[ANN] OCaml User Survey 2020

[VPhantom]

Not to go off-topic, but how could a bar chart be less readable than a pie chart? The heights are side by side, directly comparable without any processing, and directly connected to their labels instead of having to cross-reference a legend of colors.

[mjambon]

@VPhantom

Not to go off-topic, but how could a bar chart be less readable than a pie chart?

I don’t know. I have a better perception of the size of blobs than of the length of sticks. Perception is a complex topic.

And yes, we’re clearly off-topic

[cjr]

Pie charts are fine when the number of response categories is small and there’s good disparity in response rates; many (but not all) of these meet that criteria. Pie charts also don’t impose an ordering on the response categories, which can be useful.

It sounds like @mjambon has better spatial perception than most; in general, people find it very difficult to compare the sizes of pie slices. Another problem with pie charts is when they rely upon color coding (as these do). This can make them difficult or impossible to decode for people with color blindness such as myself.

I don’t think that I know of any studies finding that pie charts are easier/faster to decode than bar charts/line graphs/histograms/etc; there are studies showing the opposite. (I published a paper on set-theoretic visualization techniques last year: https://journals.sagepub.com/doi/full/10.1177/2059799119862110)

[mjambon]

I never said pie charts should be preferred over bar charts. They clearly should not. My only intent was to point out the ineffectiveness of arguments like “this looks good to me and you’re wrong if it looks bad to you because just look at it”.

[xavierleroy]

I extracted the replies to the free-form questions and made them available as a Gist. There are so many replies that I have no idea how to exploit and summarize them!

https://gist.github.com/xavierleroy/88a22b2b8e609c46bf9d23f36521a483

[gasche]

Note: I “split” the excellent discussion by @patrickoferris as a separate topic, as it was going in the (very useful) direction of discussing broadly the ecosystem, rather than specifically the survey result. I would encourage people to post here for specific details on the survey process and results, and create new topics for discussions inspired by the survey.

Let me quote below the part of @patricoferris’ post that would be most useful to anyone interested in processing the results:

I think it’s quite important to take the results from this survey and find actionable items (or highlight ongoing efforts) which aim to improve different problems for different users. As a result I have put together some graphs trying to categorise the answers based on proficiency (beginner, intermediate, advanced and expert). These groupings are quite subjective but I thought it might offer a slightly more nuanced look into what problems users had, where they come from, what they are doing and what commonalities they share. The code is here and the HTML version of the notebook here (I’m no data-scientist nor python developer if you see glaring mistakes do raise an issue!).

The new topic has excellent discussion on Patrick’s findings from the survey data.

[xavierleroy]

Here is a summary and analysis of the survey results I wrote on behalf of the OCaml Software Foundation: https://www.dropbox.com/s/omba1d8vhljnrcn/OCaml-user-survey-2020.pdf?dl=0
Enjoy!