Defining Responsiveness

When people say that something is responsive, what do they mean?

I’m not being facetious. I think many of us think we know what is meant, but when you dig deeper, there is disagreement about what it means for something to be responsive.¹

To his credit, Ethan defined responsive web design clearly with three technical pieces:

• Fluid grids
• Flexible images
• Media queries

That seems simple enough. But when it comes to defining what is responsive, things get a bit fuzzier.

The best way to understand where the differences of opinion exist is by looking at an real life example: is Google Plus responsive?

It looks like it might be responsive. The layout varies from one, two or three columns depending on the width of the page. The size of elements on the page change as well.

Let’s look at some reasons why Google Plus might not be responsive.

The first time I cited Google Plus as responsive example, I was told that it didn’t qualify as responsive because it doesn’t go all the way down to support small screens.

This is silly to me. Nothing in Ethan’s article says that responsive web design has to support any possible screen resolution.

We’re perfectly happy calling things responsive designs that have a fixed width past a certain point on wide screens. And we praise the efforts by the BBC, Guardian, People and others that build a responsive mdot site that will eventually replace their main site.

So why would the fact that Google Plus doesn’t support small screens preclude it from being considered responsive?

Google Plus isn’t using media queries the way you might expect.

As far as I can tell², none of the major layout changes you see in Google Plus are handled by media queries. They are controlled in JavaScript.

This is similar to Nathan Smith’s Adaptive.js library which puts breakpoints in JavaScript instead of CSS.

So by definition, Google Plus is not a responsive web design. It may look like one, but it doesn’t contain the three technical pieces necessary to be a responsive web design.

So I exaggerated a bit. Google Plus is using media queries. But as I said, they aren’t being used the way we normally think of them.

All but one of the media queries is looking at viewport height instead of viewport width. The one that looks at viewport width only applies at less than 340px; only adjusts a single icon; and I can’t get the rest of the page to adapt to less than 340px so the entire media query seems moot.

But perhaps we can give Google Plus a pass as a responsive web design because it technically has media queries even if they aren’t being used to make the design responsive?

Nah, that’s cheating. And even if we did let Google Plus off on this technicality, it would still leave us with designs that use JavaScript like Adaptive.js and wondering what we call them.

What happens when you use responsive web design, but add to it things that seem to be at odds with the philosophy of responsive design?

For example, do any of these change whether or not something is a responsive web design:

• Device detection used to select the best-sized source image?
• Redirects based on user agent string to a mobile site that is responsive?
• Less content on small screens with more content on larger screens? What if the developer uses AJAX to pull in the additional content? What if they use device detection to make decisions before the page loads and then AJAX after?
• A couch-mode design that only worries about resolutions from 720p to 1080i but uses responsive web design techniques for those resolutions?

These questions may seem crazy, but I’ve seen people use the presence of device detection as a reason to discount a responsive web design.

When people are looking for reasons why a given implementation doesn’t fit their perspective on what responsive web design is good for, it is pretty easy to find reasons why any given site isn’t a “pure” responsive web design.

When a client comes to us to help them make their existing site or app responsive, we know that we’re going to be using fluid grids, flexible images and media queries.

But we also know we’re going to be using much more than just those three techniques. The best responsive web designs are doing much more. And when we teach workshops or train client teams, much of what we’re discussing are the things that you do after you’ve got the three techniques down.

Which led me to the idea that there is a difference between “being responsive” and responsive web design. That responsiveness was something bigger.

So I asked Ethan for his thoughts.

As you might expect, Ethan had great thoughts on this topic. I’m republishing part of our email exchange here with his permission, but this is just a fraction of his insights. I hope he gets the time to share his thoughts in more detail.

I suggested that “being responsive” or “responsiveness” might be characterized by larger principles such as:

• Designs that utilize the size of the viewport to determine the layout.
• Designs that adjust the layout and the size of the elements in the layout as the size of the viewport changes.
• Designs that ensure content parity across devices.
• Designs that treat URLs as sacrosanct and ensure they work across devices.

Now, Ethan disagreed there was a need to revise the original definition. But he replied:

All of them—but the last two especially—seem to me to be pretty foundational tenets of good web design, full stop. If an interface doesn’t adapt to the display—be it responsive or device-specific—its utility is pretty limited; if it doesn’t honor and respect URLs—be it responsive or device-specific—its utility is pretty limited; and so on, and so on.

The web’s still digging itself out of the constraints of the printed page, in many ways, and we’re still—some twenty years on—trying to articulate the best way to design for this medium. What you’re describing feels bigger and more foundational to me than responsive design. It really feels like you’re describing the way the web wants to be: fluid, addressable, and accessible to all.

So Google Plus, to use your closing example, might not be responsive, but maybe it’s adopting a more web-appropriate design model. And maybe that’s enough.

I think Ethan is right. In the long run, “being responsive” is simply designing for the web the way it was intended.

But in the short run, I find myself struggling to describe the new toolbox and mindset that is required to do responsive web design well. And I see people talk past each other because of their mistaken belief that they have a common understanding of what responsive means.

I’m left with Justice Potter Stewart’s definition for whether or not something is responsive, “I know it when I see it”, which is wholly unsatisfying.

I take comfort ignoring the definitions and instead asking these questions about a design and its implementation:

• Does it adapt to screen size?
• Does it take advantage of device capabilities?
• Is it accessible anywhere?
• Does it work well?

For our users, those are the things that matter.

---

1. Not addressed in here is the fact that responsive is also used in the web performance community to refer to something entirely different.

2. Google Plus CSS and JS is quite complex. I’ve done my best to confirm they are using media queries only in the ways I’ve described here, but I can’t guarantee I’ve got it right.

3. Don’t get me started on the definition of adaptive design. Unless I’m talking to Aaron Gustafson, I know I don’t know what someone means when they say that.