DEVONthink + Obsidian

Let me take a stance opposing the impression that some voiced saying that Obsidian does not add a lot on top of DTP. I’m a user of DTP since like over 10 years now, and Obsidian since it was first released. These days, for day-to-day, operational work (tracking of meetings, research for my dissertation, task management) I’ve completely switched over to Obsidian. I use DTP basically only as a data grave for my emails since its search functionality is far beyond what I can do in Obsidian, and way more performant. At any point in time, I’m having like 2M documents in DTP, mostly emails, that I can very quickly search through.

So for content search, I’d use DTP over Obsidian any time. I’m not using DTTG as until now, for over 2 years, it has not been able to successfully synchronize even one year worth of emails. DTTG is a dead fish for me. But DTP is great for what I am using it for…

…which is not a lot beyond search for mail content (for document content I just can rely on Apple’s Spotlight anyway).

Enter Obsidian. Do not mistake it for a markdown editor only. That doesn’t do it justice. The idea behind is “Linking your Thinking” - you essentially create a network of your information, and it allows you to navigate that network. It is an ideal tool to stay on top of things - be it for research or for daily operational work.

One big advantage is its easy extensibility. You can easily add your own code. For example, for my daily meetings, I wrote a simple script that shows me, in a table within the current meeting note, all meetings related to the meeting I’m currently leading, past and future, depending on the invitees etc. I develop this code like once, and then have it used by templates for meeting notes. When I’m having a new meeting, I just add a note to my meetings folder, and because I can specify on a per folder basis which template I want to use, I’ll have everything available to me at that point.

In other words, on top of being a markdown editor that lets you manage your network of notes, it also is a database that you can query within your documents.

As another example, when I’ve my 1:1s with people working for me, I can easily see what were not only the last meeting notes, but also the tasks these people had to work on. Because of the query ability, you can have, in your notes, dynamically all tasks from any other note assigned to anyone you want, and click on them to then see those notes. It is very interactive.

I know I can sort of extend DTP, and I’ve done this quite a bit. But typically that are just scripts that I have to manually call, while in Obsidian, you can have them sitting right in your notes for you.

For my research, I’ve a full workflow that I integrated to have Obsidian notes feed into LaTeX. You can get those scripts here.

Just yesterday, I was able to easily connect Obsidian as a data feed for my blog. You can read about it here.

I think the biggest point for Obsidian is its community. There’s a load of extensions that people build for just about everything - even DTP. And you can easily extend it yourself.

Also, Obsidian is completely up to you when it comes to how you want it to look. Because it is all Javascript, you can change the look and feel completely by stylesheets. Compare that to DTP, where the simple request I had a while ago, to not have things in the Trash being strike-through (because I can’t read those things there just because of this), and I was just told, nope. Not happening. No configurability planned. Really? It’s annoying. Like you cannot even define default views for dynamic groups. Every single freaking one of them you’ve to add your column selection and order yourself. Seriously?

DTP is like old fashion we tell you guys how to work with our software. Obsidian is like, you tell us how you want to use it, and actually, we give you the tools you need to extend it in any way you want. That’s a competitive advantage that is probably the biggest challenge to DTP in the long run.

Obsidian vs. DTP, for me, is Sharing vs. Telling.

Also, think about how DTP forces you into their data model of “Databases.” Yes, you can index files/folders, but ultimately you’ll end up using some “Database”, within which everything has a naming convention that’s just as useful as parsing your Time Machine sparse bundle. In Obsidian, there is nothing of the sort. It’s files and folders, that’s it.

So where do I see DTP? DTP has persisted metadata about the things that it manages. That makes it fast when it comes to a gazillion files. Obsidian, at this moment, does not have that. But. But. That does not mean it can’t do it - it already has a workspace cache, etc. This is just a question of time. There is no limitation really telling it is not technically possible.

Another big advantage of DTP is the number of file types it handles. I’ve never even had the idea to start writing notes in DTP (also, because the Editor is like the worst thing I’ve seen in a long while; again, the UI of DTP is like 1990s). But managing PDFs etc., automating on top of them (e.g., all my OCR runs through DTP’s Abbyy license) is really top notch.

Maybe I’d put it this way: Static content: DTP. Dynamic stuff: Obsidian.

DTP for me is a filing system. A very clever (set of) shoebox(es) for paperwork, with excellent search functionality, plus some automation when it comes to into which box to put stuff. It’s like ELO on steroids.

Obsidian, on the other hand, is a control center, a research tool, an idea management tool, a living organism, a second brain.

DTP is my long term memory.

Obsidian is my short term memory.

Just my 2c,