Self-Hosted Main

In case anyone from Budibase ever sees this, I did want to say that I understand needing to make money from what you do. I don't have a problem with open source projects having an enterprise version with additional features as long as they don't consider basic SSO to be an enterprise feature.

Lots of other features like auditing, group and role based access controls, commercial database drivers for Oracle and others, high availability, multi user collaboration, rebranding, live deployments, project management features, etc make sense being enterprise only features.

If your self hosting and need all that other stuff then you probably should be paying for the software. Plenty of other projects like GitLab and Coder work this way and the open source version isn't limited in a way that affects hobbyist.

Budibase advertised their free/oss version as being unlimited users since the very beginning and it's a real stab in the back for anyone who's contributed to the project.

Get rid of Budibase and use Windmill.dev instead. Back when I evaluated Budibase, it was horribly buggy and unusable. Windmill.dev is full-featured, self-hosted, is pretty darn stable, and AFAIK doesn't have these user limitations.

I’m using the Budibase community plan, and my account has over 20 users. What is going to happen to my account and users?

- You will not be able to add new users (if you’re over the 20 user limit). To add new users, you will need to upgrade to a paid plan.

Maybe you are able to create a bunch of dummy users now (like 50 ghost accounts with various emails) and if you want to add a new user, you just edit one of the already created ghost accounts. Is that an option? Otherwise look out for alternatives or pay the price.

since it's open source, one could just remove the limitation.

Then it is not "OSS Self Hosted"... Thank them for the clarification !

Checkout saltcorn.com

The enshitification begins …

limited to 20 users even on the open source version

Lol

sed -i 's/20/9999999999/g' *

Well, it was predictable because they changed their pricing several times this year. With REI3 you will never face such limitations for the free version.

Its one of those firebase clones?
We used one called directus and they just changed license to BSL not allowing us to use it anymore.
Typical - build up the sheeple base that trust you then start shaving the sheep

i looked into a few of these.

some names from memory : supabase, baserow, nocodb, ragic, saltcorn.

can't immediately swear which are self hostable.

use UrlyUp for free to a local version of buddybase, https://urlyup.com

Going through the comments and alternatives, seems like the only option without limits on users if you want to use openid/oath SSO is Directus. Is that it? :/

This is shame for open source project

This is a real bummer, particularly as we started using budibase for our volunteer-run organization back in the spring specifically because the BB people promoted it as being a less-restrictive alternative to all the other low-code app builders. In fact, having unlimited users was specifically mentioned here on reddit by a co-founder a mere 7 months ago!
It's also a drag because as nice as I think the platform is, it's still buggy; it comes with the OSS territory and was tolerable because updates were relatively quick in coming, but now we can't update 🤷‍♂️.

I get that these companies need to make money but they benefit from an open-source community and the least they can do is make the self-hosted versions relatively open. Putting enterprise-level features behind a paywall is all well and good but limiting user count on a self-hosted version is very uncool, particularly given the pricing. There's just a huge gap in coverage here... clearly we're not the only non-profit/volunteer organization using the platform. We don't need enterprise features, but we have more than 20 people who need to use the app we built. It's a crazy expectation that small orgs start paying $150+/month for the platform just to get more than 20 users on a self-hosted install.

Several folks have asked why not just fork it? And I think that would be a great idea however like most of these projects, it's not fully open source and thus all the source code isn't available. Large parts of what makes the app function are only available as already packaged libraries in npm. All of that would have to be deciphered and re-written.