Questions I ask when reviewing a design | Hacker News Reader

Questions I ask when reviewing a design

37signals.com

Questions I ask when reading this blog post:

   Why is this a list of questions without any further explanation?

Furthermore, is it really good design to have a big wall of text composed of similar-looking lines? Very hard to read.

"Why is this a list of questions without any further explanation?" - because that's all I said it was. I didn't promise anything beyond a list of questions that I ask when I review a design.

I may expand on a few of these in future posts.

Are we not allowed to post brain dumps on the web?

These are great! A few more of my own:

  What does my reptilian brain "sense" in .5 seconds?
  What can be removed to improve it?
  What is confusing?
  What is offensive?
  What could make it simpler?
  What could make it clearer?
  Which entities are dead weight?
  What is too small to easily read?
  What is too dark/light/unfocused to easily read?
  What questions does it raise?
  How do I quickly and easily find answers to those questions?
  What about it excites me?
  What would it take to excite me more?
  What is it supposed to do?
  Does it do that?
  Who is it supposed to me for?
  Is it for them?
  How do I get out?
  How do I get back?
  On an integer scale from 0 to 10, what is it?
  On an integer scale from 0 to 1, what is it?
  How fast is it?
  What slows it down?
  What should be removed to speed it up?
  When I do something, how do I know it knows I did it?
  Is everything about it intuitive?
  What needs to be made more intuitive?
  How comfortable am I staying here for a while?
  How comfortable am I staying here all day?
  What else should it do without getting too complicated?
  What could be better about it?
  Would my boss understand it?
  Would my customers understand it?
  Would my parents understand it?
  Do I want the past 5 minutes of my life back?

This is like a web-app focused "Oblique Strategies". Someone should make cards! I printed them out and taped them on my wall.

I'd like to add: What the fuck?

Great list, although it doesn't include the first thing I ask myself:

Does it have personality?

What a terrible list. One of the most important things to do in a review is to keep it focused on improvement and not turn it into a kibitizing session. That's why so many sources recommend avoiding "why" questions altogether. It's nice to have "why" explained, but "what" and "how" are far more important. This list contains no fewer than ten explicit "why" questions, and even more implicit ones. Some others are just too vague or subjective ("How does this make you feel?" or "Where's the idea?") to elicit any useful response. Worst of all are the questions that embed (usually negative) assumptions. Here's a thoroughly predictable and counterproductive example of someone using this list.

"Why isn't that clear?" "Why do you say it isn't?"

Anybody who has ever been in a design review before should be running for the exit at this point, afraid of being caught in the crossfire. If you want to improve your design reviews, by all means use this list . . . as examples of questions to avoid.

It depends on how you use the list. If you use the list as a pointy-haired boss trying to micromanage somebody's work, that's one thing. But every one of those questions begins a conversation--like you mention with the reply "Why do you say it isn't?"

It's through those conversation that ensures that the best can ideas can come forth. This hinges on the relationship between the reviewer and the reviewee being solid, something that should be worked on long before the design review.

Correct. These are all conversation starters. The goal is improvement through a deep understanding of what you're building and why you're building it the way you are.

Starting a conversation is a wonderful thing, but how you start a conversation matters. Some questions naturally elicit a defensive response, especially when they make the reviewer's opinion a default and force discussion of that - implicitly rejecting the idea that the reviewee's position even deserved consideration. There's simply no excuse for a "question" that embeds a negative assumption about the work under review.

Similarly, solid relationships are well and good, but they shouldn't be abused. Maybe framing a review question badly is OK between co-founders who've known each other for years, though as a fifteen-year participant in an even closer kind of relationship I'd say it matters even then. What's OK once can wear down any relationship if done a thousand times. Even more importantly, that sort of thing just doesn't scale to the weaker relationships that are inevitable in any company larger than will fit into your bedroom. "Why isn't this clear?" will simply poison the tone not only of that review but of the entire relationship. "How can we make this clearer?" or even "Do we need to make this clearer?" would be far better.

As an example, BTW, how do you think the OP might feel about my critique here? Does it seem overly harsh, like I'm putting my opinion above theirs? Do you suppose that it might take some work for us to establish trust, after such a beginning? Yes, it would. Behavior that's OK here on HN would often be seriously inappropriate in the workplace.

Reply on news.ycombinator.com