biff VS remy

I am currently reading many books and papers and use highlighting a lot. My life would be a lot better if I could automatically extract the highlighted text. I tried several (by now outdated) solutions like Biff (https://github.com/soulisalmed/biff ), RCU and https://remarkable-web.vercel.app/ . The first two are apparently outdated, and the last I can't connect to (and all my files are >4 MB, so manual select doesn't work).

How? Do you use https://github.com/soulisalmed/biff? Thanks!

I just want to spread awareness that an amazing tool first posted here (github here) is lying in destitution since the 2.7 real highlight update.

> For example, there's no direct switching between documents.

In case you haven't seen it already: https://github.com/ddvk/remarkable-hacks

The ddvk hacks are easy to apply and reversible. They add a bunch of gestures like instantly flipping between documents. One of my other favorites is a quick swipe gesture to switch to the last-used tool.

I don't want to annotate PDFs and then only save the annotations, but it sounds like biff would help with that if you don't mind another tool in the chain: https://github.com/soulisalmed/biff

I use mine quite a bit for annotating documents and presentations. Provided the screen size works for you (you have to be comfortable reading somewhat smaller font sizes than usual), it's quite good for how I work. As another poster pointed out, it follows a different model than most PDF readers, in that annotations are done just like on paper; you don't select text and highlight, you just highlight like you would on paper. My first reaction was that this was lame, but I've come around to feeling that it's probably the right approach for me, in that my mind is not constantly switching between reading and selection modes. Your annotations do sync back to the computer, but to get annotations out in a way that's reusable in other software, you have to use third-party software like "Biff": https://github.com/soulisalmed/biff (there are others too) Another positive thing about it is that because the interface is very sparse and you're working with it largely like paper, you don't have to devote mental load to worrying about accidentally pressing buttons onscreen or about whether palm rejection is going to work or not (at least if you're right-handed). I find that freeing.

What I do is I create backups with rsync (so "low-level" backups of the actual data the tablet is using to represent the notebooks, not just the rendered pdf). Then I use Remy to browse them if needed. (Disclaimer: I'm the developer of Remy) It currently lacks a way to export/import the notebooks in native formats (that would allow you to restore archived ones through the GUI) so if you need that you need to do it manually, which requires some basic knowledge of how the notebooks are internally stored.

reMy and RCU are the best alternative desktop clients. The former has a focus on notebooks, while the latter is focused on ease-of-use and does it all (templates/notebooks/wallpaper/and much more). Both use their own rendering engine for custom PDF export options. Neither installs anything to your tablet, so they usually work through software updates. (Disclaimer: I am the author of RCU).

You can also take backups using easy, convenient, community-written software, like RCU (which I'm the author of), reMy, reMarkable HyUtilities, rmExplorer, rmAPI, and many others found in the Awesome reMarkable list.

You could use something local, that uploads over SSH instead of the Web UI. The Web UI and rM Cloud choke on files over a few hundred MBs. reMy + 2.x firmware might be what you're looking for -- it has sync capability. (I assume OneDrive has something like a shared PC folder that you can use as the target directory.)

You could try reMy, which has its own renderer. There are more rendering programs in the Awesome reMarkable list, many of which will work with 2.15 and below--just avoid anything saying 'cloud' or 'web UI'.

Mathpix has this functionality and is cheap. I integrated it into Remy, see here for a demo (although it's a bit outdated, the app has now way more features). It's not a fully fledged integration, I was planning to use it for having a search index that could match on handwriting but had no time to implement it.

If you need support for the new v3 software update, not sure. Otherwise Remy can do that for you

I use my own Remy tool (in conjunction with rsync for backups). Unfortunately it is not working with v3 just yet but I plan to eventually add support for it, once the effort to reverse engineer the new file format settles.

You're not alone :-); we started discussing it here https://github.com/bordaigorl/remy/issues/49

But parsing and rendering are two different things. RCU used to use the remy parsing, while the developer put considererable effort into the rendering himself. Yet he wasn't satisfied (I was) and started to completely rebuild it. That's where he lost interest in RCU :-(