The file is still the starting point. What should sit beside it?

Cool Workflows
,
1

Most projects I work on are spread across more than one place.

The files might be on my Mac. A project note might be in Notion, Apple Notes, or Obsidian. A task might live in Things, Todoist, or Asana. A decision might be buried in Slack or email. An AI conversation might be sitting in ChatGPT or Claude.

Each tool has part of the work, but none of them has the whole thing.

The file browser is still where I usually start. That’s where the documents, screenshots, recordings, and folders are. But the file is only one part of the project. The reason it matters might be in a task, a message, a planning document, or an AI conversation somewhere else.

When I return to the project later, I can usually find the file again. What takes time is working out what it was connected to.

Which decision led to this version? Which discussion was it part of? What did I give the AI agent? Which result did I keep? What was I supposed to do next?

The problem isn’t that any one of these tools is impossible to use. The problem is that each tool remembers its own part of the work, while the project itself is spread across all of them.

That is why I still think the file browser is the natural starting point, but maybe not the whole surface. I don’t want to replace every tool with one giant workspace. I want to understand whether other people also need a better way to get from a file back to the work around it.

When you come back to a project after some time away, how do you reconnect the local files with the notes, tasks, decisions, and AI conversations spread across your other tools? What do you wish you could see together?

3 Likes
2

This is a ancient problem that can be solved in dozens of ways. For myself, my practice is the simple discipline of keeping a single markdown file in a project master folder in DEVONthink. (It doesn’t have to be DEVONthink, that’s just my repository.)

That markdown file is a log of all active and inactive files, messages, bookmarks, whatever, that are relevant to the project. The discipline is keeping it current by being persistent: after a message is sent or a convo thread complete, for example, grab a link to the convo and paste it into the log file with a short note saying why it’s in the inventory. If the file is in DEVONthink or you use Finder, copy a link and paste in the log. Etc.

I’ve tried other methods (including Hookmark which led to no end of tsuris because it is so **** complex and fragile), but manual discipline is the best. One reason is the act of logging what you do is both a break and a forces a memory creation, and it really takes no time. Never be too busy to think about what you’re doing for a few seconds. Once a day, read the log, cleanup as needed, and feel satisfied you have a roadmap.

Katie

10 Likes
3

I will add to Katie’s @KVZ excellent response: the fewer apps you can get by with, the better. It is easier to find what you are looking for the fewer drawers you need to look in.

5 Likes
4

This is what keeps me locked into Obsidian. Yes, it took work to get it right, yes it’s Electron and yes it can sprawl out of control with gazillions of tempting plugins.

However, the fact that all the files, notes, meetings, tasks, email links et al for a project are all local means I can fiddle with Obsidian, search and analyse using DT and still have Finder access at the same time.

For me, it’s worth the effort.

4 Likes
5

Devonthink stores or links to them all

3 Likes
6

Throw this into the mix.

You work in a job where you have sensitive information on your computer and you’re locked to MS Tools on your machine. Obsidian not allowed. Bear, Apple Notes, Reminders - certainly not allowed and not even on Windows.

As someone locked into a Windows laptop, all these creative solutions look great. But I can’t use em.

Just a note to say it’s hard having your hands tied.

I go on web forums and people post their in depth Obsidian or Notion setups. What kind of job do you have where you’re not mixing patient, client, inmate data in there that shouldn’t be?

Its an often overlooked issue in the productivity space IMO. People locked on corporate machines and behind privacy legislation.

2 Likes
7

@Wonsik : Reduce the amount of used apps: Only one app for notes taking. Only one app for tasks. Find an effective structure for organising your projects, e.g. starting with @KVZ recommendation to use a log fie.

My workflow:
Things: Projects, trips, important notes which have a due date, drag&drop works for e.g. mails
Bear: saving websites, notes
2Do: common tasks, mainly not project related
BusyCal: calendar
BusyContacts: adressbook
Files: clear structure and folders - here I wait for the native version of EagleFiler.

Before I have used Obsidian, OmniFocus … but these were too complex, and I did not like the UX. I was too busy with organising my data inside these apps instead of being productive. I try to keep it simple.

1 Like
8

A printer and manila folder works for me.

2 Likes
9

I’ve come to appreciate that what I do with apps is generally more important than what apps will do for me.

For instance, The Archive, an nvAlt descendant. I suspect it’s too spartan for most tastes. On the other hand, it’s pretty amazing how much can be done with a little thought and care.

Besides, it’s pretty hard to crash a directory of Markdown files. That’s all there is to a project in The Archive.

All else is theater of the mind.

5 Likes
10

Now that is a shrewd observation!

1 Like
11

This is the answer I keep coming back to, and the part I find hardest to copy isn’t the format, it’s the discipline. One log file per project is easy to describe and hard to keep alive.

What I’m curious about: when you return to a project after a few weeks, is the log the first thing you open, or do you open the folder and only reach for the log when something looks unfamiliar? And when the log does fall behind, what tells you it’s out of date?

Your point about logging forcing the memory is the part I hadn’t thought about. It suggests the value isn’t only the record you leave behind, it’s the moment you spend deciding what’s worth recording.

12

That’s the shortest version of the answer and probably the honest one. I’d like to understand where the boundary sits for you. DEVONthink can hold a document, or a link to one. What about the things that never become a document? A decision that happened in a Slack thread, or a conversation with an AI where you kept one result and dropped the rest. Do those end up in DEVONthink too, or do they stay where they happened?

13

This is the case I don’t see discussed enough, and I don’t have a good answer for it. Almost every setup in this thread assumes you get to choose your tools.

When everything is locked down, is the workaround to keep the split and accept it, or is there something in the tools you’re allowed to use that carries more of the job than people expect? I’m also curious whether the constraint changes the problem. If the material can’t move, does knowing what’s connected to what still matter, or does the boundary make that question moot?

14

The line about what you do with apps mattering more than what apps will do for you is doing a lot of work in this thread, and I think it’s right. The Archive is interesting for the opposite reason most tools are: there is almost nothing in it to maintain.

If the markdown directory is the project, what carries the part that isn’t a note? A decision you made in a call, a reference you never wrote down, the one answer from an AI you kept. Or do those not need carrying, because you only keep what you’re willing to write?

15

The “fewer drawers” point landed with me. Several of you said some version of it and I think it’s the strongest thing in this thread. It also moves the problem: most of what I described isn’t caused by the tools, it’s caused by how many of them I decided to use.

For those of you who cut down successfully, was there a moment you decided to consolidate, or did it just happen as one tool took over the role of two? I’m trying to work out whether that’s a decision you make or something that emerges.

16

This is why I ended up printing and placing everything in one folder. If you are able to discipline yourself with each interaction with a programme/service and copy the link, hookmark, can be very good as it creates a link from almost everywhere. The issue I found was needing to log into various online services for the link to work, and what you do when you’re not at the desk, hence printing. Where this system breaks down is if you forget to add something, you then cannot trust your system as a single source of truth. This is especially so when receiving data via your mobile phone when you’re out and about. You could immediately copy the data to a project in your task manager or a link that you can then add the data to your system. A good easy to use task manager may be the best solution to ensure you don’t forget to add data.

Another alternative would be Zengobi Curio, which would allow you to track everything. You could even just take screenshots and past.

It is unlikely that one app will cover everything and apps like Slack are a pain in the butt.

Personally, I print everything, make notes on a 3x5 card so I don’t forget to add something to my system when I’m out and about. When the project is finished I scan it all in and save to devonthink and shred the paper files. This has worked best for me, but I’m an old dog who likes paper!

17

You named the failure exactly. Losing one item does not cost you one item. It costs you your trust in the list, and then you stop relying on it and go back to checking everything by hand.

That is what I keep running into. Every version of this only works if I keep feeding it, and I have to feed it at the exact moment I am busiest. So it works for a while and then quietly stops. Printing survives because it never asks you to do anything later. I do not think paper is the answer, but I think that is the clue.

I agree that one app will not cover everything, and I do not want to replace the tools you use. Each one is better at its own job. What I want is narrower: not one app for everything, but one place where the work from the tools I already use sits next to the files. Then the project does not depend on me keeping it assembled.

Between the task manager and Curio, which one actually held up for you?

18

Both actually. If i received a text, it only took a second to copy and add to my task manager.

Curio allowed me to take a quick screenshot or email url and just copy it into a workspace. Of course both these need processing, but I found that easier. I used cleanshotx for that. Quickly capture a Slack chat and paste it in.

I now do this on paper. I don’t use slack anymore and whatsapp responses needed are written down.

Actually printing is surprisingly helpful as my printer tray acts as an inbox. Plus everything is detached from apps as I have the information I need. Once archived by scanning it is a clean way to archive something as no email, Slack, etc links are needed.

19

It was solved one way, as a bunch of aliases in a single document, with Stapler by Patterson Software Works in 1992 for classic Macintosh.

And similarly in 2009 for Mac OS X, as LaunchList by Ali Rantakari.

You open the document and all your aliases are opened so you can resume your work.

In 2024 I built a modern macOS take on this concept:

1 Like
20

Truthy, but consider that if there is FOMO driving the need to record everything connected to a project, then maybe there’s a project planning deficit. Logging actions taken or messages received or Slack convos completed might suggest too much objectiveless churn. Following a predetermined action plan and recording results is more effective than rabbitholing. Returning to a well planned project after time away should mean looking up the next planned action and proceeding.

Anyway, we’ve sort of entered a new era of quick-capture where most apps have widgets or shortcuts for grabbing message threads and putting them into a log, or grabbing links or page URL, grabbing emails, etc. and putting them into a log.

Katie