Jeffrey Zeldman: Designer, Author, Visionary, and Speaker
Jeffrey Zeldman is an outspoken web designer, author, and speaker. His book, Taking Your Talent to the Web has received high accolades. Zeldman is the publisher and creative director of A List Apart, co-founder and current group leader of The Web Standards Project, a grassroots coalition fighting for standards on the web; and founder of Happy Cog, the New York City web design firm.
Tell us about your early career that obviously has some artistic make up that is often reflected in your Web designs and writings.
I was novelist who couldn’t sell his novels; a composer who barely made a living; a synthesizer player in a post-punk techno-surf band with a small but ardent following and a manager who was a member of the Communist party. I ended up in advertising, where I was able to earn a living while focusing my creativity and learning about communication, brand identity, and the gentle art of drinking before noon.
You got two Bloggies in 2001. Why do you think Blogs are so popular? Why are people interested in reading about other ordinary people?
With a few notable exceptions, awards are meaningless. I learned that in advertising, where there are a few genuine awards that everyone wishes they could earn (Communication Arts, One Club), but where anyone can win plenty of other awards. You get a regional Addy for “best use of spot color varnish in a non-standard billboard on Rural Route 7,” and you state on your resume that you won an Addy. Like I said, there are some truly meaningful awards. I didn’t win many. Nobody does. I won the other stuff.
The Bloggies are nice, though, because it’s an award from one’s peers. That has meaning to me.
I’m not sure why anyone is interested in other people’s lives, other than the fact that as human beings we are always interested in what our neighbors are doing.
We are also interested in what Tom and Nicole are doing, in what Meg Ryan is doing, and it’s kind of nice that we can transfer that emotion to our peers. Never mind Tom and Nicole, what about Meg and Jason? It’s odd but well intended, I think.
I think we all grew up wanting to be rock stars and movie stars and the Web provides the illusion that we are living this fantasy. (See Joshua Davis dot com.)
I don’t write about my life very much and I don’t think I would have won an award if I did. Mainly I write about web design. Of course, to me, web design is incredibly personal. I confine my confessional writing to a section of my site called “My Glamorous Life.” It’s an ironic title but some people may miss that.
You spoke on the topic of “What Do Today’s Web Designers Need To Know” at Web Design 2001 in Atlanta. For us unlucky folks who didn’t get to attend, what do we need to know?
Two things, really. The first is that web design is communication. Good websites are conversations. I can talk about that for a long time but not here, not now.
The second thing all web designers should know is that the way we build sites must change and will. The way we’ve been building them – to the quirks of individual browsers instead of to a common set of standards – is wrong for a lot of reasons. Suffice to say, the Web is broken, and all of us helped to break it. All of us can fix that by learning about and using W3C recommendations, but at a cost of decreased attractiveness and limited functionality in some old browsers that don’t support these recommendations. For more about that, read “Why Don’t You Code for Netscape?” at A List Apart. Pardon the plug.
The dot com bomb dust has finally settled. What do you see for the Web’s future regarding online businesses?
The Web has not been hurt; businesses have been hurt. In some ways, the Web has been helped, since it is now less likely to be cluttered by a lot of useless, ill-conceived junk. Investors won’t drop their life savings on useless junk; agencies won’t build sites that promote it. With less ugly glut to sift through, Web users may actually find sites that provide real value.
But many really good companies got hurt, many really good ideas have died or gone on the respirator, many really talented people are waiting tables instead of designing Web interfaces, and that is tragic. When it started, some said nervously, “Good, that will get rid of all the idiots and poseurs.” Instead, we saw some of our most talented and visionary people hit the wall. It is terribly sad, and a bit frightening. Two of my favorite clients – really smart folks – are bleeding right now. I not only feel for them, I worry for me, too.
Will we get through this? Sure, most of us will. How can we survive at this time? I asked a group of Web designers, developers, and content folks, and they shared their answers in ALA Issue 97. Pardon the plug.
What do you think Web consultants should do now that companies can’t pay big bucks for fancy schmancy Web sites?
What they should always have done: help their clients analyze what they have to offer, decide what is most valuable about it, and create accessible, attractive sites that communicate real things to real people.
In tandem with that, stop importing bogus business values to a medium and an industry that were never about those values. As a client, I don’t care about your suit, I don’t care about your 24-foot ceilings and reconditioned loft space and networked video games; I care about the work. Do you understand my business? Can you communicate it in a Web-specific way? Can you entertain, teach, extend my reach and my brand? Can you do all that without expecting a Hollywood budget? Then I want you to work for me. If you can afford the 24-foot ceilings, good for you; if not, I don’t care.
You’ve spent a lot of time and energy as a warrior for Web standards through Web Standards Project (WaSP) that you lead. Since much of the W3.org’s specs are written in a foreign language to most of us, what are some key things we should know? (Note: To learn about reading specs, check out A List Apart’s article on How to Read W3C Specs
I’m having them translated into English now, and will let you know when that’s done.
I *think* CSS-3 is about increased power and flexibility – for instance, making it trivial to do with Style Sheets what we now do easily with HTML table-based layouts, but have a tough time pulling off in CSS.
But I’m in the trenches with every other Web designer; what we really need now is solid support for CSS-1. With a few exceptions, we have that in the latest versions of IE, Navigator, and Opera. (The Mac browser iCab is getting there too.) But previous versions of some of these browsers are so badly broken in their CSS implementations that using even the most basic parts of CSS-1 can cause some of these browsers to crash.
This has led many developers to shelve CSS and continue building broken sites with non-standard, non-validating HTML extensions and workarounds in the name of backward compatibility. The absurdity is that browser makers have finally given us what we asked for three years ago, and now many of us are afraid to use it. It’s a cycle that can only be broken by taking bold steps. I’ve been selling my clients on the notion of “forward compatibility” via standards compliance, and I’ve been lucky enough to find clients like The New York Public Library who want what I’m selling.
Forward compatibility isn’t essential for every site – if you’re designing a movie website, with a shelf life of twelve months, it hardly matters if you use HTML or XHTML … but for most sites my partners and I get asked to design, long-term durability does matter. If long-term durability matters to your site, then you need to learn about web standards, or hire designers and developers who know about them. The New York Public Library’s Standards-Compliant Style Guide can help anybody get up to speed with a basic, standards-compliant interim strategy that looks forward without leaving anyone behind.
iPlanet has sent shockwaves by rolling out Netscape 4.79 to please businesses that won’t get with the program. Armed with this info, what should Web designers do when designing for behind the times browsers?
WaSP actually asked developers to learn about and use W3C recommendations, and to inform their audience that better browsers are readily available and may be downloaded for free. How developers choose to implement these suggestions – if they choose to implement them at all – depends on the nature of each site and its audience. We expect that developers will analyze their audience and develop their own strategies and language.
Why do we encourage users (not developers) to upgrade their browsers? For the same reason Firestone encourages drivers to trade in unsafe tires. Previous browsers were not built to support CSS-1, HTML/XHTML 4, ECMAScript, XML, and the DOM; the big two browsers, in particular, were built to compete with each other by any means necessary. Their focus was on proprietary technologies, and the market (web users, developers) didn’t tell them that standards were important. So support for standards in these browsers was often partial, often spotty, sometimes worse than no support at all. (It’s better not to support CSS-1 than to support it so badly that websites crash.)
Continuing to rely on broken browsers is not a healthy long-term strategy for Web users. Ultimately all Web developers will build standards-compliant sites, and these users will be hurt. We’re encouraging them to upgrade now, while most sites are still patched together with browser-specific code and non-valid markup.
Where can developers learn about W3C recommendations without getting a PhD? Web Review has numerous articles working designers and developers can follow. A List Apart has a DOM series, a CSS series, and other articles anyone in the field can understand and use. We even try to make it entertaining, since learning new things can be daunting. (Pardon the plug.)
There’s good stuff at Web Reference, Webmonkey and Builder. Peter-Paul Koch runs a good DOM mailing list. Apple has good stuff. And as I mentioned earlier, the NYPL Style Guide will certainly help. (Pardon the plug.)
What do you think advertisers should be doing to get our attention in an already overcrowded and overzealous Web world?
Sponsoring instead of running banners. Sponsoring design sites, community sites and forums, independent sites. I saw two of the Cluetrain guys discuss this idea at Geek Pride 2000.
Get rid of the rhetoric, which businesses often find disturbing, and you’re left with a core idea that any smart marketer will probably respond to: people use the Web to have conversations with each other. An advertiser who facilitates those conversations will be much better-received than one who interrupts the conversations.
Advertisers can facilitate conversations by providing bandwidth: this forum about child safety is brought to you by Toys-R-Us. That’s it; no sales message, no banners, no tie-ins (“Thanks for discussing your child’s well-being! Save 5% when you order now!”). None of that junk that sophisticated consumers reject in old media and resent in new media. If I’ve learned something valuable about my child’s well-being, I’ll feel kindly disposed toward Toys-R-Us for sponsoring the message board.
Similarly, this design site brought to you by Macromedia (no ad message), this XML discussion site brought to you by O’Reilly Publishing (no ad message).
Can it work? Apple and Macromedia have both done something like that with the design site K10k. It remains to be seen if other advertisers will have the guts to try this on more conventional sites. They may as well: what they’re doing now isn’t working. And what they tried a couple of years ago – big bandwidth interstitials – didn’t seem to work very well either.
You admitted in a past interview you don’t know what “great Web design means.” Then, what do you think are the most important things in a good Web site (other than Web Standards compliance)?
Sometimes great graphic design isn’t great Web design: for instance when it is inappropriate to the audience, message, or technology best suited to the audience’s needs. Sometimes what looks best *is* best. It’s great when that happens. Sometimes the best design is one that is simple, elegant, understated, low-bandwidth, and degrades well. But this kind of design never wins awards, and designers have to be willing to live with that if they wish to do the right thing on a site like that.
To me, overall, a great website is one where I feel I am communing with a mind – as I feel when I listen to music, read a novel, watch a movie. Music and movies are collaborative, as are most websites, but one still feels a one-to-one communication with a creative mind – a dialogue between viewer and author – and websites that can achieve the same thing have a greatness to them. It is easier to achieve this on personal sites than on commercial sites, especially commercial sites that have to perform 1,000 functions for 100 different types of user.
On a site that must be many things for many kinds of people, design greatness lies in the architecture. I don’t care what my neighbor hopes to achieve on the site; can I achieve what I want to? Can I do it without thinking too hard? Do I have to search? Will my first search fail? Will I be asked to learn what “Boolean” means because the site architects didn’t foresee my needs?
Complex sites should only rarely rely on back-up functionality like Search forms. No matter how many users they are intended to satisfy, they should always feel like they were designed to allow me to do what I need to. This is a very tough thing to pull off, but it often makes the difference between success and failure.
Then again, I still can’t figure out Ebay and my friend, who doesn’t understand the Macintosh file management system, can buzz through Ebay like Sherman marched through Georgia. I don’t think Ebay is intuitively structured or well-designed on any level, but it’s one of the most successful sites ever dropped onto the Web, so why are you even talking to me?
And a bad Web site?
A bad site is hard to use. A bad site is easy to use but does not communicate. A bad site is easy to use and communicates, but it does not communicate the right message to its intended audience. A bad site is easy to use and communicates the right message, but the site serves no actual human need and should not have been created at all.
A bad site commands, a good site invites. A bad site throws up barriers, a good site opens like the petals of a flower. A bad site displeases no one and therefore greatly pleases no one; a good site, like a good book, like good music, greatly pleases some while turning off others.
A bad site tells me what I already know and does not tell me what I came to find out.
A bad site repurposes non-Web content; a good site offers design and content created for an online audience. A bad site shovels text down a loveless hole of ugliness; a good site attempts to make text easy to read and attractive. A bad site does not guide or guides but provides no options. A good site guides and offers alternatives. A bad site works only one way when it works at all. A good site works exactly as its designers intended, yet I feel it works the way I want it to.
What approach do you recommend for dealing with site maintenance?
Here is what you do. As much as possible, you separate style from content. Either with Web standards, or with an ugly stupid overpriced publishing system, or with a smart publishing system, of which there are too few.
Then you train your client (if the client will be updating the site). Then you find out everything you can about the software the client will use to update the site. Is he going to take your lovely standards-compliant template and start banging out FrontPage extensions without even realizing it? Find out. Recommend. Teach.
Before beginning the site, explain what you are going to do, educate the client about the technology involved, keep educating the client throughout the process, do it again when you finish, do it again two months later. Before handing off the site, create a Style Guide that doesn’t require an MFA in Design to understand. Take the client through the Style Guide. Do it repeatedly.
After all that, the client will still mess up the site.
What are some examples of what you consider a smart publishing system?
Manilla, Blogger, Greymatter and Movable Type all seem to do a very good job at publishing specific kinds of sites. I think with R&D money and time, these tools could be expanded to do more, and to work with more complex types of sites. I haven’t worked with these tools but thousands have and I’ve seen the results. I’ve also seen a demo of some of the more sophisticated things Manilla can do. I suspect the others have tricks up their sleeves as well.
I’ve heard good things about Zope, and my girlfriend swears by it (not at it), as do some sysadmin folks I respect.
I’ve had bad experiences with expensive publishing systems used by some of my clients. These systems seemed overly complex yet terribly limited. My experience with them has been echoed by others who’ve come in contact with them – designers, builders, and clients. My sense is that these products are way too expensive and was way over hyped when they launched.
Colleagues and I have done some limited work with more traditional backend technologies like Cold Fusion and PHP. These tools seem very powerful, but they’re not something you hand to a client, they’re something you build with, and maintaining structures with these tools requires a certain amount of knowledge. I have seen clients totally take to these kinds of technologies (say, to update a bulletin board system). One problem I’ve had recently with Cold Fusion is that it really seems to want me to do everything in HTML tables and FONT tags. This could be due to my limited understanding or it could be that the software is optimized for the “traditional” way we build sites.
Any advice on how Web developers avoid link rot (outdated links)?
Don’t destroy URLs. Never, never. The only acceptable excuse for link rot is going out of business. It kills me that professionally-staffed, million dollar content sites routinely change their entire file structure, or delete articles that are a month or two old. News sites do it, software sites do it. Microsoft.com seems to have a fetish about changing their URLs every five minutes. Twenty million people have linked to this page? Good, let’s remove it. Let’s rename it. Let’s get them to perform a Boolean search. I hate this. It’s stupid. It’s beyond stupid. It’s one thing when your grandmother deletes a URL on her homepage. It’s another when a Fortune 500 company does it. I am down with Jakob on this one. I often disagree with Jakob, but on this one we are brothers.
Tell us about your book, Taking Your Talent to the Web and what you’ve learned from the reactions to the book. (Editor’s note: see our book review)
Thanks for asking. Taking Your Talent to the Web is a guide for the designer or communications professional seeking to become a Web designer, or to add Web design to a repertoire of existing skills. For instance, a book designer whose client has asked her to create a small site. She’s not going to stop being a book designer, but she wants to increase her professional range. Or an art director who no longer wishes to do print advertising, who finds the Web intriguing, and wants to become a full-time Web designer. Or a greatly experienced creative director who hasn’t a clue about the Web.
It’s based on the Populi curriculum in Web Communication Design, which I co-authored with folks from Populi and Pratt Institute of Design. Populi is a business that helps designers and programmers migrate their skills to the Web. This book is based on the Populi curriculum but it’s my book, filtered through me, for better or worse.
It’s a how-to book but it’s also a what-and-why-and-where-is-it-all-going kind of book. It talks about how this medium differs from the one they know, and how it is the same. How Web agencies are structured, and where the designer fits in. It teaches what you need to know to start working at a Web agency now (if that’s what you want to do). But it also explains how the Web is changing, and what you will need to know in the near future.
It’s a unique book that addresses the needs of a market that should have been addressed long ago, but for some reason, never was.
Now, of course, everybody is trying to take their talent OUT of the Web instead of into it, but that will change as the market adjusts.
If you go to a bookstore and check the Web titles, you don’t see books like this. You see lots of books on XML and XSLT and PHP and other stuff that’s wonderful but scary to a designer who’s new to the online world. You see lots of books on using Flash 5 and Dreamweaver and Photoshop but few books like this. It’s a needed book, I think.
I get mail every day from people who’ve read it and liked it, and that is really wonderful to me. Because of the bad economy, it seems that most readers – at least most readers who write to me – are not the initially intended audience of design professionals with limited web experience, but rather my peers (other web designers).
I find that moving and inspiring.