Opera: Single-minded about widget development

Jon von Tetzchner is the chief executive of the Norwegian browser company Opera. Although Opera first became known for its desktop product, the company has also become well known for its Opera Mini handset-based web browser.

Opera has become heavily involved in the development of standards for widgets — the lightweight, web-based applications that are starting to become prevalent on new handsets. It has also been working hard on the development of HTML 5, which has more built-in rich media functionality than the current version of the web standard.

ZDNet UK caught up with von Tetzchner at the Wireless '09 event in London on Wednesday to discuss standards processes and how Flash may soon become unnecessary.

Q: Tell us about the work Opera has been doing with widget standards.
We work mostly through the W3C, which is where the widget standard per se is being worked on. The widget standard [as far as it has been established] is more about the packaging — on the relation of how you connect to the underlying device, it hasn't been standardised fully. Bondi is trying to standardise that, and we have engaged with Bondi and with [the Joint Innovation Labs].

There are already quite a few initiatives, and there is a risk of fragmentation, and obviously our goal is always to try to make things migrate... to a single standard. Sometimes, on the way, people are eager to get started, but we try to engage as much as possible to make sure that this gets standardised in a way that works for everyone.

What does Opera gain from these widget standards bodies?
I don't want to say this is a philosophical thing for us, but we do believe the internet is too important to be limited. There is a significant risk of the fragmentation of technology. It'll be like on the PC: you'll write an application for a platform, and it will only run on that platform. We're already seeing some of this [in widget development], where web technologies may be in the mix. But you're mixing all those things in, and suddenly you have to write for the platform instead of for the technology.

Our goal is to try and make this work because we believe that's the right thing to do. We've seen the benefits of this from the PC side, where there are differences between the different operating systems, but you can still run all the applications. That's the benefit of having things standardised.

We have a lot of people that know how to write standards, how to implement standards and how to engage in the standards bodies. We have the biggest active group in the W3C to do just that. Considering that our competitors tend to be a lot bigger than us, that shows our commitment to this.

You see fragmentations — in the worst-case-scenario, you'll think they are lobbying towards a single-vendor lock-in again, and I don't think anyone really wants that.

In what way could that happen?
Clearly, if there is one vendor that wins, then [lock-in]is the potential outcome. It keeps people on their toes to have competition. If there are multiple vendors implementing the solution, then you'll see innovation. To me, again, it doesn't matter whether it's open or closed source, because if everyone was using a single piece of open source, that wouldn't solve it either.

Currently, we are seeing competition in the browser space. There are four browsers, basically; there's Microsoft — they have multiple [browsers], but let's simplify the picture a little — there's Mozilla, there's Apple and there's us. The competition there is leading to innovation and improvement in the different browsers. We've seen that so many times in history — that competition leads to better quality products.

You didn't put Chrome in that list.
This is more about the engine. Chrome is based on Safari — they're based on the same piece of code. But again, if WebKit wins the market or the Mozilla code where there are a few vendors that are building on top of it, I think that would stifle innovation in that particular field.

Tell us about Opera's involvement in the development of HTML 5.
We are very engaged in this. In some ways, we had a period of time when HTML was kind of not being worked at. There was HTML 4 and a lot of work on XHTML — these are technologies that are beautiful, but...

...you are almost restarting [with each version], and they are not totally compatible. XHTML has not really been used very much, which is sad, but that's the way it is.

For a while there, we actually worked outside, together with Mozilla and Apple, in the WHAT [Web Hypertext Applications Technology] working group. This is a group where we were basically looking at web standards. This was done because there was no other forum to do this. A fair amount of the work that we did there has then gone in to the W3C and is now part of HTML 5, then changed because there have been more people, including Microsoft, having input on that.

That's the way we like it — you like the competitors to be in there. You also have the browser vendors, you have the people who are actually writing the content and you have people [with] input who clearly think it is important to promote their different causes.

In your presentation earlier, you were talking about what HTML 5 can do as a replacement for Flash.
I think you can do most things with the web standards today. In some ways, you may say you don't need Flash. On the other hand, I like Adobe — they're a nice company. I hope they flourish and do well, so this is not about killing Flash. I think Flash will be around for a very, very long time, but I think it's natural that web standards also evolve to be richer. You can then choose whether you'd like to do it through web standards or whether you'd like to use Flash.

What we definitely don't need is more proprietary technology — that's the main thing. We have Flash. It's there — fine. Let's not get anything more.

Are you talking about Flash becoming more niche?
It's more of a choice of what you like doing.

Where's the line between what web standards can do and what Flash can do?
You can do everything, I believe, through web standards — you don't need to use something else. But there might be something where you believe Flash is better; then you choose to use Flash.

When Vodafone announced this month that it was going to launch a cross-platform app store, analysts noted it may be time for a switch back from native widgets to web-based widgets. Where do you see web-based widgets working better than native apps and vice versa?
For most things, web is better. As you're seeing the browser getting more powerful and better at executing, I think you'll find that someone can maybe code a more efficient piece of code in machine language than they can in a more high-level language. But it becomes very difficult, and you need a really good programmer to make the most of it.

The benefits of doing it web-based are just so much bigger, especially the fact that you can get something that runs across all the different platforms. Now, if you really, really want to make the most of the device, and you need to code at the lowest, lowest level, then nothing beats native, right? But it will mean that [the application] will only run on that device, and it's not very portable. So it's not a scalable model, which is why everything has moved to the web on the PC.

Palm have said their new WebOS widgets will be built on web standards such as JavaScript, HTML and CSS, but will run natively. Are they doing it the right way?
I need to look closer at it. It is a question of how pure it is. As soon as you start doing a combination of native and web, it becomes difficult. This is why it's important to standardise on those elements as well, because if you do a combination of web and native and you end up having something that is proprietary, you lose the benefit of the portability.

You can implement web solutions incorrectly. You will have a time — and this is not about Palm; they may be doing it right — where we will see a little bit of fragmentation. We're already seeing that with the iPhone and the like. I think people will move towards the centre point again, because the benefits are just too big. You want to reach everyone. The iPhone may be cool, but if it's one or two percent of the market, that still leaves a lot of people that you're not reaching.