity, or even us in several years when we have "nothing better" to do. - Eben > Erik > _______________________________________________ Devel mailing list Devel@lists.laptop.org http://lists.laptop.org/listinfo/devel