Call for Participation: Middleware Working Group

We’ve had several design issues come up in the past revolving around the middleware components of ROS 2, and so, to help address those and future needs, we propose to setup a Middleware Working Group to focus on these issues and try to resolve them, as well as help involve more people in the development of the packages which provide access to the communication features in ROS 2.

The Middleware Working Group’s mission is to identify and address issues with usability and performance in the core communication features of ROS 2, which are provided by the middleware (i.e. implementation of the rmw API) or one of the client libraries (e.g. rclcpp or rclpy).

This working group will focus on the core ROS 2 stack, including the rmw API and its various implementations as well as the client libraries like rcl, rclcpp, and rclpy as needed.

I’m still working on setting up a working group repository based on the ROS 2 TSC’s WG template, but I’d like to start hearing from people that are interested in being involved and/or helping with the working group. We’ll have open meetings, I’m thinking biweekly, with meeting notes posted after each meeting.

If you’re interested in participating please fill out this doodle for which times would work best for you and I’ll do my best to select a time that’s international friendly and also not overlapping with existing working groups:

I’d also like to take feedback on what the first topics for discussion in this working would be, so if you have ideas please let me know here, or if you’d prefer to wait, once I have the issue track setup for the working group I can post a link to that here and you may open an issue instead.

Thanks!

16 Likes

I notice you posted this today (June 16th) and some of the options I see for the meeting are yesterday and shortly after the post today. Was this meant to start next week?

2 Likes

Sorry for the confusion, I’m bad at doodle apparently. I haven’t scheduled the first meeting yet, those were just supposed to be recurring times that could work, regardless of what week it falls on. The dates are not important.

3 Likes

Just double checking on those times listed.
They were intentionally set at 10-12 pm pacific time correct?
Is the intention to alternate between morning/late pacific to ensure all individuals from all timezones can reasonably participate?

Yes. The early morning times for NA are often very late for Asia pacific areas. I was just trying to see what could work. I don’t plan on swapping back and forth but just seeing what people would be willing to do. It’s perfectly fine to decline such times.

I tried to pick times that worked for Madrid, Tokyo, and San Jose using this site: https://www.timeanddate.com/worldclock/meetingtime.html?iso=20200625&p1=283&p2=248&p3=141

So have we picked any times/dates to start?

1 Like

Out of curiosity, is there any movement on the formation of this working group? I have extensive experience working with DDS in commercial production robots and am very interested in joining and contributing.

1 Like

Not yet. I’m going to try and plan something next week, with the first meeting perhaps the week after.

4 Likes

Ok, thanks for everyone that filled out the Doodle survey.

I’ve landed on Wednesdays at 8am. My personal opinion is that we start on a two week period between meetings and see how it goes from there.

I’ve decided to schedule the first meeting for Wednesday July 29th at 8am PST and we’ll use a Google meet instance which I will either create and post here or attach to the calendar event on the “Upcoming ROS Events” calendar:

Either way I’ll announce it here.

I’ll also post an agenda for the meeting ASAP, my plan is to use our GitHub repository, which is still private pending some edits to the working group template I based it on.

If anyone has further ideas for the agenda please let me know here, in a DM, or an email for now.

Thanks!

5 Likes

Ok, it’s on the ROS Events calendar, repeating every two weeks.

The meeting link for the first meeting is https://meet.google.com/cgm-jxpp-bas. I’ll post the agenda ASAP, please send me ideas if you want them considered for the agenda.

3 Likes

@wjwwood I would like to contribute for Middleware and be part of Middleware working group. Could you please let me know if there is a process to join the working group?

I’ve posted the agenda, I apologize for it being late, but I wanted to finish the working group template before posting it.

I’ve added a link to the agenda in the meeting description, but here it is again:

And I’ve recently made the working group repository public with my first take at customizing the template for this working group. We will discuss it in the meeting, see it here:

Looking forward to speaking with folks tomorrow morning.

2 Likes

@malapatiravi for now, you may just attend the meetings, they’re open to anyone who is interested. We’ll discuss possible roles in the working group during the first meeting.

1 Like

Hello All,

iRobot would like to present a quick performance improvement proposal for the executor, and RMW.

Thanks !
-Lenny

Would it be possible to add a link to a meeting recording the agenda document after each?

Can you propose that on the agenda here?

I’ll triage it this week and let you know if we have time for that (should be possible).

@dawonn-haval Sorry, I didn’t record the first meeting. I’ll look into options for recording future meetings, though I want to talk with the working group about it first, so next meeting I’ll discuss it and let others chime in publicly or privately to me if they have an issue with recording meetings or not.

The agenda and notes are here:

For everyone who was at the first meeting, please add yourself to the list of attendees if you did not already. In the future I will try to capture the attendee list myself. Thanks :slight_smile:

From now on I won’t make announcements for future meetings, unless we make a change to the regular schedule.

Please see the agenda and meeting notes document to see what’s been happening and to propose changes to the agenda: