I went and used time machine to get an old version of the app and removed the upgrade.yml from the contents of the app package as well, so it won't auto upgrade.
This was a really bad handling of the upgrade. It is unrealistic to expect that someone with a large enough library of content locally to also put this all in the cloud for storage. I have a curated library of content so I can work offline.
The 2.0 upgrade won't and can't auto-upgrade because 2.0 is far far far more secure and we wanted to keep it that way.
2.0 is very locked up in a sandbox so it can't actually access your disk. 1.x should be secure but 2.0 radically reduces the chance of breaking out of a jail and to do automatic upgrade would have required us to relax this significantly.
You can just drag and drop your .polar directory into 2.0 and it will migrate all your 1.x data for you or you can just stick to the 1.x version I linked to above.
I would have loved to make it auto-upgrade but we just couldn't do it..
I always liked that I could keep my ebooks/Polar Libary in my Nextcloud, but now it sounds like that isn't possible anymore :-/
I tried the cloud version in the past, but it had only disadvantages for me.
1. your data is 100% NOT lost. Just pull in your local Polar directory and it'll all be in the cloud
2. we increased the storage for any existing user to 2 GB to support somewhat larger libraries
3. we released an updated stable version of the latest 1.0 release for anybody who wants to keep the local directory
Two issues/questions though:
1. How do you import flash cards into Anki? After adding the AnkiConnect add-on the documentation just says to use the "Sync Flashcards to Anki" within Tools but there is no Tools (on the Web App / MacOS versions). I can see it being done in the old version of this video: https://www.youtube.com/watch?v=F2aMWBdZUeQ&t=636
2. When importing using the Polar chrome extension I ran into a bug. When importing this page (https://en.wikipedia.org/wiki/Magnus_Carlsen) the page mark feature is completely broken. Once you drag it below the top picture it shoots all the way to the bottom, rendering the feature useless. This didn't happen on the few other pages I tested it on though.
I also think the formatting could be a bit improved -- but I'm sure that's extremely difficult. I was preferring the formatting offered by uploading the pdf created by this converter (https://www.web2pdfconvert.com/).
Regardless this seems fantastic, great job!
Where did you see that? The directions might be stale somewhere. I checked our site and it's not there. It's in the overflow menu in the toolbar (the three dots on the top right).
It only works from the Polar desktop app though.
> 2. When importing using the Polar chrome extension I ran into a bug. When importing this page (https://en.wikipedia.org/wiki/Magnus_Carlsen) the page mark feature is completely broken. Once you drag it below the top picture it shoots all the way to the bottom, rendering the feature useless. This didn't happen on the few other pages I tested it on though.
Thanks. I'll use this as a test candidate. The EPUB pagemarks have a rare but I can't really duplicate that seems to happen like 5% of the time. I might work on a different implementation.
> Regardless this seems fantastic, great job!
Thanks! Keep the feedback coming as we rapidly iterate. Usually pushing 2-3 releases per week!
Another suggestion is to gray out the 'Start Review' button if there's nothing to review. I couldn't figure out if my test cards were made correctly when it kept saying 'Review Completed!'
I now assume the reason was that the algorithm (?) doesn't have you review cards until a certain amount of time was passed so that screen just indicates there's nothing left to review.
I'd also suggest a way to test the features before creating an account. Perhaps let someone upload and edit a document and only save it once they've created an account.
The only thing stopping me from going all-in on paying for Polar and using it as the "hub" of my reading (mostly technical docs, long-form blogs, whitepapers, and textbooks) is the lack of tie-in with E Ink readers, which I vastly prefer for reading and note-taking over computer screens.
I use a ReMarkable 2 tablet for reading technical docs (nd a Kindle Oasis for casual nonfiction. It would completely "connect the ecosystem" if there were ways to sync with my readers or at least push documents to them and pull annotations back. Boox, ReMarkable, and other popular E Ink tablets have APIs, OTG USB disk access, and other ways to interchange the docs (with authorization). I'm seeing loads of hacks for them on Github, so maybe even opening a plugin system would be enough; the community would contribute to fill in the gaps.
IMO, this is the killer final piece of the puzzle and would keep my reading and note-taking in one wonderfully organized place, which is something I'm 100% interested in paying for and supporting.
> IMO, this is the killer final piece of the puzzle and would keep my reading and note-taking in one wonderfully organized place, which is something I'm 100% interested in paying for and supporting.
The plan here is to support tablets (Surface, iPad, etc) very well and have very tight integration.
Stay tuned ;)
If you'd like to look at the code.
We have a unique build environment so DM me on Discord if you want to build it... You just have to use lerna and our npm repo but should work well.
404 not found on download. Please fix. Is there homebrew formular ?
The link is: https://getpolarized.io/download/
got it. Great open source project!
This is the interesting thing about Open Source is you get interesting developments like this that you didn't anticipate!
I ended up switching away from it just because of the lack of support for EPUBs... so this is really exciting!! Can't wait to check out v2!
Since this new release has ePub support, is it possible to use it without the cloud and subscription?
Edit: From other comments here it seems like 2.x is cloud-only (and tied to the Polarized service storage). Though it says that there’s a latest update to the local supporting 1.x, I doubt if that would remain supported for longer (of course, someone could say it’s open source, so just fork it).
I wouldn’t mind paying a small license fee for the freedom of local storage or being able to use another cloud storage along with better support. Since 1GB on the free tier is not enough for me and since I have other cloud storage already available, even the Plus plan it quite expensive for me.
Subscription isn't required for < 1GB of documents.
Really enjoying the updates so far. The dark mode inverting the actual PDF to dark is a nice touch for reading (and I can toggle it off for visualization heavy portions).
I added a feature request for them to support chrome-extension here:
I'm going to submit a PR to them to support moz-extension too.
Great job in any case!
About two years ago we launched Polar on HN  and now we’re back to announce Polar 2.0!
Polar is a tool used by software engineers, students, and professionals. Anyone who wants to manage academic research, technical documentation, textbooks, or web-based reading to build a personal knowledge base.
Polar is an integrated reading environment similar to your IDE that you use for writing code. Think of it like Visual Studio or IntelliJ but for books, web content, PDFs, EPUBs, etc.
What makes Polar special is that we’ve built an integrated workflow so that you can read, annotate (make highlights, add comments, pagemarks etc), and convert anything to flashcards. You can then use spaced repetition to review your material directly in Polar or sync it to Anki.
My co-founder and I have been working tirelessly on this since the lockdown started to get this shipped.
This release represents everything we’ve learned since 1.0 and includes:
- Dark mode!! This was by far our #1 requested feature
- Both EPUB and PDF supported (originally it was just PDF)
- Web pages capture content using Mozilla readability, then saving them to EPUB
- Web capture works entirely from our chrome extension so that your cookies, etc get captured too
- Easy reading management with flagging, tagging, archiving, pagemarks to mark specific sections in the book, and more
- Improved annotations and highlights management, including area highlights, annotation tags, margin notes, etc
- One-click flashcards directly from annotations
- We’ve greatly improved our Anki sync to make it more reliable, faster, and has more features like support for cloze deletions, front and back cards, HTML, etc.
What’s more, Polar is open source so you don’t have to risk being locked into a platform you don’t control!
We want to make obtaining an education as easy as possible. In addition, it shouldn’t cost a fortune to get a decent education. Be it tuition or overpriced textbooks. This is why we are building Polar with the vision to truly democratize education.
This is still early stage for us of course. Some of the upcoming features will include mobile apps, integrations with tools like Zotero, and much more.
If you’re someone who reads a lot and likes to remember the material, give Polar a shot! For power users, we also have a premium version with up to 500 GB of storage. For HN readers, we are doing a 20% discount promotion for a year-long subscription. Use the code hackernews20 to take advantage of that
We would love to hear what you think about Polar!
Except the v2 update changes this from local or cloud storage to cloud only. So in order to use the app at all you do have to upload your documents to servers you don't control. But this isn't my biggest pain point, as long as export capabilities are comprehensive and my data is private.
But that's not the case here. For an open-source app that requires uploading personal documents to yet another USA-based server there is a surprising lack of focus on privacy. Without end-to-end encryption it's unreasonable to take the privacy risk that serious use of this app entails. And I've been burned by one too many productivity apps to believe there's not going to be a willful or accidental privacy breach here too.
A safari extension, or at least a Firefox one, would have me pulling out my credit card on the spot.
The code actually is written to be portable as a 'web extension' as both Safari and Firefox use the same general API.
I'll try to get some time and see if it works on Firefox. The biggest issue we have is testing. If something breaks on FF I want to know and we don't have a testing environment there.
Can you expand on the Zotero integration? I like Polar for reading/annotating more than Zotero, but the metadata import and bibliography functionality of Zotero is extremely useful to me. Combining these in some way would be pretty amazing!
The other is to just open the PDFs directly but I don't think we can get the metadata that way unfortunately.
We might be able to manage bibliographies inside of Polar but we don't have plugins like Zotero so that it would work in MS Word or Google Docs, etc.
Curious if it's worth moving off the convenience of 1.0 for mandatory cloud storage otherwise.
I would assume it's elastic with storage so I could just continue to use it as a reader and delete books as I finished(?)
We've also tried to make it pretty cheap so if you have a huge repo it won't be super expensive. It also won't bloat your local computer so if you have a large repository most of that data will be on the cloud.