Well, it's not every day that GitHub releases a product that competes directly with my young startup. Thankfully, it's still very DIY, which is the opposite of what my customer development has told me is important. So a sigh of relief from me.
At my talk at the Italian RubyDay on continuous integration I received huge critiques for endorsing a Java solution for ruby CI. My point was that Jenkins was far more complete ( at the time of the presentation) of any full ruby stack CI product. This news is a fine Christmas present. This tells me that people clever than me share some of my views
I don't really get the problem though. Java is trivial to learn for anyone. It's not your main product, it's not going to take a massive amount of time for a single plugin. Just write it and be done...
It might be really easy to learn. But after having programmed Ruby for any significant time Java is very unpleasant to program ;)
Also just learning basics to do simple customizations is not enough. If something breaks you will not have a slightest idea how to debug and fix. With a Ruby CI server I can just stick a debugger and walk through the code.
> Think of a ruby dev team and the wish to extend the CI server?
It's not exactly a show stopper as long as they have the sourcecode.
If the team are capable enough to code Ruby and extend the CI server, then they either already know enough Java to do so or can learn it quickly enough to be effective.
Mainly yes.. shouts like " Java sxx" " I hate Java". The most relevant may be to resource usage: Jenkins would require more resources than the production machine in mid sized project. But in a world of commodity HW I'don't think this is a real issue
