Stories about Software


You Aren’t God, So Don’t Talk Like Him

Queuing Up Some Rage

Imagine that you’re talking to an acquaintance, and you mention that blue is your favorite color. The response comes back:

Acquaintance: You’re completely wrong. Red is the best color.

How does the rest of this conversation go? If you’re sane…

Acquaintance: That’s wrong. Red is the best color.
You: Uh, okay…

This, in fact, is really the only response that avoids a terminally stupid and probably increasingly testy exchange. The only problem with this is that the sane approach is also perceived as something between admission of defeat and appeasement. You not fighting back might be perceived as weakness. So, what do you do? Do you launch back with this?

Acquaintance: That’s wrong. Red is the best color.
You: No, it is you who is wrong. You’d have to be an idiot to like red!

If so, how do you think this is going to go?

Acquaintance: That’s wrong. Red is the best color.
You: No, it is you who is wrong. You’d have to be an idiot to like red!
Acquaintance: Ah, well played. I’ve changed my mind.

Yeah, I don’t think that’s how it will go either. Probably, it will turn out more like this:

Acquaintance: That’s wrong. Red is the best color.
You: No, it is you who is wrong. You’d have to be an idiot to like red!
Acquaintance: You’re the idiot, you stupid blue fanboy!
You: Well, at least my favorite color isn’t the favorite color of serial killers and Satan!
Acquaintance: Go shove an ice-pick up your nose. I hope you die!

Well, okay, maybe it will take a little longer to get to that point, perhaps with some pseudo-intellectual comparisons to Hitler and subtle ad hominems greasing the skids of escalation. If you really want to see this progression in the wild, check the comments section of any tech article about an Apple product. But the point is that it won’t end well.

Looking back, what is the actual root cause of the contention? The fact that you like blue and your acquaintance likes red? That doesn’t seem like the sort of thing that normally gets the adrenaline pumping. Is it the fact that he told you that you were wrong? I think this cuts closer to the heart of the matter, but this ultimately isn’t really the problem, either. So what is?

Presenting Opinions as Facts

The heart of the issue here, I believe, is the invention of some arbitrary but apparently universal truth. In other words, the subtext of what your acquaintance is saying is, “There is a right answer to the favorite color question, and that right answer is my answer because it’s mine.” The place where the conversation goes off the rails is the place at which one of the participants declares himself to be the ultimate Clearinghouse of color quality. So, while the “you’re wrong” part may be obnoxious, and it may even be what grinds the listener’s teeth in the moment, it’s just a symptom of the actual problem: an assumption of objective authority over a purely subjective matter.

To drive the point home, consider a conversation with a friend or family member instead of a mere acquaintance. Consider that in this scenario the “you’re wrong” would probably be good-natured and said in jest. “Dude, you’re totally wrong–everyone knows red is the best color!” That would roll off your back, I imagine. The first time, anyway. And probably the second time. And the third through 20th times. But, sooner or later, I’m pretty sure that would start to wear on you. You’d think to yourself, “Is there any matter of opinion about which I’m not ‘wrong,’ as he puts it?”

In the example of favorite color and other things friends might discuss, this seems pretty obvious. Who would seriously think that there was an actual right answer to “What’s your favorite color?” But what about the aforementioned Apple products versus, say, Microsoft or Google products? What about the broader spectrum of consumer products, including deep dish versus thin crust pizza or American vs Japanese cars? Budweiser or Miller? Maybe an import or a microbrew? What about sports teams? Designated hitter or not? Soccer or football?

And what about technologies and programming languages and frameworks? Java versus .NET? Linux versus Windows? Webforms vs ASP MVC? What about finer granularity concerns? Are singletons a good idea or not? Do curly braces belong on the same line as a function definition or the next line? Layered or onion architecture? Butter side up or butter side down? (Okay, one of those might have been something from Dr Seuss.)

It’s All in the Phrasing

With all of these things I’ve listed, particularly the ones about programming and others like them, do you find yourself lapsing into declarations of objective truth when what you’re really doing is expressing an opinion? I bet you do. I know I do, from time to time. I think it’s human nature, or at the very least it’s an easy way to try to add additional validity to your take on things. But it’s also a logical fallacy (appeal to authority, with you as the authority, or, as I’ve seen it called, confusing fact with opinion.) It’s a fallacy wherein the speaker holds himself up as the arbiter of objective truth and his opinions up as facts. Whatever your religious beliefs may be, that is a role typically reserved for a deity. I’m pretty sure you’re not a deity, and I know that I’m not one, so perhaps we should all, together, make an effort to be clear if we’re stating facts (“two plus one is four”) or if we’re expressing beliefs or opinions (“Three is the absolute maximum number of parameters I like to see for a method”).

Think of how you you would react to the following phrases:

  • I like giant methods.
  • I believe there’s no need to limit the number of control flow statements you use.
  • I would have used a service locator there where you used direct dependency injection.
  • I prefer to use static methods and especially static state.
  • I wish there were more coupling between these modules.
  • I am of the opinion that unit testing isn’t that important.

You’re probably thinking “I disagree with those opinions.” But your hackles likely aren’t raised. Your face isn’t flushed, and your adrenaline isn’t pumping in anticipation of an argument against someone who just indicted your opinions and your way of doing things. You aren’t on the defensive. Instead, you’re probably ready to argue the merits of your case in an attempt to come to some mutual understanding, or, barring that, to “agree to disagree.”

Now think of how you’d react to these statements.

  • Reducing the size of your methods is a waste of time.
  • Case statements are better than polymorphism.
  • If you use dependency injection, you’re just wrong.
  • Code without static methods is bad.
  • The lack of coupling between these modules was a terrible decision.
  • Unit testing is a dumb fad.

How do you feel now? Are your hackles raised a little bit, even though you know I don’t believe these things? Where the language in the first section opened the door for discussion with provocative statements, the language in this section attempts to slam that door shut, not caring if your fingers are in the way. The first section states the speaker’s opinions, where the language in the second indicts the listener’s. Anyone wanting to foster a cooperative and pleasant environment would be well served to favor things stated in the fashion of the first set of statements. It may be tempting to make your opinions seem more powerful by passing them off as facts, but it really just puts people off.


I want to mention two things as a matter of perspective here. The first is that it would be fairly easy to point out that I write a lot of blog posts and give them titles like, “Testable Code Is Better Code,” and, “You’re Doin’ It Wrong,” to say nothing of what I might say inside the posts. And while that’s true, I would offer the rationale that pretty much everything I might post on a blog that isn’t a simple documentation of process is going to be a matter of my opinion, so the “I think” kind of goes without saying here. I can assure you that I do my best in actual discussions with people to qualify and make it clear when I’m offering opinions. (Though, as previously mentioned, I’m sure I can improve in this department, as just about anyone can.)

The second caveat is that what I’m saying is intended to apply to matters of complexity that are naturally opinions by their nature. For instance, “It’s better to write unit tests” is necessarily a statement of opinion since qualifying words like “better” invite ambiguity. But if you were to study 100 projects and discover that the ones with unit tests averaged 20% fewer defects, this would simply be a matter of fact. I am not advocating downgrading facts to qualified, wishy-washy opinions. What I am advocating is that we all stop ‘upgrading’ our opinions to the level of fact.

1 Comment
Newest Most Voted
Inline Feedbacks
View all comments
Paul Matovich
10 years ago

Nice post except for the fact that everybody knows that BLUE is the best color and studies by guys wearing lab coats have proven that Blue is the best. So… we can just defer to the study and no need to think it through ourselves 🙂