Building a Self-Hostable Product

> We do something on the IoT side with our local first air quality monitors. So no forced platform lock in or subscriptions.

Took a quick look at your site (from your profile). Are you charging a one-time fee for the monitor, then? How often do people need new ones? Is there data reporting that can be sent to your servers or elsewhere? How do you avoid running into a situation where you are losing money supporting a 10 year old air monitor?

> Wish you all the best with your business model.

Thanks! We've definitely found it to be a differentiator for certain types of customers.

> didn't we once call these desktop applications

I struggle with the name. Self-hostable, downloadable, BYOS (bring your own server). Desktop doesn't feel right, at least for our product, because in prod, it is almost always running on a remote server.

Little late to the thread, but after reading this I wanted to come back when I had a good minute and look at the AirGradient. I was in the market for exactly this product not that long ago and remember spending time looking at AirGradient's site. Wanted to try and figure out why I spent more than twice as much on something else. After looking, I have some totally unsolicited feedback for you (and what's the internet for after all if not random strangers giving their unsolicited opinions).

I ended up with a monitor from air-q.com (only providing the URL because there's some serious name collision there; not a recommendation and no affiliation). As far as the local first / no subscription / etc, they check most of the same boxes. The air-q can be set up and operated fully offline, stores historical data directly on the device so the phone app can still render the nice historical graphs without any cloud services, optional cloud component, etc. It's not open source and the official API documentation is gated behind purchasing their "Science" or "Industry" models, but my hard requirement was "can run fully offline and works with Home Assistant", not "I can turn this into yet another programming project".

Going back and comparing, I find myself a little confused about the particulate measurements on the AirGradient One (`/indoor/`)----the "Key Characteristics" diagram lists PM1/2.5/10 (which makes sense given that's roughly what the PMS5003 measures and I think the same or similar sensor is used in the air-q), however nowhere else in the page is anything but PM2.5 listed. I could see simplifying for the marketing copy, but even the "Technical Data" and linked spec sheet PDF seem to mention nothing but PM2.5.

I think I remember a large part of why I went with the air-q was the additional sensors, and it looks like the AirGradient might not be missing the PM1/10 sensors like I thought it was? If that's the case, I'd maybe surface that _somewhere_ obvious. I'm not saying to copy air-q's site[0], but when I was comparing and researching they made it _really_ easy with the way they list exactly the things the device measures/derives front and center in an easily-digestible way.

The other major gap I'm seeing is that there's no carbon monoxide measurement. That might be a more regional thing given how many gas appliances and fixtures and general combustion we have going on in our homes here, but I'd be hard pressed to pick up an indoor air monitor with no CO monitoring at all. I have dedicated CO alarms, but they're like a smoke detector--by the time that triggers, you're already in shit. I run both because the air-q is not a safety device, but I would like to be able to see if we're constantly floating just below a limit somewhere (especially with a small child around, because that probably means we're over the limit for her).

Totally understand another $20 sensor might not fit on the BOM, but if there were an option to buy it separately and install by opening the AirGradient and plugging a cable in or something I'd have no issue with that. (I understand I could try and tackle that myself but... I've already got a project list growing faster than I can knock things off of it.)

Anyway, again, totally unsolicited feedback, but my money's sort of where my mouth is in that I spent like $600 in this space not all that long ago. Barring the two issues above (one of which could be quite easy to fix it seems if it is just a copy issue), I think I would've grabbed two AirGradients over one air-q both on cost and on supporting open source / open projects. I may still grab one or two just to get better coverage around the house here now that I already have at least one CO monitor.

Cheers!

[0] https://shop.air-q.com/air-Q-basic-air-analyser-10-sensors

Thanks for using FusionAuth, and thanks for the feedback!