harrisonlo

It's a valid point. From a technical standpoint, I'd say over time the number of feature implementations that can be re-used across modules increases. You're right for now that nino has yet matched all the feature sets, but the path to reach there might not be as linear as you think.

It's on GCP and 99% "serverless" with TypeScript + React frontend. Flutter for mobile apps and Electron for desktop apps (planning to switch the desktop apps to Flutter too once their desktop frameworks are more stable)

Noted on the "local-first" term, thanks!

I thought about doing CRDT in the beginning, but because not every block type is text, I simplified the implementation to last write wins. Note that every paragraph is a block, so it's not like the entire page is overwritten. CRDT is still doable though.

There are real-time text cursors and block selections, but I intentionally did not implement Figma-style floating cursors that people may expect in certain modules. I thought it looks cool but doesn't offer much value, especially when you can already see what other people selected.

You can export everything in 1-click! It only supports JSON, HTML, CSV for now tho. PDF support will come at one point.

On your first point: I think the trend is more will start to care. I agree many platforms provide a lot of security features, but few allow you to encrypt with custom keys though.

On your second point: That's a good point, bad phrasing on my part. Thanks for the feedback.

I've thought about email - it'll likely be a module that syncs with your outlook or gmail. This is one area that is very hard to replace, because the cost of switching email is huge. Once the emails are synced to nino, they become blocks and can interoperate with other modules i.e. crm type of use cases

A lot of features you mentioned are actually implemented (role-based permission, cross workspace sharing, domain hosting...) and some that will come (audit logs and other enterprise things).

I get the best-of-breed argument. Nino's thesis is that people will find more value if enough tools are bundled in one place.

Thanks! Data sovereignty is also top of my mind, but I think we should approach it not with the physical location of the servers/databases, but with the ability to encrypt, thus own, your data.

Currently it uses GCP with data centers in the US, but when E2EE is rolled out, I don't think it would matter where it's located? There might be legal complications so I'm also thinking of moving the physical databases to jurisdictions like Switzerland, if it comes to that.

Given the complexity of the backend, I think a self-hosted version might not be possible in the short term. However, single tenant versions (like GitLab dedicated) is definitely doable.

Thanks for checking it out! Noted on the documentation shortage, focusing on building now but that can definitely improve. Specific import/export features are not added yet, although you can simply paste in HTML or get the output HTML from the document's public page. I'll think about adding that feature, but it'll probably just be HTML or PDF, not to specific services. The pricing is tentative, tho it is aimed towards professionals who may use it for hours a day, and not necessarily for normal consumers who may just want to take class notes. And thanks for the interest!