Dogmas Are Meant to be Broken: An Interview with Eric Reiss

Posted by

At a university where robots were celebrated and MBAs were heroes, I studied technical writing and communication design, using the combination as a way to understand how to write and design for different audiences. At that time, these studies—called mostly “technical writing”—took the form of the reticent user manual. Sometimes it took form as an instruction booklet on how to operate a washing machine. Other times, it took form of a help system on golf course irrigation. You see, the content always differed somewhat fiercely. What remained remarkably the same, however, was the need for clear ways to communicate instructions. Jargon-free text, clarity, and concision were the dogma of our days as technical writing students. Tech writing, as humdrum as it may be reputed, is rich with wildly helpful principles and guidelines.

And writing is still where we find many well-known guidelines. William Strunk, Jr., and E.B. White, perhaps best known for the staple The Elements of Style (or, in some circles, the latter for Charlotte’s Web), are the most prominent household name on word slashing and clarity. “Omit needless words,” they caution. “Vigorous writing is concise. A sentence should contain no unnecessary words, a paragraph no unnecessary sentences. …”

Since those days, I’ve been interested in finding the same kind of principles written specifically for web design. There are, no doubt, Jakob Nielsen’s heuristics. But I’ve been searching for something that allows for a bit more autonomy—guidelines that provide structure without confining. That’s why, when I saw Eric Reiss’ Web Dogma quietly hanging on the wall at the Information Architecture Summit in Vancouver, I was intrigued. This simple, typewritten list of 10 said the following:

Web Dogma ‘06

  1. Anything that exists only to satisfy the internal politics of the site owner must be eliminated.
  2. Anything that exists only to satisfy the ego of the designer must be eliminated.
  3. Anything that is irrelevant within the context of the page must be eliminated.
  4. Any feature or technique that reduces the visitor’s ability to navigate freely must be reworked or eliminated.
  5. Any interactive object that forces the visitor to guess its meaning must be reworked or eliminated.
  6. No software, apart from the browser itself, must be required to get the site to work correctly.
  7. Content must be readable first, printable second, downloadable third.
  8. Usability must never be sacrificed for the sake of a style guide.
  9. No visitor must be forced to register or surrender personal data unless the site owner is unable to provide a service or complete a transaction without it.
  10. Break any of these rules sooner than do anything outright barbarous.

Now, Eric Reiss started out as a stage director at the Danish Royal Theatre in Copenhagen gave him his first introduction to designing experiences. Since 1984, he’s been a business strategist and writer, and in 2001 founded a web design company. With training in everything from stage design to Egyptology to hypertext games to web projects, Reiss has had extensive practice in finding out what makes an experience work. Could these be the principles I’ve been waiting for? I tracked down Reiss in Vancouver to find out.

I need to know about the Web Dogma, the ten principles. What are they all about?

I don’t know whether you’ve seen the Danish movie The Celebration by Thomas Vinterberg. This was a so-called “Dogma” film. In 1995, Vinterberg and another Danish director, Lars vonTrier, got together and wrote down ten rules they called Dogme 95. This was a reaction to the gaudy Hollywood films where there was a lot of lighting, a lot of special effects. Their dogma dictated that cameras had to be hand-held, you had to use natural lighting, if you wanted to use a prop, you had to find a location where that prop existed; you couldn’t import things into the actual film.

I did some voiceovers for Lars a couple years back during post-production on Dancer in the Dark with Björk, and I asked him about his dogma and provocation. I thought, there has to be a way to do this for the web. The problem is that we have standards. And standards are very useful. We agree, for example, that in the United States, we drive on the right. But in the UK, Australia, Japan, etc., they drive on the left. Doesn’t mean one is better, but we’re at least in agreement where the cars should be on the road. So that’s a standard.

And standards are just standards, and they can change as technology changes. Best practices are guidelines; they’re not rules. It’s best to keep two seconds worth of stopping distance between you and the car in front of you, for example. That’s just good common sense. As cars get faster, these distances change, and the best practices change. But they remain only guidelines.

The trick with doing a dogma for the web was to avoid the “rules syndrome” (For example, Links should be blue.) for best practices that were liable to change as technology changed. How do you do a set of rules or guidelines that would prove helpful despite the technological advances and would also be relevant as fashion changes?

That’s how the dogma came about. It’s not particularly spectacular, and it’s not the Ten Commandments. These are simply why users find [some] sites and devices less usable than others.

Why are some sites and devices less usable than others?

For one, because today’s content management systems can be pretty sophisticated and throw up all kinds of things that it thinks are related. Because we design pages that often have, for example, a three-column format, designers and editors feel obligated to fill that third column. Often that stuff is just noise.

As sites become more complicated because the technology is available, then we need to step back and ask whether we’re doing our users a service by providing them with information that is not necessarily relevant within the context of their own mindspace.

Just to clarify then: the dogma is for designers of all kinds. It’s not for users.

No, it’s not for users at all. It’s absolutely for people who are working with web design, and I use “design” in the broadest possible sense. Could be content providers, could be graphic designers, could be experience designers, could be business people who are trying to unite business goals with the goals of users.

But why leave out the users? Couldn’t they use these principles to evaluate what is good?

Well, I think they are using them already. In many ways, they created them. They can’t use sites that violate many of the principles. And “principles” is even a dangerous word. I don’t want this [dogma] to be more than it is. It’s just a list of things that I’ve found work really well. And if I look at sites that either work or don’t work, they either follow or violate one or more of these edicts.

I’d like to go back to the first one: “Anything that exists solely to satisfy the politics* of the site owner should be eliminated.” Aren’t a site owner’s politics the reason that many people read one site over another?

Sometimes, but not always. Most of us have met executives who insist on adding content that is completely personal in nature: “I want a picture of my wife’s cat because this is really important for her.” That’s, of course an extreme example, but there’s a lot of stuff, particularly in HR sections of big, corporate websites, that is not useful to site visitors. A lot of people put information on the site in a cover-your-ass fashion. And that’s unfortunate because it distracts from whatever the main emphasis is.

I’d like to remove the clutter.

Some of the best websites I’ve seen have been ones that are truly personally motivated. This may not apply for a large company, but an “About Us”-like section is really only about a person’s politics. And site visitors can kind of gain a lot of meaning from that. At least I do.

Right, I agree with you. And I realize that there are also a lot of blogs and other websites that are absolutely politically motivated on the part of the site owners. However, their audience is also there to learn and to get a take on these people’s politics. So, in fact, the information is relevant for the user. I love political sites. I love blogs. I love that people have a forum for discussing what’s going on. Because that, by its nature, makes the content relevant.

Just because things are called “political” doesn’t make them bad. It’s when a site clutters itself up with stuff—somebody in marketing insists on including a particular photo because they dated the model once, or whatever. Invariably in any large project, you run across bits of content that serve no purpose whatsoever other than to make someone in the organization happy. This is the kind of stuff that I’d like to take out.

What do you think of the web-specific heuristics and other criteria that might exist out in the world? How does your Web Dogma fit in with those?

Heuristics are fine as far as they go. I do a lot of heuristic website evaluations, and those are based on best practices. You know, for example, if there’s some sort of a horizontal divider on the page, people have a tendency to ignore stuff above the divider. If you take search results, and pull something out as a “best bet” and put it in a box, a lot of folks will skip over it because their expectation is that they will get a list, so this is what they look for. And they’re going to skip stuff they think is marketing fluff. So, those are heuristics and those are fine. It’s also based a little bit on what’s going on at the time. What is the fashion? What are people used to? People learn skills on one website that they expect to use on another website. Is it intuitive, or is it something that requires instructions? Heuristics take all of these things into account with a value judgment to say, “This works” or “This is something that can be improved.”

My dogma is not meant to conflict with heuristics because heuristics change. I see these as being more general rules that will apply for many years. And this is the difference between the Dogme 95 and my dogma. One of the mistakes the first Dogma made is that it suddenly got very categorical: All film should be shot on academy 35 mm. That dates the dogma. And I’ve tried assiduously to avoid anything that would date it, but to find general principles that transcend technology and transcend specific media.

But in one of your principles, you use “page” and said “readable first, printable second, downloadable third.” Won’t those things date it?

As long as things are still two dimensional, the page metaphor works. No matter whether that page is a mash-up or if it’s a more sophisticated 2.0 application, it’s still something that exists in a two-dimensional form. And so until that disappears, I find “page” to be a useful term.

I think I want these principles to apply to ubiquitous computing. Using technology or format-specific terms strikes me as limiting them from being used outside the context of a two-dimensional space.

I hadn’t really thought much about that. Ubicomp does things automatically, gives me information that I want but didn’t necessarily ask for. When it starts giving me information I don’t want, then it’s time to re-read the dogma. I called this “Web Dogma” because I am mostly thinking about websites. We go on and on about folksonomies and Web 2.0, but the truth is we didn’t even really get Web 1.0 very good. There are a lot of sites that just don’t work. And there are a lot of sites that work surprisingly well despite being put together by the neighbor’s kid. It’s wrong of us to be so elitist.

Take all the zillions of websites in the world, then subtract all the websites that have been done “by the book,” and we still have zillions of sites left. Then, let’s deduct all the sites that don’t work for one reason or another. You’re still left with zillions of sites. So there’s something that indicates that the neighbor’s kid is still doing a pretty good job. He’s maybe not a trained designer, but let’s face it, I get a lot of pleasure from hobby sites put together with a stick of chewing gum that fulfill my needs. And I know a lot of sophisticated sites that fall apart because they violate these dogma.

Take the sites that are built by the neighbor’s kid or look like eBay. They’re so successful, but are poorly defined by the book definition that you’re talking about. Why do we love them?

They’re fulfilling a need. The mistake that is being made is that people say, “Ugly sites sell better.” And there’s no cause and effect here. And the people who are trying to imply a cause and effect (“Well, if we make our site ugly, it will work better”) are fighting a nonsensical argument.

Of course, they’re fulfilling a need and they’re abiding by the dogma that you talk about.

There was a lot of talk a couple years back about the back button being the Back Button of Doom because 62% of people who hit the back button are unable to complete their task. Well, my contention is that by the time people hit the back button, something else has fallen apart in the system. They are unable to complete their task. They’ve either gotten lost and need to back up to someplace where they can recover their bearings, they think they missed a link, or there’s bad navigation and they’re pogosticking. There are things that are falling apart for them, and it’s not the back button’s fault. Gee, I‘d better write a letter to Meg Whitman and say, “Your site really sucks because I have to rely on the back button.” Because on eBay, it’s much more convenient to use the back button then to use the site’s navigation. Everybody who uses eBay knows this. That does not make the back button bad.

A lot of the things that are being said in the name of best practice just don’t hold true.

Interesting. It’s almost like the Prairie architects.

Right, Frank Lloyd Wright. Fallingwater meets Cascading Style Sheets.

Yes. They used the landscape as part of their design. They didn’t try to design against it. It’s like eBay using the back button as part of their design.

I don’t know if eBay did this consciously, but that’s the way it works and I haven’t seen much in the eBay design process that has changed the way I operate, and I’ve been using the site since 1997.

What made you put forth the Dogma now?

This has been percolating for about three years now. A couple of months ago, I finally started showing it to people because I couldn’t continue in a vacuum. There were a range of comments that didn’t necessarily change the text of the dogma, but helped me prepare the answers to the kinds of questions you’re asking now. It is open to a lot of interpretation and I suppose if I were more mercenary, I would write a book and explain what I meant. But I’m kind of hoping that people will figure this out for themselves. These are pretty generic patterns. Maybe somebody else will take it up and write about it. I’d like it to be a general tool, and something that’s useful to a broader crowd.

Would you feel insulted or flattered if I compared what you were doing to Strunk and White?

Oh, I’d be flattered. I still have my Strunk and White, and it’s probably one of the most important books I’ve ever read. But I’ve also been a professional writer for 20 years now. The last dogma was stolen directly from George Orwell—it’s his tenth legendary rule of writing.

Update: After this interview, Reiss edited #1 in the Dogma, adding “internal” before “politics” to make his point clearer.


  1. Interesting article.

    There is definitely a need for practical, reusable heuristic patterns like this, and it’s good to see other people spreading wisdom as it is gained.

    Later on this year, you should consider reading my upcoming book, “Designing the Obvious”. It is all about some of the same types of ideas found here, except it focuses more on proactive, application-level patterns found in many great web-based applications and discusses how to achieve them. The simple elimination of things implemented because of internal politics and such can make an application better by default, but there are many other factors that contribute more directly to the success of a project. Designing the Obvious discusses these things in depth.

    Anyway, it should be out in September/’06, so keep an eye out for it.

  2. I love the dogmas – they are a great approach to thinking, but flexible. Great interview too!

  3. The “dogmas” are perfect. A lot of re-stating what has been said a thousand times before; but clearly, there is a need to re-state these ideas and more importantly, for them to be read or heard.

    It all comes down to a well formed information architecture and simplicity and actually practicing the dogmas…

  4. I’m a big believer in having a starting point for your thinking. It serves as a rudder when things start to get tedious or too micro-focused.

    As long as you have a standard to begin with, you have to come up with an argument to depart from it as opposed making up a rule on the spot or being at the mercy of the most charismatic person on the team when it comes to what and why something finds its way onto a page.

    It’s not going to solve all your problems and compromise is part of doing business. However, you’d be surprised how often you can close a discussion by being the person in the room with an external basis for a point of view.

  5. Eric’s dogmas are spot on. I’ll print them in 48pt Frutiger (our corporate font :-/ ) and permanently glue them onto the office wall for everyone of my colleagues to read, every single day.

    The dogmas are stating the obvious, but that’s what dogmas are for. As a consultant I feel I’m stating the obvious for my clients every day, but sometimes you need a guy with a higher oomph factor than yours truly to state it too for the CEOs out here to hear it.

    Come on – shout it out.

  6. Eric presented them to me several years ago and I fully agree to them.

Comments are closed.