xml-dev - Re: URNs as SYSTEM IDs

["Simon St.Laurent" <simonstl@simonstl.com>]

At 12:37 PM 11/22/00 -0800, Joe English wrote:
>PUBLIC ID resolution doesn't need to be universally deployed
>for it to be useful.  In many scenarios, out of all the software
>that might be used to process an XML document, only one -- the
>document's author's validator -- really needs PUBLIC ID support.
>All other document consumers should be able to ignore the
><!DOCTYPE ...> declaration entirely as long as the DTD
>makes no essential InfoSet contributions.

Unfortunately, the use of URNs for SystemLiterals makes PUBLIC a more
important option than it is when URLs are used.  Processors which lack
PUBLIC identifier processing and which receive documents using URNs as
SystemLiterals are pretty much stuck, since they can't get to the DTD at all.

_If_ the DTD makes no essential Infoset contributions, the DTD doesn't
matter at all except to provide constraints.  However, since DTDs
frequently make Infoset contributions (and the IANA registration document
sets no such limitations), processors have no way of knowing whether or not
the DTD is required for proper processing.

I guess this is familiar ground, since the option for non-validating
parsers to ignore external resources has similar ramifications, but it
seems to drive it even deeper into the heart of XML by making even
validating parsers unreliable.

Simon St.Laurent
XML Elements of Style / XML: A Primer, 2nd Ed.
XHTML: Migrating Toward XML
http://www.simonstl.com - XML essays and books