A lot of work has been done since the initial alpha release that was made. It's probably worth it, rather than wallowing in a pre-release mire forever, to just commit to a 0.1.0 final release. Especially with the recent documentation improvements in #213 I think it's safe to call the general interface stable. Originally I wanted to fiddle more with the API as part of #180 but from discussion with @braingram the conclusion seems to be that it's not worth overturning the apple cart for right now (but could be provided later as an API extension). That, to my mind, was the main blocker to committing to an 0.1.0 stable release.
If, after the initial 0.1.0 release is tagged we still encounter build issues with packagers, that can certainly be handled with patch or post-releases. Though I would probably like to revive and test at least the conda-forge build before committing to 0.1.0, just to test against, then bump that to 0.1.0 as well.
A lot of work has been done since the initial alpha release that was made. It's probably worth it, rather than wallowing in a pre-release mire forever, to just commit to a 0.1.0 final release. Especially with the recent documentation improvements in #213 I think it's safe to call the general interface stable. Originally I wanted to fiddle more with the API as part of #180 but from discussion with @braingram the conclusion seems to be that it's not worth overturning the apple cart for right now (but could be provided later as an API extension). That, to my mind, was the main blocker to committing to an 0.1.0 stable release.
If, after the initial 0.1.0 release is tagged we still encounter build issues with packagers, that can certainly be handled with patch or post-releases. Though I would probably like to revive and test at least the conda-forge build before committing to 0.1.0, just to test against, then bump that to 0.1.0 as well.