OK Let me try:
Two documents doc1 and doc2. Each document is represented by an repectively InfoSet i1 and i2. InfoSet i1 has a reference to InfoSet i2 via a key or XPointer. In a reference or linking case there are 2 infosets i1 and i2. In XInclude, wouldn't there be just one InfoSet ib which included both doc1 and doc2? Incidently, the InfoSet spec says an infoset can only contain 1 document. cheers Jim > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED] > Behalf Of Jonathan Borden > Sent: Wednesday, July 25, 2001 9:33 AM > To: [EMAIL PROTECTED] > Subject: Re: Preparing for a new update > > > Jim Tivy wrote: > > > I would suggest that XInclude is an entirely different technology - more > > like an entity expansion. > > It is fine to suggest this, but your suggestion provides no > guidance to the > characteristics of this "entirely different technology". If you read my > reply to Paul, you should note how XInclude is different than DTD entity > expansion. > > The question remains: different than what? My criterion remains > that I would > like to see a functional difference in an Infoset resulting from one > mechanism or another. If the infosets are equivalent, then differences are > merely syntactic sugar. > > -Jonathan > > > ---------------------------------------------------------------------- > Post a message: mailto:[EMAIL PROTECTED] > Unsubscribe: mailto:[EMAIL PROTECTED] > Contact adminstrator: mailto:[EMAIL PROTECTED] > Read archived messages: http://archive.xmldb.org/ > ---------------------------------------------------------------------- ---------------------------------------------------------------------- Post a message: mailto:[EMAIL PROTECTED] Unsubscribe: mailto:[EMAIL PROTECTED] Contact adminstrator: mailto:[EMAIL PROTECTED] Read archived messages: http://archive.xmldb.org/ ----------------------------------------------------------------------
