My Photo

November 2005

Sun Mon Tue Wed Thu Fri Sat
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30      

Atkins Papers & Talks

« Press Release about Tony Hey going to Microsoft | Main | Intelligence: We've Lost Our Edge »


Margaret  Hedstrom

Dan, in your talk yesterday you mentioned that cyberinfrastructure depreciates faster than bricks and mortar infrastructure, but that it is more generalizable. I thought this was a brilliant point and wondered whether this was your idea or if there are additional sources I could refer to. I would like to use this point and augment with the following observation. If one considered digital libraries, archives, and data repositories as a component of infrastructure, then one could argue that those components accumulate and appreciate in value. From the physical world we have countless examples of collections that far outlive the brick and mortar infrastructure that supported them. In the digital world, it is assumed that the systems (and infrastructure) will also evolve around the collections.

Dan Atkins

This was my observation but made in a very practical, budgetary sense. It has been my observation that universities and NSF for the most part have had difficultly setting up budgetary support for IT that adequately accounts for the rapid obsolescence of equipment. To stay at the leading edge, for example, NSF-funded supercomputer centers need constant upgrades or replacements. Optical telescopes for the astronomy community stay current a lot longer. Modern computer equipment is more like perishable fruit than traditional capital equipment.

I had not thought of your abstract or generalization but it is a great thought and I encourage you to use and develop it.

The comments to this entry are closed.