Fedora-Council/tickets ticket #527: we need to start on the survey!

Project Discussion
1

@mattdm filed Fedora-Council/tickets ticket #527. Discuss here and record votes and decisions in the ticket.

Ticket text:

2

Does anyone (doesn’t have to be a Council member) want to take lead on this?

The process is described in details in

It doesn’t require any specific skills, except doing an action on the pre-defined date (which is the hardest skill of them all).

3

Are we happy with the questions as they are?

I remember us wanting to do some tweaks during the last review, like asking questions we know the answers to to help us understand how different groups answered, and how well they’re represented. Something like that.

Regardless, we most probably want to do at least some tweaks, as it still asks about Fedora modules. :slight_smile:

4

We do need to review and update the questions.

The best way for this is to suggest PRs with changes to

5

I’m definitely interested in helping out. I’ll take a look at Pagure tomorrow :+1:

1 Like
6

I don’t have this thought in PR-ready form, but I’d like to ask a basic “AI: hate, meh, love” question.

7

I would avoid generic Yes/No question and consider something like

Which AI-based tools do you use?

With “None, and feel strongly against them” being a possible answer.

I think that the term “AI” is too generic and it will be hard to interpret the outcome. And I would be interested in seeing the range of tools people can come up with. Will it be just LLM?

1 Like
8

To me it would make most sense to start with:

  1. agreeing on what we want to get out of the survey
  2. reviewing good survey practices (like the way to know the different group representation we discussed in the last review)

Only after that we can come up with good questions that’ll get us what we want. Questions in specific formats, asking (perhaps even indirectly) about specific things.

I’d propose a live discussion first, inviting some experts (hi @gwmngilfen!), and taking it from there.

9

Or maybe it doesn’t need to be live, if we’re not in rush. But my point was: desired outcome first, questions second.

That way we’re more likely to end up with a good set of questions that fit together, and a good overall structure of the survey, so it’s easy for people to take (and even answer partially and still be useful, for example).