The results are in. The curl user survey 2017 was up for a little over two weeks and attracted answers from a total of 513 individuals. This was a much better turnout that last year’s disappointment – thank you everyone!
The 2017 survey analysis as a 40 page PDF
This year we learned that the distribution curve for the amount of protocols people use curl for looks like this:
And the interest in getting even more protocols supported is still high, if not even very high and I think the top-most requested protocol is a bit surprising:
The outcome of the survey is the analysis document in which I’ve summarized by thoughts and added a bunch of graphs and other diagrams that illustrate the numbers. In particular compared to previous’ years results. It became a 40 page thing as I’ve tried to be detailed and also somewhat elaborate on commenting and reacting to a lot of the write-in suggestions and comments!
If you want to draw your own conclusions or just verify mine, I also offer you the following source material:
- The pristine 2017 CSV file as downloaded from Google, with all the results from the survey.
- To compare with last year, I also offer you the 2016 CSV file.
- During my the work of producing the analysis document, I imported the 2017 CSV file into libreoffice and fiddled with a lot of numbers and graphs, most of that didn’t end up in the document but you can find the raw 2017 survey libreoffice calc file and verify the outcomes or the formulas used.
Regarding your remark that “As always, I question the amount of […] TLS client certificate (35.9%) users” I found that lately of lot of big institutions are dropping VPN setup to go that route.
When TLS client certificates are involved the server stack is often more complex and under more than one responsible, and in my personal experience rarely fully tested.
So when everybody tells you ‘it works for me’ but not for you, that’s when you start sending some ‘curl -E’ around.
The cross-platform availability and no-setup aspect allows everybody to visualise the issue from a simple email, this is powerful stuff.
Regarding the curl_formadd comment, I looked at
https://github.com/curl/curl/wiki/formpost-API-redesigned and I like the idea of using several related functions instead of long and complex curl_formadd calls.
I’d offer to help implement but I don’t know C. How else can I help move this idea forward?
@Andrew: nice to hear! I would urge you to bring the discussion to the curl-library mailing list. I figure we should continue to polish the API design to make sure we cover most of what we can do with the old API. Implementing it shouldn’t be that tricky once we know what we want.