START WHERE YOU ARE

Start writing a book. Choose your next step.

You do not need to read everything before writing something. Choose the description closest to the work on your desk, then make one decision you can use.

A hand rearranging cream index cards into a developing book sequence beside a blue notebook
ORIGINAL BOOKDRAFTLAB PHOTOGRAPH / ILLUSTRATIVE
An idea
Make one passage test it.
An unfinished draft
Map the work that already exists.
A revision
Choose one reading problem.
A clothbound book, notebook and index cards on a writing desk
ORIGINAL EDITORIAL ILLUSTRATION
01 / THE WRITING DESK

I have an idea, but it is still very large.

Write a temporary description of the book: who it is for, what question or experience holds it together, and what belongs outside it. Then choose a small sample that could test the direction. A sample is not a contract to keep the first version. It is evidence about what the project asks you to do. If several ideas compete, compare the actual work each would involve rather than choosing only by which title sounds most exciting.

02 / THE WRITING DESK

I want to write fiction, but the next scene is unclear.

Name what a character wants in this particular moment and what makes getting it difficult. Describe what is different by the time the scene ends. The change can be knowledge, intention, trust, or a practical circumstance; it does not need to be a dramatic event. If you cannot decide whose experience should carry the moment, test a short passage through another narrator. Keep both versions until you understand what each permits the reader to know.

An invented character holding an envelope at the doorway of a reading room
THE WORK, MADE VISIBLEA scene starts with a choice.

An invented scene: a letter in hand, an open doorway and a decision still to make.

Original editorial illustration
03 / THE WRITING DESK

My book is nonfiction and the research keeps expanding.

Choose the reader’s question before adding more sources. A chapter needs material that supports its own job, not every interesting fact you uncover. Record a source with the question it may answer, what you actually read, and what remains unverified. If research changes the book’s scope, update the plan deliberately. You can keep useful background in a separate file without requiring the reader to follow every path you took while learning the subject.

04 / THE WRITING DESK

I cannot find a routine that survives a normal week.

Look at the time and attention you actually have. A twenty-minute session may be enough to draft one exchange or summarize a chapter’s function. It may not be the right place for a complicated research decision. Name the task before choosing a word target. Try a few sessions, observe what happened, and revise the plan. A missed session does not need to be repaid with an exhausting catch-up block. Leave a clear return note so the next beginning is easier.

05 / THE WRITING DESK

I have pages. I am not sure what to do with them.

Begin with a description of what exists. Read a manageable portion and summarize the work each chapter actually performs. Record open questions without repairing every sentence immediately. Once the sequence is visible, choose a revision pass at the right scale. A structural question, a continuity problem, and a clumsy sentence need different kinds of attention. Keep a copy of the original before experimenting, so testing a change does not remove your ability to compare it.

06 / THE WRITING DESK

I want to use AI without losing the decisions.

Start with a bounded request and an evaluation plan. Ask for a few possibilities, a set of questions, or an observation about a short excerpt. Say what should remain unchanged. Check the result against the actual manuscript and your intention; a fluent answer can still be wrong. Preserve originals, record meaningful uses, and review a service’s current handling of text before submitting private material. The destination where you later share the book may have its own requirements.

Browse the writing library

A closer look