During the recent events around the 2025.1 release (and especially its accompanying 1300+ entries long behemoth of a forum thread), I stumbled over a few things in the relationship between Nabu Casa and Home Assistant that left me thinking. On the surface things should be pretty straightforward, but the backup topic surfaced what I would call unclarities and also areas of improvements. So I decided to write things down a bit.
Disclaimer: this is a simplified write-up, mostly based on my observations, and some available online sources from HA and NC. Itâs furthermore simplified by letting out the role of the Open Home Foundation almost entirely.
Full disclosure: I applied for a role at Nabu Casa in the past, and didnât get the job. So Iâm in no way a neutral person. This might impact my wording here and there, but Iâm still reasonably convinced that the underlying foundation of my writing holds up. Please let me know where it doesnât, seriously. I donât see this as a conflict of interest, as we all (including me ) want Home Assistant to be as awesome as possible.
Letâs start with the official wording found in https://www.nabucasa.com/about/:
âNabu Casa, Inc. commits time and resources into Home Assistant so this will be a shared success story. We want to improve Home Assistant, also for the people that are not customers of Nabu Casa, Inc.
-
We are contributing features to Home Assistant to make it easier to install, manage and be accessible to a wider audience.
-
We are responsible for hosting the Home Assistant Community so that it can remain an ad-free experience.
-
All integrations with cloud partners (Amazon, Google, future ones) will be contributed to Home Assistant so people can run their own. â
To rephrase and extend things a bit from my perspective:
-
The employees of Nabu Casa make sure that the foundations of Home Assistant are uptodate, to everyoneâs advantage. High-level architectural changes, data structure and management, event handling, etcetc. And they do the boring / âun-sexyâ stuff that usually nobody has an intrinsic interest in, while making sure that despite the open source nature of HA all the individual bits & pieces still align.
-
Nabu Casa doesnât create a âwalled gardenâ of features or (physical) products for HA. There are other ways to remotely access your HA instance, there are other ways to build voice assistants, there are other ways to create cloud backups, and you can run HA on any random Raspberry. There are alternatives available for everything that Nacu Casa offers. They just make things easier for the non-techy part of the users, increasing the size of the potential target group in the process.
-
Nabu Casa has an interest in making/keeping Home Assistant an important player in the smart home field, for multiple reasons, e.g.:
- continuous and growing revenue stream for NC to keep the foundation solid and to have a budget for keeping up with technical developments (voice, llm, âŚ)
- HA being important/big enough to âhave a seat at the tableâ when it comes to industry standards like Matter, and for initiatives like âruns with Home Assistantâ to be taken serious
Where I feel itâs getting tricky a bit: Nabu Casa in some areas is run like a normal tech company (surprise! ). Which comes with some âbest practicesâ that are not specific to Nabu Casa, but sometimes clash a bit with the very open-source community-driven vibe of Home Assistant itself.
Examples:
-
âbig bangâ releases for new products (to ensure omni-present media coverage as much as possible)
-
a certain level of upfront-secrecy to built the suspense needed for those releases
And one potential for conflict that is also not specific to Nabu Casa, but common for the combination âbig open source project PLUS an enterprise behind itâ:
- valid(!!!) monetary and capacity-related priorities and decisions that are sometimes primarily tied to the well-being of the company, but wouldnât be the first choice of the HA community
The whole topic around encrypted cloud backups seems to be the best example for that: making backups encrypted by default enables Nabu Casa to enhance their service offering, as dealing with unencrypted user data on your servers is one hell of a mess, effort- and legal-wise. Were encrypted cloud backups the most called-for requests during WTH months and in the feature request board? No. Was ist important for extending the business opportunities for Nabu Casa: quite certainly.
Where it gets extra tricky then is when the communication and participation processes are somehow trying to cater to both entities, the open-source project and the tech company.
The product team in Nabu Casa has taken quite some steps in 2024 to communicate a roadmap, and to provide updates. Still the topic of the encrypted backups is not listed in the visual overview published in November in https://www.home-assistant.io/blog/2024/11/15/roadmap-2024h2/, but only to a certain extent mentioned in writing further down in the article.
The voice hardware (which is not targeting specifically NC subscribers) made its first visual appearance in a reddit thread where someone had shared photos from an FCC approval document.
To not just nag, but to also get some clarity into things, and maybe to even improve things here and there:
How about more prominently communicating the primarily NC-driven developments as part of the overall roadmap? (e.g. as part of a fifth stream below âresearchâ which could be called âbusiness/foundationâ or something similar)
How about adding a more prominent âpreviewâ section to the end of the release party live streams? (which would surface more high level feedback on in-development topics before the beta phase)
How about an overall bigger transparency around how priority decisions for the items on the roadmap are made?
What are the ways that currently user input makes it into product and development decisions, and how can people proactively chime in if they want to?
Also: from what I see in live stream chats and facebook comments etc, HA/NC seems to have quite a good relationship with basically every important smart home youtuber etc. Is there maybe an approach that gives more insights to the community early on into things like the voice hardware, while still creating a concentrated media buzz for new fancy stuff when itâs ready to hit the stores?
If you made it that far: thanks for reading Curious about your thoughts. And the corrections on things I got factually wrong, please!