- From: Erik Bruchez <erik@bruchez.org>
- Date: Mon, 10 Mar 2014 11:20:10 -0700
- To: michael odling-smee <michael.odling-smee@xml-solutions.com>
- Cc: "public-xformsusers@w3.org" <public-xformsusers@w3.org>
> I have not fully researched web-components as yet but perhaps it is > the best long term solution. However by looking at the steps you list > to getting it into the standard and knowing that XForms 2.0 is likely > to be the main focus of energy currently I would have thought an > approach such http://www.exslt.org/is the simplest for now. > > For instance Orbeon, betterForm and XSLTForms[1] all support a combo > box like control. Could we not define a simple declarative way of > representing the combo box and some mini-XSLT transforms to convert it > into implementation specific representation? If all the transforms are > published in an open way would that not achieve the portability I am > seeking? If there are controls not implemented by certain XForm > processors then these would at least be known gaps and the processor > authors and others could contribute "code" to help plug these gaps. > > If there is broad support for this approach it is perhaps something I > could put my energy into. I also like the idea of agreeing on common behaviors outside of standard bodies first when possible, and standardizing later. I fear that there might not be enough players and energy available for that though :( Also I don't think that in general XSLT is the right too for the job here. Components do more than transforms. Transforms do not provide proper encapsulation, proper behavior within repeats, etc. Or if they manage to do this, that would be like reinventing a component system. But I might be proven wrong. -Erik
Received on Monday, 10 March 2014 18:21:00 UTC