- From: James Fuller <james.fuller.2007@gmail.com>
- Date: Wed, 19 Nov 2008 12:18:00 +0100
- To: "Norman Walsh" <ndw@nwalsh.com>
- Cc: public-xml-processing-model-comments@w3.org
+1 to this JF On Tue, Nov 18, 2008 at 4:23 PM, Norman Walsh <ndw@nwalsh.com> wrote: > Toman_Vojtech@emc.com writes: >> We have a number of custom XProc steps, some of which use c:result in >> the output document. Now I am thinking that the >> "http://www.w3.org/ns/xproc-step" namespace is maybe meant only to be >> used with the standard steps, and not with custom steps. I am wondering >> whether what we are doing right now is actually correct... > > I was thinking about this too, just yesterday, see pxp:unzip in: > > http://exproc.org/proposed/steps/ > >> So, say you are implementing a custom step that needs to return a URI >> string (such as in the case of p:store), what is the best thing to do: >> >> 1. Use c:result as much as possible to promote consistency >> 2. Don't use c:result at all (because it can be used only with standard >> steps) and invent your own vocabulary >> 3. It doesn't matter > > I think, on the whole, it will be easier for users if we say it's ok > for extension steps to produce results in the "c:" namespace. > > The alternative just means that users have to declare yet another > namespace and remember that some steps produce results in c: and > others in d: > >> If any of the options is preferred over the others, do we need to say >> anything about it in the spec? > > I suppose we could say something. What do others think? > > Be seeing you, > norm > > -- > Norman Walsh <ndw@nwalsh.com> | In great affairs men show themselves as > http://nwalsh.com/ | they wish to be seen, in small things > | they show themselves as they are.-- > | Chamfort >
Received on Wednesday, 19 November 2008 11:18:36 UTC