- From: Norman Walsh <ndw@nwalsh.com>
- Date: Wed, 07 Oct 2009 10:07:34 -0400
- To: public-xml-processing-model-comments@w3.org
- Message-ID: <m2ab03l2l5.fsf@nwalsh.com>
"Toman_Vojtech@emc.com" <Toman_Vojtech@emc.com> writes: > After two years of re-reading the spec, I came to the following > conclusion regarding the difference. Suppose you have the following > dependency graph of steps (Y reads the result of X, Z reads the result > of Y): > > X <- Y <- Z > > Then X is connected to Y, and Y is connected to Z. Looking from the > other side, Y has a binding to X, and Z has a binding to Y. > > Following this logic, most of the spec seems to make sense to me :) Wow. I have a much simpler notion in my head: two ports are bound together (have a binding) if one is connected to the other. The concern I have about "binding" isn't in connection with the use of the term "connected" but rather with the use of "binding" to mean "variable binding" in some places. I wonder if "connection" could be used instead of "binding" for the input/output port case... Be seeing you, norm -- Norman Walsh <ndw@nwalsh.com> | No victor believes in chance.-- http://nwalsh.com/ | Nietzsche
Received on Wednesday, 7 October 2009 14:08:24 UTC