Has anybody developed a workflow to integrate Readwise and DT3? Readwise is good to manage highlights from Kindle and Apple Books.
5 Likes
Not yet, but I keep thinking I need to develop one. Most of my Readwise highlights end up in my Zettelkasten, so my workflow at present is:
- copy a highlight in Readwise
- create a note in Obsidian (using my Zettel template), paste in the highlight and do whatever I need to do in terms of thoughts, questions, links, tags, etc.
- my Obsidian vault which constitutes my Zettelkasten is indexed in DT3 so I have the convenience of Obsidian’s markdown editor, graph view, etc., and interconnection with my DT material.
Jim, is importing directly from readwise to DT something that can be done with a script?
I would think it might well be possible. Readwise does have an API, but it’s beyond my abilities. Maybe @BLUEFROG has some thoughts.
It can’t be done directly via script. It’s technically possible the API could be used but I’d have to investigate that a bit further.
4 Likes
There is now a Readwise plugin for Obsidian that automatically creates a note for highlights. The highlights can be stored in their own folder (or not) and DT can index that folder.
Edit: To clarify, each source gets it’s own note of highlights.
This is excellent! Thanks.
This is great news! I find this super useful and really enjoy the See Also feature in DEVONthink when I browse my highlights 
Readwise has released its official plugin for Obsidian.
It works perfectly.
You just need to have this Obsidian folder indexed in DEVONthink.
Have a nice day
1 Like
I wonder whethet anyone uses DevonThink as their read-it-later instead of Pocket/Instapaper/Feedly etc. And if so, how can we then combine it with Readwise? I can’t send my annotations/highlights to Readwise FROM Devonthink?
1 Like
I export my annotations/highlights from Readwise in markdown and move them to DT3. I use Readwise mostly because I have lots of kindle books. For websites, etc., I use DT3.
I believe Jim wrote a post about how to use DT3 as read-it-later for websites, etc. In addition, there is this article: Using DEVONthink as a Read-it-later Service – The Sweet Setup
Thank you, but here I am looking for how to export/get highlights and annotations made in DevonThink into Readwise. 
@BLUEFROG I’m very interested in your workflow on Read-it-later if you have it somewhere. 
I see :). It seems you can import your from DT to Readwise is creating a csv spreadsheet with your notes or sending your annotated pdf to add@readwise.io. Let us know the outcome of your experiments.
1 Like
By the way, it seems Readwise has big ambitions: The Next Chapter of Readwise: Our Own Reading App. Let us see!
Being almost a year later since the last reply in this thread, any updated thoughts?
I am a brand new DT user and have been using Readwise with Obsidian so far. I like the idea of simply indexing my Obsidian Vault with DT (hadn’t thought about that), but it would be awesome to be able to get highlights FROM DT into Readwise, as I was considering using DT as my read-later service… but getting highlights into Readwise is important to me.
1 Like
You could give the PDF import functionality in Readwise a try. This captures highlights contained in PDFs and should at least partially cover what you’re trying to do.
DevonThink is even listed on the help page as one of the PDF-related apps that should be the most compatible (as PDF is a fickle format).
1 Like
If I was going to go the PDF route, I would guess the best method would be to save the initial page as a PDF in DT? I have been using “Web Archive” based on my thought (not sure if it’s valid) that it would still be “responsive” for viewing on my iPhone. Reading PDFs on my iPhone is not something I like to do. 
I have been using Instapaper as my read-later service/app, which works perfectly with Readwise… so I could continue with that, I suppose. I just liked the idea of all things being in DT for the future search and AI.
I’m not quite following you there regarding what you’re doing with web archives in this context, and also what you’re considering doing with the first PDF page. Could you clarify your perceived gap between what you want do and what is technically feasible?
The way I use Readwise, it’s an intermediary between highlights/clippings/snippets from different kinds of services and DevonThink (via indexing Obsidian). The goal is to get all of those highlighted key contents into DevonThink and this works well.
In principle, if a PDF is already in DevonThink, then as far as I know there’s only one reason to import it into Readwise: If you want to use their spaced repetition feature. And luckily such an import functionality exists - great.
Yeah, especially with an iPhone SE …
I wrote a script that lets me comfortably read a PDF’s content on iPhone. Due to the needed conversion from PDF to RTF(D) the result is not 100%, but it’s definitely better than reading the actual PDF on iPhone. The steps are:
- convert PDF to RTF(D)
- run script to automatically split RTF(D) into chapters
- sync
The conversion from PDF to RTF(D) is sometimes not satisfying, so I always keep the PDF in DEVONthink. The RTF(D)s that the script produces are really just for reading on iPhone. However, I sometimes keep the RTF(D)s after reading as it’s nice to have these smaller records, e.g. it’s easier to link to a relevant part (and DEVONthink’s search results are also better with smaller records). Of course the RTF(D)s contain the PDF’s page numbers, chapter title etc. But for one time reading it’s ok. It’s possible to remove the page numbers etc. (e.g. in Nisus Writer Pro), but I rarely do that.