The best way to learn design in any field is to study examples, and some of the best examples of software design come from the tools programmers use in their own work. These lessons build small versions of things like file backup systems, testing frameworks, regular expression matchers, and browser layout engines both to demystify them and to give some insights into how experienced programmers think. Please see the description of our audience for more information about what we assume you already know.
-
All of the written material in this project can be freely reused under the terms of the Creative Commons - Attribution license, while all of the software is made available under the terms of the Hippocratic License.
-
All proceeds from sales will go to support the Red Door Family Shelter in Toronto.
-
Please see the contributors' guide and our Code of Conduct if you would like to help improve or extend this work. Anyone who provides minor fixes or feedback will be added to the acknowledgments.
Note: this book was originally titled Software Tools in JavaScript. The title has been changed to better reflect its content and intention.
Greg Wilson has worked in industry and academia for 35 years, and is the author, co-author, or editor of several books, including Beautiful Code, The Architecture of Open Source Applications, JavaScript for Data Science, Teaching Tech Together, and Research Software Engineering with Python. He was the co-founder and first Executive Director of Software Carpentry and received ACM SIGSOFT's Influential Educator Award in 2020.
-
Is this done yet? Almost: what we have now is a complete, usable draft.
-
Why did you start this project? Because most books with the words "software design" or "software architecture" in their titles spend most of their pages telling readers how to describe a design, but don't actually describe the designs of real systems. Beautiful Code and The Architecture of Open Source Applications were intended to fill this gap, but were too wide-ranging to be useful as textbooks. We hope that using one language (modern command-line JavaScript with Node.js) and one problem domain (software engineering tools) will make this book more approachable to junior programmers.
-
How long did it take you to write this? An average of four or five hours a week over the course of a year. The code and point-form notes came first; once that had settled down, it only took an hour a day for five weeks to turn the notes into prose and draw the diagrams.
-
Why JavaScript?
- Forty years ago, Donald Knuth said that Pascal was every programmer's second language. The same is now true of JavaScript: it's the one language that every developer needs to know a little of.
- We hope there will be a second volume that will show readers how the components of a modern web stack work. JavaScript is the only realistic choice for the front-end tools that will be part of that.
- We wanted to learn more JavaScript than co-authoring JavaScript for Data Science taught us.
-
What about doing a Python or TypeScript version? Or a Spanish one? Translations into other languages (both human and machine) would be very welcome: please get in touch if you're interested.
-
I'm teaching a programming class: can I use this, and if so, how?
- The answer to the first part is "yes please". All of this material is covered by an open license, so as long as you acknowledge the original source (e.g., by providing a link back), you can use it in whole or in part, as-is or with modification.
- "How" depends on what you're teaching. This material isn't suitable for a first course on programming or on JavaScript, but should provide lots of material for an undergraduate course on software design or software engineering at the third- or fourth-year level. Students can tackle the exercises at the end of each chapter or write and explain small tools of their own.
- If you're interested, I'm very happy to chat. I'm also happy to give a guest lecture in your software engineering class about software design, how to run a meeting, and/or life in industry.
-
Do you want a chapter on X? Possibly: as we said above, we would like the second volume to show people how to build things like a simple relational database or document database, an HTTP server, a basic authentication and identity management package, an issue tracker, and so on. If you are interested, please get in touch.
-
Why are you using your own formatting tools? Believe me, I've tried the alternatives. Static site generators like Jekyll, Pelican, and Hugo don't support numbered cross-references, while tools like Bookdown and Jupyter Book have complex tech stacks: figuring out whether a problem is in the source document, a Pandoc template, a LaTeX template, or somewhere else entirely is very frustrating. Ivy, BeautifulSoup, and 700 lines of Python aren't ideal, but they've proven to be less painful than the alternatives.