we are delighted to announce the OCaml User Survey 2022. With this survey, the OCSF is trying to get a better picture of the OCaml community and its needs. It would be very helpful if you could take a few minutes (10 to 15) to fill the survey and share it with other OCaml programmers.
The survey is run by the OCaml Software Foundation. It builds on the previous iteration issued in 2020. The results will be published here on discuss and on the website of the OCSF. We would like to particularly thank @cjr for his help as well as everyone who commented on the previous survey. We tried our best to take all remarks into account but surely missed something. Don’t hesitate to give us your feedback (you can post here or send me a message/email).
The survey will remain opened until March 11th 2022 (AOE).
The survey has quite a few typos, I haven’t found a way to report them through the survey itself, but could you please double check it and fix the typos?
Also seems odd to have the company usage trend question be multiple choice.
And there’s no longer an active project called BuckleScript, although it’s of course still possible to use it in legacy projects. There are two successor projects to BuckleScript, the official ReScript project which is moving further away from OCaml, and Melange, which is moving closer to OCaml.
Bah, Perl isn’t in the other languages.
Also I know how I first heard about OCaml but none of the choices are even close so I chose “Do not recall”.
I agree that “Reason with/without Bucklescript” is confusing. For instance I used Bucklescript (now Mélange) with OCaml syntax, which can’t really be described in these answers so I selected none of those, just vanilla.
yes it’s unfortunate. I’m unsure whether modifying an open survey invalidates the URL though. If I can fix the typos without invalidating the survey URL I’ll do it, otherwise it’ll have to wait for next year.
let’s also move away from google forms next year
Why? It does the job. Is there a better platform?
well first of all it’s google
but a lot better can be done with a homebrew solution even if relying on google was a-ok.
with the custom route, it’d be easier to follow from the language branding/aesthetics, allow for more dynamic/interactive questions, the ability to roll out questions for random subsets of respondents, better data representation chances when the survey is over, etc…
honestly it’s in one part aesthetics and in another google mistrust.
it is unfortunate that we used GoogleForms but it was the most practical this time. I tried FramaForms but the interface was definitely clunky, there was no easy way to import the questions in whatever format (including their validation action) and adding those validation actions (like limiting the number of check-boxes or the length of text fields) was not really intuitive either. I was also impossible to randomize the choices of some answers.
Other providers have a free tier that either limit the number of respondents or do some form of data retention (impossibility to export). Their subscription model is also not geared toward our use case of one survey/year. But if there are strong feelings against using GoogleForms, we can try another provider (we are very much opened to suggestions).
The homebrew way demands a lot more work, in particular securing the web site and making sure it is accessible.
It’s just not worth the effort IMO. It’s just a survey. Google provides an easy platform. People can create a throwaway account if they’re really worried.
Let’s make it clear that you don’t need an account to take to survey – I don’t have one and took it.
Maybe people should take the survey in their browser private mode if they are worried.
I’m not sure I understand people’s concern about using GoogleForm here, is there something I should be made aware of ?
indeed, we made a point of writing a disclamer before the first question that the reference you see to your google account is only to save the progress within your account (it’s something provided by Google that we cannot disable). But one can use the private mode, hide behind a vpn or what not.
I think the problem is “just” to not empower Google more than it already is and a defacto stance to avoid their services as much as possible. I (talking as myself and not a representative of the OCSF) think it’s good to explore alternative solutions (hey I make a point of not using Chrome) and will suggest that for next year survey, but at the end of the day it amounts to the time people willing to do the ground work have.
Yep. Google will try to link your account with the survey if you are already signed in to your Google account. But you can fill the survey ‘anonymously’ if you sign out or use a fresh private mode.
I deploy my OCaml programs using 0install which is written in OCaml itself. It would really nice if someone could add this option for the next survey.
I haven’t followed what is the relationship of the OCSF with Inria lately, but FYI Inria provides a survey platform based on Limesurvey (the most advanced open source survey tool). We have used it successfully for the Coq Community Survey 2022. AFAICT, this is also the survey software used by the NixOS Community Survey 2022 (but they self-host).
About Framaforms, FTR here are my notes while looking for which survey software to use for the Coq Community Survey 2022:
Note that after reading this blog Forget about Framaforms-the-software, make room for Yakforms! – Framablog and the reference it gives on why Framasoft created a new survey software En savoir (un peu) plus sur le projet Framaforms – Framablog, I came to the conclusion that for our purpose, in terms of free software, there is no point looking further than Limesurvey.
Here is what the creator of Yakforms has to say about Limesurvey (translation mine):
Limesurvey is perfect for complex surveys, but not well suited to people who just want to create a small form in less than 5 minutes.
How did you first hear about OCaml ?
o Social media (LinkedIn, Facebook, Twitter)
Reading articles on the websites before there was social media, except for Usenet!
Indeed! I found out about OCaml in Rosetta Code and the Debian-hosted Computer Language Benchmarks Game when I was actively looking for the best language for my criteria.* I couldn’t find how to convey that in my answer so I used “I don’t recall”. I’d love for that question to be expanded for next year.
*Yes, OCaml won beating Kotlin, Rust, Go and TypeScript. I’m going on 10K lines and I don’t regret that choice.
I am outraged that Nix is not mentioned in the “Which installation methods do you use?” but npm is (and there is no “Other” option). Well, in my case I could check “Distribution’s package manager” since I’m on NixOS, but this is not the case of all Nix users, and confusing Nix for a standard distribution’s package manager is quite ridiculous anyway given its large OCaml package collection.
Other remark about:
If I was granted up to three new language features today, I would ask for (no choice means “no new feature, please!”)
Well, how do I express “I don’t know” then? I guess I’ll be counted in the “no new feature, please!” category…