Thank you for the question and the insights, very helpful!

We are still working on the monetization strategy for Colanode and there are multiple ways we can (and probably will) do it. Before answering with more details, I would like to make sure I understand your perspective correctly. For the pay-per-version approach how would that work? You want to self-host a specific version and use it? Where does our part come? Will we manage the infrastructure? Is it our infrastructure or yours and we just make sure it runs? Or am I wrong?

(might be a bit of a rant)

I personally like the approach immich takes ( https://github.com/immich-app/immich ). You pay one time fee, but basically don't get much else than warm fuzzies (it's basically a donation, but shows you a nice badge in the instance that shows you're a supporter). However, you still need to self-host the instance.

When stuff is cloud-hosted, it's acceptable that there is a monthly fee. But even then, some monetization models are really non-sustainable. I recently cancelled a subscription for a project planning app, because:

• To get one feature (Jira integration, which didn't work that well in the end), I had to get the most expensive plan.
• I needed to pay for that plan for every single user, even if those users were basically read-only. I could not add every developer that was participating since each seat was $50/month (or even more, I'm not sure).
• ... And we didn't even have that much data or activity there.

... I think this is similar to the OP who has multiple customers and wants them to have access, but if there's nothing happening in the project (for example it's deployed, done and only bugfixes and maintenance are happening every now and then), he still needs to pay a chunk of money just to keep the lights up.

I can name a couple of other projects that have poor monetization practices (in my opinion). For example for OS the model of per core/per CPU socket tends to punish light users that want to have a reserve for an occasional spike (Proxmox - I'd be paying hundreds of dollars because I have a couple of small machines running different chores at different places. I'd like to support the project by buying at least one license, but I don't want to support this kind of licensing strategy... so I didn't). Or actively preventing users from building and deploying from source (VyOS). Or having a huge gap between the free/community/opensource tier and the paid tier (again VyOS - the cheapest paid license is $2000/year if I'm reading the current pricing correctly, which is insane for a home lab router). Other commercial projects tend to force users to pay subscription for everything and convert existing users that have perpetual licenses to subscriptions and that is also unacceptable behavior (Altium Designer, Teamviewer).

When paying for cloud hosted, I think any features should be one time purchase + maintenance fee to cover the hosting expenses (I'd say per storage used, per number of items in the database, whatever causes the highest expense).

That said, I will not be subscribing to the cloud model and will only self host if and only if the app remains fully opensource. There are already too many similar apps that do have dubious monetization models - you can self host everything, but some plugins are paid and closed source. Or become closed source eventually. The app looks very promising though and I would consider one time or even recurring donation. I already do donate to a couple of opensource projects that I use on a daily basis to keep them living.

Wanted to chip in my two cents as someone who recently found Colanode, is enthusiastically setting up a cloud hosted workspace for my small town nonprofit community, and doesn't mind contributing financially towards development of the project:

I'm very willing to pay a per-instance hosting fee. If I can pay $20/month to provide 10 different tiny grassroots orgs with the digital collaboration tools they need to work together, and help you folks pay the bills at the same time, that feels like a fantastic use of $20/month.

What would immediately turn me away is any sort of per-user pricing. In the kind of community work that I do, I need to be able to invite new people in without worrying that I'm going to be on the hook for $8/month for the rest of my life whether or not they end up using it. The biggest appeal of the Colanode model for me is knowing that our costs are only going to go up if we start using a bunch of data, not if more people get involved.

My biggest fear is that investor-driven ensh*tification creeps in over time and leads to a per-user pricing scheme coming in after we've spent years building a Colanode-based community resource. I'm going to try to mitigate that risk by learning the skills I need to self-host, but in the meantime I would love to see robust export functions implemented, as well as a commitment to stay away from per-user pricing.