-
Notifications
You must be signed in to change notification settings - Fork 43
Developer Survey #85
Comments
I think a survey would be a good way to get broader input. I think this group (the modules wg) would need to put together the initial list of questions to be asked and then the user-feedback team can work with the Greg at the foundation to tweak and then have the survey run. |
As an example, its good that the questions be closed whenever possible. How important is interop between CJS and EJS to you: 1 - not at all |
For that specific question (and probably many others), we should also separately ask how important that feature is to be available by default versus "by a loader" versus "not at all" |
nit: How about "enabled by default" vs. "possible" or "opt-in"? Mostly because loader might be hard to understand for people / might imply too much. |
I think we need to be sure that we encapsulate the questions so that we are clearly trying to get either the expectation of how ESM would be implemented, or the opinion of if a feature is important. I think it is very easy to be unclear on the difference of these. Expectations can lead to opinions of if a feature is important, but they are not necessarily the same thing. You can have a neutral opinion while having a strong expectation of how things should work. Defining what interop is is a complex task for example. Consider the following points and how detailed I think we need to be in order to start to expose this difference:
Even after all of this, we might have comflicts with implementing some features such as synchronously returning a module namespace from |
Oh, and we should probably figure out what questions each feature needs so maybe that is a good place to maybe start making a list of questions and combining similar ones as we go along? Does that seem a good idea to others? |
@bmeck sounds reasonable to me. |
Any interested in reviving this effort? I independently started to work on an iterative survey building effort, which starts with the premise that while I know to some degree the composition of the participants that I am interested in surveying, I still don't know exactly what type of questions will elicit the more informative and reliable results. From there the survey will be iteratively built by getting two participants (starting with the group) to take the survey then provide recommendations and feedback (privately) which will be incorporated and peer reviewed by those who have already taken it, and so on. It is common practice that drafts would be shared openly, and obviously this will happen, but after we all had a chance to somewhat anonymously contribute to the resulting initial draft. I am looking for the first two participants, the survey is very raw and likely not on point at all, but that is naturally going to change especially if you take part. |
@sendilkumarn Are you interested? If so you can take a go at the first round and once we refine it you can also weigh in on changes as we move forward. We will still need at least one more participant to carry out the first round. |
I'm interested |
Lets do it 👍 |
@inidaname @sendilkumarn I just sent you invites to take the survey (used address listed on gh profile) Once done, we'll take it from there |
Please jump over to #173 regarding efforts to prepare Initial Draft (set to unroll for October's event) |
Here is an open link to the most recent "ongoing" draft: C1: Developers Survey |
@inidaname per this comment please advise if the |
We've discussed this at multiple meetings, removing from agenda for now. |
We've been making a bunch of claims regarding what "developers would want". While I think we all have quite a great intuition on this, perhaps we can do a developer survey to prove out some of our hypotheses before digging too far into a philosophical debate.
@nodejs/user-feedback can maybe help?
The text was updated successfully, but these errors were encountered: