Hey Martin, -----Original Message-----
From: Martin Desruisseaux <[email protected]> Organization: Geomatys Reply-To: "[email protected]" <[email protected]> Date: Wednesday, April 17, 2013 12:54 PM To: "[email protected]" <[email protected]> Subject: Re: Would like to replace servlet-api dependency by javaee-api >Hello Chris > >Le 17/04/13 19:09, Mattmann, Chris A (398J) a écrit : >> +1 to: >> >> * NetCDF CF-1.5 to ISO 19115 bindings >> - happy to help here, what do you think is the starting point? >Actually this is already implemented [1], so we just need to port that >package. One thing very welcome would be more tests... +1 OK. You OK with me trying to port this? > >> - should we connect to the NetCDF-Java library here? If so, I >> can help >This is already done actually. We may send a notice later saying that >this package moved to SIS... +1 > >> * Storage in a simple (the most possible straightforward) database >> - how about storing it in Apache OODT? http://oodt.apache.org/ >> That way we could be independent of actual database, and could >> leverage Apache Tika perhaps in there. >I was not aware of this project... The description on the web page seems >very attractive, and I would like very much some connector to it. > >I'm thinking more about a bridge in an optional module ("sis-oddt"?) >than a dependency straight in the SIS core, since I would like to keep >the amount of required dependencies low. I have the same wish for >NetCDF: provide anything related to NetCDF in a separated optional >module, e.g. "sis-netcdf". ++1 for sis-oodt. > >On the database topic, we also have an implementation for that. It is >only 5 classes (only 2 public, a reader and a writer) [2], because most >of the work is performed by more generic code. The idea was to have a >straightforward "one interface = one table" and "one property = one >column" relationship, use ISO names (not necessarily the same than >JavaBean names), automatically add comments in the database based on the >GeoAPI resource files, have the usual special handling for code list, >and I think that we will also need special handling for leveraging >PostGIS or Oracle Spatial features anyway... We did some work with OODT and SIS earlier, my GSoC student last year Ross Laidlaw (also on the SIS PMC) wrote this guide: https://cwiki.apache.org/confluence/display/SIS/OODT+File+Manager+to+SIS+Co nnection+Demo Shows how if geospatial data is managed in OODT how it could be connected to SIS. I see this as sort of "the other way around" but easily doable as well and I'm willing to help which makes it all the better :) > >Anyway, we don't need to take any decision now. I just wanted to let >peoples know what is available. We can try to finish the core metadata >first, and come back on those two topics after? Sure that's fine too. I think getting an 0.3 release out ASAP (light-weight, we can always do 0.3.1 or 0.4 shortly thereafter) would be ideal. Thanks! Cheers, Chris > > Martin > > > >[1] >http://www.geotoolkit.org/apidocs/org/geotoolkit/metadata/netcdf/package-s >ummary.html >[2] >http://www.geotoolkit.org/apidocs/org/geotoolkit/metadata/sql/package-summ >ary.html ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Chris Mattmann, Ph.D. Senior Computer Scientist NASA Jet Propulsion Laboratory Pasadena, CA 91109 USA Office: 171-266B, Mailstop: 171-246 Email: [email protected] WWW: http://sunset.usc.edu/~mattmann/ ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ Adjunct Assistant Professor, Computer Science Department University of Southern California, Los Angeles, CA 90089 USA ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
