Ergin Soysal wrote: >While playing with ADL, I found myself collecting language rules in an >app. I converted it to an editor. I hope you like: >http://soysal.com/modules.php?op=modload&name=Downloads&file=index&req=viewdownload&cid=2 >Please test it. I can continue development on interest. > > Hi Ergin,
very nice. I thought I had started up gvim when I started this app! In case you didn't know, there is a syntax mode for gvim, which does the same colourisation etc, and works in either vi or mouse mode; see http://www.openehr.org/repositories/implem/latest/publishing/tools/windows/index.html, or I can post just the gvim files on this group if anyone wants them - they're very small. Gvim you can get from http://www.vim.org/, it's a very nice editor, free, and works on all platforms. The gvim mode I developed - could be better, so anyone who wants to improve it is welcome. >And back to ADL discussion. First question here, who will use (e.i. learn) >this language in a clinic requiring an additional archetype or >modification? > I should point out that just because we have defined a formal language for archetypes, does not mean we think that this is the way people will generall write archetypes. There is an open source GUI editor nearing completion which will be available in a few weeks, designed for clinicians. It has already been tested in turkish by Koray Atalag as a matter of fact, and also demostrated to CEN and HL7 audiences. > Next question, what will be the frequency of definition of a >new archetype or modification in an health service? > Well, that is very dependent on how people apply it, but I would expect that after an initial phase of a lot of creation, the creation of completely new archetypes will probably not occur that often - but the modification of existing ones, and creation of archetypes which are specialisations of existing ones will be quite common. That is only a guess, however... > I believe, in majority >of the cases, it may be impractical to expect some stuff to learn and >remember this language. > I agree totally - the main audience of such languages is technical people and tool-builders, not actual archetype authors. > xml may have an advantage with html-like syntax. > > I can't say I know many people who think raw xml or html are nice to edit in - I think they're awful (but it's obviously a matter of personal taste!); ADL was designed to be something like a programming language - so technical people might write in it, but as I say above, the real aim is GUI tools. >But the fact that, instead of manual coding these scripts, soon, there >will be a heavy demand for a visual designer, string/term translator, >and/or script generator wizards etc. In this case, you'll need to reverse >engineer the ADL code back to load into the IDE :), sorry. > which is what the GUI editor does: it uses the parser back-end, which is already on the web, with its own technical viewer GUI. Exactly the same parser is used by the clinician's editor tool, guaranteeing that the editor must conform to the exact semantics of ADL. Since the parser is able to save in ADL, and HTML, and in the near future, in XML and OWL/RDF, the editor will be able to do this as well. Of these formats, ADL, and probably XML-schema-based XML will also be able to be read in (for the XML, we have to see yet whether it is semantically lossy or not). But at least we know that ADL works properly as a lossless format for saving and reading archetypes in. hope this helps. - thomas - If you have any questions about using this list, please send a message to d.lloyd at openehr.org
