Cart
[Home] [By Thread] [By Date] [Recent Entries]
I can speculate about why it was designed that way, but it's still icky IMO. Two things: 1) I would prefer the "stat" to be declared like a regular XMLNS 2) A PI with a name like ?Mapping is a bit grandiose. Like nobody else would ever want that name? It should be scoped IMO. > -----Original Message----- > From: Bullard, Claude L (Len) [mailto:len.bullard@i...] > Sent: Monday, July 19, 2004 7:26 AM > To: Joshua Allen > Cc: xml-dev@l... > Subject: RE: What is the rule for parsing XML in a > namespace inside HTML? > > All good. How about this: > > <?Mapping XmlNamespace="stat" ClrNamespace="WiseOwl.Statistics" > > Assembly="WiseOwl.Statistics.Library" ?> <Window > xmlns="http://schemas.microsoft.com/2003/xaml" Visible="true"> > <SimpleText Foreground="DarkRed" FontSize="14">Hello > World!</SimpleText> > <stat:PoissonDeviate Mean="5.0" /> > </Window> > > I would still expect a namespace declaration for the stat namespace. > I've no objection to the mapping PI given that it has to tie > together three pieces of information. I realize that the > colon isn't doing that much, but again, expectations are set. > Care to comment on the implications of the above with > respect to interoperability (yes, the example is a user > extension, so interoperability is dicey at best). > > len > > > > From: Joshua Allen [mailto:joshuaa@m...] > > ... going with whatever works best for > the user and can get consensus among browser vendors, rather > than doing user-hostile things in the name of purity to some > random irrelevant spec. In fact, the Hyatt discussion has > morphed in a sense into a discussion about namespaces in HTML > [1], and the remarkable thing is that *none* of these people > are suggesting the facist route of "force the whole world to > use XHTML and only then will it work!". That's a good sign, IMO. >
|
