Liveblogging a presentation at @media2007 London by Håkon Wium Lie

Lie took the stage at 2007.06.08 12:16.

Before I came to Opera, I worked for W3C. But mostly I’m going to talk about W3C-related stuff for a technical perspective and how Opera is working to support those things.

One Web: That’s really my conclusion. We should all work towards one Web, blah blah blah. Today I had to start with the conclusion, but I’m sure you agree that one Web is the way to go. You’re old enough to have survived the WAP thing. We want to make the Web available on all these devices, and that also requires work from specifications, because as more and more devices and screen sizes get to he Web, we need to modify the Web slightly as well. We need to evolve the Web as we get more videoclips on, more image types, more things that creative types like yourselves are doing. Evolve slowly, but evolve still.

Some Web devices we’ve been working on: Started in 1995 with a product for the desktop, Windows-only. Mosaic was released for X Windows. Wasn’t at Opera then. I didn’t believe anyone could compete with Mosaic; I had to eat my words there when I joined the company a few years later. Around the turn of the century, mobile devices started appearing. Opera had been a small company. We didn’t have the resources s to write more code; we always had to write lean code, and we were small enough to fit on those devices more or less by coincidence. But they didn’t have the processors to run huge code, so that was a good fit.

We also have Opera for devices, including the Wii. They’re starting to come out in the store now – you can sometimes see them on the shelf for about 10 minutes. It’s a different Web experience when you sit in your living room and you surf as a family experience. It’s really changing the way one surfs the Web, and it may change the Web as we move along.

David Storey is here. He’s Opera’s Open the Web guy; he’s working to make Web sites work well with Opera. Sometimes when we’ve called people, they say sorry, we don’t have time to fix it right now because we’re fixing it for the Wii. But it’s the very same codebase; whenever something works for the Wii, it works for everything (in Opera) as well. We only have one codebase.

The $100 laptop: It’s a wonderful machine. It’s made for children in developing countries. You think less-less-less: You want to make it as simple as possible to use. It still has the hardware to make it compelling – 200dpi screen, camera, mic, speakers, USB. It has a keyboard that isn’t so nice to type on if you’re an adult.

We are running Opera here. Opera is almost perfect for this machine, with relatively little processing power, memory, battery. It’s quite similar to the Wii. Neither has a hard drive, they have roughly the same processing power, and they have wireless connection. Opera isn’t on here by default. They rely on open-source software from the foundation, but they realize that other people will supply software for it. We’re talking to some countries now. That way you don’t have to convince users to download and install Opera.

mini.opera.com: A different kind of product, really. Mini is slightly different in codebase. We use the Opera formatter to fetch Web pages on the server side, then Opera Mini is a 100K Java program that receives the binary protocol from the proxy server. By compressing it, we save time for the download and money for the (subscribers). Opera Mini has actually enabled Web access for a lot of people. Almost all phones these days can run Java, and by downloading Mini, you can turn your phone into a surfing machine.

Specifications: When I was at W3C, I worked on HTML and CSS. Over the last year, the groups that started developing HTML outside W3C now have found their way back to work with W3C. All the right intentions were there in converting to XHTML, but it didn’t really catch on with the Web-design or -authoring community. Most people, including me, continue to use HTML 4. Some of it was POSH, but some of it was pretty awful, too.

The WHAT WG wanted to make sure that HTML 4 had a successor. We have a wonderful organization within the W3C and we want to continue working for that specification. But anyone can join the working group.

The video element: There’s a lot of video on the Web today. We all have video cameras in our pockets. Still, support for video is quite rudimentary. We let binary plugins do the encoding, mostly Flash – not available on all platforms, the quality is debatable, the code you have to insert on your pages is quite obscure. It’s time to make video a first-class citizen, on a basis equal to text. We should be able to style it, make it part of the structured documents we are authoring.

There are two problems to solve. The first one is quite easy. The video element feels pretty good. People understand immediately what it means. The hard part is to find a suitable baseline video format. There’s a lot of patents in the field as well, which, as we’ve seen in other context on the Web, can be very harmful. (Mentions Ogg Theora.) We made an experimental build of what we can do if we insert a video decoder into the browser itself, and added a simple API for JavaScript to control the video as well. (Runs demo with three panes all running video.) We can clean up the markup and make use of an open standard that everyone can use freely. I think it would be impossible to use video on a $100 laptop if we use formats that cost money.

CSS: I proposed CSS in 1994 when I came to CERN. I shared offices with Tim there and I got machine number four. What I saw missing was a way for authors to express how pages should look, not only what the structure and content should be. (Shows text-only original CSS proposal, then Zen Garden unstyled and styled.) The goal of visual diversity was achieved, plus our goal to save HTML from becoming a presentational language. Fewer and fewer people are using it that way now.

CSS3 has been split into many modules. I’m sure not every one of those will be implemented or even finished, but some are important, like media queries. (Shows his homepage, with border around it. Then handheld mode with no borders and less whitespace. Below 300px wide, the background goes black, as expressed in CSS.) The syntax is a little more cryptic than I would like, but it’s backward-compatible.

Multicolumn layout. (Runs demo in Firefox, which scales from one to two to three columns as width changes. Scales image, too.) Could be used for professional printing. People want to start with the Web format rather than starting with InDesign or whatever. (Shows paged media and generated content for paged media: Crop marks, crosshairs, columns, footnotes, table of contents with tab leaders, headers and footers.)

(Shows Cascading Style Sheets book generated entirely by HTML and CSS.) Used Prince XML, on which I am a member of the board, so I am promoting something that I have an interest in. I joined the board, though, to make sure that my favourite bugs were fixed first.

They used a book microformat, “boom,” with about 20 classnames (chapter, preface, foreword, title page, table of contents, appendix). Semantics borrowed from DocBook.

Web fonts are hopefully coming in the next 12 months. I have been predicting that for the last 10 years. I see so many good fonts out there because we cannot use on the Web. You designers sill use the fonts; you just put them in background images. Translating a site using images for text is a big job, and they don’t print easily. We’re using typically Microsoft core fonts. I think it’s one of the greatest projects that Microsoft has started, and they gave us 10 high-quality fonts, hinting and all. It was very generous. But after ten years, perhaps some of us are a little tired of seeing the same fonts all over again. We need visual variety while still sending text as text.

Web fonts was introduced in CSS in 1998 and was also implemented by Microsoft and Netscape. But they picked different font formats, EOT and TrueDoc, neither of which is important anymore. With no support for TrueType, either; they should have just supported that. No browser has implemented the current Web fonts yet. Hopefully, though, by next year we can use this. I don’t have anything to announce yet. But Prince implements this. (Shows PDF.)

Acid2: A test page when Microsoft proposed doing an IE7. They’ve been promising standards for so long now, we should help them get there, especially features for standards that nobody has been able to use because there hasn’t been support in IE.

In the ’90s, we had an Acid1 that tested the CSS box model. All browsers have bugs. If you get Acid2 right, you will immediately see the smiley face on the screen. We had to make sure that Acid2 wasn’t specifically targeted at Microsoft, because it wouldn’t be fair and it wouldn’t have been accepted by the community We worked for over a year to get it right. We have Safari, Opera, and Firefox (3) doing Acid2 well. Only Internet Explorer doesn’t. It’s sad that Microsoft hasn’t taken this challenge and gotten this right; but in fairness to their developers, they have had a lot of bugs to fix.

(Scribe ended at 2007.06.08 13:00.)

The foregoing posting appeared on Joe Clark’s personal Weblog on 2007.06.12 17:29. This presentation was designed for printing and omits components that make sense only onscreen. (If you are seeing this on a screen, then the page stylesheet was not loaded or not loaded properly.) The permanent link is:

Values you enter are stored and may be published.


Search for very early blog entries, and for anything else on fawny.org:



Other reading

Popular topics

Photographs to look atTypography; graphic design; the death of design criticismTTCCanadian EnglishAccessibility

Archives by date

Just add /year/month/day/ to the end of site’s URL, blog.fawny.org. You can add just /year/month/, or just /year/, if you wish. Years are four-digit, month and day two-digit (with padding zero below 10). For example:

Very old archives are still available.

Archives by category

Copyright © 2004–2018

You enjoy fawny.blog