[Discourse.ros.org] [OpenEmbedded] ROS 2 Layer for OpenEmbedded Linux
Víctor Mayoral Vilches
|
vmayoral
October 4 |
Happy to announce that we’ve open sourced a set of recipes to get ROS 2 for OpenEmbedded Linux.
Code is available at https://github.com/erlerobot/meta-ros2 and released under GPLv3.
Visit Topic or reply to this email to respond.
To unsubscribe from these emails, click here.
Esteve
|
esteve
October 4 |
This awesome! Any particular reason for choosing GPLv3 instead of MIT? Perhaps picking the latter would just make it easier to get it integrated into meta-ros (which is licensed as MIT, much like OpenEmbedded’s core recipes). Also, given that these are build recipes, maybe there’s not much of benefit in using GPLv3.
Thanks Victor (and Erle)!
Víctor Mayoral Vilches
|
vmayoral
October 4 |
Hey @esteve!
We don’t have any particular concern about changing the license.
We’d like to spread its use and would love to get contributions, if you and the community believe that changing to MIT would benefit those purposes… we hear you. Certainly, matching meta-ros original license makes sense as well.
Let’s see what others think about it.
Esteve
|
esteve
October 4 |
Thanks for your honest response and for considering whether to change or not the license. I don’t think there’s anything wrong with GPLv3 per se, but given that most OpenEmbedded layers (meta-beagleboard, openembedded-core, meta-96boards, etc.) have very liberal licenses (MIT/X11, BSD) I wondered if there was an actual reason for picking GPLv3.
Lukas Bulwahn
|
bulwahn
October 4 |
Hi Victor,
here just two comments:
-
Did you already have a look at the content of this pull request https://github.com/bmwcarit/meta-ros/pull/538? Is it done identical? Or can we merge those two developments to a best solution of those two in your repository?
-
Concerning the license, I chose MIT just as most layers did use MIT when I started meta-ros. I think more important than the license of the metadata, is actually this further important remark on the license: “All metadata is MIT licensed unless otherwise stated. Source code included in tree for individual recipes is under the LICENSE stated in each recipe (.bb file) unless otherwise stated.” This allows others to take your patch files and contribute and merge them back into the main development without further legal considerations.
Best regards,
Lukas
Víctor Mayoral Vilches
Did you already have a look at the content of this pull request https://github.com/bmwcarit/meta-ros/pull/538? Is it done identical?
First time looking at it. Wish this would’ve been available back in May when we started with it 
.
From quick code to the commits, i can see that the PR you refer to is from August. Our commits started however in May. I asked @abilbaotm to review the code and here’s the feedback provided:
Main differences are:
- The ament.bbclass https://github.com/rojkov/meta-ros/blob/433b6457bbac9a512097f4d9e55843ff16bb0f22/classes/ament.bbclass: Our ament.bbclass depends on the system’s ament which calls to cmake (I was expecting to change that) and the @rojkov’s depends directly on cmake and meta-ros classes.
- The rest seems rather similar.
bulwahn:
Or can we merge those two developments to a best solution of those two in your repository?
I wouldn’t say they are identical though but we’re happy to explore this in more detail. We’ll create another branch at https://github.com/erlerobot/meta-ros2 exploring the differences. Evaluate both and push the one that gives best results to our master branch. @bulwahn I’m happy giving you maintainer rights to meta-ros2.
License-wise:
esteve:
I wondered if there was an actual reason for picking GPLv3.
From my (humble) experience with licensing, private companies, for obvious reasons, tend to contribute back more when code is licensed under GPLv3. (Speaking as part of a private company) I’d expect meta-ros2 to be rather popular for quite a few private companies so the license was placed with that in mind.
That said, again, happy reconsidering the license if it creates a hurdle.