Cart
[Home] [By Thread] [By Date] [Recent Entries]
> -----Original Message----- > From: owner-xml-dev@x... [mailto:owner-xml-dev@x...]On Behalf Of > John Aldridge > Sent: 22 May 2000 12:04 > To: xml-dev@x... > Subject: Locating schemas (was Re: Irony heaped on irony) > [...] > Unfortunately, this author may have no understanding of the > environment of the _reader_ of the XML document. The schema location used by > the XML author may be inaccessible to the reader, and this reader should not have > to resort to editing the XML document before (s)he can make sense of it. > > I don't object to URI dereferencing as a fallback if the reader has not > given more explicit instructions, but the primary method should > be for the reader to "install" a schema locally, and for XML processing > software then to "know" where to find the schema information for the > corresponding namespace. I agree. In fact this is a general issue with any resource identification mechanism. We've recently seen this solved for PUBLIC identifiers using catalogs. Retrieving an XML Schema, or whatever else is at the end of a Namespace URI (Schema, documentation, or the mythical XML Package) hits exactly the same problem. Being able to "install" a schema locally would be extremely useful. We just need to define how we want the software to "know" where to find the schema. The obvious solution to me is a catalog/packaging based system. L. *************************************************************************** This is xml-dev, the mailing list for XML developers. To unsubscribe, mailto:majordomo@x...&BODY=unsubscribe%20xml-dev List archives are available at http://xml.org/archives/xml-dev/ ***************************************************************************
|
