I'm delighted to be publishing the very first update of Readwise Reader now that it's in public beta. We have some exciting updates to share including a bevy of export options, document notes in the mobile apps and browser extensions, random sort, PDF to text, and more!

If you're a new Reader tester, I write these beta updates approximately every two to four weeks covering features we've just shipped, bugs we've recently fixed, and what we intend to work on next. I also share tips & tricks to help you get the most out of Reader. If you prefer to read these in Reader, you can subscribe to the private RSS feed here.

Before getting into those product updates, I want to share that we're a little behind on in-app feedback and customer support emails due to unexpected launch volume. If you've sent us an email or left a piece of feedback and we haven't responded yet, don't worry: we will. We're working hard to respond to everyone as quickly as we can – even if it takes a little longer than usual. Once we're through this backlog, our response times should return to normal. I apologize in advance and thank you for your patience here 😳

Now let's get into the product updates!

Exporting Features 📤

As we enter public beta, it's important to us that everyone can export their data out of Reader if desired. Accordingly, we've added a bevy of exporting features with more on the roadmap.

CSV Export

Generating a CSV file of all links saved to Reader.

First, you can export a list of all links saved to Reader to a CSV file. We're still adding this to the user interface, but in the meantime you can find the option in the Command Palette (Cmd/Ctrl + K).

OPML Export

Generating an OPML file of all RSS feeds subscribed to in Reader.

Next, you can export a list of all RSS feeds subscribed to in Reader to an OPML file. As above, you'll currently find this option in the Command Palette (Cmd/Ctrl + K).

File Export

Downloading a zip file of all PDFs and EPUBs uploaded to Reader.

Next, you can export all files uploaded to Reader (ie PDFs and EPUBs) to a zip file. As above, you'll currently find this option in the Command Palette (Cmd/Ctrl + K). Note: If you've uploaded a lot of files or some big files, this option might time out. We're working on a more robust method that will generate a downloadable link from the server.

Annotation Export

Exporting annotations on a per-document basis.

Finally, you can export your annotations (ie your notes and highlights) for any given document using the export option shown in the screenshot above. You can quickly copy these annotations to the clipboard or download as a Markdown file. You can even customize the exact format of these annotations using a template. Right now this is only in the web app, but we'll be adding the mobile analog soon.

As a heads up to folks new to the Readwise ecosystem, every highlight you take in Reader instantly syncs to Readwise and from there you can configure automated syncing with your note-taking apps including Notion, Obsidian, Roam Research, Logseq, and Evernote.

Props to Bruno for taking the lead here.

Document-Level Notes 📝

We shipped document notes in the web app back in November, but then had to shift attention to more urgent matters before implementing them in the mobile app, the mobile share sheet, and the browser extension. These are all now done!

Many different ways you can access document notes in the mobile apps.

The document note in the browser extension.

The use cases for document notes are many, but one common application is to jot down a short note upon saving a new document about why you saved it or who recommended it to you.

We're now adding document notes to the various note-taking exports so you can get these into Obsidian, Notion, et al. We're also hoping to add Ghostreader at the document-level on mobile soon now that we have a destination for Ghostreader's responses.

Props to Jesse on design and Mati on execution. Lots of surface area here.

Random Sort 🎲

You can now randomly sort any View in Reader. This is great for when you're tired of seeing things in chronological order and want to spice things up a bit.

Every time you tap the Random sort or refresh, the list will randomly reorder.

Props to Adam again for adding this in on web and mobile.

YouTube on Mobile 📱

When we entered public beta in mid-December, YouTube transcript follow-along worked on only web. Mati has since added full YouTube support on mobile!

As a reminder, Reader is pulling the transcript from YouTube. If the creator uploaded a subtitles file like Erin has above, the transcript will be nicely formatted with proper capitalization, punctuation, and so on. Otherwise, the transcript will be Google's speech-to-text rendition, which is serviceable but leaves some room for improvement as a standalone document.

Experimental: PDF to Text 🩻

Problem:

The PDF file standard is basically as bad as the US healthcare system.

Solution:

Reflowable PDFs!

You can now read your HIGH QUALITY TEXT-BASED PDFs as if they were proper, reflowable hypertext documents.

While this is cool on web, it's absolutely game changing on mobile phones where letter-sized PDFs are effectively unreadable.

While this feature works well enough to justify shipping to early adopter testers of a beta product, I cannot stress enough that PDF to Text is still experimental with many caveats:

• This feature will obviously not work well on PDFs with complex layouts such as pitch decks and infographics. It needs columns of text to work.
• Embedded images and tables are part and parcel of a complex layout. These elements will generally not parse to text well.
• Many PDFs were generated poorly causing the parsed text to be sprinkled with imperfections. It will take us a long time to improve the machine learning processes that make this whole feature possible so we really hope the improved experience of reading parsed HTML is vast enough to offset any annoyance caused by the tiny glitches in the text in the short-term.
• Highlights in one format will not translate to the other. In other words, if you highlight the text version of the PDF, you will not see those text highlights on the PDF version and vice versa. Candidly, the translation would be so hard to build that I doubt we'll ever prioritize it.
• If you can't find the View as text option on a PDF in your Library, it's because we're still running a migration to generate the text on previously uploaded PDFs. You should always find it on newly uploaded PDFs, however.

Aside from making PDFs legible on mobile phones, this feature will ultimately unlock so much additional functionality such as:

• Keyboard-based reading experience on PDFs
• Properly estimated reading lengths on PDFs
• Text-to-speech on PDFs
• Ghostreader on PDFs
• and more!

Again, the above will remain limited to high quality, text-based PDFs. We'll be continuing to refine and polish this feature over the next few weeks.

Props to Bruno and Artem for leading this ambitious feature.

Minor Improvements 🦐

• The Views tab on mobile will now show a count of documents contained therein. Mati is adding a sorting option here so you can choose how you want Views to be ordered on this tab.
• Push notifications for Daily Digest are now working properly on Android.
• Reader will now follow most redirects and link shorteners (such as t.co) to resolve the proper URL of a saved document.
• There's now options to hide both the right and left sidebars by default in the Command Palette on web.
• There's now a right-to-left text option in the web and mobile Command Palettes.
• Paywalled articles from aggressive media sites should parse much much better now.

Bug Fixes 🐛

• Fixed massive glitch with Chinese characters.
• Fixed many, many issues with the browser extensions (particularly Safari).
• Fixed many RSS edge cases discovered by new users.
• Fixed many RSS cover image errors.
• Fixed many glitches with Daily Digest (particularly first time uses).
• Fixed our Twitter saving caused by Twitter pushing a change to their site.
• Fixed lots of bugs with emails forwarded to Reader.
• Fixed wrong date showing on list items when sorted by date.
• Added in better error pages on web.
• Fixed Enter shortcut not working to launch a link on web.
• Fixed bug where document notes couldn’t be written while playing a YouTube video on web.
• Fixed bug where merging highlights could cause the highlight to delete.
• Fixed styling of multi-line toast notification.
• Fixed bug where Android app would intermittently crash.
• Fixed bug where a single Reader document would create multiple "articles" in Readwise 1.0.
• Fixed bug where sharing the same article twice to Android would crash the app.
• Fixed bug on Android where TTS wouldn’t stop if the user was called.
• Fixed several instances around mobile app deep links.
• Fixed many cases where Safari extension wouldn't install and authenticate properly (though maybe not all).
• Fixed bug where some users wouldn’t have the Onboarding or FAQ doc in their Reader Library on initialization.
• Fixed bug where newly saved Views on mobile showed "Unsaved view" despite being named.

Creator Content 📼

Jeremy Caplan

I sat down with Jeremy Caplan (educator, journalist, and author of Wondertools) for an in-depth interview about Reader: how it started, how it works, and my personal Reader workflow.

A Better Computer

Matt Birchler has built an entire Youtube channel to teach people how to better use software they already own. In his recent video, you’ll learn the non-negotiable features he leverages in Reader 🙂

BearTalk

Here's what Bear thinks about Reader. Love that meme thumbnail 😂

JimmyLv_吕立青

For our Chinese-speaking friends, we hope you enjoy this in-depth walkthrough from Jimmy LV.

Readwise Official

Finally, Erin has glowed up our Readwise Official YouTube channel and uploaded a new video on how to find your ideal Library Configuration. If you're turned off by the default Inbox Zero workflow or want a more first-class Shortlist feature, we highly highly recommend you watch this brief video.

Other Resources

Because Reader supports so many different media types, sources, and workflows, it can be somewhat more complex than a single-purpose application. We're still building out our library of resources, but here are some for ease of reference in case you missed them during the onboarding process:

• Frequently Asked Questions (links to a document in your Reader account)
• Reader 101: Getting Started with Reader (video)
• Reader 201: How to Set Up Your Reader Feed (video)
• Reader 202: How to Organzie Your Reader Account with Filtered Views (video)
• Readwise Community Discord Server

Coming Up 🔜

Our main priorities over the next few months are to fix all the bugs and edge cases discovered by our new public beta testers, to make sure the web & mobile apps are mostly in parity, to deepen v1 features such as text-to-speech, PDFs, and EPUBs, and finally to smooth out the many small UX wrinkles separating this beta app from a proper version 1.0. If you prefer a crude metaphor, Reader is like an awkward teenager who just went through a hormone-fueled growth spurt and now needs to "fill out" a little bit before resuming further maturation.

One project that's especially important to us in the near-term is optimizing performance. Particularly on mobile and particularly for accounts with a lot of content. We have a pretty clear idea of what we need to do to dramatically enhance interaction times. Now we just need to take the time to refactor A LOT of code.

Farewell 👋

Now that everyone on our team is back and recovered from winter holidays, we hope to share many more updates like this one soon!

Thank you again for your continued support and please reach out any time 🙏
– Dan, Tristan, & the Readwise team