April 2024 Survey #261
April 2024 Survey #261
Conversation
|
Thanks @marco-ippolito @mhdawson. I have this started, but I think its a bit too long and users might not complete the full survey. Is there any questions that can be removed? It's best to have 10-12 questions maximum to get the best response. |
|
I know that the discussion is tense on this subject. But we should have some stats/question on Corepack. for me the two questions that are missing:
|
|
We should also look at the results from the last survey. Even though overall the answers/data was great, I know there were a few questions were the answers suggested the question was not very clear. We should identify those and make sure to tweak the related questions. |
|
i just went through last year's survey data and i recommend the following:
|
|
@mhdawson @marco-ippolito here is what i have right now: https://www.surveymonkey.com/r/Preview/?sm=azuRRBptcGtNPJ1HXU5pgFGqa9o53iCd8PmuDc0c_2FMfpPH9Tqx_2FPQV1lAeZ7BYY7 please note that i don't think we should have open ended questions. this doesn't guarantee a good answer rate. recommend that if there is one i didn't add due to that, we make it multiple choice if possible. |
I agree we should limit these, but we got a lot of very interesting responses in the last survey (several hundred) even if many people skipped (since we got a lot of responses). |
|
I remember digging through the feedback from the last one, to try to identify biggest pain points: nodejs/TSC#1445 (comment). It would be nice if we can somehow gather this information in a more structured way than a free response “what issues are you having with Node.js” open question. It was clear from the last survey’s data that many users struggle with ESM and TypeScript, and that’s probably our biggest user complaint. What’s less clear is why they struggle, which is what we need to know in order to help them. Anecdotally it seems that the largest group of such struggling users are ones building React apps with Webpack, implying that it’s something about older build tools not keeping current with modern APIs; but I don’t really know much more than that. In a perfect world I’d somehow be able to get 100 responses from people who struggle with TypeScript and ESM, and they could be categorized into groups where these 50 are on this particular tech stack and are getting this particular error; and these 30 are on this other one, etc. Then we could reproduce the problems, and try to find solutions for them; better error messages if nothing else. But I don’t know how to get anywhere close to getting data like this. Perhaps “Have you run into issues with TypeScript and/or ES modules, and if so, what errors did you encounter and what tech stack are you using?” |
|
This is the current survey draft, which should reflect all of the comments above: https://www.surveymonkey.com/r/Preview/?sm=azuRRBptcGtNPJ1HXU5pgFGqa9o53iCd8PmuDc0c_2FMfpPH9Tqx_2FPQV1lAeZ7BYY7 |
|
I'll try to gather everyone's feedback and update the survey. |
| - No | ||
|
|
||
| - For those of you wishing to use ESM in an existing application, what have been the pain points or blockers preventing you from doing so (if any)? (open question) | ||
|
|
There was a problem hiding this comment.
Can we add this?
| - If you use or have attempted to use TypeScript with ES module syntax, whether ESM dependencies or running your production code using ESM syntax, what issues (if any) have you encountered? Please be as specific as possible and please include your tech stack (build tool, application framework, testing framework, runtime customizations, etc.). |
There was a problem hiding this comment.
@GeoffreyBooth do you think we can simplify it in
have you encountered issues when using or migrating to ESM syntax (`import`/`export`)?
- Yes
- No
- If yes which one? (open question)
There was a problem hiding this comment.
No, I really want to get the tech stack information. Also lots of people think they’re using ESM when they’re really not, because the distinction is lost between writing ESM source and running code that uses ESM syntax.
Co-authored-by: Antoine du Hamel <[email protected]>
Same link as before |
On that link, two questions both ask “How do you manage the package manager for your project?” We don’t need two copies of this. I’d like this question added please:
|
The duplicate was fixed. I will let @marco-ippolito and @mhdawson chime in but i strongly recommend that we don't have that broad of an open question + we cannot add another question, as we are well over the recommended limit |
|
I suggest we find some other questions to cut, then, because I really need this information and we’ve been discussing it for the last few years, and the survey is the way it was suggested that we get this data. |
@kyliewd, is this a hard limit of the tool. If we just add the open ended question at the end of the survey, that should not affect the likelyhood that people don't complete the questions up till then right? |
Not a hard limit but you really start to lose folks after question 10 - so you're at risk for that. my goal is to help you all get the best response / completion rate here! 😄 |
|
@mhdawson if we need to add the question suggestion from @GeoffreyBooth it needs to be shortened/simplified. |
|
Here’s a shortened version:
|
|
Survey is live: https://linuxfoundation.surveymonkey.com/r/nodenext10survey24 |
|
We make the distinction between local OS vs production OS, but then the next question for architecture does not make that distinction. I think we need it there too. For example, macOS and arm for local vs Linux and x84 for production. TLDRChange this question:
Into two questions:
|
|
https://x.com/justinfagnani/status/1783231031396721131 I thought this was good feedback. Is this something we can change while it is inflight? |
|
@kyliewd thanks for all of your help on the survey so far. |
|
This survey is already out in the field and gathering responses. We will not be making anymore changes as that messes with the data. |
|
|
|
EOW Update: 494 responses. I will update y'all weekly on the numbers. Keep promoting / telling your communities. The survey will close on May 24. |
|
Survey update: We have 1,612 responses. Please take the survey + tell your collaborator friends! |
|
I told my collaborator friends and they said Taiwan and Hong Kong are not in the country list 💀 |
And yet the list includes some questionable options:
The list in the survey appears to be derived from this list of all countries with which the United States has ever had diplomatic relations, which would explain why an earlier version excluded the United States. A better list might be ISO 3166: https://en.wikipedia.org/wiki/List_of_ISO_3166_country_codes |
This has been adjusted. |
|
we are at 2,312 responses for the survey. 🎉 A little less than 2 weeks until the survey closes. Thanks for your promotion and help here. |
|
survey closes this friday! we're at 2,731 responses 🔥 |
|
Survey went out and closed, so landing this PR |
|
Are the survey data available yet? |
|
We just finished the review/sanitization of the open ended questions and have the ok from the Foundation to add to the repo. @marco-ippolito were you going to do that? |
Yes I'll open a PR |
Fixes: #250
I've copy pasted 2023 survey, feel free to add/remove question