Warning - long (and very ranty) post. All you TLDR people can just keep scrolling. Thank you.
I’ve held off on commenting until the actual Nabu Casa staff was involved in providing some feedback, if not justification for why this was done.
What stands out is a tone-deaf response to how you should not deploy breaking changes to existing functionality, then frame them as a new feature which “will continue to be improved according to feedback and usage over the releases”:
Those who have been around for more than a year are starting to see a pattern where a decision is taken behind closed doors, then all the user feedback in the world will not convince the powers-that-be not to deploy it.
We’ve seen it happen with scan interval, weather forecasts, badges and now, backups. Each time there was vocal opposition during the beta phase. People tried to warn the devs in discord, github and on this very forum before the changes were released. They were consistently dismissed and ignored.
It’s like deja-vu all over again, but given the fact this is becoming an obvious pattern even more so in the past year makes it worrying for all users, not just the ones who have been around long enough.
People with more eloquence and experience than me have already explained that this is not what MVP is about. MVP only really applies to absolutely brand new features which do not impinge on existing functionality.
Changing something as fundamental as backup (formerly known as snapshots) which existed since before year.month versioning was introduced does not qualify as a new feature.
There are multiple red flags with that part of the message.
Yes, some of us have been here long enough to remember all these breaking changes framed as improvements. Others are still suffering the consequences of those changes (I hear card-mod just broke while trying to make badges customisable again. Remind me how “beautiful” that particular clusterfudge was please).
The “listen to feedback” part worries me too. We just replied to a survey in the past month, asking us what we think about HA’s values and commitment to privacy, sustainability, and choice (my emphasis).
Instead of being asked whether I would like my backups encrypted by default, with no easy option to opt-out or decrypt them, I was asked what gender I identify as and what religion I ascribe to.
The “it’s not us vs them situation” part worries me too, because it sure as hell starting to feel this way. Unless whoever calls the shots starts listening to genuine concerns before changes to existing functionality are rolled out, then people would be right in assuming there is actually an us vs them mentality.
Yes, which is why historically, the January release was either entirely skipped, or else extremely limited in scope in the past couple of years.
What follows, however, is the most tone-deaf part of the staff replies:
Honestly, why are these questions being asked now, after the release? Shouldn’t they have been asked before such a breaking backward-incompatible change? Like, you know, as part of the survey?
Also, they are entirely irrelevant. Like others have said, just because you never needed your seatbelts or airbags in the past doesn’t mean you don’t need them to work without any extra intervention on your part sometime in the future.
Great, so let’s just frame the fact that the downstream integrations haven’t caught up (yet) as something which should be applauded. Giving back the choice to do something as fundamental as an unencrypted backup “over the coming releases” doesn’t really cut it.
In the meantime, we’ll have to create yet another automation in order to restore functionality which we had out of the box a week ago. At this point, we should be used to it I guess.