https://github.com/d3/d3/blob/master/CHANGES.md
Basic outline:
The question(s) below are based on this quote (from the repo's README, of course): "The master branch currently contains the prerelease of D3 4.0. The 4.0 API is not yet frozen and may change prior to the release.",
Respectfully, teams that are planning to use D3 for a major or long-term project would appear to be in the awkward to difficult position of having to chose between: a) committing to a stable/proven (but probably about-to-become-deprecated) codebase, e.g., version 3.*; OR b) to choose version 4, but with the cost of a changed/new API, and at the risk of un-specified (perhaps un-predictable) amounts of API instability and/or the inevitable risk of breakage/bugs that a re-write entails.
It looks like version 4 is up to alpha build 49, but I don't have a good feel for what that implies w.r.t. version for getting to a stable state (and, ideally, also to a functionally complete state).
For teams in the above situation (like mine), it would be very helpful if you could provide even rough guidance as to where version 4 stands in terms of 1) API stability, and 2) feature completeness. Ideally, it would also be GREAT to have an idea of how far out (in time) to expect a 'stable/complete' release, at least something that could be considered in 'beta' state.
If you think that's an inappropriate question to ask an open source project (just say so), I can respect that viewpoint. Otherwise, any guidance you can provide would be sincerely and greatly appreciated !
I鈥檓 planning on releasing 4.0 at the end of this month, so there鈥檚 not much that will change between now and release. Many of the modules (see #2849) have already had major releases (1.0 or higher), and so there won鈥檛 be any backwards-compatible changes in those released modules between now and 4.0. And in most cases, the ones that haven鈥檛 been released yet it鈥檚 simply because I need to write the release notes, not because I plan on changing anything. The primary remaining development effort is limited to d3-geo and d3-chord, which have not yet been fully implemented.
@courtagenjim If it's critical to develop from a stable API in the short term, sticking with version 3 is a good choice. Then if you want to upgrade to v4 at an appropriate time to get new features after the API is complete and has stabilized, you can do it on your own schedule. This reference should be useful when you make the switch:
AWESOME answers: on the 4.0 release timeframe, and on the feature completeness. Also appreciate the rapid response to my query. Kudos!
馃帀
Most helpful comment
Basic outline: