> the question I get asked seems > to just be "will this be faster?"
If you're new, you might be missing some context from previous projects. If there is a focus on performance, presumably there were problems regarding performance before. At least, it's an option worth entertaining and might be more favourable than assuming the manager is insane. Are you sure you could not take further steps to automate this build? For reference, Maven has some support for building non-JVM languages, and you can call maven from other scripts or scripts from maven. In a previous job, we had a JNI library wrapping some of ffmpeg's libs, involving using swig to generate code, and all we had to do that was out of the ordinary was to start mvn from msys instead of cmd (in Windows, elsewhere there was no difference at all). > I believe my development manager was involved in the GNU > Step project and so I believed used it because he was a fan of the > technology rather than because it was necessarily a good fit. At least > that is the impression I get. Just ask him. It sounds like you're communicating more with us than with your team. :) > The Java code in the system is probably some of the worsted I have > worked on. Hashtables and Vectors everywhere. That probably indicates its age or when the coders last deliberately learned anything, rather than quality directly. > Coding to implementations including all the collections. That isn't a big problem and can be refactored out fairly automatically. I'd suggest code reviews for most of your remaining points. > This system also rolls its own web framework and its own logging > framework. Choose a logging framework, rename your logging framework's methods to match that, then delete the jar and go through fixing the imports. :) > I would have my work cut out for me just cleaning up the Java. The > Objective-C would be more liveable if the two systems were black boxes > with a well defined boundary interface for communicating back and > fourth but I think the use of wrappers means that Objective-C classes > will just be sprinkled liberally through the Java code. This is probably your bigger problem. I'd be looking to create islands of pure Java and islands of pure Objective-C, and to stretch an analogy, form continents of each connected by satellite link. > All the Java code I have seen speaks to me of a lack of thought about > architecture or design patterns / principles. So introduce them as you see fit but measure the performance so that you are armed for when you are questioned. -- You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
