I’m pleased to announce that NLNet has accepted my grant proposal for pimsync. The main goal of this grant and associated work is to port all features from vdirsyncer which are still missing from pimsync.
For my personal use case, pimsync works fine, and it’s been a great “vdirsyncer version 2”. In particular, I find the daemon mode of enormous value. In case a server rejects a single calendar event or contact, pimsync will also continue and sync everything else. These are the two main features which have made a difference for me. I can mostly run it in the background and forget about it
For other use cases, some features are still missing, and I’ll be working on addressing these in the coming months.
By the end of this project, all users should be able to move to pimsync. Of course, nobody’s force to migrate, and users happy with vdirsyncer can continue using it, although support will wind down substantially. That said, I hope the improvements in pimsync are attractive for most use cases.
With the Python ecosystem being in a perpetual state of change and backwards incompatibility, I suspect that in a few years vdirsyncer will gradually become harder to run: functionality or libraries on which it relies will slowly stop working on newer versions of Python. To be frank, I find this quite disappointing, even if I’m sun-setting the project.
The man page pimsync-migration(7) (also available online) contains a
list of all missing features. If you’re using any functionality which is missing
in pimsync and also missing from this list, please reach out by opening an
issue or send an email to the list so I can bring it into the
roadmap.
Beyond porting vdirsyncer features, I don’t have any major plans for pimsync, and I expect it to enter a long period of stability, with minor fixes. I do have several other related projects (with good synergy but low coupling) in mind.