The Myth of Architect as Chess Master


(5 comments)
January 6th 2020


For software architects (especially at bigger organizations), it's not that uncommon to get a request like this from a manager or director:

There's this tough problem the developers have been struggling with for weeks (performance, or scalability, or whatever), so can you quickly look at it and show them the best practice, quick fix, etc. that solves it all?

Reasonable ask? No, not really. The expectation is that the architect is a chess master. It's just like that movie we've all seen. There's the young or disheveled or [insert surprising adjective] savant who, walking through central park mid-dialogue with someone else, glances for a nano-second at a pair playing chess and declares "bishop to queen 4, check mate" and then keeps walking. Minds blown.

But why shouldn't an architect be able to size things up as quickly as a chess master? Both an architect and a chess master have accumulated thousands of hours of experience in their disciplines. Both are analyzing current states of a "board" and then evaluating the trade-offs of different next moves or options.

The difference is that chess masters are always playing the exact same game. Rooks always go forward-backward and side-to-side. Bishops always move on diagonals. And because the rules don't ever change, it's easier to take a mental snapshot of a board and instantly pattern match back to some other past experience - e.g. "this is exactly like the game I played/studied four months ago".

Any given software system, on the other hand, may look the same as countless others you've worked on in the past - I mean, hey, it's all ifs and elses, classes and objects, REST and SOAP calls, etc. - but just a little bit below the surface we realize the pieces and the rules are very different. In my system the rook can jump the first 2 squares but in yours the queen has 3 lives. Each system we visit has different intricacies and rules we must learn anew.

This immense variability and complexity is baked into even the simplest software system - it's in the business domain, in the frameworks/techologies used, and in the implementation itself - and so it takes time to grok all these nuances and details before we can build the right mental picture that allows us to reason about it effectively. Even for those with years of experience this is true. Sure there are common patterns, styles, tools, specifications, etc. that, if you know them, can help in terms of more quickly sizing up a system, but almost never do these alone allow you to pull off that type of chess savant "mate in one" mastery. Real-world software solutions take time. Architecture takes time.

This is why I think it's important to think about software architecture as a process and not a destination. Being an architect (or playing the role of an architect) doesn't mean simply divining the "right" solution or recommendation and handing it off to the developers. Anyone can whip up a bunch of boxes and lines that seem plausible, but do they really apply to the game of "chess" you're playing now, or the one from your past project or from that popular blog post? To truly know the right next "move" requires first understanding the rules of the game you're in. What are the business drivers, risks, technical constraints, legacy components, integrations, possible options, trade-offs, and so on. These are the details that take time to identify, clarify, and understand, and they are subtly (and sometimes not so subtly) different from project to project and organization to organization.

In my opinion, the expectation for the architect should be to setup the framework or process by which good architectural "moves" are made (Michael Keeling describes this well in Design It). Thinking an architect is going to stroll into a meeting and find a check mate lurking in plain site is sadly unrealistic.

I believe that software development is fundamentally about making decisions, and so this is what I write about (mostly). In 2018 I started Highline Solutions, a consulting practice that helps companies with architecture, devops, and full-stack development. I have two degrees from Carnegie Mellon University, one in Information and Decision Systems and one in Philosophy (thesis). I live in Pittsburgh, PA with my wife and 3 energetic boys.

Got a Comment?


Comments (5)

January 10, 2020
💯 — this is especially difficult for non-technical folks to fully grok or understand.

have any thoughts on how to level-up non-technical leader(ship)?
Ben
January 13, 2020
@John - thanks. I agree - very often it's the non-technical folks who tend to believe the architect is the chess master, and have an expectation that they can/should be able to find quick/easy solutions to hard/complex problems.

In terms of non-technical leadership, that's a good question. I think (and others have written about this) an "empathy" mindset is an important asset to architects - making sure to see things from the point of view of other stakeholders, and appreciate their goals and concerns. Also, it's helpful for architects to not see *themselves* as chess masters who have all the answers, but rather just more experienced/travelled technical leaders who can help facilitate better solutions from the team. Some of this is just personality...but I think some can be cultivated in an organization. In Sam Newman's "Microservices" book, he has a great initial chapter about seeing the architect as a city planner. Anyway, not sure if that's helpful :)...but thanks for the comment.
raghava
January 23, 2020
> What are the business drivers, risks, technical constraints, legacy components, integrations, possible options, trade-offs, and so on.


Absolutely, this!

There's also a tendency - to over-complicate and over-engineer, which probably is the result of a subconscious drive to value enriching ones' resume over driving business value. This just gets aided by the onslaught of overwhelming choices in terms of tools for various shades of a problem.

A true architect will be very mindful of the vagaries - mostly present, and build a well-proven process that could be used to adapt to the uncertainities of the future.

As said by the champions, "the process is the product/platform".

Thanks for a good post!
Carlos Balbuena
January 23, 2020
I think problem is that there is not a consensus of what the sw architect does.
Ben
January 24, 2020
Thanks @raghava and @Carlos! Agreed on both accounts. As an industry, we do tend to favor flashy but complicated solutions for the sake of chasing the next new thing. Also, agree that the role of architect is not always clear.