I wanted to give my $.03 worth since I've been quietly listening to this. Let's discuss another perspective of this not only from the Committers' point, but the documentation and end user point.
As one who has edited and re-written lots of our documentation, I can say that IMHO documentation will become more frightening than ever. Someone or some tool will need to be instituted to make sure each Async release has all the pertinent documentation. This could mean that we officially will release documentation so often, that the release numbering on the documentation will become unwieldily or be confusing to the end user. On the end user point, I want to only install twice a year when new releases come out. (I usually install 1.x.x not just 1.x.... ) I have only three windows per year that I can shut stuff down and perform installations where I have to perform Postgresql and DSpace database changes. Those are frightening as we all know. If I have to pick and choose which release or a release with a new component that may or may not be part of a main release, or, worse yet, no further "main releases", I'm going to looking for a new solution for DSpace. (Not me personally, but perhaps my institution, or another institution). When installing Open Source becomes too complex, we begin to remember the old saying "Open Source is not free" and think that perhaps it is time to migrate to a commercial product where the vendor does all the work. Is this something that we want to force onto a community such as ours? I can buy commercial IR software, there is no work at all, but just the expense of buying it. I am released from upgrades, database maintenance etc.. But now I am limited by what I can do with the software. Dspace is really quite easy to install now--even the source version. It's really a balancing act, or as I like to say "a change of politics" when you think about software options. I can choose Open Source--customizable--complete control---but a chance that it will become too difficult to install/upgrade maintain, or I can choose commercial software--expensive--limited--ease of maintenance-update. As for the Ant/Maven issues, I have no problem at all with those. They work well, and compilers do not scare me--it's actually fun to watch your work compile. (Especially if you have customized it and want to see the end results!) So, go ahead and ant/maven-ize away. A third way, which is one that some of our commercial vendors use, to to have a source release (as we do now), a non-source release (as we do now) and several (one two or more) additional optional releases that add options, but are not incorporated into the main release. (Examples: Oxygen, Atlassian, OpenSuSE, Fedora (Redhat), etc. though some are commericalized...) So, get your guns out now, and shoot me. :-) Jeffrey Trimble System LIbrarian William F. Maag Library Youngstown State University 330.941.2483 (Office) jatrim...@ysu.edu http://www.maag.ysu.edu http://digital.maag.ysu.edu ""For he is the Kwisatz Haderach..." ------------------------------------------------------------------------------ Achieve unprecedented app performance and reliability What every C/C++ and Fortran developer should know. Learn how Intel has extended the reach of its next-generation tools to help boost performance applications - inlcuding clusters. http://p.sf.net/sfu/intel-dev2devmay _______________________________________________ Dspace-devel mailing list Dspace-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dspace-devel
