As someone who loves DT and found that their usage of Readwise Reader has increased over the last few months, I thought I’d reach out to people in a similar situation to try to better understand how you’re using both tools to compliment each other.
How do you differentiate between the two? What goes where? How do the two work together, if at all?
Thanks a lot in advance,
I got an extended two months from CEO to try, but I wasn’t able to find a way to combine both, apart Readwise didn’t add anything new to the combination of Inoreader + DEVONthink.
For me, I use the Readwise plugin to get highlights into Obsidian. I index my Obsidian Vault in DEVONThink for enhanced searchability.
You might enjoy the previous thread on Readwise’s Reader and DT:
My answer hasn’t changed much since that thread:
Most webpages etc. just go straight to Reader (using their Safari integration). If I know in advance I want a copy of the page for DT, I will write it to PDF (how I save webpages) at the same time and file it myself.
Non-webpage PDFs go to DT first. I move them to Reader if and when I am ready to read them and if I don’t think I need an annotated copy of the PDF (see note below)*.
I have the “mark as read” setting disabled in DT so files are marked as unread until I manually change them, so I don’t lose track of anything new.
If something I read in Reader is worthy of going into DT and hasn’t already been added, I will do that after I’ve read it (usually this is webpages that turned out to be valuable - I go to the original webpage and print to PDF then).
Reader now supports export of highlighted PDFs, which was still pending when I wrote that comment. But I’ve not actually been using it much so it didn’t change my workflow as much as I thought it would.
I have a little niggle at the moment - I mentioned in the first paragraph that if I know I want to keep something before I’ve read it, I print to PDF in DT at the same time as saving to Reader. This is getting confusing as I’ve had a couple of instances where I’ve duplicated a file in DT (saving it before I read it, and then saving it again after reading). It’s my own fault - I have a tag on Reader for files I’ve imported to DT, but I rarely tag things as I save them, and then I forget. I’m wondering whether to “ban” importing web articles until I’ve definitely read them, in order to resolve this, but I haven’t made a decision yet.
The other issue I have, not really due to either app, is the huge amount of unread reading I have pending 
Reader has over 2000 unread items and DT is just under that currently. DT of course doesn’t care and works exactly the same as usual. Reader however has moments where it hangs, and crashes are not uncommon. I know they’re rebuilding the app to resolve these issues with large libraries, but in the meantime it is frustrating. I have found though that the apps are worse for this than browser, so I’ve switched to reading in browser. It’s not ideal on iPad, but until the issues are resolved it’s preferable.
I’ve also tagged all content in Reader by theme and set up views on the Home Screen to show content by tag. I maintain a view of untagged items waiting for processing as well. In DT, I move files to their correct group, and I found my lack of filing in Reader stressful. Reader’s ‘groups’ don’t match DT exactly as there are things I like to read that never enter DT (and so don’t have a corresponding group), but where they match I’ve identified them the same way for ease of navigation.
I am a longtime DT user and new to Reader, which I LOVE. The parsing of webpages, the growing PDF features, the cross platform compatibility with browser extensions, the ability to highlight the web when I’m on my crappy locked down Windows boxes at work, and the incredible desktop and mobile reading experience with keyboard shortcuts.
I’m in a similar boat as far as trying to work out how to integrate with DT. The Readwise highlights/notes to Obsidian vault indexed by DT hack is OK, but ultimately I will want a way to archive my Reader library in DT, not just highlights. They have fuller export options coming, and now you can request a batch PDF/epub export.
I haven’t decided about my pdf reading flow yet. You can read and highlight pdf’s in Reader, and then download annotated pdf manually, but I honestly prefer how with DT desktop and mobile I can natively read/annotate but also open in other pdf apps with all changes saved back to the synced file. Bookends integration with DT is also great.
One thing I don’t love about reading/annotating pdf’s in DT on Mac is the Annotate > Note popup window is very busy and cluttered. I would love a cleaner sticky-note type option like on DTTG. Maybe I’m missing an option?
Academically what I am currently doing is trying a switch from a pdf journal article flow to saving new articles as full-text webpages/epubs to Reader, as my eyes don’t love reading multicolumn articles on iPad. As an aside, for us presbyopic folks, I have found no better pdf text reflow than the free Acrobat iOS Liquid mode - it blows all the others away including Reader’s pdf->text. Sadly, no text reflow option I have found allows highlighting/notes to be added to the pdf (might not be possible).
Will follow this thread with interest, and have definitely requested full export from the Reader team. Since Reader works offline I wonder if DT could index that cache? I can’t find it though.
And if the DT crew gets curious about Reader/Readwise integration, they do have a Readwise API currently that supports creating, fetching, and updating highlights on behalf of the user. Readwise API | Readwise
It would be great, as a start, to sync Readwise highlights/notes to DT without doing the Obsidian intermediary.
I don’t know what they are planning as far as Reader API …
Thanks!
I’ve been using them side by side with the aid of another Mac app: Hookmark. Whenever I save PDFs into DEVONthink, I also upload a copy to my Readwise Reader. I use Hookmark to link them together for easy reference. For the most part, I find that Readwise is great for highlight and note extraction (which I send to Obsidian.) I like to keep the DEVONthink version unaltered and use it as my home base file which links via Hookmark to Readwise, Bookends, and other needed files.
I am currently using Reader and bring the annotations into DevonThink via an indexed Obsidian vault.
This is not all bad. This is the way my Kindle highlights and notes come in.
If a pdf or webpage is that important to me to keep in DT, I’ll archive it and tag it as such, but like my kindle highlights are in my Kindle, I know annotations from PDFs in Reader are in Reader.
I’ll be watching this thread to see if any other workflows emerge.
I saw this on Reddit today. It appears that the Readwise Reader app might eventually have local files you can access. If this becomes the case and I understand it correctly, then this might open the door for DevonThink to index your Readerwise Reader resources.
Perhaps I’m losing something, but I don’t see any big difference in handle annotations in Readwise or DT/DTTG. The only handicap I see is in DT/DTTG you have to take the work to extract the annotations from the documents (and sometimes not even that with the help of @ryanjamurphy PDF extract script).
Done that, DT/DTTG is far, absolutely far better than Readwise in managing those annotations. R presentation is beautiful, but it only has a way to work with it, and the limit of DT is your imagination.
I am considering trying out Readwise and Reader.
Since Reader has an API with an endpoint to watch for documents (either all new documents or new documents by tag) it should be possible to add those documents to a Dropbox or Google Drive folder and then index that folder to get the items into Devonthink.
Both Zapier and Make.com allow access to any REST API. For those who want a somewhat similar solution a Make.com module is available by a third party for a 1-time payment:
Brett Terpstra has written a script that saves Readwise highlights to DEVONthink as annotated Markdown files:
Thanks for the link!
Did you manage to get this to work? When I tried, it failed with the error
95:185: execution error: DEVONthink 3 got an error: Invalid argument in (-50)
93:183: execution error: DEVONthink 3 got an error: Invalid argument in (-50)
It’s probably user error, of course. I’ve asked a question on Brett’s forum, but wondered if you’d had to do anything out of the ordinary to make it work.
Thanks.
I’d run the script in script editor to see where exactly it failed. That might help to decide on the question of “out of ordinary”.
It’s a Ruby script actually.
I’d have thought that the error message
93:183: execution error: DEVONthink 3 got an error: Invalid argument in (-50)
comes from the AppleScript part.
According to Error Numbers and Error Messages, it indicates a parameter error in an operating system call – perhaps in the do shell script call?
There are some tools to debug Ruby scripts:
Edit This part in the AS code looks fishy to me:
if theAnnotation is not missing value then
update record theAnnotation with text "#{annotation}" mode replacing
else
set annotation of theRecord to create record with {name:((name of theRecord) as string) & " (Annotation)", type:markdown, content:"#{annotation}"} in (annotations group of database of theRecord)
end if
I can’t find an update command in the scripting dictionary of DT 3.9.8, which is the latest version (to my knowledge). No idea if that is causing the error.
Yep, this was my mistake, I developed the script against the beta and didn’t realize the update command was not available in v3 (I’m new to DT scripting). Will update the script accordingly.
Apologies, I did not try it myself as I‘m not using Readwise Reader (I have only PDFs and am importing directly from the PDFs containing the annotations). But I still found the script very interesting and worth mentioning it here.
I’m glad you did, thanks! I’m sure that if there is an issue that isn’t my fault, then it will be fixed shortly and it does look very useful, so thanks for raising it.
Just to update this thread.
There was a problem with the script working with DT3 on my system, but Brett has kindly fixed it, and it should now work for everybody.
It’s an excellent script and well worth checking out. Thanks to Brett for developing it and being so responsive!