Survey results: How software ecosystems deal with breaking changes

[Chris Bogart]

 Last fall we announced on this list a survey about how and why breaking
changes are handled differently in 18 different software ecosystems.  We've
just submitted a paper to a conference about the results, and we've also
set up a site (http://breakingapis.org/survey) where you can compare how
different ecosystems perceived their values, practices of upstream and
downstream developers, and the health of their ecosystem.

Results showed some stark differences between software communities, for
example: Making updates available to end users quickly was highly valued by
Node and Ruby communities, but was considered less important to Eclipse. (
http://breakingapis.org/survey/values.html#rapid?ecos=Haskell_Stack_Stackage_,Node_js_NPM,Ruby_Rubygems)
The
Perl and Eclipse communities value stability most, while Hackage/Cabal
valued it least. (
http://breakingapis.org/survey/values.html#stability?ecos=Eclipse_plugins_,Haskell_Cabal_Hackage_,Perl_CPAN
)

Ecosystems have very different ways of dealing with versioning; for example
in Go it's common for people to sometimes make small updates without
increasing the version number. (
http://breakingapis.org/survey/upstream.html#mon_versionless_updates_eco?ecos=Go,Maven,Node_js_NPM)
 In Rust, developers more frequently chip in and help with maintenance of
upstream packages than in NuGet.(
http://breakingapis.org/survey/downstream.html#sync_liaison_code?ecos=NuGet,Rust_Cargo)


There are a lot of other results on the linked site, and we’re interested
in your impressions of the results.  Do the results make sense to
you?  What answers would you have expected?  Do you think the differences
are intentional?  If you have any thoughts about it I’ll try to keep up
with comments here, or you can also send us comments through the website.
The anonymized raw data  is also available here:
https://doi.org/10.1184/R1/5108716

We want to sincerely thank the large number of people in the Python
community who responded, and we’re eager to hear what you think!

Chris, Anna, Jim, and Christian