Mike Purvis

@gavanderhoorn We've done some work on it, but it's not in a publicly-releasable state. If you're interested to take a look and see if it's a design direction which interests...

Another interesting angle might be exposing multiple handles of different types for a single joint. So, for example, register the usual VelocityJointInterface for your wheel, but also a EffortJointInterface. Let...

For interested parties, here's a very simple example of the third proposed scheme for implementing this: https://github.com/ros-controls/ros_control/compare/melodic-devel...mikepurvis:parent-interface Most of the code is changing resource claiming to use `std::enable_if` so that...

@jim-rothrock Have you been satisfied with [steered_wheel_base_controller](https://github.com/wunderkammer-laboratory/steered_wheel_base_controller)? Would you consider contributing that to ros_controllers?

@ipa-mdl Thanks for looking. I agree on some kind of test; I think in the past when working on this, I got hung up on what exactly such a test...

@nalt That looks great! I tried validating all our launch files with this just to see how they fared: ``` find /path/to/ros -type f -name "*.launch" -exec xmllint -schema roslaunch.xsd...

@paulbovbel Are you aware?

The issue with entry_points performance is discussed by @jbohren in #297. (upstream issue: https://github.com/pypa/setuptools/issues/510)

What is the output of `catkin config` in this scenario? Are you using the merged or linked develspace?

Your description sounds very much like the installspace mutexing problem (#378, #391). I would have thought a linked develspace would be safe from that, but perhaps not. In any case,...