<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0//EN" "http://www.w3.org/TR/REC-html40/strict.dtd"><html><head><meta name="qrichtext" content="1" /><style type="text/css">p, li { white-space: pre-wrap; }</style></head><body style=" font-family:'DejaVu Sans'; font-size:9pt; font-weight:400; font-style:normal;">On Saturday 25 July 2009, Martin Renold wrote: > The saving application dictates which fallback to use. Maybe the user > wanted a different fallback. For example, the user might end up with a > color-corrected layer but would rather want the original back. That's implementation detail in the application, nothing in your proposal indicates that. > Does anyone think this is a bad idea? Anyone got a different idea? I don't think that having a tag decided the use of the next tag is a good XML design (but then I am not an expert at XML...), I am not even sure we can enforce that in a schema. But I think the <try> syntax is a bit complicated. It might get overkill in a cascade of fallbacks. What I had in mind was something like that: <text name="rendered text" > <content> some text </content> <projection src="data/text_fallback.png" /> </text> or <projection name="text" src="data/text_fallback.png" > <text> <content> some text </content> </projection> It's worth to note that the projection is also usefull to just be quick to load, just imagine you have a stack of a lot of filters and only one layer, it could take a while to recompute everything, while with the projection, you can show that directly (and eventually reprocess in the background just to be sure). > Is there some standard way to do this that I don't know about? Good question :) I don't think ODF or SVG have a fallback mechanism. -- Cyrille Berger </body></html>