- From: Andy Seaborne <andy.seaborne@talis.com>
- Date: Sun, 08 Nov 2009 21:34:22 +0000
- To: Kjetil Kjernsmo <kjetil@kjernsmo.net>
- CC: public-rdf-dawg@w3.org
On 08/11/2009 20:23, Kjetil Kjernsmo wrote: ... > Possibly a more important argument for it, is the discussion about > transactions we just had: Having two operations may or may not be atomic, > UPDATE will certainly be. Now, if this is impossible to implement as atomic > for everyone, it is of little value, but how hard can that be? :-) I don't see who we can insist on atomicity of a single operation as being required although we can encourage it. The range of systems I think we must target includes simple processors for single request usage. Maybe a system that reads in files, changes them and write them out could be useful but it's not atomic without a presuming various details - be nice in shell scripts though (it can be made atomic but isn't automatically). Service Description would be the place to give the supported features. Andy
Received on Sunday, 8 November 2009 21:34:59 UTC