A second brain you don't have to build first.

A second brain app without setup sounds like a contradiction, because the genre grew up around configuration. Most people who try to build one never get to the part where they use it.

The pattern is consistent. You read something good about PARA, or zettelkasten, or the method a particular writer swears by. Those systems are serious work by serious people and they do function, which is exactly the problem: they function for someone who has already internalised them. What you get on day one is an architecture. You design a folder scheme. You argue with yourself about whether a note is a project or an area. You pick a tag convention, then rename half the tags a week later because the convention was wrong in a way you could only discover by using it.

At some point you notice you have spent four evenings on the container and have written nothing to put in it. The system became the project. Then a busy fortnight arrives, you stop opening the app, and when you come back the structure is stale enough that maintaining it feels like homework. This is not a discipline failure. It is what happens when a tool asks for decisions before it asks for content.

The smallest thing that actually works

Strip a second brain down to what survives contact with a working week and three behaviours are left.

A page for every day, so writing never begins with a decision about where. Links between things, so what you wrote finds its way to what it was about. To-dos that get collected instead of buried, so the actionable parts of your writing do not go missing inside paragraphs.

That is the whole method Remlo ships, and it ships as the default rather than as a configuration you assemble.

Today’s page exists when you open the app, with the cursor in it. Links work in both directions, so a project page lists every note that named it without you filing anything. To-dos written inline anywhere in any note are collected on the To-dos page, and ticking one anywhere moves it to Done. Habits are included, with a momentum chart, because the thing you are trying to do every day belongs next to the day.

No vault. No plugins. No taxonomy to invent. There is no setup screen because there is no setup.

Seven days, concretely

Here is what a week looks like when the system is that small. Say you are weighing up a piece of work, the Kestrel job.

Monday. A call. You open Remlo on the phone while walking back and write into Monday’s page: the budget they said out loud, the deadline that sounds impossible, the name of the person who will actually approve it. You type “check whether the timeline clears the Ravnes deadline” as a to-do in the middle of that paragraph, because it occurred to you mid sentence. You do not open a task app. You do not file the note.

Tuesday. Nothing. This is the day that kills most systems, and it costs nothing here, because there is no inbox filling up and no processing step you have skipped.

Wednesday. You write a proper page for the Kestrel job, because now there is enough to think about. In it you link back to Monday. From then on the two are connected in both directions: Monday’s page points at Kestrel, Kestrel lists Monday. You also write down the rate you would need, and a doubt about their approval process that you would otherwise forget by Friday.

Thursday. The to-do from Monday is sitting on the To-dos page with everything else you have written this week. You check the Ravnes deadline. It does not clear. You tick the to-do and write the conflict into Thursday’s page, naming Kestrel again, so the Kestrel page collects Thursday too.

Friday. They email asking for an answer. You open the Kestrel page. It has the rate, the doubt, the deadline conflict, and every day that mentioned the job, in order. You wrote all of it in under a minute at a time, in the middle of other things, and you never once decided where anything should go.

That is a week. It also works when the week is bad, because nothing in it requires maintenance. Six months of this leaves you with something worth searching.

What you can safely ignore

Tag taxonomies. A tag scheme is a prediction about how you will want to find things later, made at the moment you know least about that. Links carry the same information and are made at the moment you know most: while writing the sentence.

Folder hierarchies. A note that concerns two projects has to live in one folder and be missing from the other. Every hierarchy eventually runs into a note that belongs in two places, and the workaround is always more filing.

Plugin lists. Reading about other people’s setups is genuinely enjoyable and produces no notes. If a system needs eleven components to work, the failure mode is that you spend your attention on the eleven components.

Daily and weekly reviews. Some people run them well and get a lot from them. They are also the first thing to go in a hard week, and a system that depends on them is a system that breaks precisely when you need it. Backlinks and a dated page keep working while you are ignoring them.

Note titles. A good title is a small act of writing, and demanding one before you can save anything is how apps stop people mid thought. Most of what you capture is a line, not a document. Put it on the day, name it later if it turns out to deserve a name.

Inbox processing. Any design where capture goes into a queue that you sort afterwards depends on you sorting afterwards. Nobody does that at week six. A dated page has no backlog because it was never a queue.

The first week, honestly

Nothing here compounds on day one. That is worth saying, because the promise of a second brain is usually delivered as a transformation and what actually happens in week one is that you write some things down.

The change arrives around the point where you look something up and it is there. Someone asks what you agreed with a client in April and you find it in fifteen seconds, complete with the reason. That moment is what converts people, and it needs a few weeks of ordinary writing behind it, which is precisely why the setup phase matters so much: every evening spent on structure is an evening not spent accumulating the material that makes the system worth having.

What you give up

Remlo is small on purpose, and that has costs worth naming.

There is no query language, so you cannot write a page that assembles itself from conditions across your notes. There are no databases, no relations, and no custom properties. Remlo runs on iPhone, iPad, and Mac only. If you want to design the system yourself, Obsidian is a better home for that and will stay one.

What you get in exchange is a second brain that is running on day one, because the only thing you have to do to start it is write today’s date, which is already written.