How the stories are made
I finish the draft on Wednesday, in one sitting, before I try to explain it.
Each week I gather material, write a complete tale in a single afternoon, then spend Thursday and Friday making the sentences carry their own weight. The illustration and the endnote come last, so that neither is allowed to decorate a story that has not yet closed.
Wednesday
The one-sitting rule
A story that still needs another day to find its ending is a story I have not yet written. I would rather send a smaller complete thing than a handsome fragment.
I keep Wednesday for the draft and I keep it whole. The morning is for rereading the week’s notes until a shape appears: a person who wants something awkward, a place that will not stay still, a turn that can be reached in twelve to twenty minutes of reading. Then I sit down and I write until the last sentence is on the page. I do not break for other work, and I do not leave a scene to be “finished later”. If the sitting fails, the story does not go out that Sunday. I would rather miss a week than send an instalment that still owes you an ending.
The cost is real. Some weeks the tale is shorter than I hoped. Some images that looked promising on Monday refuse to become a plot by noon. I cannot chase every elegant sentence because the afternoon has a close, and I have to choose the line that serves the turn rather than the line that merely sounds well. There is also a gain that I will not trade away. A draft written in one sitting already knows its own length. The beginning remembers the ending. Readers who join in June are not walking into a corridor with no door. They receive a finished object, and they can stay or leave without being punished for arriving late.
I learned this the slow way. Serial chapters taught me to borrow attention from next week. That debt sits badly with me now. A week is long enough to build a whole shape if I protect the sitting, and short enough that the story still has heat when it reaches your inbox on Sunday morning. The rule is simple on purpose. It keeps the subscription honest, and it keeps me from falling in love with a fragment.
Sources
Where the material comes from
Places I actually use
The National Library Board on Victoria Street, for maps, street directories and the quiet work of checking what a name used to mean.
The National Archives of Singapore on Canning Rise, for port ledgers, shipping notices and the dry sentences that hide a life.
The Lee Kong Chian Reference Library, for newspapers and the small advertisements that tell you what people feared they might lose.
Maxwell Food Centre, the five-foot ways of Telok Ayer, and the lanes around Kampong Glam, where I walk after reading and listen for a phrase I did not go looking for.
I do not begin with a theme. I begin with a scrap that will not sit still: a line in a harbour record, a rumour about a ferry that asked for coins no longer minted, a household rule overheard at a coffee shop and then immediately denied. I copy by hand. Typing too early makes the scrap look like a finished sentence, and I need it to stay awkward long enough to become a story rather than a lecture. When the libraries close I walk the streets named in the notes. The walk is not research theatre. It is how I find out whether the air of a place will carry the turn I have in mind, or whether I have been decorating a rumour with geography it does not deserve.
Three kinds of material feed the week, in proportions I try to keep honest. Archival records give me furniture: dates, cargo, a clerk’s irritation, the price of a crossing. Oral fragments give me voice: the way a story is told when the teller is not trying to be literature. Pure invention fills the gaps, and there are always gaps. I would guess, across a year, that less than half of any instalment is taken straight from a source, and that the rest is built so the source can stand as a tale rather than as a footnote wearing a costume. I say this because folk material is easy to misuse. It is a starting point. It is not a licence to reprint.
I keep a working notebook for each month. The pages are ugly on purpose: lists, crossed names, a question written twice because the first wording was too tidy. If a scrap still interests me after two weeks of neglect, it has earned a Wednesday. If it only interests me when I am performing diligence, it stays in the notebook. That is the habit. It is slower than mining the public domain, and it is the only pace at which I can stand behind what I send.
Thursday
Revision: sound, logic, cuts
Thursday is not a second writing day. The story already exists. What I do is listen to it, test it, and take away what it does not need. I work in three passes and I stop when the third pass starts putting ornaments back in.
The first pass is for sound. I read the draft aloud in an empty room, at the pace a careful reader might use on a Sunday morning. Sentences that look balanced on the screen often snag in the mouth. I listen for a run of similar lengths, for a joke that arrives too early, for a name that is doing decorative work. If I cannot speak a paragraph without glancing back, the paragraph is not finished, however pretty it seemed at five o’clock on Wednesday.
The second pass is for logic. Folk-shaped stories can hide a hole behind atmosphere. I ask whether the turn is earned by what the character can actually know, whether the object that matters was present before it was needed, and whether the ending answers the want that opened the tale. I am not trying to make the story realistic. I am trying to make it fair. A reader who gives me twenty minutes should not discover, at minute nineteen, that I have changed the rules to escape a corner I wrote myself into.
The third pass is cuts. I take out the sentence that explains the mood the previous sentence already made. I take out the clever aside that belongs in a letter, not in the tale. I take out any line that is there because I was proud of finding it in an archive. Provenance belongs in the endnote. If a historical detail does not change what a person does in the scene, it is furniture I have dragged in from another room. After the cuts I leave the pages alone overnight. Friday morning I read once more, only to check that I have not cut a load-bearing beam, and then I stop. Endless revision is how a complete story becomes a serial in disguise.
Friday morning
Illustration: one plate, ink and wash
Each instalment carries a single plate. I draw in ink and wash, on paper, and I scan the result for the e-mail and the web page. There is no gallery and no sequence of spots. One image has to stand beside a finished story without arguing with it.
I make the plate after the prose is closed because the picture should answer the tale, not audition for it. If I draw first I start writing toward the image, and the week begins to serve the wrong master. Ink and wash suit the work: they dry at a human speed, they forgive very little fussing, and they look like something that could have sat on a desk rather than in a campaign. I choose one moment that is not the climax. A climax drawn in full often steals the turn from the reader. Better a doorway, a coin, a stretch of water, a street at the hour before the strange thing happens.
I send only one plate because more pictures start to narrate. A second image wants a third, and soon the instalment is a picture book that happens to contain prose. The story is the object. The plate is a companion, and companions should know when to be quiet. Readers who cannot see the image still receive the instalment in full. I write a short description of the plate in plain language and I place it as alt text and as a caption. The description names what is on the paper. It does not hint at the ending, and it does not pretend the drawing is a photograph of a real event.
If a week’s drawing fails, I do not pad the e-mail with a stock picture. I either redraw or I send the story with a written plate: a paragraph that tells you what I would have put on the paper. That has happened. It will happen again. I would rather be plain about a missed drawing than decorate the letter with an image that does not belong to the tale.
Friday afternoon
The endnote
Every story ends with a short paragraph of provenance. It is not a bibliography and it is not a making-of. It is a record of what I used, so that I cannot hide behind atmosphere when the material was thin.
The endnote names the kind of source without turning the tale into homework. I might say that a crossing fee comes from a harbour notice, or that a household rule was overheard and then denied, or that the island in the story is a rearrangement of two real shores. I say when I have invented the rest. Readers who want only the fiction can stop at the last line of the tale. Readers who care about the grain of the week can see what was found and what was made. That distinction matters to me more than it used to. It is easy, with folk-shaped work, to let a borrowed image do the labour of originality.
Writing the endnote keeps me honest in a practical way. If I cannot describe the source in a few sentences, I probably did not have one, and the story has been leaning on a mood. If the endnote wants to be longer than the tale, I have written an essay and dressed it as fiction. I cut the endnote until it is a companion paragraph, then I stop. It goes out with the instalment, in the e-mail and on the page, in the same type as the body. I do not bury it. A subscription that rebuilds old material owes you that small daylight.
Paid readers who take the annotated archive receive a little more: a note on what changed between Wednesday and Friday, and where a cut was made. That is still not a workshop. It is a record of the week, kept so that the finished story remains the main object and the process does not become the product.
Limits
What I will not do
These are working rules, written as I mean to keep them. They are here so that you know what the subscription is, and what it refuses to become in a thin week.
I do not use generated prose. Every sentence in an instalment is written at this desk, and I will not outsource the voice to a model, even for a first pass or a “clean-up”. If a week is too thin, I skip the Sunday sending rather than fill the gap with a machine that has never sat in the National Archives or walked Telok Ayer at closing time. The point of a one-person subscription is that a person made the thing you are reading.
I do not recycle public-domain text and pass it off as new work. Folk tales, port records and old news are material. They are not a quarry of free paragraphs. When I rebuild a story I change its shape until it can stand without the original beside it, and I say in the endnote what I began from. Reprinting a tale with a fresh title would be easier. It would also be a different job from the one I asked you to pay for, or to try as a sampler.
I do not leave serial hooks. An instalment has a beginning, a turn and an ending. If two stories share a street or a rumour, I say so in advance, and each still closes. You can miss a month and return without needing a map of unfinished business. I also do not write subject lines designed to provoke a click. The title of the story is the subject of the e-mail. If the title cannot carry the sending, the story is not ready, and no amount of urgency in the inbox will make it so.
I do not offer manuscript feedback, collaborations, or guest posts. The reading desk answers questions about plans, delivery, permissions and the work already sent. It does not become a general workshop. That boundary protects Wednesday. Without Wednesday there is nothing to subscribe to.
If the method suits you
The week ends on Sunday at 07:00 SGT.
That is when the finished tale, the plate and the endnote leave this desk. The sampler is free and arrives once a month. The weekly plans are for readers who want the sitting every Sunday, with a clear way to cancel from any e-mail.