Being new to DT, maybe I am the one who isn’t understanding, but I will explain my use case pertaining to this.
My point was trying to use DT as a read-later service, instead of Instapaper (I’ve read articles and seen YouTube videos describing this use case). For my “read-later” use case, I throw a ton of stuff in there that I may eventually throw away if it serves no lasting value to me. If there are nuggets that I want to resurface and/or expand on/investigate further, Readwise is a great tool to help me see these again via the highlight import.
Instapaper works here and the highlight import is completely automatic. I am sure I can also figure out how to export things from Instapaper to DT if I want to keep the whole thing, but was hoping I could do it all in DT.
BTW, one thing Instapaper can’t seem to handle are pay-walled sites I belong to. Saving something to Instapaper from those sources doesn’t provide the whole thing, but DT can save a web archive/PDF for offline viewing.
Finally, my reason for web archive rather than PDF is for being able to view those things easily on my iPhone (again, as a read-later use case). The web archive keeps the responsive web design and adjusts for the iPhone’s smaller screen. A PDF is static in its page layout and requires pinch zooming and scrolling around. Not a fun experience (IMO).
Again, once something passes the first-pass read as something I want to hang on to, it would certainly go in DT.
Hope this helps explain my thoughts and reasoning. Also hope I am incorrect in some of this and you can help me find a better way!
Thanks for clarifying @BabblingBafoon, I understand where you’re coming from and will add my two cents. No right or wrong here in any case
There are indeed, in general, many advantages to storing everything in DevonThink compared to using Instapaper. So if you were to use DT for storing read-it-later material, the decision needed is whether these advantages outweigh the downside of having to upload PDFs manually to Readwise (as a direct integration doesn’t exist).
However, taking a step back here, I think it’s not 100% clear that DT is the best place to store only potentially relevant Read-it-Later materials. That sounds like it might result in a lot of clutter, which then also shows up in search results and, in general, needs managing.
How about the following workflow (if that’s not what you’re already doing):
Review the highlights you’ve taken in DevonThink periodically (they get indexed from Readwise via Obsidian). If a highlight still appears valuable upon review then add the source article in full to DevonThink for future reference and searchability.
Create a smart group/smart rule workflow for PDFs stored in DevonThink that have new highlights (added in DEVONthink). Periodically import them in batch to Readwise so that you can use the spaced repetition (since I understood this is sth you wish to do).
As for reading PDFs on iPhone, the superficial downsides are obvious compared to reflowable eBook formats, but PDF has so many other upsides (especially in DT) that I just deal with it tbh. After a while I totally got used to reading PDFs horizontally on a small screen. I’ve read accounts from others in the forum going the “adaptation” route with this as well.
I use DT as my read-it later service. If you are concerned about clutter, set up a database exclusive for this. I put pretty much everything in the inbox to sit initially. I am sometimes surprised when I go back a day or two later that what I was pretty certain I wanted to read, is no longer of interest to me and so I will delete it. Documents I do want to read, then I will put into my main group and mark it to show up inmy Devonthink reading list. Of course this shows up both on my computer and devonthink to go.
At this point, it is usually in a pretty large pile as my reading is always that way! As it sits there, sometimes I will still decide to knock stuff off of that list and delete it due to moving onto other topics. If I do read it, then I will annotate it in most cases and when finished I will remove it from my reading list. At this point I will rank it with stars and write in the finders comments box why I think it is important to keep this document. This gives me a summary when I go back to the document to remind myself, rather than having to skim the document to refresh my memory.
This discipline really helps me focus on why I am keeping certain things. Then over time if I find that I keep referencing a document, I might increase the number of stars on a document because it is proving over time of becoming very important to me. This means the lower-ranked files in my database start to go to the bottom when I sort according to ranking. The idea is the gold always floating to the top, but with that said sometimes I will go through the lower-ranked files to see if I am missing anything or on rare occasions, delete something.
Keep shifting for gold, having it come to the top of the most important stuff. That is my workflow Of course, there is also the AI and search which may highlight something you are missing and need to elevate. I am starting to play with reminders to get prompts to review important documents, but like so many things in Devonthink, it takes time to evolve your workflows while actually doing some work and I haven’t quite got to that yet.
Off topic, I have in the last year become a very big user of Obsidian. I set up pages over there primarily of my own writing and link to Devonthink files which I think are really relevant to areas of focus using the PARA method… the ‘R’ of resources being in Devonthink. This is yet another way to not lose sight of the most important nuggets of gold as we swim in all this information. I am also doing a level of capture from key documents in Devonthink, forming my own version of Zettlekasten, though this is still very much in its infancy.
@ docrameous, thanks for this really great explanation. I am just starting to dive into the PARA method myself, and you have given me really good food for thought about using DT as its own read-it-later service, especially when combined with the fact that I can save stuff to read directly into Markdown (makes it easier to highlight later).
Question: What does Obsidian do for you that makes it worth having a separate app, as opposed to just implementing PARA within DT? I would love to hear your thoughts.
I’m using PARA in DevonThink. I don’t care for Obsidian. Too finicky, too reliant on unsupported 3rd party apps. Ugly working environment. Poor support of images, pdfs, etc.
I do use Craft in addition to DevonThink for specific purposes and notetaking. It has linking as does Obsidian. It’s a beautiful app that I enjoy working in. However, no other app I’ve encountered has the feature set of DevonThink:
Complex, AI-assisted searches
Built-in OCR and search of pdf text
The best web clipper I’ve found
Easy drop-and-drag import and export, eliminates an export step or import process
My databases correspond to PARA – Projects, Areas, Resources, and Archives. This is the organization I have in my Finder/iCloud. Each database indexes each of these categories.
PARA is the overall file organization. I use the same organization in Craft.
How do you benefit from tagging files with organizational categories? I can’t think of any use I’d have for wanting to see every file in the “Areas” category. The PARA system is intended to serve as file organizational categories, not keywords or topics.
It seems like just a lot of extra work tagging that would have no function for me. Having the PARA categories helps to know where to put files from a functional standpoint. Within those categories you have topical folders and files. To me, if I did tagging, that’s where it might be helpful.
The benefit to having each of the 4 categories as a separate database is purely to limit which category you search in, and also limit data corruption in the unlikely event. You could put them all in one database if you wanted; they’d have the same file organization function either way. It’s really just a personal preference.
As PKM goes, one is constantly evolving and changing…
Since this thread was last active, I have fully integrated PARA into DevonThink. Before this, I was mainly using DevonThink for archiving, but now I am actively managing materials in DevonThink with this entire structure.
Obsidian is still my ‘external’ markdown editor. It feeds the bulk of my note-making into DevonThink and is structured according to PARA as well.
I agree with the sentiment that Obsidian has not been pretty, with that said version 1.0 was a huge step forward with Apple conventions. Also to make it really fly according to my use case, I have had to really work on developing templates that I like. This is definitely not for everyone.
I looked at Craft. It is a beautiful app for sure and would suit me in many ways, but I am not keen to pay $60 / year indefinitely for notemaking, plus I can’t integrate it in the same way I do Obsidian/DevonThink.
