The Phoenix Project: A Novel About IT, DevOps, and Helping Your Business Win Kindle Edition

The Phoenix Project is a business book masquerading as a work of fiction. As soon as I started evaluating The Phoenix Project in the context of a business book, rather than fiction, I started enjoying it a lot more, gaining much more insight into the concepts presented, rather than focusing on the unsophisticated prose and highly contrived scenarios. The "plot" serves as a vehicle for the authors to explore and pontificate on the business ideologies presented throughout. The main character, Bill, is mentored through each of these concepts, implementing them for the benefit of the reader to see the dramatic impact each can have on a business. For the most part, this is well executed, but it can become tedious when the characters "furiously take notes, so that they can google the terms later." Yes, the authors really write this, banging the reader of the head with, "Hey Dummy, are you paying attention? This is important."

One of the best parts of the book was actually after the fictional plot ended, and the authors took the extra time to collect the concepts presented and revist them in a brief summary at the end. I hoped for this, as going back through The Phoenix Project to find each in turn, and learn more would have been difficult. If I could suggest to the authors an improvement, it would be to break the 4th wall during the fictional telling of the plot, and provide footnotes or references to the reader referencing where in the summary section they can read a bit more about a given concept. Nothing would be lost by this approach, as the reader quicky realizes the point and purpose of the plot anyways.

Overall, The Phoenix Project provides an easy way for non-business people to get their feet wet with process improvements, without making the explanations and concepts too burdensome. I find myself quoting passages with co-workers who've also read the book, realizing that all too many of the scenarios presented are real life problems we face everyday (though perhaps less severely than in the Phoenix Project.) As such, I know I've already taken something away from the book, even if I won't have a chance to master every improvement, or even experiment with them all. It provides a new way of thinking about software development, and all the organizations it impacts. Four stars.

So... DevOps.

DevOps is this mythical assembly line of progress that gets code from point A to point B in record time, and by record time, I mean no time. Coded, auto-tested, out the door, bing, bang, boom. The book was an entertaining read (and/or listen) and the authors cleverly couch the concepts of DevOps into a story about a failed delivery system (The Phoenix Project) built ala WaterFall, versus a new system hastily assembled (heh, see what I did there?) on an impromptu assembly line and delivered in record time, performing brilliantly - when compared to the failed behemoth Phoenix Project.

So... here's the problem with DevOps, and the problem with this book.

Companies with a nightmarish legacy code base (delivered or not as-of-yet out the door) that are attempting to build DevOps are using the same people that created their behemoth nightmare in the first place. As does the company in the hypothetical story in the book. Worse. The CEO in the book is a horribly bad boss, making wrong decision after wrong decision (I am sure to ramp up the tension to illustrate the saving graces of DevOps). So, I can tell you definitively that ANY company with leadership like that would lose their brilliant techs almost immediately. The market is in desperate need of brilliant techs so there is zero possibility that they are going to stick around a cess pool of politics when they can get a signing bonus and a raise from a company already doing it better and faster, and all without the drama. I'm just saying. And in order to pull off a DevOps operation, you absolutely need brilliant techs. You need tight, well executed product code with sufficient testing hooks so you can automate as you go. So you need brilliant QA that can understand and/or code the hooks right along-side the devs.

In short, you need a whole lot more than a single Brent. You just do. And to imagine that there are a room full of brilliant techs writing broken down shoddy bloated code for Phoenix, but then can turn around and write the brilliantly architected code you need for the DevOps project... well... I am able to suspend belief when called upon, but this was more like taking it out back, shooting it, and burying it six feet under. I'm just sayin'.

And with that being said, I am a huge, ginormous fan of DevOps (and the book was a FUN read / listen, thus the 4 stars instead of 3). It just takes an incredible talent pool to pull DevOps off and books like this makes companies think they just have to implement this process with the talent they have and poof! instant quality software that can be delivered instantaneously! Woohoo! <sighs> They would be better served facing the reality of the mess they have, caused by (possibly) a lack of reasonable process, yah, but almost certainly by lack of talent as well. That's basically how they got where they got.

Off soap box now. Doing the tango. Eating Oreos. Feel free to join!