Actually, no, as my experiences of the pain of versioning OWL in a
normal VC system predate OBI by someway. But the value I see in tawny
wrt versioning would be relevant for OBI also.
My main motivator from OBI would be the ability to build patterns; there
is quite a lot of boilerplate in OBI; with tawny, you can build and use
patterns alongside the rest of the ontology freely. This can be for
complex patterns, but also small things such as defining new syntax for,
for example, OBIs annotation properties.
I think also, a programmatic environment could help with the release
process, again because it could help with automation.
Having said that, I don't think shoe horning tawny into OBIs existing
process would be terribly helpful; tawny may have been motivated by
thinking what tools would *I* want, if I were to develop OBI. But, that
doesn't mean it's the right tool for OBI.
What might be useful though, would be tawny's unit testing. This could
help to check that OBI is consistent, that specific classes are
reasoning the way you expect. In short, you could use it to drive
continuous integration; actually, OBI is already on a CI server ---
https://travis-ci.org/phillord/tawny-obi --- although not with any
useful tests yet.
On re-reading the email, I worry that the statement reads pejoratively;
it was not meant so. OBI has demonstrated the short comings in many
tools; tawny is my attempt to ask, can we steal the tools we need from
elsewhere, rather than develop them?
Phil
Bjoern Peters <
bpe...@liai.org> writes:
> Quick question: When you refer to OBI being a motivating factor, were you
> mainly referring to the support for versioning?
>
> - Bjoern
>
Phillip Lord, Phone:
+44 (0) 191 222 7827
Lecturer in Bioinformatics, Email:
philli...@newcastle.ac.uk
School of Computing Science,
http://homepages.cs.ncl.ac.uk/phillip.lord
Room 914 Claremont Tower, skype: russet_apples
Newcastle University, twitter: phillord
NE1 7RU