Hi all, every now and then I get an email about what can be improved with the CDK project. In fact, there is a lot that can be improved. Rarely, however, these emails are backed up with actual help making those things happen, and it seems that in most cases the proposer backs out when asked to talk to this and that CDK developer to coordinate things. That seems often too much work.
A few months ago I sent around some 'open positions' to which I had no response. In the past 5 years at least, have of my CDK work has been mere project handling: making releases, getting people to talk to each other, applying patches, setting up build environments, etc. Rajarshi has been most helpful here, in particularly by setting up the Nightly environment, aggregating the various source code testing environments. Others, like Nina and Sam have helped out by setting up Maven repositories, and even less visible has been the work by, for example, Debian developers to create Debian packages. Since a year or so, the core developers decided to move from Subversion to Git, which was also done to reduce the work load on the core CDK development team, and at the same time we introduce patch reviewing, that brought in some shared code responsibility: the reviewer make himself a bit responsible by approving the code, while at the same time the author knows he better gets his code in shape before dumping it onto the larger project. This was not always appreciated, but did do wonders for the stability of the library. In fact, you can now fairly safely download a version from master, and it will in fact compile (this was not always the case in the CDK history). Now, there are many things the CDK needs: * an up to date Eclipse OSGI repository * an up to data Maven repository * more patch rewriting of recent publications (Padel, and various others) * more unit testing (many parts of the CDK are yet untested, and may get accidentally get broken; that too used to be quite common, unfortunately) * documentation, etc.. Yes, these are all important and all needed. Now, who is going to do that? Right now, the core development team around me, Rajarshi, and Christoph are working really hard to secure funding (and Christoph has in fact various people working in his group using the CDK), and we actually manage to get some time here and there too do some some actual coding. But... but we are under-resourced, and us three heavily depend on the larger CDK community with developers jumping in and out (and this list has become really long; I had hoped that CDK News would have become a good platform for people to show how they contribute, but is unfortunately not used enough, again likely due to time constraints... e.g. where is that CDK News paper on the Maven repositories...). Therefore, I strongly invite people who have ideas on how the project can get improved to think on what they can contribute in terms of work hours, and then start talking to people who have been contributing to the CDK project in that area. This is NOT that their input is not appreciated, as it very much is! But I would much rather see they did not communicate these insights to me personally, but to the cdk-devel mailing list, so that the appropriate people can respond. In fact, despite the CDK being under-resourced, the project is still very much alive, due to YOU, due to the community contributing in many different ways. There are still open positions in the project: * a second release manager (so, we'll have one for stable, one for master) * a new CDK News editor * a patch lieutenant (keeping a branch with reviewed and approved patches, periodically for the release managers to pull in) And, you can always at any time: * set up a Nightly build server for one or more branches * review and test patches * port code written by others which is not yet integrated into the main library * update 3rd party projects to use the latest CDK * write unit tests for code that is untested With this email I hoped to have indicated they the CDK project is aware it is not perfect, and that it knows where things should be improved. At the same time, I hope to have explained a bit what the things are the CDK could really need help with, and also hope to have clear that anyone is most welcome to help out, and shape the future of the project! With kind regards from a sunny Uppsala, Egon Willighagen -- Dr E.L. Willighagen Postdoctoral Researcher Institutet för miljömedicin Karolinska Institutet Homepage: http://egonw.github.com/ LinkedIn: http://se.linkedin.com/in/egonw Blog: http://chem-bla-ics.blogspot.com/ PubList: http://www.citeulike.org/user/egonw/tag/papers ------------------------------------------------------------------------------ The modern datacenter depends on network connectivity to access resources and provide services. The best practices for maximizing a physical server's connectivity to a physical network are well understood - see how these rules translate into the virtual world? http://p.sf.net/sfu/oracle-sfdevnlfb _______________________________________________ Blueobelisk-discuss mailing list Blueobelisk-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/blueobelisk-discuss
