Wherever I may Roam

Poetic license on the title aside, I’m doubling down on Roam.

“Sometimes you have to let things go…” and so on, but that’s what I did earlier.

I spent a year trying out all kinds of alternatives, and figuring out their pros and cons.

In the end, I concluded, empirically-but-subjectively — this being for me and my use cases, yours may be different — that Roam is uniquely well-suited to my way of thinking.

Since then, things have gotten better, with two big changes:

no limit on new graphs, and
the availability of encrypted graphs.

Previously, I was reluctant to use Roam for “work stuff” — now I know I don’t have to worry

Previously, I was reluctant to adopt separate graphs for separate mental silos, because “hey, I don’t want to waste my last remaining graph on that!”. Now, that isn’t a concern.

There will always be other products around, and people will continue to make decisions based on a variety of factors, and they should.

I would like to see a couple of features (again, this is from my personal experience, and I have no expectations of seeing them fulfilled anytime soon).

It would be good to have the ability to have different graphs open in different tabs. Not a very big deal because I don’t switch that often, and if I did I could use different browsers, but … it would be nice to do this easily.

Next up: more spaces in the document — or at the ability to carve out at least one additional space to split nodes on to. I understand this is hard and must be introduced carefully, if at all. If a general solution is too hard, three is a better number than two (before anyone points it out — yes, in a way, the sidebar is a third space, but c’mon).

I will say that for a product that affords a dizzying amount of creative flexibility as Roam does, you need a spare aesthetic and I commend the development team for having the discipline to stick to it.

Roam stays

Much as I tried to quit Roam previously, I’ve been back to using it the last quarter.

It’s good enough, and “doesn’t suck“.

It is more fluid and immediate than the alternatives for what I need, when I need.

So, it stays in my toolkit.

(in the context of an end-of-year “app house-cleaning“, where I try to rationalize what I need, what I’m using, and what I can get rid of)

Roam isn’t there yet

I’ve observed Roam periodically slowing down to a crawl, with individual keystrokes (yes) taking a second to be played out, filters taking several seconds to process “selections”, and … I haven’t even started using queries yet.

Part of the trouble might have been my own over-enthusiasm a few months ago, when I went all-in with it, getting into a “mind-meld” with it.

This was really great for a while, too, but that meant that when it abruptly became unusable, it was really painful.

Meanwhile, a few other things have been evolving.

Obsidian largely “just works”. I’ve been trying out Obsidian Publish, and it’s headed in a direction I like, and am willing to pay for.

DevonThink, my “bucket store” for a few years now, has begun to add first-class support for Markdown, explicitly keeping compatibility with Obsidian and Roam.

Tinderbox has always really had what I needed (super-flexible notes, and programmability, albeit in a “perl-ish” syntax), and every release improves support for markdown, zip links (those [[...]] references that are nice to use within markdown for making quick links (and backlinks) between notes).

OmniOutliner is … not evolving features, but it’s a robust cross-platform native outliner app, supporting rich embedding and at least theoretically allowing arbitrary levels of scriptability in JS.

Now I’m still a “Believer” in Roam, but not all use cases need “multi-player” (and for the ones that do, there really isn’t any other good alternative).

For now though, most of my use case are very much “single-player”, and for that it’s very hard to beat native apps (especially the non-electron ones).

I do have to get over the hurdle of using un-popular apps, non-mainstream apps, apps that aren’t “in the news”, etc. but … it turns out they’re just as solid, and there are enough users to get assistance and share and discuss concerns, and so on .

So, I’m not giving up on this “mind-meld” level of personal interactivity with apps that help me think, or break down things, or just keep stuff around for me to look at later — I’m just planning to do more of that with some of the local, native apps mentioned above.