on behalf of the OCSF, I am happy to announce the (belated) report on the responses to the OCaml Users Survey 2023. We apologize for the delay in evaluating and are committed to run the OCaml Users Survey reliably in a yearly fashion from now on.
Without further ceremony, here is the link to the report:
We welcome any feedback on the report, discussion around the responses and numbers (please share your interpretations and opinions), or other commentary.
Thank you to everyone who participated in the survey to help us better understand the state of OCaml and its ecosystem!
As @nojb clarified on the other thread: I am not working exclusively for Tarides, but helping the OCSF run and evaluate the survey.
As a volunteer at an unrelated non-profit org, I also understand very well what a conflict of interest is.
If you have any specific reason to assume that the survey is being evaluated or run in a way that disproportionately benefits Tarides’ interests, please bring up the specific concerns, so we can act on them.
The OCSF as a neutral instance runs the survey in order to provide the entire OCaml community with insights. That’s also why we are having the discussion about this years’ questions in the open, so that all companies and individuals in the community can contribute and help make the questions useful.
Never too late :) It is actually interesting seeing how much the ecosystem changed on its own in comparison with the user sentiments in the report at a longer time period than just 12 months.
It seems a lot of the ecosystem work was well-tuned to user needs already. Although there are recurrent lacks for which there doesn’t seem to be a lot of effort allocated, like debugging/printing. Typed-effects are surprisingly very in-demand. I went on and read the free-form replies and there are a lot of insights there. Notably, many seem to have requested better dune docs and that did improve since.
Very thankful for the work done to aggregate all the user responses, though I have to say, the discussion/summary at the beginning was a somewhat impenetrable wall of text with somewhat repetitive points, hoping for a more focused survey discussion this year! Again thanks for the effort <3
Maybe the right way to do this in these times we live in now is to just aggregate the data and let everyone mostly draw their own conclusions. Or just provide at most 2-3 pages of discussion.
Some might have the luxury of time to go and read everything like I did, but I imagine the majority would actually want summarized and cutting insights.
The discussion just repeated a lot of what the text underneath each section was revealing, with numbers. Maybe overall trends and sentiments are sufficient? I don’t really know. And to the possibility of sounding like I’m backseat-driving, I will refrain from making suggestions on how it “should” be done when I’m just not aware of the process that goes into this. This is just my reaction as a reader.
Maybe the right way to do this in these times we live in now is to just aggregate the data and let everyone mostly draw their own conclusions.
At these times as in with the help of LLMs? Personally I don’t really trust LLMs to give me deep insights from longform text. They’re too lossy, too verbose, and too shallow. I am much more trusting of conclusions made by someone who read through the text, with all their potential biases and everything! Still a better deal to me.
This is super helpful feedback, thanks for that. Also yeah, we’ll all be opinionated, and at least for myself, my world has been shaken with LLMs suddenly appearing more capable at some things (by far not all) than I am.
So definitely helps to hear that people are still reading things in a traditional way and that the amount of time spent on making analyses is worth it. Just in a more condensed fashion.
