Thursday 31 March 2005
Håkon Wium Lie is the CTO of Opera Software and in 1994 proposed the idea of CSS. Håkon is as deeply involved with the Web and with CSS as anyone can possibly be. Recently, he contacted WaSP to ask whether we could host the Acid2 test, which we agreed to do. Our role is to help build, publish, and promote the test for all browsers for CSS 2.1 compliance. Acid2 will be a free and public resource for any browser or user agent developer and any web developer as we all move toward improving CSS support and fixing existing bugs in our work.
In this interview, I ask Håkon to share some of his experiences and insights into the history of HTML and CSS, the challenges and triumphs at Opera, mobile devices and microformats, and the reasons standards mean so much to the current and future Web.
MH: Håkon, many people know you as CTO of Opera Software. But the critical work you’ve done goes back to the early days, and many people working with Web standards don’t realize that you are essentially the “father” of CSS. Can you talk a little bit about those early experiences, and how they’ve shaped your thinking both for today and for the future?
HL: I stumbled across the Web in 1992. I joined the
www-talk mailing list in September that year while working for Norwegian Telecom Research. This was before Mosaic and when HTML was a text-only language. Still, a critical mass of talented people saw the beauty of the underlying system. When Marc released Mosaic in early 1993, pictures entered the web and it suddenly became much easier to demo. Even managers could see the potential!
However, pictures were also a threat to the web. Designers started to encode text in images in order to achieve certain fonts or other special effects. In order for HTML to remain a logical markup language (as opposed to a presentational language) a style sheet language was needed. So, the motivation for developing style sheets was twofold: we wanted to give authors the presentational effects they craved, while stopping HTML from sliding down the ladder of abstraction to become a presentational language.
One important observation I made along the way was that both image and text were important for the web. Images are more appealing aesthetically, and most people will prefer a visually rich presentation to a sparse text-based one. Text, on the other hand, can be processed in a meaningful way by computers. Computers can search text and analyze its content. Google and friends have shown us wonderful things to do with text and I think there is much mileage left.
“Most people will prefer a visually rich presentation to a sparse text-based one. Text, on the other hand, can be processed in a meaningful way by computers.”
MH: The Opera browser has not been without its struggles, despite best efforts to keep it standards-aware, low-cost with a free ad-supported version, and very lightweight. Can you point to the features in Opera that you think are particularly strong and those that have remained problematic?
HL: My favorite feature is OperaShow, which instantly turns Opera into a PowerPoint-like presentation engine. It’s fully based on standards, and it extends the reach of the web from a scrollable canvas to a paged presentation. Also, I’m very proud of the work we have done to display web pages on small screens. So Opera covers the whole range, from big to small screens.
“Along the way we have struggled much with “Dynamic HTML”. There were no standards to guide efforts in the beginning and we had to reverse-engineer many pages. Then the DOM came along and things started to improve.”
MH: Many people are unaware of Opera’s reach. One example is the rendering engine in Macromedia Contribute. While many observers feel Opera has failed as a browser, the point could easily be made that Opera simply has reached a different audience, one that is in essence hidden from the general user but very obvious when one looks at your strategic partners. Could you comment on Opera’s involvement as integrated software within other applications?
HL: We think of Opera as a success on the desktop, after all we’re the best-selling browser there. And, we have about 20 million users, I believe. Recent innovations in Opera8 on the desktop are support for voice input/output and native support for SVG.
But, I didn’t answer your question. Indeed, our rendering engine is used in several authoring systems. It’s a great way of making sure pages are tested in Opera right from the beginning. A big announcement in coming up in April, and I look forward to saying more about it at that time.
MH: Taking the previous question to another level, one of the things that has long interested standards-based developers such as myself is the goal of extending our reach to alternative devices: PDAs, cell phones and so forth. Opera appears to be doing amazing things in this area. Do you think this is the “next big thing” for Opera and for those of us in development at large?
HL: I think the mobile area is very important. It’s important for the web to escape Microsoft’s grip on the desktop. And, it’s important for the mobile telecommunications industry to embrace the web to ensure there is interesting content on the wonderful devices they are making.
For Opera it’s an enormous business opportunity. Already, there are more Opera browsers on mobile phones than there are Microsoft browsers.
MH: As a pioneer of the Web, what’s your opinion as to how it should progress? Set aside any preconceived ideas such as ‘semantic web’ or ‘convergence.’ What I would like readers to know about is the vision you personally hold for the Web as it will become rather than as it is today.
“I plan to spend the rest of my life on the web.”
being developed, and that’s how the semantic web will be built, I believe.
Finally, I hope that future web formats can represent more than flat documents. I’d like for 3D models of all sorts of stuff to appear, starting with spare parts that I need. Along with 3D printers, this will enable localized production of stuff which is much healthier for the environment than moving stuff around. The web globalized information, and I hope it will also localize production.
MH: Thanks so much, Håkon.
Note: This interview is cross posted at WaSP. Thanks! -mh