Nathan Brooks

> The reality of what is slow and what is fast in computers is much harder to understand because of how modern CPUs work. In my experience memory locality is...

> Something as simple as the joint trajectory execution should return a failure when the end state of the robot after executing the trajectory is not near the end state...

On another sidenote (I will work with Denis to open a separate discussion on this), I recently learned more about some of the arm controller APIs, and ideally, we would...

I thought that might be the case - I will update the PR accordingly! I do have a question related to this I can put as a feature request if...

How would you configure a client-server discovery model with unicast? Maybe I missed something in the documentation, but I did not realize cyclone supported that! Both of those proposals sound...

Will be addressed in port of https://github.com/UniversalRobots/Universal_Robots_ROS_Driver/pull/431

Tagging @tylerjw and @vatanaksoytezer to relay how ROS 2 solves this and see if it is applicable for ROS 1.

I like this idea. Having some more semantic knowledge of what they want to do will give us a better ability to detect errors in the configuration and merge things....

Thanks for bringing this up, Doug! I made a lightning talk on my DDS/QOS learnings for the February MoveIt Community meeting. You can see the full recording [here](https://www.youtube.com/watch?v=WajZmurQYMc&ab_channel=NathanBrooks). Here is...

> It looks like the joint_states topic can/will be published as rclcpp::SensorDataQoS() in Humble Could you provide a link to this? I've only seen this [RSP PR](https://github.com/ros/robot_state_publisher/pull/179), which just changed...