Thanks for writing up your ideas Martin. I will respond a bit more clearly another day.
My initial thought is "evolution" not "revolution" ... le me explain. It is nice to clean up modules one by one starting from the bottom of the stack. That is a good direction; and something I have been working on (see the gt-api module for my progress). But we have a solid need to support the existing applications and development teams we have today - as such we need to provide useful services at all points in GeoTools 3 lifecycle. But that discussion on maintenance costs also shows one thing - the current situation has too much code for too few active developers and must be changed. So how about this for another idea - cut the code down to *just* what is used. There are many other good related ideas on the table - chunking the library up into larger modules (this is sane from a product standpoint anyways); it could be that GeoMatys only wants to depend on a single GeoTools module and ignore the rest? What I want to avoid is a repeat of the Gomatys / GeoServer experiment where I think you ended up rolling your own WMS? Remember evolution not revolution. I would rather see Geomatys take part in a GeoTools 3 time line that involves everyone; than see an experimental GeoTools 3 branch followed by a migration plan - we did this experiment with GeoSolutions if you recall; and while the result worked out in the end I really missed working with Simone on trunk. In terms of numbers if a branch approach takes 1 year; and a working on trunk and collaborating takes a year and a half I would consider the money well spent. Yes I know I am spending your money by asking for a longer time line; and It may feel like others are not on board because GeoMatys just took on the OSGeo providence review; and GeoAPI involvement probably feels similar. But what do you (or anyone) get in trade? Shared maintenance costs is the usual benifit (we may complain about lack of module maintainers - but Andrea and Justin have been doing a very nice job of hunting down bugs release after release). Access to experts (this is a big one for me - working with Martin and Andrea for example have allowed me to take on much bigger problems then I would alone). Finally I am hoping for documentation, publicity, product positioning (this has been me recently; but I am hoping OSGeo involvement will help). And finally we get access to new toys out of RnD projects (WPS is a good example of this recently; improvements to ISO Geometry interfaces, new feature model etc...). I could go on - but let me offer a small reminder to everyone. We are changing from a "project focused on RnD" to a "product focused on ability". We may be stuck in some heated discussions about what balance we want to strike for the next five years (it makes sense we are changing our focus). Please recognize that we have stuck a pretty good balance for the past five years; some decisions we made will not be correct given our new focus. Some actions we have each taken will also look insane given a new focus. Personally I am enjoying this change of focus; everyone is saying similar things from all directions. We get caught up in the details but everyones heart is in the right place. Finally I need to know if this discussion is holding up anyones work? I personally am not going to be writing down much in the way of direction; or thinking very long and hard about this stuff until after graduation requirements are met. In terms of shifting our focus to a "product focused on ability" the first step to me is to set the product up as part of the OSGeo foundation; giving it a firm legal and marketing base. All the best, Jody ------------------------------------------------------------------------- Sponsored by: SourceForge.net Community Choice Awards: VOTE NOW! Studies have shown that voting for your favorite open source project, along with a healthy diet, reduces your potential for chronic lameness and boredom. Vote Now at http://www.sourceforge.net/community/cca08 _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
