"Eating your own dog food" I've bundled a NLP appliance containing UMLS and cTAKES and I'm migrating services to the AWS cloud. Throughout this process I have extensively documented the steps and I made a VM target for Ubuntu 13.04.
I would have used the iDASH appliance but it is no longer maintained and was way out of date. From conversations and experiences witth iDASH and NCBO bioportal I strongly support a VM option for cTAKES downloads. Suggestions #[ 1 ] Public demo server (instance of VM) For new users, this is the "hook" that motivates them to get involved BEFORE licensing and install. http://bioportal.bioontology.org/annotator #[ 2 ] Downloading the public demo server is power ! http://nlp-ecosystem.ucsd.edu/ http://bioportal.bioontology.org/virtual_appliance #[ 3 ] Example code for each cTAKES component is game changing cTAKES has an impressive set of components, but no way to try them out. http://ctakes.apache.org/components.html Whereas you can play with the components of Bioportal, for example http://data.bioontology.org/documentation I'm sharing these experiences and suggestions because I truly do believe they are crucial and likely more important than any level of new-features that can be added. Hoping this message finds you well. I will soon circulate a copy of a VM to play around with and get feedback. Thoughts? --AndyMC
