On Wed, Dec 10, 2008 at 2:40 AM, G T Smith
For instance relational database concepts have been around for some time for categorising data and while it makes sense to organise computing executable structures in trees, why are we still organising our documents in this manner?[1]
Why? Because file systems have proven themselves more reliable than database oriented approaches, and a close copy of what people do in the physical world. We store documents (contracts, deliverables, schedules) in our files organized by projects, in folders, in drawers. Its simple, and even though finding ALL schedules is a bit of a mad-thrash, its do-able. Nothing but mindless duplication of documents allows us to look one place to find all schedules, or all deliverables across the entire engineering division. If we could trust databases not to come totally unglued losing ALL OUR DATA we might keep out files in such a structure, and be able to find it 50 different ways without duplication. But there is nothing in the real world that comes close to that. Further, storage and retrieval are not desktop elements, and unless/until all users use KDE it would be inappropriate to build this functionality only into the desktop.
The perceived need for indexing systems such as beagle must in part come from the recognition that the current way of doing things is not entirely effective as an organisational mechanism. If we change the way personal documentation is perceived as been organized this implies a change in how that documentation is accessed, i.e at the desktop. Rather than in individual applications.
The arguments you are making for building this technology into the desktop are would be more properly focused on the file system. Building such technology into the desktop is cart before the horse. -- ----------JSA--------- Someone stole my tag line, so now I have this rental. -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org