Survey: How will we communicate in the future (Mailing lists, forum, ...)?

Dear RIOTers,

we currently evaluate new communication channels to improve discussions within the RIOT community. We consider introducing a forum, implemented in Discourse (https://www.discourse.org/).

As your opinion matters, we kindly ask you to participate in the following brief survey, which we hope will help us to decide on the next steps:

The survey consists of max. 10 questions and should not take more than five minutes.

1 Like

Are you planning to publish the results?

Good question. Currently, @waehlisch has access to the raw data. @waehlisch, do you think we could publish them in one shape or form?

1 Like

It took me a bit to compile them, but here are the results:

Have you used a forum (e.g., Discourse) before?

82.1% replied Yes, 17.9% replied No

Have you used a mailing list?

92.3% replied Yes, 7.7% replied No

Have you been aware of the RIOT mailing lists before this survey?

92.3% replied Yes, 7.7% replied No

To which RIOT mailing lists are you subscribed?

Have you participated in discussions on the RIOT mailing lists?

47.8% replied Yes, 22.2% replied No

When considering discussions related to RIOT, which communication channel do you prefer?

Would a forum increase your interaction with the RIOT community?

Please rate the following tools based on your preference for chat-based communication about RIOT

1 = least preferred … 5 = highly preferred

Which social medial platforms are you using regularly?

Other

Word cloud rendered from answers below

  • mastodon
  • Gitlab
  • Matrix, Slack, Telegram
  • none of the above
  • irc

Any additional feedback, either related to this or any other topic related to RIOT?

Word cloud rendered from answers below

  • The RIOT community should not get lost in tools.
  • We really need better on boarding materials for absolute beginners. Documentation is excellent but not clear where to actually begin with a sensor. Drivers/Examples/Tests distinction isn’t clear in the documentation.
  • I have not selected media platforms that I use for personal/family things only.
    • the users mailing list should be removed, devel and maintainers are enough. Users of RIOT are developers. - I don’t understand the purpose of the last question. Are you seriously considering using Facebook or Snapchat or TikTok as communication tools ? - A nice feature of Slack and Mattermost (I don’t know if that exists on the others) is the possibility to configure channels linked to GitHub activity: any new issue, PR and comment sends a new message in that channel. That’s a better alternative to emails generated by GitHub. The same thing exists to link Discourse to Slack.
  • Discord
  • For chat-based communication: Telegram.
  • I think it will be bad if discussions are fragmented.
    • It’s a ironic that a survey about an IPv6-capable OS is not reachable over IPv6. Please fix this, it’s 2020! - Why are the chat-centric social media platforms (Matrix/Slack/Telegram/WhatsApp/Discord/etc) missing? - Telegram: 2 stars; Discord: 1 star :wink:
  • Question 9 feels really out of place.
  • I use GitHub, but really not for its social media aspects (Yes to bugtracking, No to following developers, setting my mood emoji or logging in to stuff via GitHub). On the topic of mailing lists vs. forums, entries in mailing list archives are notoriously hard to find in search engines – that’s a shortcoming that could be fixed individually, but if a forum it is, one fewer item to address.
  • would be nice to have discussions visible to anyone (default) without needing to login/signup
  • I think mailing lists lead to more focused and higher quality interaction than the alternatives.
  • mailing list have the advantage of being archived correctly and survive whatever happens :). Forums, and basically any web thing, is lost after some time. I can see the benefit of having async and sync solutions coexisting (mails + irc/matrix/whatever), but not really 2 async solutions.
  • gitter to merge with matrix IMHO IRC over matrix would be fine for most needs try: https://riot.im/app/#/room/#freenode_#riot:matrix.org I am RzR there
  • When I asked some technical questions, some mailing lists’ members answered it, even if I was a RIOT beginner. It is too good for RIOT being well-used.
  • :slight_smile:

All in all, the results speak all for our ongoing migration work.

The usage of chat platforms already reflects our provided chat options: Most seem to prefer IRC and Matrix and the least number of people seem to dislike them.

Regarding social media outreach, we might want to think about doing more on Youtube (67,65% of the people who answered use it regularly), Reddit, and LinkedIn (the latter two have similar usage as Twitter where we already are quite active). Hopefully the #riot-os tag on Stack Overflow will increase our activity there to fit the usage by our community. For the platforms named it might also make sense to provide OAuth login for this forum.

yes, this was the reason why i started to collect more videos and extend the playllists: https://www.youtube.com/c/RIOT-IoT/playlists

i thought we had a discussion to enable all platforms that are supported via OAuth. i don’t remember any reason against that.

I actually also thought about more community stuff, like making short videos (but who has the time for that…) or using the community tab for announcements. There are also stories, if we want to be goofy :clown_face:

Yes, I was stuck a bit in the pure analysis of the Survey, but keeping things as open as possible, makes more sense.

I am a bit curious about the preferences of the 11% who are not on the mailinglists and the 22% who didn’t participate in any discussion there. (I hope the 11% are not included in the 22%)

Not quite sure how you mean that. Do you want the “When considering discussions related to RIOT, which communication channel do you prefer?” data for those 11% who are not on the mailinglists and those 22% who didn’t participate in any discussion there? On “I hope the 11% are not included in the 22%”: Why shouldn’t those two sets overlap btw? You can after all not participate in any discussion while not being subscribed to a mailing list pretty easily :wink:.

Thats what I meant, just making sure, thx :sweat_smile:

Yes, I would like to see that. I mean these are the people that might be more involved on other channels than the mailinglists, aren’t they?

Also interesting is if a forum increases their interaction with the RIOT community? and their prefered chat-based communication.

But only if the data is any meaningful, I don’t know how many people answered the survey.

90 people answered the first three questions, the remaining questions (so including the questions you were asking about) were answered by 68 participants. The current usage (users on the forum, subscribers on the mailing lists) are:

So I think we can say that the survey is meaningful.

First the 11% (7 participants) who are not subscribed to any mailing list:

Preference between mailing list and forum for people not subscribed to any mailing list

Potential for increase of interaction with the RIOT community for people not subscribed to any mailing list

Preferred chat-based communication for people not subscribed to any mailing list

For the 22% who never participated in a discussion on the mailing list:

Preference between mailing list and forum for people who never participated in a discussion on the mailing list

Potential for increase of interaction with the RIOT community for people who never participated in a discussion on the mailing list

Preferred chat-based communication for people who never participated in a discussion on the mailing list

For the potential of of interaction with the RIOT community I also found the inverse sub-set interesting

Preference between mailing list and forum for people subscribed to any of the mailing lists

Potential for increase of interaction with the RIOT community for people who already participated in a discussion on the mailing list

1 Like

Those 22% amount to 20 participants btw, so a bit more representative than the participants not subscribed to any mailinglist.

I moved this topic to the community category now, as the original call for the survey might have been an announcement, but now that the results are also posted and discussed here, it fits more into the community related stuff.