I wish I could leverage such polished interfaces for my research group. But, we have lot of contracts that bind us to keep our research data in house. We cannot simply "run something in the cloud".
So, Jupyterhub and manual tinkering to get such polish for now.
While I can't say anything definitive or give a timeline we do want to support research groups like yours. Ideally we'd be able to have our paid offering for companies using Nextjournal in private subsidize our open science/source offering.
We also definitely want to open source parts of our product but we haven't figured out what parts (or everything) and under what license.
Our priority is currently on providing a useful hosted product and become sustainable. It's certainly also interesting to see how e.g. metabase is doing it the other way around, open source first without a hosted product but I guess I'm a bit scared of not being ready for developing Nextjournal in the open at this point in terms of bandwidth and keeping things backwards compatible.
Completely fair. Your hard work and your product - you should turn it into a sustainable business as you see fit. I was only casually commenting to share my personal thought that it would be a great time saver for me as a sysadmin to "just" use something like Your product instead of spending hours tinkering with Jupyterhub to get it to work well. I'm not entitled to anything.
From my past experiences there are a lot of enterprises that are rightfully scared to let their employees use such a service and open up data regardless of what promises a SaaS company makes. There is general assumption that if the SaaS company fucks up, all we get is a "we take our security very seriously...." blog post.
So, making your product work within a corporate network without "call home" is a great advantage and immediately expands your target audience with some potentially big pockets.
You might want to check out VizierDB (a project my group is working on). It's a self-hosted multi-language notebook with versioning, branching, and snapshots; as well as a spreadsheet-like editor and provenance-based data annotations.
This is something that I want for personal use. I want to be able to control my data, and so hosting an instance (managed and paid for is not a problem) is desirable to me.
So, Jupyterhub and manual tinkering to get such polish for now.