Wow. Brilliant post. I’ve been trying to point out these exact arguments to people at work (we’re a web development studio who sometimes get saddled with graphic designers that design and don’t think) but it’s hard to get the point across, almost as if it’s a new paradigm to discuss “design” as something which requires thorough understanding of interaction models and, for instance, the capabilities and limitations of programmers’ toolsets.

I often get asked “what’s so great about Amazon anyway?”, or I get to hear that “Gmail is really ugly”. And it seems like no one really comprehends the strides Facebook is making.

Thanks for this. I’ll be referring to it in future when I need to make a point.

I think perhaps both Armin and Khoi are lamenting the visual, aesthetic aspect of web design – not its functionality. Visual gratification is certainly one portion of the equation. No matter how well it works – it can still be uninspiring. Craigslist. Google. on the low end – MySpace. How much more enjoyable would they be if they were beautiful? Function is great – Accords, Camrys work great – but if that’s all car designers could build or dream up – they’d shoot themselves. Many people would want to follow. We need the occasional Mini Cooper, Plymouth Prowler, or Saab.

I’m not saying someone should give up function for aesthetics – but should we aspire to make everything we build beautiful. There’s a reason we have art museums in culture – so everyone can enjoy beauty and thought-provoking creations. Humans are drawn to beauty – we are visual beings. If Google and Craigslist were more beautiful – if they but a smile on my face – my day would be a little better. MSNBC just made a stab at it, and my Amazon is getting prettier, too. Google has it right with it’s holiday logos. A little bit of whimsy. I love the little mountain chateau I saw at the top of Amazon today – it made me smile.

if you note that the essential element of a web site is not how it looks when it’s sitting there, but how it looks when you move through it, I’d say that static screen shots are never going to look like much of anything compared to their print companions.

as a further complication, the really super good designs let you interact with them so quickly that even video doesn’t do the trick. watch someone navigate through too much email with pine (an iconic design for a text mailer) and you won’t be able to follow what they are doing – the screen moves too fast to make it a video narrative.

The most widely used user-interface of the 21st century is a single text box with a submit button. Mindblowing.

This is a great post. It’s well-written and clearly makes the point it’s trying to make. However, I’m not entirely sure I agree with that point.

You are pointing to examples of interaction design. I think you’re right — there are, indeed, canonical examples of interaction design, and you’ve pointed a few big ones out.

I think what Armin is talking about is visual design, which is a separate (but equally important) aspect of web design as a whole. It’s difficult to find canonical examples of visual design on the web. The few I think think of are dead or no longer relevant (k10k comes to mind). This is the nature of our beast — our industry moves so fast, it’s unlikely anything will be around long enough to really become significant in the way that Armin is talking about.

One problem in our industry today is we don’t have a good way of talking about and critiquing web design. There are so many different disciplines and factors that go into creating a great site. Google has great user interaction, but poor visual design. Other sites might have great visual design, and poor accessibility. Yet others might have great copywriting and poor information architecture. Holistic critiques are difficult because we’re not all looking at the same factors. I think eBay sucks, because I’m a visual designer. Someone else might think it’s great, because they’re an interaction designer. I think many Flash sites are beautiful and inspiring because I’m a visual designer — others might think they suck, because they’re web standards and accessibility gurus.

~bc: I would say that you offer a version of the “why not make it beautiful?” argument.

To which I cannot disagree fully, but only say again that beauty is subjective, and your beautiful Google may be someone else’s visual nightmare.

And, to your point, I’m sure we have some beautiful search engine out there that is the Cooper Mini of the search world (ask.com is nice)…notice how they went super sparse on their homepage…certainly influenced by the canonical Google.

Your view is as extreme and one sided as the purely graphic design perspective. The leading sites have failed to benefit from and incorporate basic concepts of graphic design, just as graphic designers have failed to adapt to the interactive needs of practical web applications.

The future is a coming together and transcendence.

BTW, your use of “canonical” is a bit off, better “archetypical”.

To appreciate how poor the graphic design capabilities of Google are, try “iGoogle” and some of its “themes”. I haven’t seen stuff that bad at a major site since the early 00′s.

Look at Google Blogger’s themes, for more crimes against nature.

There’s lot’s of opportunity for graphic designers on major web sites. Recommendation: leave behind Flash, and work within the AJAX/CSS/XHTML/XML landscape. And let’s try and get some better fonts licensed for downloading and embedding.

I think its way too early to make calls on what is and what isn’t canonical web design. Interactive design has only been around for a short time. So while something spectacular might be amongst us, it might take time for everyone to acknowledge it.

I haven’t thought enough about it but is a problem with the visual aspect of web design the lack of resolution independence? Perhaps if it were easier to make a coffee table book of web sites it wouldn’t seem like the web is bereft of good design.

Thanks for the shoutout, Josh. We’ll keep working hard. Love your blog, keep up the great work!

David Lifson
Personalized Recommendations
Amazon.com

As always, keep it simple, stupid.

But let’s add accessibility and clarity.

This works for both print and Web.

And this is all that is necessary.

This is really interesting question indeed and something worth thinking about.
Here are my takes on the issue.

First of all, we should look web design as it’s own discipline. Not try to look rules of graphic design (or something else) and make judgments based on those, because web design functions different –as Jeff Croft pointed out.

It’s too early to judge
We are really just starting to learn how to crawl. Like the cave man, we’ve discovered a clay, a handy material which we can manipulate easily and create different things. It takes time to master the technique well enough to build a hut. It will take even more to learn what makes a good hut and what is the function of it. This knowledge has nothing to do with clay it self. It’s something that we cannot grasp yet.

Cave man next door might have accomplished better functioning door or a fireplace. So we compete with him and copy things to be at least equal. Still, both of our huts are vulnerable for fall rains spring floods.

It will take time until we understand and learn to take an advantage of these phenomenons. It took quite a while for Egyptians to learn how to take the most out of the flooding Nile. Like the fall and rise (and fall) of MySpace and Facebook, the shifts of people between different services. We really have no idea of these kinds of movements. Right now the best we can do is to make a weak prediction that Facebook might lose it’s users sometime in the future. But what then?

To get some idea where we are, we could look something close but more mature. What are the archetypes or canonical designs of software? …Photoshop…Word…OS X? Is there any? How about computers them self? …Lisa? For me it seems that even the parents of web design don’t seem to have strong and clear archetypes.

And then there is our wacky version number 2.0 which implies that we are just meters from the starting point. For every medium, it has taken long (or really long) time to get to the point when archetypes arise: in painting, architecture, printing press, television etc. Even though development is speeding up, web design is young.

We don’t know the rules
If our neighbors hut has a good looking door or a better working fireplace, we can always copy it. For something to be canonical, it has to be something that cannot be copied. We can create an exact duplicate of it, but it still won’t be the same thing.

I see this that archetypes is not about the craft, but about the values. Something that has reached the absolute beauty of its time. It has all the right things in a balance. This is why a duplicate isn’t the same thing: it doesn’t represent the purity of those values, it isn’t an invention. You can always follow those values and try to reach the same idea of beauty.

Above was pointed out that we don’t know the rules. If we don’t know what we’re going after, it’s impossible to reach that absolute beauty and perfect the values in the design. We’re popping out new disciplines, niches and titles every once in a while. This is a sign that we’re going forward. We’re keeping our eyes open and trying to find the rules which makes things function. We’re getting closer to the values but they are blurry still.

Colosseum was and is considered the best representation of the roman architecture of it’s time. It presented perfectly the design and architectural thinking of that time. Still, the idea of beauty and architectural rules that the whole building founded on –wit it’s pylons and arches– is hundreds of years older than Colosseum. It took centuries for someone to perfect those values. Colosseum became THE image of how to build right kind of buildings. Architects all around Europe came to study it. It was and is truly a canonical design.

Where is our history?
Things don’t become canonical right after their creation. It takes usually time for the culture to evolve and/or someone to discover it. It takes truly a strong person and odd times to declare something your own canonical.

In order for someone to discover our work, it has to be seen and discovered. We have to have some reference how things have evolved and why this is phenomenal.

Web is destroying it self as it evolves. Unlike cities that expand, we keep building upon the same plots, demolishing things that were there before. We have really weak trace of history. Our work is wiped away in a second.

We should have internet museums (or way-back-machine on steroids) where we could cut through time and see how things are evolved. That way we could start seeing influences and turning points. That way someone could discover our work.

Archetypes kill creativity
I don’t think it’s bad thing that we don’t have something to point and say “that’s the absolute beauty”. This would mean that we would limit our self into thinking that “that’s the way things are and should be done”. It’s truly a richness to argue whether Amazon and Google have good design.

When Christianity started to take a strong ground in Rome, it shifted the art from rich Greek style to more iconic and simplistic. People were no longer presented as individuals with detail but instead as a symbol of a human. No longer were pylons graved full of ornaments but instead they had icons of the rich leaves. Christianity made also the knowledge of surgeries disappear. The thinking was, if a person got ill, it was an intension of the god and it should be interfered with. This was of course a shift of culture but it also meant that certain things were considered “right” and “pure” hence things were done in that “right” way. It took centuries until Charles the Great who loved ancient Greek and Rome brought back richness in art and architecture –and again thinking and culture shifted.

So as long as we don’t consider anything canonical, our mind is free to play with ideas, create different wide variety of stuff and really push things fast forward. Lets look back what we’ve done, lets keep our minds open and let’s not try to claim that we know everything. Maybe then we’ll accomplish something canonical.

When Google started back in the nineties, the homepage was designed by its programmers. They needed something simple, a logo and a searchbox. In that period of the nineties animated gifs and bold 3D faux-look text treatments were common. The logo of Google has not changed since that nineties. Though on special days like christmass the Google logo was given a christmass treatment eg with reindeers and santa claus. You could argue that stylistically the Google logo looks dated but again …. The Google logo and the simple search box as ‘a brand’ is recognized by millions of Internet Users instantly. So why change it? You could, but Google does not need to. The user design was first tackled , than the looks.

Very well put Joshua. I learnt quite a bit, thanks. It’s actually quite an old argument too; going back many decades you can apply just about every aspect of your argument to the industry I work in, architecture.

A bit of an aside:

I know everyone loves to hold up Massimo Vignelli’s New York City subway map as a great example of design, and perhaps it was a great work of art, but as maps go, it was terrible. Geographical accuracy was absolutely thrown out the window, giving commuters no sense of how long their journeys might take. Central Park was square. The colors were atrocious. The thing was generally a pain in the ass to use.

As an object that not only had to be looked at but also used, I think that the Vignelli map fails to be a canonical anything, except perhaps a canonical example of how graphic designers all too often forget what’s important in what they’re designing.

I actually liked your article a lot, then I realized I disagreed with it entirely. Armin’s question is aimed, to some extent, at the superficiality of the thing, the presentation, not the doing.

For example, the Farnsworth house is an icon of modern design. Mies van Der Rohe built this lovely modern house on a 60 acre rural estate outside of Chicago. It brought the outside in and managed to deflect a million criticisms of the austereness of modern design as championed by Mies. It is an icon.

But the doing, the living in the house, is apparently not so great. Drafty and with numerous other problems, as a house it is kind of a failure. But its influence from an aesthetic perspective is huge.

So, sadly, even for functional arts like web design and architecture, it sometimes is just the superficial…

Oh, and the NYC subway map? Can you figure out which train will stop at a sometimes express, sometimes non-express stop? What about late at night? It is a reasonable example of design, but not at all as brilliant as it is made out to be, I think the Kick map is better.

Two things:

(1) Even though “interaction design” or “user experience” are terms usually reserved for the websites we build, they are just as applicable to the graphic designs we make. We interact and have experiences with magazines, brochures, catalogs, books, logos, advertisements, commercials, billboards, typography, etc. And we’ve been having these interactions and experiences long before the Interweb ever came around. The particular ways in which we interact and respond within these experiences are only evolved, more direct, instant. No longer do we write a letter to the editor, hoping it will be answered and maybe published (of course, this still exists) — we comment on the blog, leave product feedback, etc., as fast as we can type it.

(2) This dichotomy of distinguishable graphic designers vs. anonymous web teams, is a bit of a misconception as well (this is my own perspective). Paul Rand designed the IBM logo, but it was the production workhorses that actually brought it to life: people in factories, print shops, executive meetings, production artists (unfortunately we hardly ever know these people). In the same light, Jason Santa Maria designed the AIGA website. But it was a team of other developers who brought it to life, and all of those people can (and have been) distinguished just as Paul Rand was. The application is different, but it’s the same process.

Isn’t the NYV subway map basically a re-working of Beck’s tube map (London) from the 1930s?

Isn’t the NYC subway map basically a re-working of Beck’s tube map (London) from the 1930s?

I hate the logo as well, always have. Whenever I see it I think about how I’d love to change the G, if anything the G. It’s the ugliest serif G I’ve ever seen. It’s obscene. That’s all they’d have to change, the G. It’ll become a monolithic brand on it’s own because of the power and popularity of Google. But otherwise, damn, please just change the G.

I completely disagree with your whole angle, Joshua. I think you’re unintentionally and needlessly degrading the practice of graphic design by making this a graphic vs. interaction debate. Armin is championing *graphic design*. Period. And that objective is, despite your arguments about the importance of good interaction design and powerful functionality, *also* a good thing.

I wrote more:
http://tinyurl.com/2yxah7

The idea that “aesthetics are subjective” and therefore it’s pointless to discuss them is a red herring. Armin begins his post by citing five examples of designs that are universally applauded on aesthetic grounds. He asks why there’s no equivalent on the web, and you dodge the question — first by changing the topic to interaction design, then by dismissing aesthetics as “subjective,” once in the text and once in the comments. But if they’re subjective on the web, they’re subjective in print too … and yet Glaser’s Dylan, Seven, et al are in “consistent use as exemplary cases.”

Your points on interaction design are acute, though.

Three examples of canonical web design (as opposed to interaction engineering), outside of the best in e-commerce, hipness or utility, off the top of my head: suck.com, nytimes.com (the latest iteration i mean), espn.com. There are others.

This might seem like a pointless comment, but I felt I had to tell you that you are 100% correct in everything you say.

To reduce web design to simple graphics is to misunderstand the point of web design entirely.

weird you use New Yorks subway map as an example when the London Underground map is far more successful and influential

The original K10K epitomized and exalted a particular movement of the time in web design. And although pixel fonts and pixel art are out of favor aesthetically now, it doesn’t diminish the enormous impact which that site had at the time.

In fact, I can’t think of a site which had so broad an impact purely as a piece of design. That site has appeared in practically every book about webdesign (or the history of) ever written.

Canonical? meh…Perhaps. Seminal? Definitely.

“a new paradigm of social design”

You go to such excellent lengths to clarify the meaning of ‘design’, then throw in a dead phrase like that…

A good post, definitely gives food for thought on the differences in thought processes between graphic and web design – although the two can never be mutually exclusive of each other.

It’s a shame that you only used Americanised examples though, Vignelli was only 5 or 6 when the first London Underground Diagrammatic Map was designed by Harry Beck—a design that still is used now and is still clearer than the Vignelli version.

Thanks for this elaborate post. I strongly agree with the notion of process design as an integral part of web design. I touched on the same topic
a few days ago (see link on my name above – it’s in German I’m afraid).

PS: shrib.com is my own exercise in applying good process/web design – let me know what you think

The argument of what constitutes great web design too often boils down to form vs function. The side you favor is often biased by your background – graphic design or programming. In the real world it inevitably is compromised by customer requirements (like SEO) and tastes. What designer hasn’t run into a client with no design background who wants to change colors, make the logo bigger, and move things around? What developer hasn’t had to add features because the client wanted them. An amusing example: what would Google look like if it were optimized to rank high on Google?

Good design works. An efficient, easily navigated web site is good design. A publication that is easily read and well-organized is good design. A silent, highly skewed propellor is good design. Added value can be beauty, excitement, and emotional content but it might not make a good design “better.”

Style and design are separate. Design is about ‘working well’, style is about ‘looking good’.

A web site can (should) have both, but it is far more critical for the site to work well then look good. It’s like a car, at the end of the day, you want one that gets you home. But, you are happier if it also looks good.

I agree, Joshua. The comparison benchmark for websites should be products, not graphic design output. Compare Google to iPod, Motorola Razr or Dyson, not a logo or a poster.

I think the fixation on the Google page is a bit extreme. It may be a good design the way a milk bottle is the embodiment of simple functional beauty. But its beauty comes from use and familiarity, and I would venture to guess that most people don’t really care one way or the other how the Google logo looks. Anyway, Google’s whole approach is to seem like it’s not trying to market anything.

i love this post and the ensuing thread.

even though web design/development teams know their separate roles, many companies hiring people to create their sites fail to recognize the roles of each. they see them as the same. this sometimes is where the gap is never bridged, where the attention to design is left behind.

reviewing sites using a critical response method would take the subjectivity out of it–educators use it often to help qualify and quantify students’ work.

and to add to your points, i think it’s safe to say most of the time, what pleases the designerati community doesn’t actually coincide with what the real users are looking for.

That certainly doesn’t mean thing has to be ugly to be able to succeed, just that visual design in web, as much as we like to disagree, maybe a tad bit overrated.

Print is hard, Web is harder.

While we can certainly learn a lot from print, and we should, there is a lot more to it on web.

I guess we all agree that form should follow and elevate function(interaction).

Web is just different media, where function is more important then visual design (Isn’t that with all thing that should be used). It’s possible to drive an ugly car, but you cannot drive a car without an engine. But having a beautiful car and usably designed car with an engine is what we all want I guess.

the prettiest and most visually interesting websites are often the most useless and frustrating – ie all the flash websites. flash websites are canonically the worst. the best websites are the ones where you get the info you want easiest and fastest.

breathtaking insight, i know.

The Vignelli map is a hideous monstrosity that manages to plagiarize Beck’s wonderful London map while stripping it of any elegance.

Excellent views my friend.

We can now actually segregate web design into various sub-fields/sub-domain

CSS, Markup, Usability, Accessibility, Conceptualisation, the list goes on…..

A web site can (should) have both, but it is far more critical for the site to work well then look good. It’s like a car, at the end of the day, you want one that gets you home. But, you are happier if it also looks good.

The Vignelli map is a hideous monstrosity that manages to plagiarize Beck’s wonderful London map while stripping it of any elegance.
and to add to your points, i think it’s safe to say most of the time, what pleases the designerati community doesn’t actually coincide with what the real users are looking for.

That certainly doesn’t mean thing has to be ugly to be able to succeed, just that visual design in web, as much as we like to disagree, maybe a tad bit overrated.

sohbet

ArkadaÅŸ

So, as a designer, do I worry that Google has a lousy logo? No…and I don’t think many web designers do. Most web designers know that the value of Google is in its utility, not its appearance. Can it still be canonical? Absolutely.
Absolutely, I dont worry too.
Oyunlar

“The most widely used userinterface of the 21st century is a single text box with a submit button.”
No grafics, no colors – amazing

thank you

Google’s new favicon as precursor to new interface update? Even if Google’s interface would change – at the end the current one works because the users get best search results by using the search engine with the “lousy” logo.

I agree “The web is not suffering from a lack of canonical design. It’s just that canonical design on the web isn’t as glamorous as some want it to be.”

its very good porject.

great project thanks

thanks fort the post.

great job.

Who cares if someone likes or dislikes the logo ? its personal taste. what you should be asking is , does it serve its purpose?

A logo’s design should identify the company to users who cannot read or speak the language – if any child who couldn’t read saw the Mc Donalds logo, they KNOW its McDonalds.

Same goes for google – You KNOW its google without having to read it. It works because the design identifies the brand, (like IBM, Meryll Lynch, Apple’s logos) without speaking or directly telling you – its a ‘visual landmark’, literally.

Thank you for your site. I have found here much useful information…

oyun oyna, kral oyun
I think the fixation on the Google page is a bit extreme.

its very good yeni oyunlar

Very interesting perspective! I think Rahul said it best.