Good evening.
I'd like to start off with a brief aside and say that it's already been a wonderful experience, spending the day with higher-education folks like yourselves. As a former university Webmaster who's since sold his soul to The Man (at least I got a good price!), it's been real pleasure to spend time talking with people who are more concerned about doing the right thing than the expedient thing. After a couple of years of talking to commercial Webmasters, most of whom are barely aware that there are browsers besides Internet Explorer or operating systems other than Windows, returning to my roots, even for a day, has really been a major recharge. So I thank you for giving me that boost.
Despite the Robert Frost allusion contained in the title, please rest assured that I've sworn myself to avoid forcing the text of my address to fit the poem. (As a historical note, I originally considered titling the address "Two Links Diverged on a Yahoo Page," but then quickly came to my senses.) My intent tonight is to talk about the process of pioneering and intellectual exploration, not some crusty New Englander trying to decide which woodland trail he likes better. In fact, I will go without quoting so much as a line of the poem, so everyone can relax their cliché shields. If I'm successful in my efforts, this address should be a simple reflection on my own Web journey and a few thoughts on where I think the Web is headed.
Having graduated from Case Western Reserve University in the spring of 1992 with a Bachelor of Arts degree in History, and minors in English, Artificial Intelligence, and Astronomy, I did what every other liberal-arts self-styled wannabe Renaissance Man of my generation did: I got a job working with computers. At the time the CWRU libraries were ripping out their dumb-terminal card catalog system and replacing them with 486 "smart" terminals. So I became a hardware jockey, with occasional bits of scripting and programming on the side. And I started work toward a Master's Degree in English, which seemed like a good idea at the time. (Looking back, sometimes it still does.)
It was somewhere around the middle of 1993 that a co-worker of mine, Jim Nauer, told me that there was this new thing called the "World Wide Web" and that it was an important new development in information technology. His opinion was that it should be explored by the library as soon as possible, and put to use. "It's like Gopher on mega-steroids," was how he put it. This was probably not the best way to describe it. I'd used Gopher and found it awful-- the text and images had to be accessed separately, it was hard to navigate through the menus and submenus, and it was especially hard on the eyes. It was actually a lot like the Cleveland Free-Net, only less interesting. The idea of that experience multiplied wasn't exactly doing it for me.
Then, in late 1993, Jim showed me Mosaic. I sat there for at least an hour, clicking and looking and being awed, until Jim finally kicked me off his computer so he could get some actual work done. I knew within moments of my first click that he'd been right all along, and that I should have paid more attention to him. And I knew, without question, that life would never be the same-- for me, or for anyone in the field. We could serve up library information, yes, of course; but more to the point, we could serve up anything and link it all together.
Of course, I'm sure most of us had that feeling. Remember it? That deep sense that you were encountering something truly important, something that would matter in ways you couldn't predict, but you nonetheless knew were there? I sometimes wonder if Gutenberg and his assistants had the same feeling. (There, I've fulfilled my Gutenberg reference requirement.) There was a road stretching in front of us, and we could follow it just to see where it went. The mere act of starting on that road was exciting enough, like headed to Disneyworld as a child, thrilled to be starting because somehow you knew, deep down, that when you got there it would be so much better than anything you could envision that you were seriously concerned about your sanity when all was said and done.
Obviously, that's no longer the road not taken-- just about everyone headed down it at some point, fortunes were made and lost, the shape of commerce changed, laws were written and challenged and made obsolete almost as fast as the technology itself. William Gibson once described a fictional city as being "like a deranged experiment in social Darwinism designed by a bored researcher who kept one thumb permanently on the fast-forward button." The Web was a lot like that, except it was as much an economic experiment as a social one. Whether or not it succeeded depends very much on where you stand.
But let's get back to that late-1993 moment in time. Based on that single exposure to the Web, despite its being so tiny and crude compared to what would come later-- like so many of us, I dived in without hesitation. Most of my co-workers felt the same way, so we set up a prototype Web server on an early Apple UNIX server and showed it to the head of the department. He told us not to waste any more time on such silliness. At that point, we really had only one choice: we went over his head and showed the same demo to the VP of information services. He gave us the go-ahead.
I had the most time on my hands, and I also had some experience with graphics editors, so I took charge of the content and design side of the project. This was early enough that we didn't even have tables with which to design-- all we had were the elements found in HTML 2.0. From that, I learned about document structure and its importance. Being a university Webmaster reinforced the lesson: what was most important was the content and its structure. The look was secondary-- important, but still secondary. We had to be able to support access via Lynx as well as Mosaic. Even in early 1994, cross-browser concerns were a part of my life.
I'm sure that for most of you, the same held true when you started. Even if you didn't have to worry about cross-browser compatibility at first, some irate faculty member or group of students would make your life really difficult until you made it a part of your job. That's where I think those of us in academics ended up on a far less-traveled but ultimately more enriching road, because on the commercial Web the big question was whether or not a Web site worked in the Big Two Browsers. Period. After that, it didn't matter. Mosaic? Who cares? Emacs-W3? What did text editing have to do with Web browsing? Lynx? Oh yeah, we have those-- they're the blue underlined things you click on, right?
Meanwhile, in higher education, we argued with the public-affairs offices, won a few, lost a few, and generally kept trying to serve up content that any browser could access. This actually put higher-education Web sites ahead of the curve, because it turned out that the jump from accessible markup to CSS-driven design is a lot smaller then trying to convert table-based markup to CSS and accessible markup.
I'll skip ahead to the spring of 1996, long after I'd abandoned the English degree to spend time being a Webmaster. Another co-worker, Peter Murray (of library automation fame), and I were presenting a paper at the Fifth International World Wide Web Conference. It was at that conference I saw one of the first public demonstrations of CSS, and it seized me. I knew this was what I'd been waiting to see, and as soon as I got home I tried out some CSS. It fell flat on its face, when it wasn't crashing the browser I had, IE3. This was before Navigator 4.0 came out, remember. So I set up a basic test suite to figure out what I could and couldn't do.
I quickly discovered that the road I'd chosen was choked with potholes, bugs the size of Rhode Island, and severe limitations on what I could do. It didn't matter. I'd been so compelled by what I'd seen that I pressed on anyway; somehow I could taste the possibilities and, by dint of working for a university, I had some freedom to explore what I thought was right and not have to worry too much about the Almighty Bottom Line. By necessity, I became an expert on what was and wasn't possible in the browsers of the day. In order to figure out what was and wasn't a bug, I had to teach myself the theory behind CSS. As I did so, I started publishing charts and articles to explain to other people, so they wouldn't have to suffer the same frustrations I did. The rest, some of my readers and students have claimed, is misery.
I have to say, however, that every time I use CSS, I remember why I fell in love with it in the first place, and what keeps me coming back. A couple of weeks back, I was trying to get two independently written stylesheets to play nicely together. One of them had been written with a fairly different document structure in mind, and at first there was a fair bit of layout havoc. All it took was a few judicious edits, some extra rules, and a few edits, and suddenly the page's layout was fine. And all I had to do was juggle the styles themselves for a bit. I didn't have to restructure a set of nested tables when the headings were indented too far, or not far enough. I just fiddled with a couple of numbers, and the problem was fixed. It took me less than half an hour. And if I wanted to completely rearrange the look of the page, it would take me no more than a couple of hours of changing the CSS. I wouldn't even have to touch the document's structure. The power is so subtle and overwhelming that it's like air: no matter how little I think about it on a daily basis, I couldn't function without it.
Anyway, back to 1996. At the time, there were very few people choosing the path of CSS and structure. The Web revolution was in full swing, and practically every design firm in existence was following the tables-and-GIF road that David Seigel had charted. It's not hard to guess why: it worked. Browsers would do what you asked, provided you knew how to ask. Legions upon legions of Webmasters schooled themselves in convoluted strategies in pursuit of the goal of every marketing department ever: "Make it look the same in all browsers." That this was effectively impossible didn't seem to deter anyone.
However, consider what it gave us: some truly innovative design approaches, not to mention some really sick and twisted markup. In spite of the limitations of the medium, designers and artists bent it to their will and their whims. Actually, I feel that these tricks were so interesting because of the limitations HTML imposed; I've often thought that art can be maximally effective only when it does have limitations, both real and perceived. The glass sculptures of Dale Chihuly are so striking because they seem impossible; their very nature, by challenging what we think glasswork can be made to do, makes them more beautiful. The ability of some artists to paint in such a way that the subject seems to glow with an inner light impresses us far more deeply than a flat rendering of the exact same image. We appreciate the beauty of the scene, but we also wonder at the technical achievement, and that wonder deepens the mystery and appeal. I've long believed that in an unrestricted medium, the dominant experience will be tedium.
So where is all this headed? I feel that the marriage of structured markup, CSS, and DOM scripting-- what used to be called "DHTML," but seems to be undergoing a name change-- will enable a real improvement in interface design and content accessibility. There will be room for artists to really stretch their wings and create truly beautiful designs that leave behind the boxes and grids that tables imposed. The data jockeys will get to harvest the contents of those artistic sites and make it usefully available in other media-- search engines, databases, syndication channels, and so on. It won't be automatic, by any means, and it won't always be easy. But the possibility is there, and the barriers can be low enough that people without degrees in computer science or graphic design can make their own forays into both sides of the equation.
As an example, I recently spent some time looking at a dynamic-programming example of a card game, where each card was nothing more complicated than a DIV element with some minimal content. Using CSS, each DIV was turned into a reasonable facsimile of a playing card. The card states (face up/down) were changed dynamically, based on user input, as were the positions of the cards. Of course, it's a cool demo, but what use is the content? Imagine that instead of playing cards, each DIV contained a set of search results. The user would be able to shuffle each "card" in and out of a stack, thus being able to see the full details of the search result-- things like match relevance, the context of the matched terms, the URL of the matched document, and so on-- while the stack shows the title of each matched document.
It isn't a necessary way to present such information, of course, but wouldn't it be cool? Given such a setup, it would even be possible to position the "cards" on the screen in such a way that they physically mirror their places on the Web site, as sort of a visual roadmap. That way, a user can immediately see if a number of hits are clustered in a certain area of the Web site, thus letting him jump straight to that area. Even better, such an interface would still contain all of the information of the original search. This would be displayed in a traditional list fashion by older browsers.
And that's part of what makes the current changes so exciting. Thanks to the strides browsers have taken, new advances in visual design can come hand in hand with an improvement in content accessibility, instead of a loss. Styling raw HTML (or even XML) with CSS means that the content is still highly accessible to those who can't see, for example, or to those who browse with a text-based client like Lynx. Or even for those who are still using older browsers that don't understand the latest standards. Rather than getting nothing, these users will get the content. It won't be as pretty, but it will be accessible to them.
Have modern browsers given us the keys to the world, and removed all design barriers? Of course not! As technology evolves, the boundaries of what we can do are pushed back; HTML tables and sliced-up images give way to CSS-driven designs that are far more flexible and capable than anything that came before. And yet, there are still limits, and in pushing at them, we find art. In a way, the current era of CSS-based design reminds me of those first two years of the popular Web, when everything was new and we constantly redefined our concept of the impossible.
When I look at the state of Web design today, I sense a vast region that has yet to be explored by anyone. I can feel new design frontiers opening up all around us, and almost make out the road that stretches ahead of us. I want to get moving down that road as soon as possible, because I want to see where it goes.
A year ago, anyone working in Web design would have confidently told you it was impossible to get text to flow along an irregular outline, or indeed to create any sort of non-rectangular text flow, or to create subtle translucency effects, or to get elements to appear and disappear without the use of Javascript. We now know all these are possible, and I get the very definite sense that there's much, much more we don't yet know. This is what's so compelling about the Web: after almost nine years of experimentation and evolution, it still has the capacity to surprise, to delight, to astonish.
For myself, I don't expect corporate Web sites to push back the frontiers of design. Corporations are, of course, more worried about not turning away customers, about catering to every browser known to their marketing departments. They stick to tables and spacer GIFs because those are known to work in older browsers. If there's going to be a real exploration of advanced design techniques, and innovation in how those techniques are used, it's going to be on the site of independent publishers, and of higher education.
Of course most data has to be accessible, but there is opportunity to experiment with new ways to deliver that data without bankrupting someone. With the focus on luring new students every year-- the exact demographic that's likely to be using the most recent Web browsers-- university Web sites are a natural ground for interesting new interfaces to arise. It's not only good PR, but it's a fascinating intellectual challenge: exactly the sort of thing universities like to foster. It's one of the things I really do miss about working for a university.
In 1977 Cecelia Payne-Gaposchkin received the prestigious Henry Norris Russell Prize from the American Astronomical Society, in part for her groundbreaking 1925 dissertation on stellar absorption lines and the composition of stars. This dissertation was not accepted at the time it was published, but later turned out to be exactly on target and was hailed as a pinnacle of 20th century astronomy. In her acceptance speech, she said (in part):
"The reward of the young scientist is the emotional thrill of being the first person in the history of the world to see something or to understand something. Nothing can compare with that experience..."
That thrill is within reach, even on today's web, for all of us; I have felt it myself, when I first envisioned what's coming to be called the "complex spiral" effect. Anyone can experience the same thrill, by finding a previously unknown way to use the technologies we have already available to achieve a startling effect. The key, of course, is to look where others aren't-- finding that untaken road and then finding out for yourself where it leads. Once you have seen something truly new, report back! Let us know what you found, and how to get there. It's the surest way to a web that's more useful, more interesting, and more compelling for everyone.
Thank you for your attention, and more importantly, for everything you're going to do in the future. I can hardly wait to see it.