The Quiet Power of a Simple Tool

The literature of productivity has a vocabulary problem, and the problem is that almost every word in it is about extraction. Tools are evaluated on throughput, on time saved, on friction removed, on the number of operations they can compress into a smaller number of operations. Users are described as beginners or as power users, which is a scale with only one direction in it. Workflows are optimised. Systems are stacked. Somewhere in the middle of all this the actual experience of using a thing has quietly disappeared, replaced by a measurement of what the thing produced, and a person who simply enjoys opening a piece of software has no place in the discourse at all.

This is a strange omission, because enjoyment is a considerably better predictor of whether a tool will still be in use in three years than any efficiency claim. People abandon efficient systems constantly. They abandon them because the system requires maintenance, or because it nags, or because returning after a gap means relearning something, or because using it feels like reporting to a supervisor. What survives is what a person reaches for without deciding to, which is a description of habit rather than of optimisation, and habits form around things that feel good to use rather than around things that measure well.

There is a specific and underdescribed pleasure involved, and it has almost nothing to do with capability. It is the pleasure of opening something, finding what you were looking for, adding what you came to add, and closing it, with nothing else having happened in between. No prompt about a feature. No suggestion that you might be using it incorrectly. No indicator implying that somewhere there is a fuller version of yourself who has explored more of the menu. Just the work, and then not the work, and the tool that was present for the duration and made no claims on anything else.

That experience is harder to build than a feature list, and it is rarely anybody’s stated goal. It requires restraint that runs against every commercial incentive, since a product that stays quiet cannot demonstrate its value and a product that never interrupts cannot report engagement. It also requires a kind of respect for the user’s attention that is easy to describe and difficult to sustain across a roadmap, because every individual decision to add a small prompt is defensible and the accumulated weight of those decisions is what drives people away.

There is a second element, less about interface and more about ownership, and it turns out to matter enormously to people who would never use that word. A tool that lives on the machine, in a file the person can see, writing into a folder they chose, occupies a different psychological category from a service they subscribe to. It cannot be repriced. It cannot be discontinued. It does not require an account that could be suspended, or a company that might be acquired, or a connection that might not be there. People describe this in the language of possession rather than of architecture, and the sentiment is not naive: it is an accurate summary of a real difference in who holds what.

Put those together and something appears that the productivity conversation has no term for. Not a system, not a workflow, not a stack. Something closer to a familiar object kept in a familiar place, opened most mornings, containing what it contained yesterday, and asking for nothing. That description would sound underwhelming as marketing and is, for a large number of people, the highest praise a piece of software can receive.

Two people who found that experience independently, in professions with nothing whatsoever in common, discovered it by accident over baklava on a Tuesday evening.

Cardamom on Queen Street

Felix Brennan had come to Toronto for a reunion he had not particularly wanted to attend and had enjoyed more than he expected, which is the usual pattern with reunions and one he had failed to learn in forty years.

He was walking back toward his hotel in the late afternoon, along a stretch of Queen Street that was mostly closed for the day, when he stopped without having decided to.

It was cardamom. Unmistakably, and coming from a doorway two shops ahead with the light still on inside.

He stood on the pavement for a moment feeling slightly foolish, the way a man in his seventies does when he catches himself following a smell, and then went in anyway because Ellen was four hundred miles away and nobody was watching.

The bakery was warm in the way that only a room that has had ovens on since five in the morning is warm. There was flour on the floor near the back, which had clearly been swept twice and had returned. A glass case along one side held considerably less than it had held at nine. Behind the counter a man in his early forties was wiping down a steel worktable with the unhurried motion of somebody who had been on his feet for eleven hours and had reached the part of the day that was his own.

‘We are closing,’ Amir Hassan said, without any particular urgency.

‘I am aware and I am going to ask anyway,’ Felix said. ‘Something with cardamom in it. I have been following a smell for a hundred metres and I have my dignity to consider.’

Amir laughed properly at that.

‘Baklava,’ he said. ‘With cardamom in the syrup, which is not traditional and which I do anyway. Sit down. I will make tea, and you will tell me why a man your age is walking down Queen Street sniffing the air.’

They sat at a small table near the window while the light went orange and then grey. Felix explained the reunion. Amir explained the bakery, which had been his father’s and which he had run for twenty-two years and had changed in about nine ways his father would have objected to and two his father would have approved of.

‘And you are retired,’ Amir said.

‘I am retired and I am writing a memoir, which is what retired professors do, and which everybody assumes will be finished considerably sooner than it will.’

‘How long so far?’

‘Two years and a bit. About two hundred pages of book out of about eleven hundred pages of material.’ Felix said it without complaint. ‘The material is the difficulty. Forty years of it in boxes.’

Amir set down the tea.

‘How do you keep it?’ he asked. ‘The material. I am asking selfishly because I have a version of the same problem.’

‘Ah.’ Felix considered whether to give the short answer or the long one and chose the long one, because the tea was hot and he had nowhere to be. ‘It is one file.’

Amir stopped with his hand on the pot.

‘One file.’

‘One file, opened in a browser, connected to a folder on my machine, and every note I have made in two years lives in that folder.’ Felix was watching his face. ‘You know it.’

‘My daughter set it up for me eighteen months ago,’ Amir said, ‘and she said exactly that sentence, in exactly that order, and it is the only reason I agreed to try it.’

Felix put down his cup.

‘Well,’ he said. ‘This is the most improbable thing that has happened to me all week, and I attended a reunion.’

A recipe for baklava and a chapter about a woman named Harriet

They fetched laptops, which took some negotiation, because Amir’s was in the office at the back under a stack of invoices and Felix’s was at the hotel and he had to be persuaded that walking six minutes was not an imposition.

He came back with it and a small bag of pistachios that Amir had insisted on and which he had accepted after the customary resistance.

‘You first,’ Felix said. ‘I want to see what a bakery looks like inside a computer.’

Amir opened it and turned the screen.

‘Four pages,’ he said. ‘Recipes. Suppliers. Daily. Business. That is the whole top of it and it has not changed in a year.’

‘That is fewer than mine.’

‘It was more at the start. Yasmin built me a beautiful structure with branches everywhere and I told her it looked like somebody else’s cupboard, and we cut it back to the four things I actually reach for.’ He opened Recipes, which had sub-pages beneath it. ‘And then it grew again underneath, where it belongs.’

He opened a recipe and Felix leaned in.

The body was a proper document. A heading at the top. A list of ingredients with the quantities in bold, which Amir explained was the only formatting decision he had made deliberately, because at five in the morning a number is the thing you need to find and everything around it is context. Numbered steps below, with the ones that had timing in them also in bold.

‘Why bold the timings?’

‘Because those are the ones you cannot get wrong,’ Amir said. ‘Everything else has tolerance. A minute matters.’

Beneath the body were three collapsed bars.

‘Sections. Ingredients, which is the list in a form I can scale. Procedure, which is the long version with the things that are not steps, like what the dough should feel like. And Variations, which is what I do differently for the family on Beaumont who want everything darker than anybody else in this city wants anything.’

Felix laughed.

‘And files inside the sections?’

‘Photographs, mostly. What it should look like at the halfway point, which is the thing I cannot describe in words and which somebody covering for me needs to see.’ Amir shrugged. ‘And on the supplier entries, invoices, photographed with the phone. That was Yasmin’s idea and I resisted it for two weeks and it has since saved me a considerable amount of money.’

‘Now mine,’ Felix said, ‘and I want you to notice something while I show you, because I have already noticed it and I am going to be insufferable about it in a moment.’

He opened his.

Eight pages at the top level, most of them decades of a career, one of them the manuscript, one called Loose ends. He went down into the manuscript, into a part, into a chapter, and opened it.

The body was fifteen thousand words of prose. A heading structure through it. The occasional block with a different appearance, which he explained were places he was unsure of something and which he cleared before a chapter was finished.

Beneath it were sections.

‘Research notes. Draft text, which is where a scene lives while I am still arguing with it. And Photographs, with the actual images attached, because this chapter is about a woman named Harriet who hired me and fought me for six years, and I have four photographs of her and one of them has her handwriting on the back.’

Amir looked at the screen for a while.

‘It is the same,’ he said.

‘It is exactly the same and that is what I wanted you to see.’ Felix was visibly delighted. ‘A page for the large thing. Entries for the units you actually work on. Sections inside them separating the kinds of content that must not be mixed. Files attached to the section they belong to rather than in a pile.’

‘Your kinds are research and draft and photographs.’

‘And yours are ingredients and procedure and variations, and neither of us was taught this, and neither of us copied the other.’ He sat back. ‘I have spent forty years arguing that the structure of a discipline reflects the structure of its problems. I did not expect to find the argument confirmed by a baklava recipe.’

‘I did not expect to find it confirmed by a memoir,’ Amir said. ‘Although I will say the memoir has fewer photographs of dough.’

‘My memoir has considerably fewer photographs of dough and it is poorer for it.’

The little star, which turns out to be the front door

‘Tell me how your morning starts,’ Felix said. ‘Not the baking. The screen.’

‘The screen is about ninety seconds and it is the same ninety seconds every day.’ Amir demonstrated. ‘I open it, and I open this panel, and there are six things in it.’

The sidebar showed a short list, each item starred.

‘Favourites,’ Felix said.

‘Favourites, and I thought the star was decoration for about four months.’ Amir was cheerfully unembarrassed. ‘A little star. What is that for. And then Yasmin put it on my four base doughs and on the daily list and on the supplier I am currently unhappy with, and now those six things are the first thing I see and I open one of them within four seconds of sitting down.’

‘Which six?’

‘Four doughs, because everything I make comes out of one of them. The Daily page, because that is where today’s note goes. And whichever supplier is currently a problem, which changes about twice a year and which I move deliberately.’ He tapped the screen. ‘That last one is the one I am proudest of, because it means the thing I am currently avoiding is in front of me every morning whether I like it or not.’

Felix made an appreciative noise.

‘You have weaponised it against yourself.’

‘I have put the difficult thing where I have to look at it,’ Amir said. ‘Which is the same technique my father used with a note taped to the till, and which I am aware is not sophisticated.’

‘It is not sophisticated and it works, which is the only test.’ Felix opened his own. ‘Mine is three.’

‘Only three?’

‘Three, and they change roughly every fortnight, and that is the entire discipline.’ He listed them. ‘The chapter I am writing this week. The chapter I am about to write next, so that it is warming up somewhere in the back of my head. And one entry that is not a chapter at all, which is a list of things I have decided the book must say and have not yet said.’

‘And when the chapter is finished?’

‘I unstar it, which is a very small ceremony and which I have come to find genuinely satisfying,’ Felix said. ‘There is a moment where I remove the star and the panel has one fewer item in it, and it is the closest thing to a milestone that this project has.’

Amir considered this.

‘Nobody else knows when you finish a chapter.’

‘Nobody else knows when I finish a chapter. There is no editor, no deadline, no colleague waiting.’ He said it without self-pity. ‘Which means every marker of progress in a project that will take four years has to be invented by me, and a star coming off a list turns out to be sufficient. I would not have predicted that.’

‘I have the same with the label I use for things I am not sure about,’ Amir said. ‘When the count goes down by one it is a good day. It is a stupid little number and I look at it.’

‘It is not a stupid little number.’ Felix said it firmly. ‘It is the only feedback either of us gets. A man baking bread has customers who tell him. A man writing a memoir has nobody at all for four years, and if I could not see that something had moved I would have stopped in the first six months.’

They sat with that for a moment while the light finished going.

‘What I notice,’ Amir said eventually, ‘is that both of us have made the tool into a door rather than a filing cabinet.’

‘Say more.’

‘A filing cabinet you open when you need something. A door you go through every day whether or not you need anything.’ He tapped the sidebar. ‘Six things. I could remember all six. I do not open this because I have forgotten them. I open it because opening it is how the day starts.’

The thing that hands you something you did not ask for

‘Now I am going to show you the one I actually came here to talk about,’ Felix said, ‘and you may have never opened it.’

‘Show me.’

He pointed at a widget low in his sidebar. It held one entry title and a fragment of its opening text, with a small marker beside it.

‘It picks a note at random from everything I have and it refreshes every hour. There is a little indicator when it changes.’

Amir squinted at it.

‘Why?’

‘That is exactly the right question and I asked it too, for about a month, during which I ignored it completely.’ Felix turned the screen slightly. ‘Because search finds what you are looking for. And the entire problem with forty years of material is that most of it has fallen out of my memory so thoroughly that I cannot look for it, because I do not know it exists.’

‘So it shows you things you forgot you had.’

‘It shows me things I forgot I had, at a moment I did not choose, and about nine times out of ten it is a receipt for a projector bulb.’ He grinned. ‘Genuinely. A receipt. And I read it, and I write one line saying it is a receipt, and I move on, and the whole ritual costs eight minutes.’

‘And the tenth time?’

‘The tenth time it handed me a letter from a man who died young, which I had scanned mechanically over three evenings and never read, and there were four sentences on the second page that became an entire chapter.’ Felix said it plainly. ‘I could not have searched for that. I did not know it was in the house.’

Amir was quiet for a second.

‘Try mine,’ he said.

‘You have it?’

‘I have it and I have never once looked at it deliberately.’ He clicked, and the widget changed.

The entry that came up was called something entirely unremarkable, a note from a supplier page seven months earlier. Amir read the preview, read it again, and then put his hand over his eyes.

‘What.’

‘He told me in the spring that the price on the semolina was going up in the autumn and that I should call him before it did to fix a rate,’ Amir said. ‘I wrote it down. That is the note. And it is autumn.’

Felix started laughing.

‘Did you call him?’

‘I did not call him. I have not thought about that conversation since April.’ He was laughing too, in the specific way of a man who has just been saved money by a machine and is not certain how to feel about it. ‘The computer is reminding me to do my job.’

‘The computer is doing nothing of the kind. The computer picked a note at random.’

‘Which is worse,’ Amir said. ‘If it had been clever I could resent it. It was not clever. It was lucky, and I was the one who wrote the note, and I am the one who forgot.’

They talked about the mechanism for a while, and Felix made the argument he had evidently made before.

‘The reason it works is that it is stupid,’ he said. ‘If it selected intelligently, it would show me what I appear to care about, which is what I already know I care about. Random selection is indifferent to my current interests, and my current interests are exactly the problem.’

‘Because you have already thought about those.’

‘Because I have already thought about those, and the things I have not thought about are, by definition, the things nothing would surface if it were trying to be relevant.’ He shrugged. ‘I have started my working day with it for two years. Most days it is a receipt. I would not give it up.’

Amir was already doing something.

‘I am starring that supplier note,’ he said, ‘and I am putting the star on so I see it tomorrow morning, because if I close this laptop now I will forget again and that would be genuinely humiliating.’

‘That is the correct response and it is what I do.’ Felix nodded. ‘Anything the widget hands me that turns out to matter gets starred immediately, because it is random and it will not necessarily come back, and I have learned not to trust my intention to remember.’

Two men looking at a chart of their own persistence

‘There is one more and it is not useful,’ Felix said. ‘I want to say that before I show you. It does nothing. It changes no decision.’

‘Then why show me?’

‘Because it made me sit down once.’

He opened the analytics panel. Numbers appeared: how many entries, how many files, how much space. Beneath them, charts.

Amir looked at Felix’s.

‘That is a lot of entries.’

‘Three thousand and some, which I did not know until I looked, and which represents forty years of paper being turned into something reachable by an old man in a farmhouse over two winters.’ Felix said it evenly. ‘And the chart underneath is activity by day, and it goes back as far as I have been doing this.’

‘It is very consistent.’

‘It is almost embarrassingly consistent, and that is the thing I was not prepared for.’ He tapped the screen. ‘I did not feel consistent. I felt like a man who works when he can and loses whole weeks to nothing. And there is the record, and it says I have worked on this nearly every day for two years, and the weeks I remember as lost were four days.’

‘Memory is not fair to you.’

‘Memory is not fair to anybody about effort,’ Felix said. ‘It records the failures with great precision and the ordinary days not at all. That chart is the only honest account of what I have actually done, and it has kept me going on at least two occasions when nothing else would have.’

Amir opened his own without being asked, which Felix noticed and did not comment on.

There was a pause.

‘Eight hundred and forty,’ Amir said.

‘Entries?’

‘Entries. And nine hundred and something files, most of them invoices and photographs of what things should look like.’ He was scrolling slowly. ‘I thought it was a few hundred. I genuinely thought a few hundred.’

‘Over how long?’

‘Eighteen months.’ He sat back. ‘Which is about forty-five a month, which is a bit more than one a day, which is what it feels like, and yet I would have told you I was bad at this and did not keep it up properly.’

‘Where does that belief come from?’

‘From the days I do not do it, which I remember, and from Yasmin telling me at the beginning that I would not stick with it, which she said kindly and which I have carried around for a year and a half.’ Amir laughed at himself. ‘And there is the number. Eight hundred and forty.’

‘Show her.’

‘I am going to show her tomorrow and I am going to be unbearable about it.’

Felix pointed at the other charts, which broke the material down by label and by page.

‘This one told me something too, and it is smaller and I found it useful,’ he said. ‘My material is not distributed the way I thought. I would have said the manuscript was most of it. The manuscript is a quarter. Three quarters is archive, which is correct and which I would have denied if you had asked.’

‘Why does it matter?’

‘Because I have been telling myself I am behind on the book, and what the number says is that I have spent two years building the thing the book is made of, which is not the same as being behind.’ He shrugged. ‘Nothing about my week changes. But I stopped feeling one particular way about it, and that is worth something.’

Amir nodded slowly.

‘My father would have hated this,’ he said. ‘Counting things.’

‘Would he?’

‘He would have said the bread is the count.’ Amir smiled. ‘And he would have been right about the bread and wrong about the notebooks, because he kept thirty years of them in a cupboard and nobody, including him, ever knew how many there were or what was in them.’

A thing you have rather than a thing you rent

The tea had gone cold twice and been replaced twice, and the street outside had emptied into the particular quiet of a commercial road after seven.

‘Can I tell you the part that actually matters to me,’ Amir said, ‘and which is not a feature?’

‘Please.’

‘This is mine.’ He put two fingers on the closed laptop. ‘The file is on that machine. The folder is on that machine. Nobody can raise the price of it. Nobody can decide to stop offering it. There is no account for anybody to suspend and no month in which I forget to pay something and lose eighteen months of recipes.’

‘You have had that happen.’

‘Not with recipes. With the accounting, years ago, and it was resolved and it was a fortnight of genuine fear.’ Amir said it plainly. ‘I run a bakery on margins that are thinner than anybody outside this trade believes. Every subscription I have is a small permanent claim on money I have not earned yet, and every one of them can change its terms and I will find out in an email I do not read.’

‘So it is partly economic.’

‘It is partly economic and it is mostly something else that I find harder to say.’ He thought about how to put it. ‘My father owned his tools. Actual tools. The proving cabinet is from before I was born. The scales were his father’s and they are wrong by four grams and I know it and I correct for it. Nothing in this room was rented.’

Felix set down his cup.

‘And a subscription is a different relationship.’

‘A subscription is a relationship where somebody else decides how long it lasts.’ Amir shrugged. ‘Which is fine for many things and is not fine for the place I keep twenty-two years of what I know.’

Felix was quiet for a moment before answering, and when he did it was with the specific weight of somebody who had lost something.

‘I have had it happen,’ he said. ‘Twice, professionally. A company that made a citation and reference system, which most of my department used for about six years, and which was acquired and then discontinued, and there was an export process which mostly worked and which produced a file nothing else could read properly.’

‘What did you lose?’

‘Annotations, principally. Six years of margin notes on several thousand references, which were the actual intellectual content and which the export treated as decoration.’ He said it without drama, which made it worse. ‘I still have the references. Anybody can get the references. What was lost was what I thought about them.’

‘And the second time?’

‘A writing application, less consequential, and I had learned by then and had kept copies.’ Felix turned a hand over. ‘Which is the point. After the first, everything I do is arranged around the assumption that the software will eventually go away.’

‘Which this survives.’

‘Which this survives, and I want to be precise about why, because it is not a promise anybody made me.’ He counted it out. ‘My material is a folder. Inside it there is a file holding the entries and the structure, in a format that is plain text. There are separate files holding the bodies of my notes, in a format I can open in anything. There is a folder of the actual images and documents I attached, in their original forms. And there is one file that is the application itself.’

‘You have opened them.’

‘I opened them in my first week, out of a suspicion I could not have named,’ Felix said, ‘and I found my own words sitting there in ordinary text. Which settled something.’

‘I did the same,’ Amir said. ‘Exactly the same, in the first fortnight, and I could not have told you what I was checking for.’

‘You were checking whether it was really there.’

‘I was checking whether it was really there, and it was, and I have not looked since.’ He laughed. ‘Yasmin thought it was very funny.’

‘Yasmin is young.’

‘Yasmin is twenty-four and has never in her life had a thing taken away by a company,’ Amir said. ‘She will. And when it happens she will remember that her father opened a folder to look at his own files and she thought it was quaint.’

Felix laughed properly at that.

‘The word I would use is possession,’ he said. ‘Which sounds grand for a folder on a laptop. But there is a difference between something you have and something you are permitted to use, and everybody feels it even when they cannot articulate it.’

‘And it changes what you put in.’

‘It changes what I put in enormously.’ Felix nodded. ‘There are things in my vault I would not put into anything I did not own. Not because they are scandalous. Because they are mine, and I want the arrangement to reflect that.’

A cupboard of notebooks and eleven cardboard boxes

‘I want to ask about your father,’ Felix said. ‘You have mentioned him three times and each time you have stopped.’

Amir turned his cup a quarter turn.

‘He kept notebooks,’ he said. ‘By the till. Thirty years of them, in a cupboard in the back, and I have them still and I cannot read them.’

‘His handwriting.’

‘His handwriting was a private language and he knew it and he did it deliberately, I think, because a man who writes his supplier prices where the supplier can read them is a man with a problem.’ Amir smiled without much in it. ‘So I have thirty years of what he knew and I have almost none of it.’

‘Have you tried?’

‘I spent a winter on it, about six years ago, with a magnifying glass and a great deal of tea, and I got perhaps four things out that I use to this day.’ He held up fingers. ‘A trick for keeping a proving room stable in February. What he paid for pistachios in a year I could date from the price. The name of a supplier’s father, which meant something when I met the son. And a note that said, in words I could read, do not agree to wedding cakes, which he had underlined twice.’

Felix laughed.

‘Did you agree to wedding cakes?’

‘I agreed to wedding cakes for three years and then I stopped, and when I stopped I found that note again and I sat down on the floor of the storeroom.’ He shrugged. ‘He was right and I could not read him in time.’

Felix did not say anything for a moment.

‘I have eleven boxes,’ he said. ‘Which is a smaller number and a similar problem. Lecture notes across four decades, correspondence, photographs, drafts of things that were never published and things that were. All of it mine, all of it in my own hand or my own typing, and until two years ago all of it as unreachable as your father’s cupboard.’

‘Because of the volume rather than the handwriting.’

‘Because of the volume, which produces the same result by a different route.’ He turned a hand over. ‘Thirty years of notebooks nobody can read and eleven boxes nobody can search are the same object. What matters is not whether the information survived. It is whether it is reachable, and unreachable is functionally identical to lost.’

‘That is a hard sentence.’

‘It is a hard sentence and it took me about a year to be willing to say it,’ Felix said. ‘Because I had spent forty years believing that keeping things was the responsible act, and what I have had to accept is that keeping without arranging is a kindness to yourself that costs the next person everything.’

Amir was quiet.

‘My daughter can read mine,’ he said. ‘That is what you are saying.’

‘I am saying that in about nine seconds you understood the point I spent a winter arriving at, and yes.’ Felix nodded. ‘Yasmin can open your folder. She can read your recipes in ordinary words, in a form nobody has to decipher, with the photographs sitting where they belong. Which is a thing your father could not give you and you have already given her without meaning to.’

‘I did it because I could not read my own handwriting either.’

‘The reason does not matter. The outcome is the same.’ He picked up his tea and found it cold again. ‘Mine goes to my daughter as a folder and a sealed envelope with instructions, which was the most uncomfortable afternoon of my retirement and which I would recommend to anybody.’

The page that is not about work at all

‘I have a page I have never told anyone about,’ Amir said, ‘and I am going to tell you because you are leaving on Thursday and you live in Vermont.’

‘That is the most honest reason anybody has ever given me for a confidence.’

‘Under Business there is a sub-page called Failures.’ He said it plainly. ‘Nine entries. Everything I have done wrong that cost real money or real trust.’

‘Give me one.’

‘A cake I agreed to make for a wedding I should have declined, which I have already mentioned and which is entry number two.’ He counted on his fingers. ‘A supplier I stayed with for four years past the point where I knew. A member of staff I did not let go for eleven months after I should have, which was the most expensive thing on the list and not in money. A croissant batch I ruined by trusting a thermometer that lied, which is trivial and which is on there because the lesson is check the instrument.’

‘Why keep them?’

‘Because I made every one of them at least twice before I wrote any of them down,’ Amir said. ‘That is the whole reason. Not sentiment. Twice.’

Felix set his cup down very deliberately.

‘I have that page,’ he said. ‘It has a different name and it is the same page.’

‘What do you call it?’

‘Loose ends, which is a cowardly name and I know it.’ He was rueful. ‘Sub-pages under it. Things I got wrong professionally, which is longer than I would like. People I have not forgiven, which is shorter than it was two years ago and I attribute that entirely to having written it down. And one that is simply my wife’s name.’

‘What is in that one?’

‘Things I intend to say to Ellen and have not, and things I have said and want to remember saying.’ He said it without embarrassment. ‘It is the only page in the entire vault I have ever considered deleting and I never will.’

Amir nodded slowly.

‘Neither of those is what the software is for.’

‘Neither of those is what any software is for, and both of us built them anyway, which I suspect is the most interesting thing about this entire evening.’ Felix leaned back. ‘Nobody designed a place for a man to keep a list of his own mistakes. We each made one, in a tool intended for recipes and research, because we needed somewhere to put it and there was somewhere to put it.’

‘Does yours help?’

‘It does something. Whether it helps is a question I would want more evidence for.’ He thought about it. ‘What I can say is that the material exists either way. The version I have written down sits on a page. The version I have not written down sits in me at four in the morning. Those are not the same and I know which I prefer.’

‘That is exactly it,’ Amir said. ‘My nine entries do not stop me making the tenth. But I am not carrying the nine any more.’

An extra box, and a request

It was past eight when Felix looked at his watch and made a noise.

‘I have kept you three hours past closing.’

‘You have kept me three hours past closing and it is the best evening I have had in about two months,’ Amir said. ‘Sit for four more minutes while I box something.’

‘I do not need anything else.’

‘It is not for you.’ He was already at the case. ‘You mentioned a wife who is four hundred miles away and who did not come to the reunion, and a man who arrives home from a trip with nothing is a man who has learned nothing in forty-one years of marriage.’

Felix opened his mouth and closed it again.

‘Forty-one is correct,’ he said. ‘I did not tell you that.’

‘You told me you had been married forty-one years about ninety minutes ago while explaining why you press the lock button when she comes into the study.’ Amir grinned. ‘I listen. It is most of what I do.’

He boxed a dozen, then added two more without saying anything, then tied it with the string in a way that suggested he had done it perhaps three hundred thousand times.

‘Ask me for something,’ he said, handing it over. ‘Because I have given you something and it makes the evening uneven.’

‘That is not how gifts work.’

‘That is exactly how gifts work in this shop and you are not going to win.’

Felix considered.

‘Then here is what I will ask,’ he said. ‘Write down the semolina thing tonight rather than tomorrow. Not the star. The actual note about calling him, with the date on it, before you close that laptop.’

Amir laughed.

‘You are asking me to do the thing I have been avoiding.’

‘I am asking you to do the thing you were saved from by an accident, and to not rely on being saved twice.’ Felix was smiling. ‘And I will tell you what I want in return, which is the real request.’

‘Go on.’

‘If the book ever exists, I would like to sign a copy for this shop.’ He said it slightly awkwardly, the way people do when they mean something. ‘Not for you. For the shop. Somewhere behind the counter where nobody looks at it.’

Amir was quiet for a second.

‘My father kept a photograph of the first oven behind the till for thirty years,’ he said. ‘Nobody looked at that either.’

‘Then it will be in good company.’

‘It will be in good company and you will be here in person to put it there,’ Amir said, ‘and I will make the cardamom syrup and you will bring your wife.’

They exchanged details on paper, because neither of them suggested anything else, and Felix walked back to the hotel with a box under one arm through a street that had gone properly cold.

A tool you stop noticing

What that conversation illustrates is difficult to express in the vocabulary the productivity industry has built, because none of it is about efficiency and all of it is about relationship.

Two people arrived at structurally identical systems without ever having discussed it: a page for the large thing, entries for the units they actually work on, sections inside those separating kinds of content that must not be blurred together, and files attached to the section they belong to rather than accumulating in a pile. One of them was organising baklava recipes and supplier invoices, the other forty years of academic correspondence and a memoir. The convergence is not a fact about software. It is a fact about how work is shaped, and the tool’s contribution was to have a form flexible enough to hold both without either person learning anything specialised.

The features they each described as most valuable were also, notably, the smallest ones. A star that produces a short list, opened first thing every morning, functioning as a door rather than as a filing cabinet: six items for one of them, three for the other, none of which either would actually forget, and both of them opening it daily anyway because opening it is how the day begins. Removing a star when a chapter is finished, which is the only milestone available in a project with no editor and no deadline and no colleague waiting. A widget that hands over something random once an hour, which is a receipt nine times out of ten and occasionally a letter that becomes a chapter or a supplier conversation that was about to cost real money.

The analytics panel is perhaps the clearest example of value that has nothing to do with productivity. It changes no decision and produces no action. What it does is present an honest account of sustained effort to somebody whose memory of that effort is systematically unfair, recording failures precisely and ordinary days not at all. A retired professor discovering he has worked nearly every day for two years, and a baker discovering he has made eight hundred and forty entries in eighteen months after being told at the outset that he would not stick with it, are both encountering the same thing: evidence that contradicts a belief they held about themselves. No gamification mechanism produces that, because gamification rewards behaviour rather than reflecting it.

Underneath all of it sits the property both of them reached for the word possession to describe. A file on a machine, writing into a folder they chose, holding entries in plain text and note bodies in readable files and attachments in their original forms. Neither of them uses that arrangement because they understand its architecture. They use it because they have each been on the wrong end of a company’s decision, or expect to be, and because a tool that cannot be repriced, discontinued, or suspended occupies a different category from one they are permitted to use. Both had opened the folder in their first fortnight to check that their own words were genuinely sitting there, neither could explain at the time what they were checking for, and both stopped worrying afterward.

Which is the actual measure worth using. Not how much a tool can do, but whether a person opens it without thinking about it, finds what they came for, adds what they came to add, and closes it having thought about nothing except their own work. A baker checking a dough. A memoirist continuing a chapter. Neither of them considering the software at all, which is the highest thing that can be said about it.

FAQs

Why does enjoyment matter more than efficiency when choosing a tool?

Because enjoyment predicts whether something will still be in use in three years, and efficiency does not. People abandon efficient systems constantly, usually because the system requires maintenance, interrupts them, nags about features, or makes returning after a gap feel like relearning. What survives is what somebody reaches for without deciding to, which is a description of habit rather than optimisation. Habits form around things that feel good to use, so the pleasure of an uneventful session is a better indicator of long-term value than any measurement of time saved. The useful question when evaluating anything is therefore not how much time it saves but whether opening it is pleasant, since the second determines whether the first will ever be realised.

What does a genuinely frictionless session look like?

Opening something, finding what you were looking for, adding what you came to add, and closing it, with nothing else having happened in between. No prompt about an unexplored feature, no suggestion that the tool is being used incorrectly, no indicator implying a fuller version of yourself exists elsewhere in the menu. This sounds unremarkable and is unusually difficult to build, since every individual decision to add a small prompt is defensible while the accumulated weight of those decisions is what eventually drives people away. Most people cannot describe the experience except by contrast, noticing it only after using something that constantly asks for attention and then returning to something that does not.

Why do people in completely different fields arrive at the same structure?

Because the structure reflects the shape of the work rather than the design of the software. A page for the large stable thing, entries for the units that actually get worked on, sections inside those separating kinds of content that must not be blurred, and files attached to the section they concern. A baker organising recipes and a memoirist organising chapters produce identical diagrams with different labels, having never discussed it. The convergence suggests the arrangement is what the problem looks like once solved rather than a convention anybody was taught. Anybody unsure how to organise their own material can take some comfort from this, since it suggests the correct shape is discoverable through use rather than requiring expertise.

What belongs in sections rather than in the main body of an entry?

Anything that is a genuinely different kind of content about the same subject. A recipe separates the scalable ingredient list from the procedural detail and from the variations made for particular customers. A memoir chapter separates research notes from draft text from photographs. Keeping them apart preserves a distinction that prose would blur, and attaching files to the specific section they belong to means locating a document becomes a matter of knowing what kind of thing it is rather than remembering what it was called. A reliable test is whether merging two parts would make either harder to read or harder to trust, which is usually obvious once the question is asked directly.

How should formatting be used by somebody who is not a designer?

Sparingly, with each choice carrying one consistent meaning. A baker who bolds quantities and timings does so because those are the elements that must be found instantly at five in the morning and the ones with no tolerance for error, while everything else is context. Applying that single decision consistently across every recipe produces a document that can be scanned under pressure. The principle generalises: pick one or two conventions, apply them without exception, and resist adding a third until an existing one proves insufficient. The failure mode is never too little formatting; it is a convention that has quietly acquired a second meaning and therefore stopped signalling anything at all.

What is the favorites panel actually for?

Acting as the front door rather than as a retrieval mechanism. Most people who use it heavily keep between three and six items there, all of which they could recite from memory, and they open the panel every morning anyway because opening it is how the day starts. The distinction is between a filing cabinet, which is opened when something is needed, and a door, which is walked through daily. Once a small set of items becomes the entry point, the rest of the system organises itself around that habit. Keeping the list short is what preserves the effect, since a panel containing twenty items has become another list to read rather than a door to walk through.

What should go into a short favorites list?

The handful of things opened most often, plus one deliberate inclusion. Practitioners commonly star their core reference material, the place where today’s work is recorded, and something they are currently avoiding. That last item is the interesting one: placing an uncomfortable task where it must be seen every morning is a small self-imposed control, and people describe it as the entry they are proudest of. The technique is not sophisticated and predates software entirely, but it works for the same reason a note taped to a till works. Reviewing the list every few weeks and removing anything that has stopped being opened keeps it honest, and takes under a minute.

Why does removing a favorite feel significant?

Because in long solitary projects it may be the only available milestone. Somebody writing a book with no editor, no deadline, and no colleague waiting receives no external signal that anything has been completed, and a project measured in years requires markers of progress that the person has to invent. Unstarring a finished chapter, and seeing the list shrink by one, provides that signal. It sounds trivial and practitioners describe it as genuinely satisfying, which is a reasonable indication that the need it meets is real. Anybody working alone on something long should deliberately construct a marker of this kind, since the absence of one is a common reason such projects quietly stop.

What is the random note spotlight and why does it help?

It surfaces one entry chosen at random from the entire collection, refreshing hourly, with an indicator when the selection changes. Its value is that search finds only what somebody can describe, while the central difficulty in a large archive is material that has fallen out of memory entirely. There is no query for a thing you have forgotten owning. Random surfacing addresses that directly, and the fact that it makes no attempt to be relevant is what makes it work, since relevance would return only what the person already knows they care about. Its indifference to context is therefore a design property rather than a limitation, and attempts to make such a mechanism smarter would remove the thing that makes it valuable.

Is it worth engaging with the widget when most results are trivial?

Yes, and the triviality is the price of admission. Most days produce something unremarkable: a receipt, an administrative note, a routine record. Reading it anyway and writing a line about it costs a few minutes. Users who engage only when something looks promising stop engaging entirely, because promise is not visible from a title. The ones who treat it as a fixed daily ritual are the ones who report the significant rediscoveries, since they kept the channel open long enough for the improbable item to arrive. Writing a single line about whatever appears, even when it is plainly unimportant, is what keeps the habit intact on the days that produce nothing.

What kinds of things does random surfacing typically recover?

Anything written down at a moment when its importance was clear and then forgotten. Practitioners describe recovering correspondence they scanned mechanically and never read, notes from conversations that became chapters, and practical commitments with real financial consequences. One reported having recorded a supplier’s warning about a price increase, along with an instruction to call before it took effect, and being handed that note back by the widget in the relevant season having not thought about it for months. The person wrote the note; only the timing was luck. The pattern is consistent: the person captured the information at a moment when its importance was obvious, and the only thing supplied later was the timing.

What should be done when something surfaced turns out to matter?

Capture it immediately rather than intending to act later. Random selection carries no guarantee of returning to the same item soon, and intention to remember is unreliable, particularly at the end of a working day. Starring the entry so it appears in the daily list, or writing an explicit note with a date attached, converts a fortunate moment into something that will actually be acted upon. Practitioners who have been saved once by an accident generally conclude that relying on being saved twice is not a strategy. Doing it in the same minute, before closing anything, is the discipline, since the intention to act tomorrow is exactly what failed the first time.

What is the analytics panel for if it changes no decisions?

It provides an honest account of sustained effort to somebody whose memory of that effort is systematically unfair. Memory records failures precisely and ordinary days not at all, so people routinely believe they have been inconsistent when the record shows otherwise. Seeing entry counts, file counts, and an activity chart spanning months or years frequently contradicts a belief somebody held about themselves. It produces no action and changes no plan, and practitioners describe it as having kept them going during periods when nothing else would have. Looking at it deliberately during a low period, rather than only in passing, is when it does the most good and when people are least likely to think of it.

How does this differ from gamification in other applications?

Gamification rewards behaviour by adding artificial signals such as streaks, badges, or points, which are designed to influence what the user does. An activity chart reflects behaviour that already occurred, using only the record the person generated by working. The difference matters because manufactured rewards eventually feel manipulative and lose their effect, while a plain count of what was actually produced remains credible indefinitely. Nobody feels congratulated by a number they generated themselves, which is precisely why the number is believable. It also means the figure remains meaningful over years, whereas artificial rewards need constant escalation to keep producing any effect at all. Anybody who has felt vaguely manipulated by a streak counter will recognise the difference immediately on encountering a chart that simply reports.

What do the label and page distribution charts reveal?

Usually that material is not distributed the way its owner assumed. A memoirist may discover that manuscript work represents a quarter of the archive while three quarters is source material, which contradicts a belief about being behind on the book and reframes two years as having built the thing the book is made from. A small business owner may find that one category dominates their record keeping. Neither observation changes a week’s schedule, and both change how somebody feels about their own progress. Checking the distribution once or twice a year is enough, since it changes slowly and the value lies in correcting an assumption rather than in tracking a trend.

Why do people describe a local tool as a possession rather than as software?

Because the distinction they are drawing is accurate. Something living on the machine, in a file they can see, writing into a folder they chose, cannot be repriced, discontinued, or suspended, and requires no account that could be lost. A subscription is a relationship where somebody else decides how long it lasts, which is acceptable for many things and uncomfortable for the place where somebody keeps decades of what they know. People express this in the language of ownership because that is genuinely what differs. The sentiment is not naive: it is an accurate summary of who holds what, expressed in ordinary language rather than in architectural terms.

Is the concern about losing access to a tool realistic?

It is common enough that many long-term practitioners have experienced it. Products get acquired and discontinued, export processes that mostly work produce files nothing else reads properly, and what survives is frequently the least valuable part. One reported case involved a reference system where the citations exported cleanly and six years of margin annotations, which were the actual intellectual contribution, did not. People who have lost something once tend to arrange everything afterward around the assumption that any given tool will eventually go away. Anybody who has not experienced it yet is advised to arrange their material as though they eventually will, since the cost of that precaution is close to nothing.

What makes a local vault survive the disappearance of its software?

The formats involved. A folder containing a file that holds entries and structure as plain text, separate readable files holding note bodies, and attachments stored in their original forms remains usable with ordinary tools regardless of what happens to the application. That is why the arrangement survives an acquisition, a discontinuation, or a change of direction: the material was never inside the program. The application is a convenient way of working with files that would still be readable without it. Verifying this once, by opening the folder and reading a note in a plain text editor, converts an architectural claim into something personally confirmed. This is also why backup is straightforward, since copying the folder produces something immediately usable rather than an archive requiring restoration.

Why do people open the vault folder in their first weeks?

Out of an unease they usually cannot articulate, and it settles the question permanently. Practitioners describe opening the storage folder within the first fortnight, without any particular reason, and finding their own words sitting in ordinary text alongside their attachments in ordinary formats. Neither could explain at the time what they were checking for. The answer is generally whether the material was really there, and having confirmed it once by looking, most people never check again and stop worrying about it entirely. Anybody feeling that unease and unable to name it should simply look, since the check takes two minutes and settles the question for good.

Does ownership change what people are willing to write down?

Substantially, and this is the practical consequence of the psychological difference. People report keeping material in a tool they own that they would not place in a service they subscribe to, and the reason is rarely that the content is scandalous. It is that the material feels like theirs and they want the arrangement to reflect that. Since the value of any personal archive depends on what actually gets recorded, a tool that increases willingness to write is producing content rather than merely storing it. This is why the psychological property is not merely pleasant but practically consequential, since an archive is only as good as what somebody was willing to put into it.

Is this approach only suitable for people with simple needs?

The distinction is not simple against complex but self-contained against dependent. The two practitioners described here maintain hundreds of structured entries with sections, attachments, labels, and years of accumulated material, which is not a simple system by any reasonable measure. What is simple is the relationship: one file, one folder, no account, no subscription, no connection required. Capability and demands on the user are independent properties, and the combination worth seeking is high capability with low demands rather than either alone. Judging a tool by how much it demands, rather than by how much it offers, tends to produce better long-term choices than the reverse.

How do the small features end up mattering most?

Because they attach to daily habits rather than to occasional tasks. A star that produces a short list is trivial in isolation and becomes the entry point to somebody’s working day. A widget offering a random entry is a novelty until it becomes a morning ritual performed for two years. A count of entries is meaningless until it corrects a belief somebody held about their own persistence. In each case the mechanism is negligible and the position it occupies in a routine is what generates the value. Anybody evaluating a tool should therefore pay particular attention to whatever they would touch every single morning, since that is where the actual relationship forms.

What does it mean for a tool to become transparent?

That the person stops thinking about it. A baker checking a dough temperature and a memoirist continuing a chapter are both thinking about their own work rather than about software, which is the condition every tool should aim for and few achieve. Transparency requires stability, since a changed interface forces attention back onto the tool; restraint, since prompts and suggestions do the same; and reliability, since anything that fails becomes the only thing the user can think about for the rest of the session. The clearest sign that it has been achieved is that somebody cannot easily describe the software they use every day, because they have stopped noticing it.

What is the best way to evaluate whether a tool suits you?

Notice what happens during an ordinary session rather than counting what the product can do. Does anything interrupt. Does anything need attention before work can begin. After a two-week gap, is anything different. Does returning to it feel like resuming or like re-learning. Those questions are answerable within a fortnight of normal use and predict long-term outcomes far better than a feature comparison, which measures what a product contains rather than what using it is actually like. Deliberately leaving it untouched for a fortnight and observing what returning feels like is the single most informative test available and costs nothing to run. Trusting your own reaction during that fortnight matters more than any review, since the person best placed to judge whether something is pleasant to use is the one using it.

Where should somebody start if they want this kind of relationship with a tool?

With a small number of top-level categories matching how they already think about their work, and nothing else designed in advance. Elaborate structures built at the outset tend to be abandoned because they reflect somebody’s idea of organisation rather than the actual shape of the work. Starting with three or four pages, adding structure underneath as it becomes necessary, and starring the handful of things opened most often produces a working arrangement within a week that will still be in use years later. Resisting the urge to design everything in advance is the main discipline, since a structure built before any real material exists is nearly always the wrong one.

What is the difference between keeping material and making it reachable?

Practically none, from the perspective of anybody who needs it. Thirty years of notebooks in illegible handwriting and eleven boxes of well-organised paper that cannot be searched are the same object: information that survived and is functionally unavailable. Unreachable is not a lesser form of lost, and it produces identical outcomes. Many people who have inherited a predecessor’s records discover this uncomfortably, having assumed that the act of keeping was the responsible part. Keeping without arranging is a kindness to oneself that transfers the entire cost to whoever comes next. Recognising this early changes what somebody does with an accumulating archive, since arranging is the part that determines whether keeping was worth anything.

Why does a searchable archive matter for succession specifically?

Because the person who inherits it did not write it and has no memory to compensate with. A private shorthand, an idiosyncratic filing scheme, or handwriting only its author could read all work perfectly for one person and fail completely for everyone else. Material stored as ordinary text, in a folder that can be opened, with attachments in their original forms, transfers without deciphering. People frequently arrive at this arrangement for entirely selfish reasons and only later realise they have solved a problem their own predecessor could not. Writing a short note explaining the structure alongside the material makes the transfer easier still, and takes a few minutes to produce.

What happens when somebody tries to recover an unreadable archive?

Usually a great deal of effort for a small return, and the return arrives too late to matter. One reported case involved a winter spent with a magnifying glass extracting four usable items from thirty years of a parent’s notebooks, one of which was a warning that would have prevented three years of unprofitable work had it been legible at the time. The lesson people draw is not about handwriting but about timing: information that cannot be reached when the relevant decision is being made has not been preserved in any useful sense. The effort is rarely wasted entirely, and it is almost never proportionate to what could have been achieved by arranging the material while its author was still available.

Should personal material live in the same system as professional records?

Frequently yes, in a separate branch. Practitioners in very different fields independently create pages for material that no software was designed to hold: lists of mistakes that cost money or trust, grievances being worked through, things intended to be said to a family member and not yet said, professional decisions still being reconsidered years later. None of it is what the tool is for, and it gets built anyway because the person needed somewhere to put it and somewhere existed. Keeping it in a distinct branch preserves the boundary while ensuring it actually gets written. Naming that branch plainly rather than euphemistically also helps, since a page somebody is embarrassed to open is a page that stops being written in.

Why do people keep records of their own mistakes?

Because mistakes repeated a third time are the expensive ones, and the pattern of repetition is invisible without a record since each instance feels like bad luck at the time. Practitioners commonly report having made every entry on such a list at least twice before writing any of them down. The value is not sentimental and not punitive. A short account of what went wrong, written at the time, converts a scattered set of frustrations into a reference that can be consulted before a similar decision is made. Writing each account at the time, while the details are accurate and uncomfortable, is what makes the collection worth anything years later.

Does writing something down actually change how a person carries it?

The consistent report is that it changes where the material sits rather than resolving it. Something written has a location, and material with a location competes less for attention than material being held. Practitioners describe lists of grievances shortening over years, and describe being aware that recording a mistake does not prevent the next one while noticing that they are no longer carrying the previous nine. The material exists either way; the distinction is between a version that sits on a page and a version that surfaces at four in the morning. Whether that constitutes help is a reasonable question; that it differs from not writing it down is not seriously disputed by anybody who has tried both.

What kinds of entries do people say they would never delete?

Almost always the personal ones rather than the professional. The page holding things intended for a spouse, the record of mistakes, the account of a situation that was never resolved: these are the items people describe having considered deleting and decided against. It is worth noting because it says something about what these archives become. A system adopted to organise recipes or research ends up holding the material its owner would least willingly lose, which was nobody’s plan at the outset. It also suggests that any tool intended for professional use should be judged partly on whether somebody would trust it with material that is not professional at all.

How should somebody plan for their archive being read by others?

By deciding in advance which parts are intended for whom and writing that down alongside the material. Professional records generally transfer with the responsibility they belong to. Personal branches usually should not, and a request to that effect placed with the instructions is the honest arrangement, since it cannot be enforced once credentials have been handed over. Practitioners describe the exercise as the most uncomfortable afternoon of a retirement and recommend it anyway, largely because the alternative leaves the decision to whoever finds the machine. Doing it while nothing is urgent produces a far better arrangement than attempting the same decisions during an illness or a departure.

Does a tool like this suit somebody who describes themselves as bad with technology?

It suits them unusually well, and the population using it this way is larger than most people assume. The setup consists of opening one file and choosing a folder. Formatting is done with a toolbar whose buttons behave as expected, and most long-term users rely on a handful of them. Nothing has to be installed, no account exists, and there is no version to keep current. The judgment required is entirely about the content, which belongs to whoever knows the subject rather than to whoever knows software. Many of the most committed long-term users describe themselves this way and mean it, having abandoned several more elaborate products before finding something that asked nothing of them. Being outside computing has no bearing on whether somebody can decide what is worth writing down.

How long does it take before somebody trusts a system like this?

Commonly around five or six weeks, and most people can identify the moment. It tends to be the first return after a genuine gap: a busy fortnight, a trip, a period when nothing was opened. Finding everything unchanged, with nothing queued and nothing demanding attention, does more for confidence than any amount of correct behaviour during daily use. Until that point people generally maintain a parallel record out of caution, and the quiet disappearance of that habit is the real marker of adoption. Knowing the moment is coming makes the early weeks easier, since the sensible response to initial scepticism is to keep the parallel record and allow it to become unnecessary rather than forcing the issue. Nobody should be talked out of caution about their own material.

What does a mature personal vault actually look like after a couple of years?

Uneven and unremarkable. A few top-level pages matching how somebody already thinks about their life or work, deeper structure under whichever branch gets used most, hundreds of entries of varying quality, and attachments accumulating quietly in the background. A handful of items are carefully written and most are not. Nothing about it looks impressive, which is the point: the effort went into capture and placement rather than into curation, and the result is that finding something is fast. Anybody expecting to build something elegant tends to be disappointed, and anybody who tries to find something in a well-used vault is usually surprised by how quickly it goes. The unevenness is a sign of genuine use rather than of neglect.

Is there a risk of putting too much into one system?

The genuine risk is the opposite, since material left outside a system is material that cannot be found. Volume itself costs very little once contents are searchable, and the small amount of discrimination worth applying is best done at capture: a descriptive title, a section that says what kind of thing something is, a short line explaining why it mattered. Attempting to curate afterward is demoralising and rarely completed, while a light habit applied at the moment of writing survives indefinitely. The asymmetry decides it: an unnecessary entry costs a few seconds, while material left outside the system costs whatever it would have been worth when somebody needed it and could not find it. Erring toward capture is the cheaper mistake by a wide margin.

What should somebody do if they have tried and abandoned other systems?

Examine why they stopped rather than assuming a lack of discipline. Most abandonment traces to specific causes: the system required maintenance, it interrupted, it changed underneath them, returning after a gap meant relearning something, or it implied they were using it incorrectly. Each of those is a property of the tool rather than a failing of the user. Choosing something that stays quiet, stays stable, and asks for nothing addresses the actual cause, and people who have abandoned several things often stay with such a tool for years. It is also worth writing down what went wrong with each previous attempt, since the pattern across three or four abandonments usually points at one specific property to avoid next time. That short exercise saves considerably more time than it takes.

Why do two people in unrelated professions find so much to discuss?

Because the underlying problems are identical and the surface vocabulary is not, which makes each of them a useful outside reviewer of the other. Somebody who knows nothing about a field cannot be distracted by whether the content is correct and will notice structure, habits, and assumptions that a peer would read straight past. Practitioners consistently report learning more from a conversation with somebody in an unrelated field than from an equivalent hour with a colleague who shares every assumption they have. Seeking such conversations out deliberately is worth the effort, and the exchange tends to be genuinely mutual rather than one party instructing the other. Both usually leave intending to change something.

What is the most encouraging thing about this kind of system?

That it accumulates value without anybody having to try. Somebody writing a few entries a day for eighteen months produces several hundred without ever feeling productive, and discovers the total only by looking. The archive gets better by being used rather than by being tended, which is the reverse of most organisational schemes and the reason these survive. The effort required in any single session is a few minutes, and the result after two years is something its owner would not willingly lose. This is also why the advice for beginners is simply to start rather than to plan, since the accumulation does the work that no amount of initial design could achieve. Two years of unremarkable daily use beats any elaborate arrangement built in a single weekend.

How many top-level pages should somebody have?

Fewer than they initially build. Practitioners who last tend to settle on three to five, matching categories they already use in conversation about their own work, with depth added underneath as it becomes necessary. Elaborate structures built at the outset usually reflect somebody else’s idea of organisation and get quietly abandoned, since the person opening them does not recognise their own thinking in the arrangement. Starting narrow and letting the tree grow from actual use produces something both more accurate and more likely to survive. Adding a page is trivial and removing one after it has accumulated material is not, which is a good argument for waiting until the need is obvious rather than anticipating it. Most people find their tree settles within about three months of real use.

What happens when somebody sets up a system on another person’s behalf?

It works best when they build almost nothing. The common failure is constructing a thorough structure that reflects how the helper thinks, which the eventual user recognises immediately as somebody else’s arrangement and stops opening. A small number of pages matching categories the person already talks about is enough, and everything added afterward by the user themselves tends to be better, because it reflects the real shape of their work. It also produces ownership, which determines whether the system survives a busy month. Checking back after a month to see what the person has added themselves is far more useful than any amount of instruction, since what they built reveals what they actually needed.

Is it worth keeping a short daily log?

Almost always, and its value comes from accumulation rather than from any individual entry. A few lines a day recording what happened, what sold or failed, what somebody said, looks like nothing for months and after a year or two becomes the only record of the seasonal shape of a business or a project. The practice long predates software; the difference is that a written log can be searched, read by somebody other than its author, and survives the person who kept it, which a private notebook generally does not. Keeping it deliberately brief is what allows it to survive, since a log requiring ten minutes gets skipped on precisely the busy days most worth recording. Four lines is enough.

How should somebody handle the material they inherit from a predecessor?

By extracting what can be extracted early rather than intending to work through it eventually. Inherited archives tend to be unreachable in some specific way, whether through handwriting, volume, or an idiosyncratic scheme, and the effort required grows as memory of the original context fades. Anything recovered should be brought into the current system in ordinary language rather than preserved in its original form, since the objective is that the next person after this one does not face the same problem. Doing a small amount of extraction while somebody who knew the author is still available multiplies what can be recovered, and that window closes without warning.

Why do people underestimate how much they have written?

Because memory records failures precisely and ordinary days not at all, which produces a systematically unfair account of one’s own consistency. Somebody writing a few entries most days for a year and a half genuinely feels like an inconsistent user, remembering the days they skipped and not the hundreds they did not. Seeing an actual count frequently contradicts this by a wide margin. The effect is common enough that anybody doubting their own persistence should simply look at the number before concluding anything. Telling somebody else the number is oddly satisfying, and several practitioners describe doing exactly that to the person who once predicted they would not keep it up.

Does anybody else need to see this material for it to be worthwhile?

No, and that is worth stating plainly, because the productivity conversation frequently assumes an audience. A record kept entirely for its own author is doing real work: it holds what would otherwise be carried, corrects assumptions, and answers questions years later. Two practitioners describing their most valued pages both named material nobody else will ever read. The audience question matters for succession and for handover, and it has no bearing at all on whether the daily practice is worth the few minutes it costs. Recognising this removes a common source of guilt, since people frequently feel their notes are not good enough to justify keeping when nobody was ever going to read them anyway.

What makes somebody keep using a tool for years rather than months?

Stability, silence, and the absence of any requirement to maintain it. Tools that survive do not change underneath their users, do not prompt about features, and do not make returning after a gap feel like relearning something. Every one of those properties is unglamorous and none appears in a feature comparison. Long-term users asked what they like about something they have used daily for years frequently struggle to answer, because the honest response is that it stayed out of the way, which is difficult to phrase as praise. When evaluating anything new, it is worth asking specifically what it will do when you ignore it for three weeks, because that answer predicts the next three years.

Is there a downside to a tool that never suggests anything?

Discovery happens more slowly, and people find features by accident after weeks of not knowing they exist. Practitioners routinely describe having ignored something useful for months before pressing it out of curiosity. Whether that is a cost depends on temperament: the alternative is a product that prompts, and prompting is the single most common reason people abandon software entirely. Most long-term users conclude that slower discovery is a favourable trade, and that anything genuinely needed is eventually found through ordinary use. Asking somebody who has used it longer is the cheapest remedy, and is roughly how most features in this category actually spread between users.

How does somebody explain this kind of tool to a friend?

Usually in one sentence about where the material lives rather than in a list of capabilities. Descriptions that begin with features tend to produce polite indifference, while a plain statement that it is one file, opened on the machine, writing into a folder the person chooses, tends to produce an actual question. That framing works because it answers the concern most people have and cannot articulate, which is who else has access to their material and what happens if something changes. Offering to show them the folder is usually more persuasive than any further explanation, since seeing ordinary files answers the question directly. Small differences of this kind accumulate into the reason somebody is still using something years later.

What should somebody expect in their first month?

Awkwardness, some structure that turns out to be wrong, and at least one feature discovered by accident. Most people build a hierarchy in the first week that they revise within two months, which is normal and not a failure. The habits that matter, capturing quickly, attaching files to the part of the record they concern, and starring the handful of things opened most often, all establish themselves within a few weeks. Anything not established by then can be added later without restructuring anything. Treating the first arrangement as provisional from the outset removes most of the frustration, and generally produces a better structure faster than careful planning would.

Does using something like this actually reduce stress?

Practitioners consistently describe a reduction, though they usually notice it retrospectively. Anybody holding a set of obligations, prices, dates, or details in memory is running a continuous background process and checking it out of anxiety, and that consumption is invisible until it stops. Once the material is somewhere trusted and always available, the checking ends without any deliberate decision. What people report is not remembering more, but having stopped carrying something they had not identified as a burden. The disappearance of that background checking is usually noticed in retrospect or pointed out by somebody else, which is a reasonable sign that the burden was real.

What would you say to somebody who thinks their situation is too small to need this?

That scale is the wrong criterion. A person with two hundred items benefits modestly and a person with fifteen thousand benefits enormously, and the person with fifteen thousand almost always wishes they had started when they had two hundred. The habits cost nothing at small scale and compound, while adopting them later means working through a backlog. Nobody regrets having recorded things earlier, and the most common expression of regret among people with large archives concerns the years before they began. Starting with a single page and a handful of entries costs an afternoon and establishes habits that will still be paying off in a decade.

What is the single sentence to take from all of this?

That the highest thing a tool can achieve is being forgotten while it is in use. A baker checking a dough and a memoirist continuing a chapter are both thinking about their own work rather than about software, which is what every product claims to want and very few manage. Everything else, the structure, the capture habits, the small features that become daily rituals, exists in service of that condition. When it is achieved, nobody talks about the tool at all, which is why it is so rarely described. Which is also why recommendations in this category tend to be quiet ones, passed between people who noticed the same thing and had no vocabulary for describing it.

Why is this kind of tool so rarely recommended enthusiastically?

Because the experience it provides is difficult to describe as praise. Somebody asked what they like about software they have opened every morning for two years will usually say something about it staying out of the way, which sounds like faint approval and is meant as the highest compliment available. Enthusiasm attaches to novelty and to capability, both of which are visible, while reliability and restraint are visible only by their absence. The result is that the tools people keep longest are the ones they mention least. Anybody looking for a recommendation should therefore weight quiet, long-standing use more heavily than enthusiasm, since the two tend to point in opposite directions.

What is the honest summary for somebody deciding whether to try this?

That it will not transform anything, and that this is the point. What it offers is a place to put things that stays where it is, opens without ceremony, holds what it held yesterday, and asks for nothing in return. Whether that is worth adopting depends entirely on whether somebody currently has such a place. Those who do not tend to recognise the description immediately, and those who already have one, in whatever form, do not need convincing about any of this. The cost of finding out is an afternoon and a folder, which is a low enough price that hesitation is usually about something other than the effort involved.