Diversity is a good thing. Systems-level research is at a woeful state right now because the status quo is Good Enough for a lot of people. Thus, the number of

Diversity is a good thing. Systems-level research is at a woeful state right now because the status quo is Good Enough for a lot of people. Thus, the number of interesting ideas to try, to fail, to succeed is a trickle to what it should be. Database-based and object-based filesystems are little more than research projects because of the hard to shake belief of what stored items should look like and what their properties are. We need to work on projects with little to no preconceived notions, so we can start poking at things, figuring out what can be done better. We indeed have a working model, but I'm certain we can do better than just working.