Adrian, I was very careful to preface my comments with an acknowledgement that I wasn't speaking from a position of great expertise or authority. I think the tone of your response demonstrates a lack of maturity and professionalism.
Honest and open communication is important in any community. People should be able to question established practices without having to worry about getting blasted out of the water. Obviously my idea had some merit, as Matin acknowledged in his first response, but not enought to make it worth the switch. Landon On Wed, Feb 4, 2009 at 2:04 PM, Adrian Custer <acus...@gmail.com> wrote: > Hey, > > Honestly, you are so far off the mark that it's not worth discussing. > > Geotools needs a build tool. You say a Java project doesn't need to have > a build tool and experience says otherwise. Not using a build tool would > be plain idiotic. Any reasonably sized project has a build tool. So look > around, notice that it is so, and ask yourself why. > > Geotools cannot afford to have more than one. That's experience from any > number of projects. Every effort to mix tools that I have ever seen has > been a total fiasco. Very tempting for the short sighted: "Hey, we can > do that in three lines of code with tool xyz." Very painful over the > long term. Look around, see that very few projects mix build tools, and > ask yourself why. > > Which one to choose? Doesn't matter, Geotools chose. Much like choosing > a language determines the character of a project, choosing a build tool > is something that happens along the way. Changing is very costly and > only done for *really* good reasons. The only other tool with the power > of maven in common use that I know of is Ant. Several projects use Ant > with some success. When you compare the complexity of the OpenSSO system > based on Ant with the complexity of a maven based system, you quickly > see why projects are regularly moving to maven. But again, it doesn't > matter, Geotools chose maven. > > Maven is not cast in bronze, merely carved in stone. For it to be worth > replacing, the replacement would have to be absolutely fantastic, widely > known, with repositories all over the place. So if you find something > that good, that widely supported, and that commonly known, by all means > suggest it. > > > So, if maven is a blocker for you, no problem. Grab a mercurial/bzr/git > copy of the full repository and hack away using whatever you like. Free > software for free people. Want to mix in another language? Great, it's > your code, do with it what you will. But, *if* you want to throw your > code into the common pool, then you have to do the work to play by the > same rules that everyone knows so that no one has to learn the > particular rules you happened to choose to work with. > > And with that, you have wasted enough of my time for today. > > Have a good evening, > --adrian > > > And no, "maven rocks" is not a feeling---it's observation gleaned from > the experience of all the other tools I have come across. I hate maven > just like I hate all build tools---getting code to build is a pain all > around. Maven rocks because it is far less painful than any alternative > I have ever worked with. > > > > > ------------------------------------------------------------------------------ Create and Deploy Rich Internet Apps outside the browser with Adobe(R)AIR(TM) software. With Adobe AIR, Ajax developers can use existing skills and code to build responsive, highly engaging applications that combine the power of local resources and data with the reach of the web. Download the Adobe AIR SDK and Ajax docs to start building applications today-http://p.sf.net/sfu/adobe-com _______________________________________________ Geotools-devel mailing list Geotools-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/geotools-devel
