Software is Like Cars
From Zanecorpwiki
Criticism is a hard thing to deal with. The first thing one learns is to disregard malicious or patently ill-informed criticism.ref group=notesThough how to keep a conversation productive when someone has made hopelessly naive statements can be a challenge./ref Taking constructive criticism well is not a hard trick either; it only requires focusing on the goal rather than the self.
I like to think that I am pretty good at taking criticism, but there is one formulation in particular that really pushes my buttons: Why would you do that? It's just like X.
When I share an idea and get this response, it's one of the few things that actually annoys me. In fact, it evokes an emotion somewhere between annoyance and anger. I try to be suspicious of my own bias, so this raises a red flag in my mind and I've spent a lot of time searching out the character flaw that doesn't let me hear that something's already been done.
I don't claim to have any shortage of character flaws but after many years of struggling with how to deal with this particular criticism, I've determined that--as a general proposition--this response more often than not demonstrates a misunderstanding of the software market. I call it the Cars Criticism.
The car market is filled with literally thousands of products which are all essentially exactly the same. The one line description of every car in existence would be a four wheeled personal transport device. They all have seats and controls in the same basic layout. They all have the same basic shape. The same basic mechanics. This has all been true for almost a hundred years. Pick any car in the world and assume it didn't already exist. Now imagine pitching that car to anyone in the world with any knowledge of cars. The response will surely be, Why would you do that? It's just like X (and Y, and Z, plus 997 other examples).
Even at the extremes, Bugatti Veyron and Geo Metro are far more alike than different. There's not a single mechanical or design feature that really sets the two part, and even when considered as a whole, the two actually perform about the same. In fact, the Geo will get more people to a destination faster considering it has four seats and you're limited to 30-60 MPH anyway.
Yet the car market continues to turn out hundreds of new models each year that are only ever slightly different than last year's models. Not only their own models, but everyone else's. The 2009 Chevy truck line is not very much different different from the 2008 Chevy line, but it's also only fractionally more different from the 2008/9 Ford line. For the most part, you could take one manufacturer's car and label it as another and most people wouldn't even notice.
The difference is rooted in market maturity. Software has, until recently, existed in a Gold Rush style market. There was so much that was not-yet-done, so much space not-yet-explored that it just made sense to stake a claim and see what panned out. It was risky, but the payoff was potentially large so that's the kind of business people have focused on. This kind of market and business does not fit the car metaphor.
As certain sub-markets started to mature and specific domains clarified, it started to make sense to offer different versions of the same thing. Developers began creating small business versions where the market was dominated by enterprise and vice versa. This period ushered in a new and even more significant boom in software because investors could understand it's just like X, but we've changed a handful of things to make it work in a currently unserved market.
In other words, where there are only two seat coupes, it's easy to make a case for trucks and vice versa. This is the sweet spot market--the market is easily understood, and so is the new product.
Software--spurred by the inherent ease of delivery--has continued to mature rapidly. Consumers have become very sophisticated. Software is now (in many domains) like cars.
Why? Because when your market is sophisticated and delivery mechanisms well advanced, you can exist in super-niche markets. A single paying customer can be enough to sustain development and they might prefer your product for something as urbane as the color scheme. When you consider the variations in lexicon, interface, models, and feature mix--all of which allow software to be far more varied than cars--it's sheer madness to listen to a condensed description of a thing and respond with the Cars Criticism.
Notes
references group=notes /
