My recent experience of publishing 30 atomic essays in 30 days¹ failed in one way: I only got 25 articles out. I succeeded in most other ways and will apply my learnings from this time to how I work on and publish Gradual Modularization for Ruby and Rails².

Developing Ideas In Public

The core tenet of writing and publishing atomic essays is to gather feedback on thinking and writing early and using social signals as a way to steer towards what resonates with others. This is in stark contrast to typical big book projects. For these there is little early feedback gathering. Typically a few folks read drafts and give feedback. Most folks get the entirety of the work only when it is done.

With Gradual Modularization I am working through ideas for how to improve application architecture and our ability to shape it deliberately while reducing risk. There are so many directions where the emphasis could ultimately be, why not engage with readers even earlier and even more often to allow them (you!) to influence not only the end product but also the journey.

I am going to make the following three changes going forward.

Publish even smaller slices

Every time I add content to the book (or change existing one), I am going to publish it if there is at least one complete idea in there.

For example, this morning I added a section about isolated databases for packages into the Gradual Modularity chapter. It complements other boundaries enforcements but has one super interesting property that is worth sharing (and talking about!) early: If a package can completely enforce a property of "don't use anything but the exclusive database for this package" then it is really close to being deployable as a separate application.

Create a conversation every time

I will also talk about every change I make and every addition on twitter. This will create a placeholder for the conversation I would be excited to have with you about the new content.

Once a month I will also summarize the additions in an email to existing readers.

Reduced level of editing for initial publishing

Editing and proof-reading by humans creates a very good feedback loop that always critically improves my writing. It also typically adds at least 7 days to anything I try to publish.

For the atomic essays, I relied only on the spell and grammar checker built into Google docs. It still catches a lot of problems and creates a feedback loop of about 5 minutes. This allows me to get content out much faster. This is the editing process I will adopt for the book while it is in progress as well. When the book is complete I will have it edited professionally as a whole.

Feel free to engage with me any time there are any issues with what is coming out!

Except where otherwise noted, content on stephanhagemann.com is licensed under CC BY 4.0 by Stephan Hagemann