This post is for you if:
- Projects that excite you are growing to be a burden on your to-do list
- You have a nagging sense that you’re not making the most of the ideas you have every day
- Your note- and idea-system has grown to be an unwieldy beast
Years ago, I ready David Allen’s “Getting Things Done”. One of the core ideas is to write down everything, collect it in an inbox and sort it once a day.
This lead to me writing down tons of small tasks. I used Todoist to construct a system that worked for me — and rarely missed tasks.
It also lead to me getting a lot of ideas, that could sprout a bunch of tasks. But ideas are much different from tasks to me. They’re something that I might, but probably won’t complete. Something that may come in useful in the future. And plain fun to come up with.
I stored them in Todoist as well, but recently I’ve started considering whether that’s wise. They started weighing on me. It became a growing list of possibilities, many of which I’d never finish. The task at the top of my list became the top of my priorities, simply because of its location.
There must be a better way.
But what might that look like?
The ideal idea-management system
1. Separates ideas from commitments
I want a system that separates obligations from ideas. Form follows function, so I’d prefer something that doesn’t structure ideas in lists. This rules out Todoist completely.
2. Shows you the right ideas at the right time
Even with complete foreknowledge, finding the perfect schedule might be practically impossible. In contrast, thinking on your feet and reacting as jobs come in won’t give you as perfect a schedule as if you’d seen into the future-but the best you can do is much easier to compute. — Algorithms to Live By
Most of us live dynamic lives where priorities change often. Your children start a new hobby, you’re handed a task at work, or you can finally work on your passion-project. Your idea-management system should reflect this. It shouldn’t just show you your most recent idea, it should make it easy to find ideas associated with whatever you find most important right now.
Avoiding lists makes it more likely that you take action on ideas that matter to you right now. You don’t skim from the top, you go for the area that matters and find ideas related to it.
If you sort ideas around a central node, you can pin-point synergies and conflicts. If you’ve taken notes on 4 different project-management systems, you want to see them all when you need them.
3. Doesn’t distract you with ideas that you can’t execute
You don’t want to waste time considering ideas that aren’t important right now. Sometimes you’re missing resources, or you’re waiting for some dependency.
Project/idea-lists are terrible at this. As you skim through them, a plethora of memories activate, most of which are irrelevant to what you end up doing.
4. Allows you to break down ideas into smaller parts, and re-combine them as needed
Tiago Forte’s Intermediate Packets inspired this. Many of our ideas can work in exactly the same way.
5. Has low overhead
You want to spend as little time as possible sorting and searching through your ideas. This should be a no-brainer. You want it to be easy to inter-link ideas and to add reference material. And when you get an idea on the train, you want to offload it without wasting time.
Time for action
Okay, Martin, I’m sold. But folders, task-managers, outliners like Workflowy and Dynalist — they’re all hierarchical!
You’re right, and until last week, I didn’t know what other solution there could be. But now there is. Roam.

Roam is different. It makes it trivial to link- and back-link pages. It creates a new page just by linking to it. And it back-links as well! When you link [[Self-determination theory]] to [[Motivation]], the motivation page will show a link to Self-determination theory in its footnotes.
This is tremendous. It creates a clear divide between commitments and ideas. Commitments belong on lists, ideas in dynamic networks.
When you get a new idea on a motivation tweak, you add it and link it to [[Motivation]].
Most of the time, everything is going well, so you don’t need it right now. But 6 months later, you’re assigned a grind of a task. You decide to read up on on Motivation, and voilá, in the footnotes is a link to that idea you had that might help you now.
Not only that, all your other ideas on motivation are there, for you to synergise or compare/contrast.
And you’re less distracted. You don’t have to take action on an idea in fear of forgetting it. Nor are you presented with ideas only because they’re recent. You have a need, [[Motivation]], and you’re presented with ideas on that topic alone. No distraction.
Roam also allows you to link to any bullet-point in any other note. Say you want your collaborators to identify with the core values of your projects. Why not embed that idea you encountered 3 months ago from Organismic Integration Theory? In this way, you can re-use sub-ideas from other major themes in any of your other projects. And if you find out that idea didn’t work? You add a note, and that note propagates to any other places you’ve referenced the idea.
Roam becomes your second brain. You draw associations, and Roam remembers. You want to look something up, and Roam shows you what you’ve considered relevant in the past.
How do you avoid losing track of important projects?
I advocate for using Roam as an idea-management system, not a project-management system. If you have an obligation, by all means track it in a list-style way that you review.
But if it’s an idea, you don’t want to spend time thinking about it when you’re executing something else. You want focus and produce, and to save the idea for when there’s time and a need. Don’t just be efficient, be effective.
There is nothing so useless as doing efficiently what should not be done at all — Peter Drucker
The Nitty Gritty, A Recipe for Implementation
For each idea that may turn into a project, I create a new page in this format:
“PI: Description”, eg. “PI: Research How to Effectively Integrate Motivations”
I have 3 prefixes:
- PI for “Project Idea”
- WO for “Working On”
- AP for “Archived Project”
In each of these pages, I spend ~5 seconds referencing concepts where I may want to encounter the project. For this one, [[Motivation]], [[Self-Determination Theory]] and [[Organismic Integration Theory (OIT)]].
I also add any references I may want to read, and whichever ideas I’ve already had about the project.
This makes it trivial to pick up the idea when I have the time and need, and to execute it efficiently.
~
These posts are about getting ideas into the wild, having other people criticise them, making them better and connecting with like-minded people. So feel free to let me know what you think 🙂
I appreciate your time.
Originally published at http://martinbern.org on October 18, 2019.
Yes, that's why I did try it. And found that I could neither get my data into it nor out of it while preserving structure or markup semantics. That's kind of a killer, since style issues I could theoretically fix with a user style sheet. Plus, having your own unique markup language means that I can't even write in it without conversion.
I'm not sure if you're speaking to me or someone else here. I was interested precisely because I understand what your tool would offer me, if only I could use it. It reminds me of some of Ted Nelson's innovative hypertext designs, like ZigZag. Or good ol' Houdini from my DOS days.
I appreciate what it could do for me in the realm of thought; my issue is that I'd like to be able to use it to help write things, things that will be shared in media that is not your program (e.g. blog articles, books, wiki-like websites, etc.) Which means I must be able to not only bring in text I already have, but bring out text I produce.
I spotted what was unique from your not-very-detailed screenshot and jumped at the chance to try it. I'm an extremely long-term power user of outlining, idea management, and thinking-enhancement software of all varieties, for some 30+ years, so I have a keen eye for novelty in such systems.
If I didn't understand what it could potentially do for me, I wouldn't have been so bothered by the interop and usability problems, nor would I have bothered to try it in the first place. I actually explored a fair handful of features that would have had me trying the tool more seriously, if I had any certainty of getting my data out again, in a useful format. (For example, in order to take real advantage of your transclusion feature, I'd have to be able to export usable markdown versions of the documents where said transclusion was occurring.)
On top of that, I took some time to reflect whether there was anything I was currently working on that could benefit from the tool without being able to take structured/formatted content back out, or having things go out but not able to come back in after outside editing.
And I concluded that this wasn't the case for anything on my "front burner" projects at the moment, and I can't currently afford the time to play with it for back burner projects right now. I do expect to revisit this decision at some point, but from my experience I would expect the big payoffs from your unique features to only happen with days, weeks, or even months of continued use. In the short run, without a lot of data in it, it might as well be Dynalist plus bidirectional wiki links, and I don't need that enough to spend the time learning or working around its particular quirks.
All that being said, I'm a little personally irritated that you not only jumped to the conclusion that I didn't get what your product does (or the implications), but that you also wrote off nearly all of my feedback as "aesthetic" concerns. If you're not concerned with feedback from people whose visual impairment is as lightweight as mine, I hate to think what you'd do with feedback from people with more serious vision problems. (Aside from you not acknowledging the many issues with editing, markup syntax, and interoperability that I took the time to write up for you and share, instead of saying "ho hum" and moving on with my day.)
As another software developer and entrepreneur, I consider even negative feedback from potential users or customers to be very helpful, especially if it lets me see their first impressions. That kind of information is something money can't buy, even from a dedicated UX tester... and most people won't take the time to give it to you. They'll just move on.
I thought about just moving on yesterday. And then I thought, "eh, true first-impression UX feedback is valuable, I'm sure they'll appreciate a write-up."
Oops.