- From: Al Gilman <asgilman@iamdigex.net>
- Date: Mon, 19 Nov 2001 09:22:11 -0500
- To: Martin Duerst <duerst@w3.org>, "David G. Durand" <David_Durand@brown.edu>(by way of Martin Duerst <duerst@w3.org>), uri@w3.org
At 04:27 AM 2001-11-18 , Martin Duerst wrote: > >However, it became also very much clearer to me (I already thought >so implicitly before) that in a web context, centered on concepts >and digital resources, where copying and moving things around is >virtually instantaneous and extremely cheap, the distinction between >names and locations is not necessary at all, and usually confuses >more than it helps. > AG:: Excellent and thoughtful post as ususal. On the other hand, the appearance that all this data speed means that the transport lag can be ignored is not to be trusted. As regards applications of the Web of Services that is coming over the hill at us, there is no such thing as a negligible lag in accessing data. There is a BIG community doing BIG computations and they alone generate a large market for staging functionality to get the fixed-name data to the variable place where it is needed as an articulated separately managed step in the process. This is a large enough an endemic enough problem such that the Global Grid Forum is forming a working group to address this issue in particular. <http://www.globalgridforum.org/groups/WG.html#RDA>http://www.globalgridfo rum.org/groups/WG.html#RDA Al > >Regards, Martin. >
Received on Monday, 19 November 2001 09:17:06 UTC