In Search of Certainty: The Science of Our Information Infrastructure

by Mark Burgess

Quite soon, the world's information infrastructure is going to reach a level of scale and complexity that will force scientists and engineers to approach it in an entirely new way. The familiar notions of command and control are being thwarted by realities of a faster, denser world of communication where choice, variety, and indeterminism rule. The myth of the machine that does exactly what we tell it has come to an end.

What makes us think we can rely on all this technology? What keeps it together today, and how might it work tomorrow? Will we know how to build the next generation--or will we be lulled into a stupor of dependence brought about by its conveniences?

In this book, Mark Burgess focuses on the impact of computers and information on our modern infrastructure by taking you from the roots of science to the principles behind system operation and design. To shape the future of technology, we need to understand how it works--or else what we don't understand will end up shaping us.

This book explores this subject in three parts:



Part I, Stability: describes the fundamentals of predictability, and why we have to give up the idea of control in its classical meaning

Part II, Certainty: describes the science of what we can know, when we don't control everything, and how we make the best of life with only imperfect information

Part III, Promises: explains how the concepts of stability and certainty may be combined to approach information infrastructure as a new kind of virtual material, restoring a continuity to human-computer systems so that society can rely on them.

Genres: Nonfiction, Science, Technology, Computer Science, Technical, Engineering, Computers

472 pages, ebook

First published July 30, 2013

About the author

Mark Burgess is a writer with many interests. His books span from fiction to hard science, and he actively writes philosphically in his blog about all aspects of modern information culture. He is an active public speaker on the international conference circuit, and is engaged in promoting science to a wide audience.

He is the Founder and original author of CFEngine. He was senior lecturer and then appointed full professor of Network and System Administration at Oslo University College from 1994-2011. He was the first professor with this title, and is largely responsible for defining the field. Mark Burgess obtained a PhD in Theoretical Physics at Newcastle University, for which he received the Runcorn Prize.

Reviews

[Ozzie Gooen · Rating 4 out of 5]

I got around half of the way through, then gave up. Found it packed with intellectual references, but a challenging read (trying to get to the point) for sure.

I think there's a big space of books of "smart people trying to describe things using all the tools they know of, but definitely having a difficult time to do so", with a bunch of common topics (in this case, context seemed really important).

[Lorin Hochstein · Rating 3 out of 5]

A thought-provoking but often frustrating read.

Burgess lays out a conceptual framework for buildilng reliable IT infrastructure using promise theory: a bottom-up approach to constructing systems that can heal themselves in a manner akin to biological organizations.

The author offers a lot to think about, the presentation of the book hurts his thesis. The author attempts to weave a narrative that includes many differnt concepts: from Shannon's information theory, Feynman diagrams, cellular automata, W. Edward Deming, and much more. However, he tells the story at too high a level of abstraction for a reader new to these concepts. In particular, he relies too much on analogies and not enough on specific IT-based examples. Burgess wrote the venerable CFEngine configuration management system that employs these concepts, and perhaps CFEngine users would have an easier time understanding how these concepts apply on the ground, but as someone new to these concepts, I often struggled to see how they would apply in practice. The high level nature of the writing also made the book into a slog. I had to force myself to push through to to get to the end of it.

[Matt · Rating 4 out of 5]

Fascinating book with a lot of important and interesting points. Not great on the readability front. If you are working on/with a complex distributed system, this has some good ideas that are worth wrestling with. Recommended.

[Ivano · Rating 5 out of 5]

This is an excellent, if somewhat longwinded, read on scientifically approaching the topic of technology infrastructure and interaction as of today.

It discusses an impressive spectrum of topics across physics, information technology, computer science, formal logic, showing a deep insight of the current shortcomings and challenges we're facing.

He doesn't get too deeply technical throughout the book, it's mostly a storytelling approach to discover a path that the author had followed throughout his research history. During this journey he points out discoveries and pitfalls on current research, and outlines his own approach to modelling complex dynamical and semantic systems with a theoretical framework of his own, i.e. Theory of Promises.

At times I felt overwhelmed by the number of references to scientific notions and research results.
I also often felt puzzled while trying to follow the author's reasoning when some of the concepts were introduced and discussed with what I felt was a lack of explanatory examples.
I may be possibly missing the necessary background knowledge to completely understand the connections.
It maybe that a single read is not enough to actually absorb all the content presented in the book.

I would suggest other books from the author if you want to get detailed explanation of the theory exposed, in particular Thinking in Promises.

Overall I heartily recommend this book to everyone interested in modern software systems and their growing impact on our lives.
I will certainly follow Mark Burgess work in the future.

[Roger K. · Rating 4 out of 5]

This is an insightful book that will make you rethink how we manage infrastructure and changes.

Burgess covers a lot of background before proposing his approach. While many of the topics (history of science, quantum physics, information theory, etc) will be familiar to those reading this, it was a nice summary of current thinking. He then bring all these threads together well in the last third of the book.

I would summarize Burgess' approach as DevOps + Antifragile. The book's theme of creating stability at higher layers by embracing instability at lower levels screamed "ANTIFRAGILE" to me at every turn. Oddly enough, though, Taleb is not cited and the author seems unaware of the linkage.

One warning for anyone reading this book: the typography is terrible! To Burgess' credit, his Web page cites that he struggles with typographical errors. Even so, if anyone edited this book, it does not show. I didn't mind the grammar and spelling issues (as a somewhat reformed grammar nazi), yet the number of repeated words, left out words, and sentences that just made no sense made me want to throw this book against the wall about a dozen times.

Even so, this book is worth working through. I am hopeful that this will drive further research and conversation in this important area.

[Dave · Rating 3 out of 5]

Thought provoking (my copy is now full of post-it flags from all the interesting passages I wanted to come back to), but ultimately Mark Burgess was trying to do too much in this book. He spent a ton of time trying to hang together physics metaphors for large scale computing systems as a way to make the case for Promise Theory, the modeling approach he has been developing. As a former physics guy myself, I appreciate where he's coming from and I fall back on some of the same tools (phase transitions, equilibrium states) when I think about infrastructure, but I'm not convinced that's useful beyond just being a personal habit of mind. There were lots of citations to academic papers in here - perhaps some of those make explicit how promise theory modeling is actually isomorphic to physical structures.

I was most excited to learn more about promise theory, in which systems are modeled using voluntary commitments from individual actors (human or machine) to act in specific ways conditional on promises being kept by other agents, but I felt starved for practical examples of what a promise-based model looks like in practice. I got the sense that there are conceptual similarities with agile software development, pull-based workflows, DevOps, and actor-based concurrency, but it never really gelled into a unified framework.

[Peter Sellars · Rating 4 out of 5]

This is heavy but enjoyable reading. It is very science focused and the first section of the book can be hard work - but getting through it provides essential concepts useful in the later sections. Promise Theory as a means to build better infrastructure gets a good run out - although once again at times this is quite hard work for someone lacking the science background. Understanding the concepts behind the CFEngine tool was really useful and will encourage me to read some other papers produced by the Author on Computer Immunology and Promise Theory. If you are interested in building better infrastructure...you need to read this book!

[Alain van Hoof · Rating 5 out of 5]

The least technical book of Mark, and therefore more easy to read. Still a lot of connections to other fields than IT like physics, building to a statement that System administration is still at the beginning of a revolution.

[Stephie Williams · Rating 3 out of 5]

I found the first two parts to be more interesting than the third. Overall is was a decent book, but it lag in the end. There was also a severe editing problem. I found so many mistakes in grammar, it became quite tedious at times. Just who I wonder is responsible for.this.