- From: Norman Walsh <ndw@nwalsh.com>
- Date: Tue, 24 Jul 2007 10:18:13 +0100
- To: public-xml-processing-model-wg@w3.org
- Message-ID: <87zm1m40tm.fsf@nwalsh.com>
/ Jeni Tennison <jeni@jenitennison.com> was heard to say: Fixed. | Technical comment: | | Paragraph just after Figure 4 says "When a step is used in a pipeline, | connections are made to all of its inputs and outputs." But non-primary outputs | don't have to be connected, right? Same comment further down where it says "All | of the declared outputs of a step must be connected." I don't think this is true. | | Editorial comments: | | I know what you mean by "When an input accepts a sequence of documents, it may | have one or more bindings to any of these locations." but I think it would be | better phrased as "When an input accepts a sequence of documents, the documents | can come from any combination of these locations." | | On error ports, it says: "These messages appear on a special error | output port defined (only) in the catch clause of a try/catch." It's a bit | confusing to say that the error output port is defined in the catch clause of | the try/catch. Isn't it actually that every step has an error output port, but | it can only be bound to the input of the catch clause of a try/catch? | | Section 2.2.1 External Documents: I wonder if the last couple of paragraphs | should go in an informative note? Be seeing you, norm -- Norman Walsh <ndw@nwalsh.com> | Fish die belly-upward and rise to the http://nwalsh.com/ | surface; it is their way of | falling.--André Gide
Received on Tuesday, 24 July 2007 10:46:06 UTC