Patterns of Software: Tales from the Software Community by Richard P. Gabriel

by Richard P. Gabriel

A wide-ranging set of essays by a long-time Silicon Valley insider. Gabriel muses on software development, factors that have led to the success or failure of software companies, the nature of successful programming languages, and more. Evidently, Gabriel is currently writing poetry, and his concern for language and the well-turned phrase shows up in this book as well.

Genres: Programming, Software, Computers, Computer Science, Technology, Nonfiction, Technical

Paperback

First published January 1, 1996

Reviews

[Zachary Siegel · Rating 4 out of 5]

This book is really a collection of a few sets of related essays. Richard Gabriel begins by analyzing the beautiful work of the architect Christopher Alexander, attempting to relate his work on architectural patterns to the realm of software engineering. Gabriel spends much time narrating his time in graduate school and his experiences with the company Lucid which he founded.

I was specifically fascinated by Christopher Alexander's work, so much that I went and read The Timeless Way of Building (by Alexander) before having the chance to finish this book, and Gabriel's extended thoughts continue to heavily influence the way I think about abstractions and developer experience.

[Algirdas Raščius · Rating 4 out of 5]

Do not get tricked by the title of this book. If you are looking for patterns that would improve you code, design or your development process, you will be disappointed. This book is essentially what it's subtitle says: tales from the software community. Nevertheless it was a pleasure to read.

Book contains essays on wide selection of topic including discussion on what "quality without a name" as defined by Christopher Alexander is; what we as software engineers typically miss to understand when talking about patterns; how to objectively define beauty; what "habitability" of software system is and how it may be more important than pure mathematical elegance. Author also presents kind of memoirs from his carrier in academics and as a founder of a failed software company. He describes what and why went wrong with his company. Although the book was written a long time ago (even before burst of dot-com bubble), lessons learned from it can still be useful today. Finally author presents idea why "worse is better" in software business - why it makes business sense to start with small product quickly instead on spending lots of time and investors' money before initial product launch.

[Mark Seemann · Rating 2 out of 5]

This book is a collection of essays, originally published elsewhere.

The first half of the book is mostly concerned with an exegesis of Christopher Alexander's original works on pattern languages. Mostly, these essays relate more to the philosophical and building architecture aspects of Alexander's writing, than they relate to software development. I must admit that I had trouble connecting with the material, in part because, to its credit, it talks about the limited success Alexander had with carrying out his ideas in reality, but mostly because the prose is too hard to read. More about that later.

The second half somewhat abruptly shifts to a personal account of Gabriel's struggles with taking a university degree and running a company. This half is much easier to read, although I was puzzled about the relevance of the contents.

In the last essay, Gabriel returns to more abstract concerns, but this time specifically about software development. He discusses a concept he calls worse-is-better, which resembles the idea about a Minimum Viable Product. Quite a progressive idea, considering that that essay was written in or before 1996.

While there are some gems of insight here and there, the book seems only superficially relevant to me, and the prose is hard to read. It could have used some editing, or, at the very least, some punctuation. Here's an example from p. 225:

"So the very investment you might have thought was the way to put distance between you and possible competition by making the effort for entry too great is the very impediment to modifying your product to be what customers need."

I had to read that sentence four times before I understood what it said. Would it be too much to ask for at least a comma or two?

Except for the personal narrative, most of the book is written like that, so it's difficult to extract meaning from it. A few sentences I simply had to give up on.

[Bugzmanov · Rating 3 out of 5]

This is an odd one. It's a collection of essays from the 90ies (before blog-posting became a thing), but it's not typical collection of unrelated random stuff, the essays are clearly clustered around 2 topics and have continuity and coherency inside a cluster.

First big topic is pretty deep reading of Christopher Alexander's work. I liked this part. The main take away is the despite idea of software patterns originating in Alexanders work, it's pretty disconnected from what Alexander envisioned patterns to be. In software the idea of a pattern ended up meaning a precooked isolated pieces of design. While the main body of Alexanders work was about finding a harmony in architecture, a quality without a name, and one of his thesis was exactly the opposite - that it's impossible to find a harmony in modular architecture.
The connection of this part of the book to "Patterns of Software" can be found on some meta level, but it's not clearly expressed and left as an exercise to the reader. My conclusion: if you're interested in becoming more familiar with Christopher Alexander philosophy, but don't have time to read originals (and my god those are thick books), this might be pretty good intro.

The second cluster is autobiography of Richard P Gabriel: school, university, entrepreneurship. If you're interested how startup scene looked in 80ies and early 90ies, this might be of interest. But the writing style felt a bit off to me in this part. I understand and appreciate that some of it is just "sweet memories of the past" that have high nostalgic value to the author.. but as a reader this leads to just unrelated paragraphs of text that leads nowhere and isn't connected to the plot line. Things that just happened. There was a conference, employees were wearing leather jackets and jackets had company name made in specific colors. Cool. Why am I reading this?
And while Richard Gabriel coined the term "AI winter" not much was said about AI research in the 80ies to be honest. I feel like you can have better understanding of events and causes by reading a wiki article.

Overall this is an odd book that I kinda-sorta enjoyed but I don't see me recommending this one to anyone I know. :shrug emoji:

[Mrinny · Rating 3 out of 5]

Actually I would consider this two books in one.

The first part which I would give five stars to, is about the inspiration that the author got from the Architect Christopher Alexander. I happened to read this book during an exceptionally painful time in my software development journey and the book simply reminded me of why I love what I do.

The second section, I'm sure a lot of history enthusiasts would love. With an autobiography like story of the authors' journey through Lisp, C++, Lucid and other languages for AI etc. and the business aspects that taught him that being 'smart' has more to do with worldly skills that simply being a good programmer. He describes some valuable lessons learned during his arduous journey from which I admit, I gleaned a lot of wisdom.

Yet, I would rate the second half not so generously as the author often describes his life experiences in excruciating detail for my tastes. Having said that, I am sure a lot of nerds out there would love to read about the exciting times of AI during the early 40's through the 80's.

Let me be very clear though, I deeply respect a person to be sharing such experiences in a book. It takes a lot of courage and desperate need to not see other's suffer with the same problems, for someone to write such a book. I am sure that's the reason it does so well. It resonates with life's experiences even today.

Bottom line: You will end up wanting to read all of the architect Christopher Alexander's work after reading the first half of the book and get a good idea of the 'Quality without a name'. The second half definately has some broad strokes on the characteristics of good software design and I would recommend the last few pages even if you skip the entire second half of the book.

[Jordi Puig · Rating 3 out of 5]

Un llibre que en realitat son molts llibres, ja que cada secció parla de coses prou diferents entre sí. Tot i així veus que el fil conductor son les idees de l'autor. No he acabat la última part però el dono per tancat i m'emporto coses positives. Intento fer un resum/review de cada part:

Pattern language: les fumades del Christopher Alexander portades al mon del software. Hauria preferit no intentar portar-ho a la programació, ja que la història d'aquest arquitecte és prou interessant de per sí. A més, moltes idees dels anys 90' ja no apliquen tan avui en dia.

Languages: No té tan interès aquesta part, simplement la reflexió que els llenguatges de programació son eines i cadascun té les seves característiques.

What We Do (writing broadside): En resum ve a dir que tothom ha de practicar d'escriure. Aquesta part és la que més m'ha inspirat per escriure i aprendre a fer-ho millor. Bé, per començar a fer-ho (estem en ello).

Life of the critic: autobiografia molt entretinguda. És un home que li han posat pals a les rodes a la seva carrera acadèmica, i que sempre ha tirat endavant fent grans esforços durant un període de temps concentrat. Més raons per voler fugir del món acadèmic :)

Recomanable si us agrada alguna cosa intensa i més filosòfica/reflexió sobre idees.

[Cees · Rating 5 out of 5]

I like the book because I like books that make you think. In the first part, Gabriel basically combines a critique of the idea of Pattern Languages with questions around how to map it to the craft of software development. The second part is like a separate book, talking about his early career and the slow rise and hard fall of Lucid; I found it less interesting than the first bits, but still a good read because it is an interesting story and it is well told. And it's an anthology, so it's fine.

It is meandering, philosophical, and a bit random at times. That helps make you think. I like books that make me think rather than present solutions, and I really like books that make me think hard about my craft in a way that books on programming languages, algorithms, or even most books on "agile software development" don't do. Five stars because of that.

[Ian Bird · Rating 5 out of 5]

The first half of the book can mostly be skipped, imo. The real heart of the book is in Richard Gabriel’s autobiography, one of the rare windows into the mind of someone at the absolute pinnacle of academia in computer science. So few people have put that world into paper, and fewer still with an insider perspective. Gabriel’s skill at writing made it all the better, too. Sections of prose were especially poignant, and the chapter covering Gabriel’s academic career was excellent.
This is probably the most well written book on the topic of computer science I’ve ever read.

[Leo · Rating 2 out of 5]

From the comments and foreword I wasn't really interested in second part, and as first (essays on patterns / pattern languages) couldn't captivate me, I've decided to stop at that point.

Essays were just too theoretical and far from real world for me, and when I mentally tried "discussing" some ideas with author - I wasn't even sure if I got the idea he was trying to express correctly.

[Ian Mihura · Rating 4 out of 5]

Half the essays in this book are very good. It's overall not too technical: explores abstract concepts of day-to-day (patterns, germ propagation, markets) as they relate to programming. Interesting ideas that prompt the reader to digest them: at best, the writer gives you half-chewed ideas.

A bit outdated, too much talk about Lisp and C, UNIX and other 1990's things.

Worth the read.

[Luís Soares · Rating 2 out of 5]

This is not a book about patterns of software. It's a book about tales. It is worth it only due to the last chapter, "MONEY THROUGH INNOVATION RECONSIDERED" where iterative development is described as "piecemeal growth.

[Roman Bly · Rating 3 out of 5]

Personally, nothing new for me. And some things are a bit outdated, given the age of the book.

[Michiel · Rating 4 out of 5]

Loved the philosophical way of writing. Great insights in patterns, abstraction, and compression.

[Rauno Villberg · Rating 3 out of 5]

Conflicted on this one. Loved the first half, the software development / architecture essays, but largely left cold by the second half; essentially a biography of the author.

[Riley · Rating 5 out of 5]

A must-read for Richard P. Gabriel fans, and those interested in the history of commercial Lisp. The first part, an exploration of Christopher Alexander’s work to software, stands out on its own. Dorian Taylor is currently working on a book in the same vein, based on Alexander’s later work in The Nature of Order. After, RPG goes into his autobiography going through academia the long way and founding a Lisp company in the midst of the 80s AI hype cycle (he even claims to have coined the term “AI winter”), which pivoted into a C++ IDE before shuttering. Maybe it gets a little too personal at times, and the later part of the book was poorly edited, but overall I found it a fascination look into the life and mind of someone I admire and a time I wish I could have experienced.

“Software is a plant”

[Reynard · Rating 4 out of 5]

Interesting analogy between software development and architecture (with its habitability and "quality without a name"). I totally agree with his views (and suggestions) on technical, scientific writings/communications. And I especially enjoyed his autobiography, a tales of failures and struggles :-)

[Dagbedji · Rating 4 out of 5]

Great introduction to the similarities between the work of Christopher Alexander on patterns in architecture and similar applications in software development. This is quite a good read for anybody interested in some of the topics addressed by Richard Gabriel. It is a key book for any software developers interested in improving the quality of their design and code.

[Luke · Rating 3 out of 5]

Fine essays from a creator of Common Lisp and author of "Worse is Better" on software as a habitable environ, a mid-90s reflection on the influence of Alexander and patterns that rings through today. A critique of American Capitalism applied to any pursuit of beauty (the quality without a name) through too, or the inevitability of human pettiness?

[João Sampaio · Rating 3 out of 5]

Maybe a little too much poetry where there's actually none. But there are some useful lessons here, specially business lessons.

[Two Readers in Love · Rating 3 out of 5]

Software engineers who are fans of Christopher Alexander's "A Pattern Language" will find this interesting.

[Thomas · Rating 4 out of 5]

The first parts of the book are very good. The rest is a fine read.
I loved the preface by Christopher Alexander.