A great idea and a possible base for other features as you say.

This discussion should probably be on the Forge Side of OATSsoft but I'm wondering why you decided a client solution would be the way to go here? I agree extension rather than plugin as it offers more scope.

As I understand it Firefox extensions are just XUL overlays and I've been wondering for a while about the suitability of XUL (http://www.mozilla.org/projects/xul/) for creating rich client accessibility tools using a nice declarative tool. It has accesibility features and you can do both client apps (Firefox) or web applications. (and for conceptcoding XUL has some RDF support http://xulplanet.com/tutorials/mozsdk/rdfstart.php). Then again I'm wondering if we're rapidly moving to web applications, but may be restrictive for OATS. Would be a great research project - think I'll do a PHD ;-)

Perhaps of use for this is the excellent greasemonkey extension(https://addons.mozilla.org/firefox/748/) to give easy addition of symbols to existing sites.

as far as i know (and i'm learning it at the moment) xul describes the interface components of the extension, you use javascript/java to provide the functionality.. but i may be wrong.

i see ccf serverside only if author burden is reduced to a minium, i.e doing it using semantic web technologies, and expecting authors to add loads of rdf annotation for each word is a hell of a lot to ask, when we're (accessibility advocates) struggle to even convince them that an image alt tag is worth doing. The automated tools, are not available yet either... the solution symbered done works, but are going to have to do that for anysite that wants this capability - again who's going to do it / pay for it. So a nice short term solution is to provide it as an extension to already excellent browser and its free.

andy

> as far as i know (and i'm learning it at the moment) xul describes the interface
> components of the extension, you use javascript/java to provide the functionality..
> but i may be wrong.

That's seems to be the case (I added more info to the forge forum, sorry no links working here in comments). XUL definately looks good for projects that need a Visual UI (i.e not fo VI). Seems the files that describe the look and code can be locally installed or pick up from a server and thus dynamically generated; that adds loads of flexibility. I'm learning to seem interesting that we both stumbled across it at same time.

Eitehr way you could perhaps make it social software and let users share their symbols like tags. Though that may cut against the grain of onotologies.

Hi, just a brief update to say i've finally got round to playing with this again. I'm also working on some new CCF demonstrators which this will be able to interact with. I've been working on some .net central ccf components - a sqlserver database (containing around 9000 concepts (based on the standup-widgit ontologies), bliss ontology of around 3000 symbols, and a some asp.net web pages. The latest and most exciting addition to this core is an .net webservice that allows you to interact with the ccf from your own applications (web/desktop). So rather than foxysymbols do all the work it will exploit the web service as could any one else.

I hope to get some basic functionality working soon, i'll then try to upload what i have so we can develop it together.