Textus: An open-source platform for collaborating around collections of texts
textusproject.org
textusproject.org
Seriously, I spent a lot of time on this "digital humanities" (how I hate the term) stuff in the latter part of the last decade, and I came away from it with one major takeaway: when the grad students who are working on and advocating for a DH project move on to their next gig, then it's up to a professor or two to constantly recruit the next generation of grad students to replace them and make this piece of software Their Thing that they can put on a CV and get tenure with. But it never really pans out, even if there's some money behind it, because if you want to stand out in academia and move up then you have to do Your Own Thing and not Somebody Else's Thing.
Making matters much worse if the fact that few professors outside of CS departments are equipped to understand why what you're doing is The Future and you deserve to be hired into a humanities department for it. And to the profs within a CS Dept., it's really clear that while what you're doing may matter for the humanities you're not solving a Hard Computing Problem so shop it somewhere else please. So the project eventually goes in the archives, or becomes open-source abandonware. But one or two cohorts of grade students get a few journal articles out of it in specialized DH organs that nobody reads, so all is not totally lost.
What I'm describing here is endemic to the academy, so it's not like you can fix it by waving the "open source" wand over a project, or throwing money at it.
Anyway, I won't ramble more because I once gave a talk on this topic called "Why Most Digital Humanities Computing Projects are a Waste of Time". You can see it on YouTube: https://www.youtube.com/watch?v=7OeC8qGT8DI
Again, the TL;DR is this: if you want to collaborate and annotate texts with the ability to cite specific versions, track changes, etc., then use a commercial or open-source version control platform that is actively maintained.
Still, I'm unaware of any VCS that actually does half of what many of people want to do — in diachronic linguistics you often have to deal with variation in source texts that /isn't/ semantically important (Romeo & Juliet v. Romeo & Iuliet, for example), and then doing n-way diffs between various copies of the "same" text. You don't want every single i/j or u/v to be flagged up, because the signal:noise ratio would be astronomical. Lots of different groups have lots of very subtly different requirements, and basically nothing off-the-shelf works well.
However I don't know of anything with the advanced functionalities you describe.
Edit: I found https://github.com/okfn/textus-viewer and https://github.com/okfn/textus-wordpress from 2014 so somebody is still working on it.
Textus are targeting text, hypothes.is are "building an open platform for the collaborative evaluation of knowledge" -- i.e. annotating the web.
"Re other Open Annotation, we obviously are very excited about annotatorjs (which we created) and hypothes.is. I would reiterate that for static texts in the Textus structure, the whole point is you can do annotation in a much easier and nicer way (you don't need to run off addressing into the DOM) and that's what Textus' viewer takes advantage of. "
We're in the process of migrating some of the older content to this URL and would love any interested parties to contribute their skills.