Poll Results: “Sites” vs “Apps”

Published by Chris Coyier

Time to wrap up our last poll on how useful it is to distinguish between "web apps" and "web sites".

The exact question was:

Is it useful to distinguish between "web apps" and "web sites"?

First, the data:

  • 28% of people voted: Nope. It's all just the web.
  • 72% if people voted: Yep. They are different things with different concerns.

A visual breakdown:

See the Pen SVG Doughnut chart with animation and tooltip by Chris Coyier (@chriscoyier) on CodePen

The vast majority of people think it is a valuable distinction to make.

Let's kick off some armchair analysis by sampling some opinions from around the web. Jeremy Keith wrote about this very subject somewhat recently. He's of the opinion that the distinction isn't valuable:

Why?

Why do you want to make that distinction? What benefit do you gain by arbitrarily dividing the entire web into two classes?

He argues that not only isn't it useful (and used as a term because it's simply fashionable) but that it is actually harmful (e.g. assuming some technology/advice/approaches don't apply to you because of this label).

Some folks think the distinction does have some value. Billy Brown says it might be useful for thinking about the future, although doesn't mention exactly how. "Armstrongest" says it is useful simply because human beings need to categorize things by our nature in order to make sense of the world, and that useful properties of that categorization emerge.

Defining the Difference

Setting aside the argument on if the distinction is useful or not, how might we define the distinction.

Here's some people's thoughts:

  • Kory says it might be scope. A "five page website" vs. an app with "dynamic data and user input". Kevin agrees and says web apps "require a non-trivial amount of code" around user interaction.
  • Russel Heimlich says "If it requires JavaScript to work then it is a web app, otherwise it’s a website." Mat Marquis says it is excuse for that.
  • Ginestra says it is about how users access it and what their expectations are.
  • Paul Wallas says it is about time spent. Less than 5 minutes, website. More than 5 minutes, web app.
  • Ben S says the distinction comes between a front and back end angle. More front-end, website. More back-end, web app.
  • Covarr says it's about uses. "Perform tasks", web app. "Provide information", website.
  • Joshua says search bots are a part of it. "Crawlable resource", website. "Personalized for your consumption", web app.
  • Steve says it is about navigation. Normal of things like the back button, website. Customized approach to moving between screens, web app.
  • Jeremy Worboys says web apps content is created by users where websites content is created by the site workers. Jeremy Castelli agrees.
  • ernO says if the page never scrolls, it's a web app.
  • Colin Léger says if you can print the page and get the same content, it's a website.
  • Sean says it is about cost. Web apps are simply more expensive.
  • Michael Chang says it is about public vs private. Websites are public while web apps exist in a private space.

Other things I've heard:

  • Content strategy is only important for websites.
  • A website has an RSS feed; a web app has an API.
  • It is important for investors (i.e. web apps == make money)
  • It is important for hiring/recruiters (i.e. people who make web apps are different than people who make websites)
  • It is "get stuff" vs. "do stuff".
  • You login to web apps.

Almost none of the points above ring true for me. All I see are exceptions and gray area.

I was kind of hoping we could get somewhere close to a solid distinction between these two classifications, but I don't think it's going to happen. There is very little agreement and heaps of opinion.

This leads me to think perhaps the "web app" moniker (certainly the newer of the two) is simply just a fashionable term. We like the sound of it, so we use it, regardless if it truly means anything.

One thing that almost rang true to me while discussing this was the hiring aspect. If you look at all the web workers in the world as a whole, I feel like there is one group of people who work on a short term basis on a website and then move onto something else. Agency and freelance work would be a big part of this. Then there is this other group of people who work on a long term basis on website. They work at LinkedIn or something and they've been there working on it for 4 years. It's tempting to think the short-term stuff is more websites and the long-term stuff is more web apps. So then if you were hiring for one or the other, you might look at someone's history and talk to them about which they tend to prefer and that might influence the hiring situation, as some people are better suited to one or the other.

The more I think about that though, the less true it seems. I think the long-term vs short-term is more relevant than what "kind" of site is in question.

Gray Gray Gray

A restaurant site that has just photos and a menu on it is just a website right? What if it has a Google Map embedded on it? All the sudden you can scroll around and pan and zoom and get directions. Is the restaurant site a web app now? Is it a website with a web app on it? Is componentizing like that still useful?

What if you could leave a comment on the site? That's a web app now right because there is user generated content right? Or are comments too commonplace to get that distinction. What if you had to create an account before you could comment? What if the menu items had aggregate ratings stars on them? Still too simple? Isn't Twitter just comments, accounts, and meta data?

But Twitter has an API right so it's definitely a web app. And the experience while logged in is customized to you. But you can also search it for content, I thought that was websites you can do that on.

What if the restaurant site was actually primarily a blog where the chefs wrote up recipes and cooking tips and wrote about the restaurant business. That's a website right? What if there was a paywall to read that content because it was so valuable to other restaurant owners? An agency was hired to custom develop that paywall and there was user accounts with access levels and admin and Stripe integration and whatnot. Surely that's a web app. What if they used used WordPress and a plugin? Which is it then? And speaking of WordPress, the admin area is a web app right? But what you build are websites? Can a website be both?

There is just nothing but questions, exemptions, and gray area.

If we could pin down a super accurate definition that we agreed on, even then it might not be particularly useful. And since we can't, I argue it's even less useful.

Yet, the poll numbers show otherwise. Most of us are convinced that it is important and useful to distinguish between these things. I guess more investigation needs to happen.

The Shared Importance

Again, even if we could draw a perfect line between web apps and websites, wouldn't these things still be true?

  • They should be fast.
  • They should be secure with any data they store.
  • They should be accessible.
  • They should offer a good user experience.
  • They should have appealing design.
  • They should meet the goals of the site owners.
  • Good people would be valuable to have on the team.
  • They are built with the foundational technologies that power the web and that browsers understand.

If so much of the fundamental parts of web(thangs) are shared, isn't the distinction not useful?

Wrap Up

I know I spent most of the time disagreeing with the data here. Perhaps some more conversation will help. Let's hear those thoughts.

New poll soon.