OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]
RE: [xml-dev] Nested Documents (was: XML 2.0)

Perhaps it is bizarre.  This note here is all redundant information (to
some extent) but I couldn't tell whether your comment it out of
disagreement or out of frustration that it seems that a step backward is
being taken.  If it's the latter, then skip this message.

However, I don't think I've heard a good solution on how to ensure that
you have everything.  Yes, it's a good idea to want the network to
ensure delivery of everything but I don't think XML is tied to protocols
that guarantee delivery.  By allowing multiple root elements, either the
network needs to guarantee delivery or the something in the content
needs to indicate either the number of root elements (an extension of
the prolog as already suggested by someone) or by indicating the end of
the stream.  If XML 2.0 (or whatever version) supports multiple roots,
it seems that it should also support a mechanism to indicate this.

The logfile use case is a great one.  But I would be pretty upset if a
hospital record got dropped.  Or in the environment I work in, it would
not be a good thing if I couldn't guarantee that a caselaw document
didn't get dropped.  I don't know the actual likelihood that the network
would fail right in between two root elements but it's a situation that
should be considered...shouldn't it?


-----Original Message-----
From: Michael Kay [mailto:mike@saxonica.com] 
Sent: Friday, February 29, 2008 4:29 AM
To: Johnson, Matthew C. (LNG-ALB); xml-dev@lists.xml.org
Subject: RE: [xml-dev] Nested Documents (was: XML 2.0)

> I like the idea of the 
> processing instruction.  Although it seems that it should be 
> a standardized one (e.g. <?xml:docclosed ...?>) that all 
> parsers can look for. 

I'm afraid I find this all completely bizarre. I thought the days when
put "****" at the end of a data file because you couldn't rely on the
system reporting that you'd hit the end of your reel of paper tape were
about 40 years ago.

Michael Kay

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index]

News | XML in Industry | Calendar | XML Registry
Marketplace | Resources | MyXML.org | Sponsors | Privacy Statement

Copyright 1993-2007 XML.org. This site is hosted by OASIS