| [recommenders-dev] What's going on at Code Recommenders? A brief status report |
Hi there, the last officially announced drop was quite a while ago now. I would like to give a brief update on what's going on since most of the development discussion happens on IRC or face-to-face - the code recommenders mailing list is a silent place :) Currently we have 17 people working on various parts of the project. Most of them (all except Johannes and Eric) are students doing a one semester hands-on or thesis. Johannes and Eric are working on a novel stacktrace search engine that helps developers to quickly find resources like posts in forums, stackoverflow.com or mailing-lists related to these stacktraces. Ingo Jakobs and Marc-Christian Schulze are working on further improvements of the JDT based chain completion engine. See http://code-recommenders.blogspot.com/2010/12/how-do-i-get-instance-of.html for a reference on this topic. Katrin Schützkowski works on a local version of our code search engine for Eclipse. Niklas Lochschmidt is improving the code search engine user interface and its query builder. Kristijan Mandunic is working on new strategies for creating code summaries. A related blog post to this subproject is here: http://code-recommenders.blogspot.com/2010/07/why-is-google-codesearch-not-google-for.html Mateusz Parzonka and Fabian Richter are working on the Clean Code Method Sorter as mentioned in the blog here: http://code-recommenders.blogspot.com/2011/04/clean-code-method-sorter-for-eclipse.html Thomas Schlosser and Gerrit Freise will extend the JDT to support more flexible configuration of the Eclipse Content Assist lists as mentioned here: http://code-recommenders.blogspot.com/2011/04/at-lunch-with-jdt-team-zurich.html Michael Kutschke is working on our own implementation of Bayesian Networks for java so that we can get rid of the existing native implementation with Eclipse-incompatible licensing we are currently using. Florian Jakob and Andreas Kaluza work on the idea of a community drive code template store. Sebastian Proksch is working on new recommender models that leverage new information like how a reference to an object was obtained and evaluates the benefits of clustering in terms of performance and quality of the recommendations. Dennis Saenger currently evaluates a novel approach to data collection which may supersede the existing data collection concepts. Andreas Frankenberger is working on the round-trip of sharing knowledge data, mining new, updated models and delivering them to the clients. He will evaluate this approach for Android where is is creating a new reference set for. See http://code-recommenders.blogspot.com/2011/04/say-hello-android.html for details. Stefan Henß is working on the extended javadoc client during his Google summer of Code project which is - more or less - described here: http://code-recommenders.blogspot.com/2010/03/problem-of-incomplete-javadocs.html But I guess he will send some more details on this in a few days. Quite a lot what's going on - but not too much to see at the moment. Anyway, there will be a lot to report at the end of this summer... :) Best, Marcel -- Eclipse Code Recommenders: |