Show HN: Polar – an offline web browser with annotations and tagging

Right now if I want to play through a game with a walkthrough (usually HTML or text only), I load it into my tablet and leave it in a browser window. This is an inefficient way of doing things, because the page generally sits there for about a month as I'm slowly progressing through the game, and very often the browser will forget its place if the tab loses focus or if the tablet goes to sleep.

This will also probably be much better for going through Programming Books than my normal e-reader app.

I feel like there was a missed UI opportunity with the pagemarks and annotations: When you add your page marks, the percentage bar at the top increments, but there is no easy way to jump to where you were last reading. Likewise, if you add multiple pagemarks, if e.g. you've read multiple sections, there is no way to know this from the progress bar.

Instead of highlighting the progress bar at the top, I would just highlight the scrollbar on the right. This would (1) free up space at the top, (2) make it trivial to jump to where I was reading last, because I would just click on that spot on the scroll bar, (3) make it obvious if I have marked multiple page marks, and (4) open that pattern up for other uses, such as marking on the scroll bar where I have made annotations (which are presumably very important passages).

I am thinking of the scrollbar in an IDE. For instance, in VS Code, I can see all the lines with errors etc directly on the scroll bar, and I can click on those marks to instantly jump to that section.

We created a web based multiuser web annotation and discussion system, for inline annotations around ‘99-00, then the dotcom bubble burst and we had to shut down. I have been looking for something along those lines since then. There have been several attempts by others but it seems that the alternatives (PDF, screenshots, email etc) are hard to displace.

Regardless of whether it ends up being something I start using, I'm very glad to see people working in this direction. Thanks a ton for sharing.

One question - since this is Electron, it's based off Chrome's engine, right? Does that mean when you talk about adding 3rd party extensions you're including Chrome extensions in that? Having a hackable browser that I can directly mess with gets rid of the need for me to have a lot of extensions -- but there are a still a few (uMatrix, uBlock Origin) that would be a huge benefit. Even when I'm just downloading a document for offline access, I still don't really want to have a bunch of trackers in the stuff I access.

I've got some ~150 tabs I'm holding onto just because I'm not ready to archive them into bookmarks yet, just waiting to be moved into this thing. If this thing works as well as I'm imagining, this is really going to fix my shit (and probably save me tons of ram)

Edit: A couple issues I've noticed from a few minutes of usage however, at least from my interest of using it as an archive: there doesn't appear to be a proper way to retrieve the original url of a document (important for sharing, especially to non-polar users), and its not clear if the polar archival format stores it in the first place. And ofc, a convenient way to find the polar archive, for sharing to polar users

There also doesn't appear to be a proper way to edit & save a polar archive (important for archiving; cleaning up). You can't edit properly pre-fetching the webpage, because of the webview/js trick you're doing, and you can edit but not save when viewing the archive post-fetch. It would also be a useful bandaid until readability or something is attached properly.

Since these all appear to be absent from roadmap, I'm wondering if the project has (my) interest as a shareable archival system, or is it intended primarily for personal usage? In terms of my own interest, I'd be looking to replace bookmarks & tabs with this: the offline archive has both value for speed/convenience, but also the defense against site-decay; but bookmarks/tabs also operate as my reference for sharing, and the site would be preferred, with polar second, and webarchive last.

With the current (roadmap-planned) setup, I'd need to maintain both systems: bookmarks for (slowly) decaying, shareable archiving, and polar for personal, long-term archiving/reading. And probably a third system for proper, shareable, long-term archiving (ie webarchive).

Thanks for the hard work!

We will double down and investigate web standards once Polar is a bit more solid. Right now it's just extra work when I haven't validated that Polar is going to be useful to 100s of thousands of people.

We're going to re-write the viewer backend at some point and once that's done I think ePub will be straight forward.

We will ship a chrome extension so you can do one page sync between Polar and your browser.

We're going to use Firebase for the auth and sending of this data between instances.

> - The "inspect element" menu in the editor seems a bit unnecessary, as I don't really think of PDFs as HTML. I guess this may be nice for some folks who are familiar with HTML but for me its a bit confusing.

I will probably remove it by default and add a 'developer' mode which turns these on.

> - I had some performance issues when resizing the page on a longer pdf (140ish pages), crashed and then wouldn't reopen until I restarted the application

That's weird. It's usually pretty solid for me.

They are apparently broken due to a bug in Ubuntu's snap but I don't have a work around.

Could you try to install canberra-gtk-module and let me know if the problem goes away?