Completed:
- Re-orgnaized entire project into .xsl files corresponding to NeXML elements, all imported into a master one
- More robust tabular output now, especially for <<otu>> elements
- Realized that tab delimited format requires de-normalizing to a certain extent
- i.e.I think I have to repeat group metadata for each member of a group within ISAtab
- But the ISA configuration will determine this, so I'm going to shift to tht kinda.
- Now construing each <<meta>> element as a 'cell' in a table
- Able to correctly label column headers & align data
Not Completed:
- Verify ISAtab files. As noted before, verification of ISAtab output requires a custom configuration.
- Integrate with ontologies. This isn't planned, but I'm keeping an eye on it because I might need to. Likely, though, ISA's functionality will be sufficient
Next Steps:
- (!) Update project plan & documentation (behind)
- Create simple shell script to pass variables to Saxon (almost done)
- Create simple ISA configuration for NeXML (in progress)
- Meeting with Rutger Vos this week or next RE NeXML content (email out)
- Continue to refine/tweak XSLT
- (next week or after) Simple jar interface combining Saxon and my XSLT.
- (next week or after) draft NeXML-MIAPA checklist for community approval
- (next week or after) incorporate draft NeXML-MIAPA checklist into ISA configuration
Basically, I've been learning about ISAtab and NeXML quickly to the point where the way I had the work broken up before doesn't work. Also there are dependencies on things like the ISA configuration that I'm now factoring in to my schedule I didn't get a chance today to update that plan, but it's first on my list tomorrow. I also got clear guidance from Hilmar last week to make sure that I do simple prototypes early on to feel out any problems ahead of time. Hence the shell script I wrote today & the planned Java.
Comments, questions and suggestions welcome.
E