kg4zow

joined 1 year ago
[–] kg4zow@alien.top 1 points 11 months ago (1 children)

For me it's more about understanding what's going on inside the tablet. The only real "change" I've made in the software was renaming the "Life/oragnize" template category to "LifeOrg" in order to make room on the screen for a new "Custom" category, which I use for the custom templates I use - and that was done by editing a JSON file, not by monkey-patching the xochitl executable itself (which makes me nervous).

Other than that, everything I've done is writing programs to interact with the tablet (backups, list files, delete files which aren't cleaned up on tablets that don't connect to a cloud account, etc.), and writing documentation about what I'm learning along the way.

[–] kg4zow@alien.top 1 points 11 months ago

If you're just looking to make a backup of the entire tablet at once, and you don't need the backed-up files to be usable for anything on the computer (other than being able to restore them back to the tablet when/if needed) ...

  • You can use rsync or scp to just plain copy the files from the tablet. My rm2-backup script does this, using rsync, plus it stores the backed-up files in a way that files which didn't change from the previous backup are not copied again, they are "hard linked" to the same file in the previous backup - so if you've only edited one document, it might only download 15-20 files, but the directory it produces will still be a full backup of the entire tablet.

    The Filesystem page on the same site has more information about how the tablet stores documents, and why you can't "just download them" in a format that the software on your computer can deal with.

  • You can use RCU to back up individual documents to .rmn files, which can later be uploaded back to a reMarkable tablet.

    I've heard that RCU also has a command line interface to do this without having to use the GUI, but I haven't used it myself - mostly because it takes forever to start. Just running ./RCU --help takes 16 seconds on an M2 MacBook Air, which IMHO is "not great but not horrible" for a GUI program, but is a deal-killer for command line utilities which might be called from a script. (I understand why it takes so long, and I don't disagree with doing it that way, it's just ... as a user it's kind of irritating.)

  • I am working on a Perl script which can download the same .rmn files that RCU creates, and which has an option to download all documents at once. It isn't ready yet, mostly because I haven't had time to concentrate on it, but I have gotten as far as producing files that RCU is able to upload into a different tablet. (Plus it doesn't take 15+ seconds just to start up.)

    I want to finish it, then I want to re-write it in Golang so people don't have to deal with figuring out how to install Perl modules just to use it.

If you need the backed-up files to be usable (i.e. all files converted to PDF, with pen strokes "burned into" the PDFs and therefore not edit-able if the PDF files are uploaded back to a tablet), you can ...

  • use the built-in web interface to export them as PDF files.

  • use the reMarkable apps, if the tablet is connected to a cloud account. (I'm pretty sure they can export documents to PDF, but I've never used them so I can't say for sure from direct experience.)

  • use RCU to download them to PDF files.

[–] kg4zow@alien.top 1 points 11 months ago

The problem is the reMarkable cloud, which is actually hosted on google's EU cloud. If you never connect the tablet to a cloud account, paid or free, then the data never leaves the tablet.

With that said, reMarkable claims that their cloud service is compliant with GDPR, and is willing to execute a BAA with you, and assume shared liability in case of a data breach in their cloud servers. That doesn't guarantee that a breach will never happen - only that if a breach does happen, and it turns out to be because of an issue with their cloud service, they agree to split the liability with you.

I don't know if reMarkable also has a BAA in place with google, in case a breach is caused by an error on google's part. For me, the fact that the data isn't encrypted is enough to make me not connect to the cloud in the first place, so I have no reason to ask them about it ... however if you do ask them, I'd be very interested in their response.

[–] kg4zow@alien.top 1 points 11 months ago

The only "search feature" I use is using the "marker" tool to write titles across the tops of the pages, and eyeballing them in the "page selector" tool to find what I'm after. The way the reMarkable software handles actual text is horrible.

As for storage ... it depends on how you use it, but pen stroke files don't use a whole lot of space to begin with. I don't use the reMarkable as an "ebook reader", although I do have about a dozen PDFs that I refer to during the day for work. I also have about 120 "notebooks". The tablet's total usage is 647MB, and 574MB of that is the original PDF files I had uploaded. This is about 120 regular notebooks plus about 10-12 PDFs.

The cloud service is a mirror of what's on your tablet. It is NOT A BACKUP, and it's NOT long-term storage.

In particular, if you delete a document from the tablet, it will be deleted from the cloud as well. (And if you use a reMarkable app, or the web interface, to delete a document from the cloud, it will be deleted from your tablet as well.) This is why I find their "unlimited cloud storage" claims to be misleading - each account's cloud storage is limited to the capacity of the one tablet linked to that account. Both rM1 and rM2 have about 6.5GB available storage, so ... each cloud account has about 6.5GB available storage, no matter how they market it.

If you want to make "real" backups, there are a few choices:

  • Use the built-in web interface to download the documents as PDF files. Note that when you do this, any pen strokes will be "burned into" the PDF, and if you later upload that PDF back to the tablet, those pen strokes will be part of the background and you won't be able to edit them.

  • Use something like RCU to download the notebooks as .rmn files. You won't be able to do anything with them on the computer, but you'll be able to upload them back to a reMarkable tablet in their original format, and the pen strokes will be edit-able.

[–] kg4zow@alien.top 1 points 11 months ago (1 children)

The computer thinks it's a USB ethernet adapter. The reMarkable tablet runs a DHCP server and, when you plug in the cable, it assigns a 10.11.99.x IP to that interface on your computer.

Be careful with third-party integrations. They work by storing a copy of your credentials for (google, dropbox, or microsoft) in reMarkable's cloud servers. The tablets don't talk directly to those services, they only talk to reMarkable's cloud servers, and those talk to the outside services.

If you think your large corporation doesn't like non-encrypted USB sticks, they're really not going to like you handing your username and password to an outside company like that.

[–] kg4zow@alien.top 1 points 11 months ago

I've been using Linux almost every day since 1993, so SSH isn't "foreign" to me - in fact I've written some of my own scripts to work with the tablets.

With that said, I also paid the $12 for RCU. I can upload custom templates by hand, but RCU makes it so much easier. (And yes, it's a third party program, the developer is a lot more responsive than anybody at reMarkable, and he's been known to read this sub.)

I haven't paid for anything other than RCU itself. I've created my own templates, in fact the original reason I started the remarkable.jms1.info was to share the information about how to make your own templates. Most of the other content came later, and I have a backlog of things I want to write about but haven't had time. One of things is the rm2-cal script, which generates simple but (I think) useful "daily calendar" PDFs, similar to what I've seen other people selling on etsy. (Functionally similar, but not as pretty - I have the artistic talent of a sheet of drywall.)

[–] kg4zow@alien.top 1 points 11 months ago

If it doesn't have to be new, there may be people in this sub who have one and are looking to sell it, and might be willing to give you a better price.

I also remember seeing where somebody started a second sub which is nothing but people looking to buy and sell used reMarkable tablets and accessories. I didn't think to write down the name, maybe somebody else will see this message and reply with a link?

[–] kg4zow@alien.top 1 points 11 months ago (1 children)

Templates are image files which are used as "background images" for pages in notebooks, i.e. documents which are not PDF files. In a notebook, you can choose which template you want to use for every page, including using the same template for multiple pages. The reMarkable software comes with a collection of templates.

PDF files can be duplicated, but individual pages within a PDF file cannot. Pages within a PDF can be re-ordered, and blank pages can be inserted into the document. When you're viewing a PDF file, the original page from the PDF is the "background image" for that page. (The same is true of EPUB files, which are converted to PDF when you upload them to the tablet.)

Custom templates are not supported by reMarkable (the company), but there are ways to upload new templates to the tablet and "trick" the reMarkable software into using them. This page (which I really need to update) has more detail about this.

If you want to upload custom templates but you're not comfortable with using SSH and editing JSON files, you'll probably want to look into a third-party solution. RCU is what I use, it's not free but it's not expensive, and it does a lot more than just custom templates.

[–] kg4zow@alien.top 1 points 11 months ago (1 children)

Thanks for letting me know.

So it sounds like the documents are converted the first time they're opened under the new software, rather than as a batch process when the new OS first boots.

If you had a document that worked in 2.x and didn't open under 3.x, was it still usable after downgrading back to 2.x? If so that would seem to confirm the theory.

[–] kg4zow@alien.top 1 points 11 months ago

RCU can also upload "splash screens" (including the sleep screen, power-off screen, battery dead, overheated, and others). It uploads the files to a part of the tablet which doesn't get erased by the OS upgrade process, and "links" them to where the reMarkable software expects to find them. The first time RCU connects to a tablet after an OS upgrade, it will detect that there are splash screen images which aren't "linked", and will offer to "re-link" them automatically.

RCU can also be used to upload custom templates, and it also uploads them to the document storage filesystem and "links" them to the directory where the reMarkable software looks for them, and will offer to automatically "re-link" them after an OS upgrade. Plus it can upload and download documents, either as a PDF (the same way the built-in web interface does it) or as .rmn files, which can later be uploaded back to a reMarkable tablet with the pen strokes still edit-able.

RCU isn't free, but it isn't expensive. I paid $12 for it, and I'll be happy to renew it after a year (although I would also be happy to pay a higher price for "updates for life").

[–] kg4zow@alien.top 1 points 11 months ago (5 children)

You can't, at least not safely.

  • In 2.x, each document had a single .lines file containing the pen strokes for every page in the entire document.
  • In 3.x, the pen strokes for each page within a document are stored in separate .rm files.

The first time 3.0 boots up after being installed, if it finds any .lines files from 2.x, it will convert them to a directory full of 3.x format .rm files. Once this happens, if you were to reboot into the 2.x software, it wouldn't be able to read the pen stroke files. (Full disclosure, I've never used 2.x at all. My understanding is based on a description of the file format and the fact that reMarkable wouldn't let me upgrade from 3.0 directly to 3.5, it made me install 3.2 first.)

With that said, given the fact that the 3.5 software can't work with files from 3.0, I really doubt that 3.8 will be able to work with 2.x files at all.

I am willing to be proven wrong about this, but until/unless that happens, PLEASE BE CAREFUL. You mentioned having RCU, I highly recommend using it to back up of every document in your tablet to an .rmn file on your computer, before doing anything else. This way if I'm correct about the mess it's going to make of things, you'll at least be able to do a "factory wipe" under 2.x (which clears the user storage area) and then use RCU to restore your notebooks. (And if not, you'll only be out the time it took to make the backups.)

[–] kg4zow@alien.top 1 points 11 months ago

Guilty as charged.

Thanks for the mention, I'm glad you find the site useful. I just wish I had more time to work on it - the list of things I want to cover is probably as long as the list of things I already have covered.

 

One of the pages I've been meaning to write for remarkable.jms1.info is a list of what I've seen as "issues" with reMarkable, including the software on the tablet, the cloud service, and possibly with the company themselves.

A few days ago I sent them a problem report because one of their support pages was giving out incorrect information, i.e. that pages could be freely moved between PDFs and notebooks. I wasn't sure if telling them about it would result in the page being corrected, but it confused me when somebody replied with directions on how to find the "text tool" on the menu, and a link to a page about gestures ... neither of which had anything to do with my original report. I responded to that with what amounted to a copy of my original report.

I didn't hear anything back from that, but earlier today I got an email asking me to take a survey about my experience. I checked the web page and found that it had been updated, and while it's now technically correct, it's not very clear.

This prompted me to finish my "list of issues" and add it to the site, so that I could include the link in my survey response.

If anybody's interested, here's the link I sent them.

 

I'm working on a script to automate downloading firmware images. At the moment it seems to work for "production" images, but ...

  • I don't have a tablet enrolled in the beta program
  • because the beta program requires that the tablet be connected to the "reMarkable cloud"
  • which I don't trust, because it's hosted by google and the files are not encrypted.

So I have a request, primarily for people who (1) have tablets enrolled in the beta program, (2) are not afraid of the command line, and (3) use macOS or Linux (because I don't know if the process described below can work on windows or not).

This page contains a walk-through of how to manually capture the upgrade request from a tablet, save it to your computer, and then "re-play" that request to the reMarkable servers. reMarkable's response contains two string fragments which can be joined together to produce the actual URL of the firmware image file.

My request is for somebody to follow the first part of the directions on this page, and capture the request that their tablet sends when it asks reMarkable if any new firmware is available. If you want to follow the rest of the process and also capture reMarkable's response to that request (i.e. 03-response.txt), and even download a beta firmware file, that's fine ... but for now all I'm looking for is the 01-req.txt file itself.

Note that the 01-req.txt file will contain your tablet's serial number and machineid value. If you're not comfortable sharing this information, you can edit the file and replace the values with xxx or something ... just be careful not to change anything else. The other UUID values (other than appid) appear to be randomly generated, feel free to obscure them if you like but I'm not aware of any risk to your privacy if you don't.

At this point I'm trying to figure out how a "request for a beta version" is different from a "request for a normal version", i.e. if there's something different about the request, or if reMarkable has a list of specific serials and/or machineid values which are allowed to request beta versions. And if it's not tied to serials or machineid values, I plan to make the script able to download either the normal or the beta firmware.

ALSO ... and again this is only if you're comfortable sharing, please send me a copy of the /home/root/.config/remarkable/xochitl.conf file from your tablet. Feel free to obscure values like DeveloperPassword, IdleContact, and IdleName, as well as everything under the [wifinetworks] section.

If you are willing to send any files, please send them to ...

  • Keybase: jms1 (preferred)
  • Email: kg4zow@mac.com

Thanks in advance.

view more: next ›