Sep 7th, 2010 |
Kaitlin PikeOpera CTO Håkon Wium Lie on the Future of the Web (Spoiler: It’s Not Dead)
Having worked everywhere from CERN with Tim Berners-Lee and Robert Cailliau to W3C (where he invented CSS) to Opera Software, Håkon Wium Lie has thought a lot about the Web. And at his upcoming sessions for Web 2.0 Expo New York, he’ll share some of these thoughts on CSS3 and the future of Web styling, what to expect from browsers in the next five years, and 10 things you likely never knew about Opera.
We recently interviewed Håkon about his upcoming appearances at Web 2.0 Expo New York, as well as what he thought about the controversy Wired drummed up in its recent article, which claimed the World Wide Web and browsers were in decline.
“I think Wired is wrong when they say the web is dead,” Håkon said. “The web as we know it is going to be around for 50, 100, maybe 500 years. It’s a revolution just only comparable to Gutenberg’s printing press. Browsers may change, but many of the underlying concepts and technologies will be with us for many years. My prediction is that a typical computer 500 years into the future will be able to read web pages written today.”
Håkon said applications as we know them will become “a footnote in history.”
“I think all application development in the future is going to be using the web as a platform… practically speaking all of it. What people get will be based on web standards – HTML, CSS, JavaScript. It really doesn’t make sense to do the kind of binary-for-one-device-only application development that we’ve seen because you want to reach everyone. You don’t want to just reach the ones who can afford the fanciest phone. Of course you want to reach those as well, but those devices can use exactly the same languages that the web itself is based on.”
“Maybe it won’t look like a browser or feel like a browser, but it’s going to be a browser underneath because it’s based on the same standards that browsers support. And there is a browser running in the background, even though it looks like an app.”
Håkon said the platform the web provides for application development is getting richer. For instance, we’ll have more styling in CSS3, more tags in HTML5 (like the video tag), and we’re already seeing geolocation rise in popularity. On a social level, the Web reaches more people today than ever before. Opera Mini and Opera Mobile, for example, give anyone in the world – from Malaysia and Indonesia to Turkey and Egypt – who owns a phone access to the online world.
“I think it’s important to think of the future [of the Web] not just 12 months to come, but really think this is the place where we plan to spend our lives, and we better make it a nice place to be,” he said. “We need to make sure we get these standards right and not make any short sighted compromises. In due course we can clean up some of the mistakes of the first decade of the Web. But I think we have to think of this not just decades either but centuries.”
The Future of Video in Browsers
A long time Web standards activist, Håkon said despite MPEG LA’s recent announcement that it will not charge royalties for H.264 codec, he still believes Google’s WebM will become the de facto standard. Opera was the first to release WebM support, he said.
“It’s not exactly clear what the provision is,” Håkon said, referring to MPEG LA’s news. “I haven’t seen any text that defines what kind of use that would be considered free, and I don’t think they would allow browsers like Opera and Firefox to freely add decoders. The web has always been based on freely usable formats, and video should be no exception. I think that WebM will be the format of choice for the web.”
The Future of Web Styling
CSS3 will bring the Web features such as rounded corners in boxes, shadows on boxes, small animations, and small transitions. “Rounded corners were part of CSS in 1995, but I thought they looked so 70′s so they were removed”, Håkon said. “Many of the new effects can be achieved with JavaScript or Photoshop today, but CSS takes such effects easier.”
“That’s really what CSS is about. It’s about reducing the need for images on the web and doing styling in a little language instead. It’s going to be a very visual presentation that one – what’s possible to do today and what’s possible to do tomorrow.”
Web fonts will make a strong comeback as well, he said. “Web fonts is a favorite of mine. It’s something we had in the specifications very early on in CSS2 in 1998. And it was implemented at the time by Netscape and Microsoft, but they chose to support different font formats and none of them chose to support TrueType, which even then was the most common format.
“But now we have support for TrueType and open type fonts on the Web and there’s even a new wrapper developed by W3C called WOFF, which adds compression to those well known formats. I think we’re going to see the face of the Web, the look of the Web, is going to change in the year to come when people start realizing ‘Wow, I can use fonts on my web pages I’m no longer stuck in Verdana Land.’ We can have amazing creativity unleashed upon us.”
With CSS3, for example, designers can use fonts to make logos (instead of relying on images) and have that logo appear in every browser without fear of it looking different each time.
Questions from the Web 2.0 Expo Audience
We asked our Twitter followers and Facebook fan group to come up for some questions for Håkon to answer. Here are the questions and his response. (Questions slightly rephrased for clarity.)
Would he consider expanding CSS3 coverage to device touch & motion events?
“I think so. One of the things we started out with in CSS1 in 1996 was to have support for things when you clicked upon them… and then we added things when you hovered over something with the mouse. You could attach styling to all these events. And I think it makes sense now that we have these new types of interactions to attach styling to those as well. I’m not entirely clear what kind of features one would be looking for, but it’s certainly something I’d be interested in discussing with [her]. [She] should send me an email.”
Would he seriously advance audio tag in HTML5?
“I think audio and video are very natural tags to have in HTML and we’re seeing strong support for them. And I think that by using the audio tag, for example, you can build a music player in HTML which is something I think people should be able to do.”
Do you think Opera will ever become an open source browser?
“It’s a question we’ve asked ourselves a number of times. There are good reasons I think for open sourcing projects. I think that we’ve seen many open source projects. I’m using Linux here myself. Most of the software I use except Opera and Prince, a CSS-to-PDF converter for printing, is open source these days. At Opera we haven’t found a way of being able to share our code. We’d very much like to share our code because we’re very proud of it. The business model just hasn’t made sense for us yet. We have found no way of both publishing our code and allowing them to reuse it without paying us anything… Our current revenue model at least wouldn’t be compatible with that model. So unfortunately we haven’t been able to do so. At the same time though I should say that we’re kind of part of a big open source movement because when you download a Web application or an HTML page or a CSS style sheet you get the source code for all this. By making the Web reach further and to more devices we actually are part of making open source, source as in documents and Web apps, available to users.”
How will you market Opera in the United States to increase its adoption?
“It’s hard to compete with Google’s marketing. We rely on friends telling friends about how wonderful this browser is… I don’t think you’re going to see Opera commercials on TV in the time to come. But I think you’re going to see innovative features come out, things that people want to have in their browsers.”
What do you think of Apple abandoning Flash in favor of HTML5, CSS3, and JavaScript in iOS?
“I think we’re seeing that HTML5 and CSS3 and JavaScript – the Web platform as we call it – will be able to do what propriety languages have done in the past. And as such I think the Web platform will be the place where people gather, and when they want new features, fancy things, sound video, they’ll find it there. The Web, with its open and freely usable formats, will be where the action is.”
~~
Kaitlin Pike is the Community Manager of Web 2.0 Expo. She can be reached @w2e or @kcpike.
Thanks to Jason La, Bess Ho and Abrar Qureshi for their questions!
Register here with code webny10scm4 to save 25% on Web 2.0 Expo conference passes.
> Håkon said applications as we know them will become “a footnote in history.”
I’m a huge opera fan, but he seems to be blind there. What’s his take on the Google Wave debacle?
I didn’t ask him about Wave :-/ Sorry.
What makes you think he’s wrong?
Hi Kaitlin, thanks for the answer, i was surprised to see in my email inbox that there actually was one.
To give you a straight and easy answer: The main difference between a web application and a native os application is performance. Browsers cannot feasibly make this difference up.
Here’s why:
First off: For every application you run on a machine, you incur a performance penalty for every abstraction layer you put between it and the hardware. A native application only has to deal with the OS. An application making use of a third-party gui toolkit has 2 layers. Browsers, due to their complexity, already are forced to those two.
However browsers add even more complexity, you have the abstraction layer of the browser’s own rendering engine, AND the HTML/CSS/JS parsing engine.
So web apps already have twice the abstraction layers to go through over native apps.
This was *extremely* noticable with Google Wave, as it ran very sluggishly, no matter on what browser, device or OS. I think this is also the main reason for lack of adoption and finally, its death.
Now, you could say \but computers are getting faster!\. Yes, they are getting faster in new ways that are quite difference from the past twenty years. The past twenty years marked a rapid increase of performance yield from a single core, which is great for programmers because it means their stuff goes faster without them having to do *anything*.
However in the past two years it has become obvious that that development has hit a roof. Clock speeds are not improving anymore. Speed nowadays is gained more by optimizing the instruction sets of CPUs and most importantly: Adding cores.
However, in order for an application to make use of more cores, it has to be written in a certain way. And quite frankly: Last time i checked there was no working generic implementation of threading in Javascript.
As such, web applications are stuck on the performance front. (As long as they are tied to browsers.) It would require massive advancements in technology to make writing fast web apps in browsers *easier* than writing them natively. And unless this actually happens, there is no way that native applications will ever be pushed out of the way on a big scale.
If you’re going to be at the show in NY I recommend trying to talk to him about these points. He was a great interviewee/very thoughtful and seemed open to challenges like this.
As for writing native mobile apps… It’s pretty annoying (understatement) creating/maintaining/even using different versions of an app for each mobile OS. That’s why I’m a fan of his idea of the future.
Thanks for the suggestion. Seeing how I’m in germany though, this is actually quite outside of my financial range.
As far as mobiles go: I agree, going with web interfaces can be much faster there, since most of these apps have a very small and focused domain of use. I wrote a diet tracker for myself for the explicit reason as a web app so i can access it from anywhere with any device.
On the other hand, especially on mobile devices performance is a heavy restriction. Could you imagine a 3D gps guidance application in a web browser on a cell phone?
re “native vs web” battle:
As a software developer I will always vote for “native” – hey, it’s cool, new platform/language/features coming up every month to experiment with!
As an application owner, I may choose one or another platform based on a number of criterias besides internal beauty of the source code or instant response time.
There are hundreds of thousands of desktop/mobile apps and millions of websites, so it is clear what platform serves business/practical needs better.
Besides, web browsers act more and more as OS, not as just a simple HTML renderer.
The line between “native” and “web” gets blurry ( obvious example – Web OS!).
In the end, it’s all about the purpose of your app and whether you need to push hardware to the limit or you need
a “standard” form-based app to interact with your users.
I’m surprised to see Håkon conclude that if the source-code of something is available for everyone to look at, due to the way it’s being deployed (Web), it’s therefore also “open source”:
“At the same time though I should say that we’re kind of part of a big open source movement because when you download a Web application or an HTML page or a CSS style sheet you get the source code for all this. By making the Web reach further and to more devices we actually are part of making open source, source as in documents and Web apps, available to users.”
Yes, you may look at things and learn from them, but 99% of the stuff out there is still copyrighted without a permissive copyleft-license to go with it, meaning you _can not_ use it without breaking that copyright. Håkon should be careful about feeding the mentality that just because something is on the Web it’s fair game for people to use as they see fit.
@Christian Walde – he’s talking about what’s going to happen in the next 500 years. I think it’s a bit early to give a verdict on Hakon’s eyesight.
“Massive advancements in technology” you say? I think research has been increasing somewhat exponentially for the past 100 years so it’s probably safe to assume that there will be massive advances until 2500.
@Tomis
Even if handheld devices become comparably powerful to current desktop computers there is still the little problem of the limit of speed of electricity in wires (c*.66), as well as in air (c*.96).
What this means is that, when you’re communicating with a place that’s 1500 miles away from you, the best you can hope to get for a round trip time is 16 ms when communicating entirely through air or 24 ms through copper wire.
Both of these are actually noticable and things are bound to only be much higher in a real life situation.
These are things that could not be overcome by mere technological advances. They’d require massive changes in the scientific field of physics itself.