After the IDL 8.0 overview at the IDL User Group meeting last week, Bill Okubo also asked for a list of features IDL users needed. I think I have all the ideas people gave (along with some pertinent links to proof-of-concept ideas and background material):

1. better map projections
2. geo-browser export (i.e., to Google Earth, etc.) ([exporting from ENVI](http://groups.google.com/group/comp.lang.idl-pvwave/browse_frm/thread/7f26747c807ffc34/dd652e5fd5fcd804?lnk=gst&q=kml&pli=1), [discussion of doing it in IDL](http://groups.google.com/group/comp.lang.idl-pvwave/browse_frm/thread/7f26747c807ffc34/dd652e5fd5fcd804?lnk=gst&q=kml&pli=1))
3. stereo display ([simple anaglyph system](http://michaelgalloy.com/2006/06/16/anaglyphs-mggr3dconverter-and-mggrwindow3d.html), I also have code for using Alioscopy's autostereoscopic monitors from IDL object graphics which I have not published yet)
4. a better widget system ([a way to make IDL bindings for Qt?](http://michaelgalloy.com/2009/09/11/pyside-qt-for-python.html))
5. easier runtime distribution ([`MAKE_RT` über-installation](http://michaelgalloy.com/2009/08/04/make_rt-uber-installation.html))
6. better handling of large data sets
7. better support for databases
8. integration of GPU and parallel processing (Tech-X does both [GPU](http://www.txcorp.com/products/GPULib/) and [parallel processing](http://www.txcorp.com/products/FastDL/index.php) in IDL)
9. OPeNDAP support ([I have something coming soon for this!](http://www.txcorp.com/products/RemoteDataToolkit/))
10. easier interface to gridding data
11. ability to read Matlab ".sav" file
12. better volume rendering

Thoughts? Did we miss something crucial?

*Full disclosure: I work for Tech-X Corporation and have worked on GPULib, FastDL, and the Remote Data Toolkit.*