Wooooot +1000 Andrea
I think we must go back to reason, calm down and define a long terme strategy in order to work all together. I'm not efficient enough technically speaking, to give you some solutions but I'm sure we're going right to the split if we don't take the time to define a strategy and create a synergy for the future. We're all under pressure with our projects and have no time to stop now and discuss, but what about defining a rock solid policy for GT 3.0 and take the time to submit propositions to avoid problems we're encountering now. I think it's important to define this for GT3 wich is not on our actual focus and prevent us from emotional overflow due to deadline pressure. What I can suggest for this kind of discussion is to find where each group can bring more hands to complete others weakness. For me the goal of this collaboration will be to harden geotools genericity in unsing it in differents projects, and push up performances and code quality in having more than one pair of eyes looking at the code. One concrete example of discission for me in GT3 is GeoAPI and using standards API in Geotools. It's a strong need for us and we can spend more effort than others to be sure GT will continue in that way. For now, it's clear we've to deal with the different approach that each project has taken, and two solutions are facing us : 1) finding a non-blocking solution that satisfy everybody (it seem's annotations could let us enough room for that 2) starting with a DVCS system and split everything into alternative solutions tainted by each group for it's needs. My 0,02€ ... Vincent. Le 25 juin 08 à 20:36, Andrea Aime a écrit : > People, can you take a step back from the monitor > and look at the thread again? It's starting to look > like a kindergarden fight. > > Let me spend a few words, hoping I don't offend anyone, > or at least, that I offend everyone in the same measure ;) > > We have basically one group of people that orbits around > Eclipse, the open source community, and a set of solution > that have been developed over time and that have proven > to work (GeoServer, uDig) > The other group believes more in "Sun standards", has > quite a good development funding to change the status > quo, and it's targetting the latest OGC specs. > > I see the fight on the XML stuff as something really > simple: > a) the existing XML parser can do lots, yet it's rooted > in some Eclipse xml technologies, it's "big" (jar wise), > and to be used to the fullest _requires_ you to use > Eclipse EMF. Of course you can develop even without > it, but it becomes painful, there is no tool support, > so I personally don't consider it viable > b) JAXB has apparenlty evolved a lot, and makes lots of > sense to people used to work in a Sun environment > and NetBeans in particular (frankly, would you use > a technology that forced you into a specific IDE > _other_ than the one you chose to do any effective work?) > > The divide goes along other routes too, geometry model > choices, streaming vs in memory, the very way development > is done (practical and time constrained vs takes what it > takes but certain "rules" cannot be bended). > > So what do we want to do? Shall we fork? Ah yes, the world > really needs yet another java GIS library, everybody is > waiting! > > I would urge everybody to calm down and be pragmatic. > We can work togheter, but to do so, everybody has to > understand that it takes patience, and that certain > decision one make affect others and may waste their > time, so one should be prepared to help along if this > is needed. Imho the thing is simple: whoever wants to > push change should take the onus or fixing whatever > breaks in other modules as well, unless the other > maintainers agreed otherwise. > At the same time, when a GSIP is approved, everybody > must try to be patient and understand there will be > some rough time after big changes. Today is Geomatys > doing changes, tomorrow it can be TOPP, GeoSolutions, > or Refractions. > > I understand it takes a quite a bit of extra effort > to work togheter, but I also believe that it's > unavoidable in such a big and diverse library, > and that the result is worth the pain. > > So please, put aside arguments like "my way > or the highway" and try to work toghether for good. > Bitching away is not constructive, it's just a way > to waste time and leave a bad taste in everybody's > mouth. > > Cheers > Andrea > > > > > > > ------------------------------------------------------------------------- > Check out the new SourceForge.net Marketplace. > It's the best place to buy or sell services for > just about anything Open Source. > http://sourceforge.net/services/buy/index.php > _______________________________________________ > Geotools-devel mailing list > Geotools-devel@lists.sourceforge.net > https://lists.sourceforge.net/lists/listinfo/geotools-devel > ------------------------------------------------------------------------- Check out the new SourceForge.net Marketplace. It's the best place to buy or sell services for just about anything Open Source. http://sourceforge.net/services/buy/index.php _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
