This is seriously cool and might do the things I wanted Perkeep to do[1] and more.

[1]: but never could get it to because documentation seems to be close to non existant.

This was my first thought exactly.

I _love_ the idea and tech behind Perkeep. But holy shitballs the documentation and UX are ...imperfect.

This might be the midway between "bunch of markdown files" and Perkeep for me.

Hmm... It seems that I should finally stop using (good but too generic) DokuWiki engine in favour of this.

What interests me the most is an advertised ability to bridge self-hosted Wiki (did I get it right?) for taking any kind of notes with web-bookmarking and ability to perform a good-quality search over the _content_ of bookmarked pages. This is where I'm having a real pain. There are so much great useful stuff on the Internet, but it so hard to find some very specific thing that you know exists and you've even bookmarked it long ago... So if it'd solve that pain and also features a comparable to DokuWiki notes engine... I think, it'd be an absolutely great thing! I'll give it a try.

BTW: does it support importing existing bookmarks from Firefox?

Is there a privacy statement? Given open source PIMS that variously phone home (when started, or when visiting assorted websites, etc), it's now something I look for, right after self hosting.

Oh there's nothing of the sort with Archivy. No information is sent back and there isn't even a server where it could be sent to really.

I haven't looked into redacting a privacy statement, might be a good idea :)

This is a great project!

I think you should put in a simple privacy statement that comes down to "Archivy won't make any external requests" if it's going to be as simple as that, and then stick to it like glue. If it's not that simple, be accurate to whatever room you'd like to leave yourself, and then stick to that.

Not having a statement like that leaves the future open for a future git pull to change the landscape. Technically the statement won't stop that, but it'll keep the project accountable.

FWIW it's way more important for this sort of thing because it's a stack you have to invest lots of time and notes into and buy into long term. Were I going to run a personal wiki, I'd be very interested in knowing I could likely use it for a long time.

May not be archivy directly. But for example if the web pages use static assets like google fonts, jquery, they will send external requests thus leaking the presence of the self hosted server. Or may be some javascript plugin that phones home.

Any plans to allow references / citations?

cite-proc or bibtex style reference processing would make this very useful for keep a research/science log

Is there a specific, technical, reason, why you don't use an XML database with, let's say, an XQuery frontend for storing and retrieving document data?