S60 Open to new features

Choice is a funny thing. Offering "choice" to customers is a common argument used by companies to justify various kinds of competitive behavior. And "choice" is generally considered a good thing; as Barry Schwartz says, it's the official dogma of western industrial societies that the way to maximize freedom is to maximize choice.

Schwartz is not against choice as such; he's arguing that choice is not an unalloyed good and can lead to paralysis.

In a similar vein, Håkon Wium Lie (the chief technology officer of Opera) recently wrote in CNet about Microsoft's approach to standards. Talking about an open letter from Microsoft, he notes that The authors argue that consumers want several standards from which to choose and suggests I don't think so. Consumers never wanted the choice between VHS and Beta, and mobile telephony in the United States was hindered by customers having to choose between competing standards.

I think this hints at a key to the paradox, and has a lot to do with where we need to take our browser in the future. People do like choices, but nobody wants to choose about some things. You can choose from a vast array of stereo components, as Schwartz points out, but you want to be able to plug them all together easily. It's good to have a choice of devices that play movies on your television, but incompatible formats like VHS and Betamax (a more current incompatibility is between HD DVD and Blu-ray) don't make anything easier.

Browsers offer some degree of choice. On desktop systems, you can choose tabbed browsing, lots of integrated features, speed, and so on. On handhelds there are choices of layout, navigation, and, again, speed. But I'd bet not a single reader of this blog would opt for the choice of a mobile browser that could only open WML pages.

The best place to build choice into browsers -- and by extension, into the web itself -- is not at any underlying level of interoperability. The browsers you can choose among should all open "web pages", not just some other kind of documents, whether those documents are especially designed for special sorts of devices or not. Because if people choose a browser that just opens some other kind of documents, they're simply going to need TWO browsers, one for "some stuff" and one for "the other stuff". It's too easy to imagine the "two" becoming "three", "four", or more.

We're working hard to make the S60 browser one of the best choices. After all, paralysis only sets in when there are many choices and they're not so different that one stands out. When our browser is clearly the best, we're doing our part to solve the world's problem of "too many choices"!

And yeah, I know, we also need to make our browser something you can more easily "choose" in the sense of just downloading the latest version to your S60. It's just that there are so many things we could choose to do first, it gets so hard to pick one... ;-)

By the way, if you like the Schwartz presentation in the link above, here's a longer presentation he made at Google.

"Simplify, simplify."
-Henry David Thoreau

"Everything should be made as simple as possible, but not one bit simpler."
Albert Einstein

Simplicity is one of the ideals we try to attain when designing the browser user interface. A reaction I'd like to evoke in browser users is "that was easy!" In general, easy is often associated with simple.

It's not always easy to pin down "simplicity" though. For one thing, it's very subjective. When you're accustomed to doing something in a particular way with specific tools, you come to regard it as simple. Introduce someone else to the process, though, and they may have a very different opinion. Not because you're smarter or more capable then this neophyte (it goes without saying that you are, of course, but one tries to be nice :-), it's because you have different knowledge.

The knowledge you rely on to determine what's "simple" includes things you may not regard as explicit knowledge. Hand your mobile to anyone you know and ask them to call a number. Most of them will find it "simple". Hand the same mobile to someone who's never seen a phone and the task isn't simple at all. The task is exactly the same; it's the user that's different. Although if asked to list all the things you know, you might not include "how to use a mobile phone"; it's something we tend to assume.

We design for simplicity, then, for a particular collection of users. Mobile phones are inherently complex devices, and some of the tasks users want to do are also very complex, so sometimes we "hide" the complexity, or put it somewhere else. John Maeda in his book The Laws of Simplicity talks about this, and says (simply!) that "some things can never be made simple." But many things can, and as Rob Tannen notes at Boxes and Arrows: This does not mean that things cannot be made simpler.

Given time (which means time left alone, and stop bothering me about those TPS reports) designers can often do a pretty good job with simplicity. So what's the problem? Why are S60 devices widely regarded as the opposite of simple? There isn't just one answer, of course, but a big one harks back to where I started: when you design for simplicity you aim at a particular collection of users. S60 has an astonishingly wide array of users, and while S60 devices may be marketed to them on the basis of fairly cohesive groups with a lot of similarity, they're not (yet) designed that way. You'll be able to immediately think of some ways each of these would be unique: a kids' browser, a university students' browser, a European professional's browser, an Indian entrepreneur's browser, an African health care worker's browser, and a Chinese urban dweller's browser. I can too, but even working at The Browsary we can't create all those products.

Not yet anyway. Once, the story goes, the Model T Ford was available in whatever color you wanted, as long as you wanted black. Today cars are nearly made to order; one car is likely to be quite different from another even while both are clearly identifiable as cars. I think the days of "idustrial-model" software -- one version for everybody, limited customization that you have to do yourself (and it's not simple!) -- are numbered. It's actually fairly curious that the industrial model came to be applied to software, where it (arguably) doesn't apply particularly well.

Want to call us and order a browser with the UI and features you like? I'd love to take that call and fill that order -- just wait a while; I have these TPS reports that need cover sheets...

Thanks to Rachel for pointing out a column about mobile browsing in today's Washington Post.

A lot of the web is all about removing context. Digital content can be reused -- "repurposed", in the clumsier corporate argot. Here is a good cinematic treatment of the idea. This is certainly what browsers are about. If a web page can be said to have "a design", browsers don't really present that design so much as present an interpretation of the ideas behind the design.

If anything, this is an even stronger characteristic of browsers, like the S60 browser, that are designed for devices with unique characteristics. In our case the display is small and the input is limited. But you could imagine something using, for example, a monochrome vector display that would still be recognizable as a "web browser". In that case the characteristics of the device would be unique in a different way. You could even imagine a web browser that used a really unusual display mechanism. How about using a web page as a map for planting a flower garden? You plant the seeds and a few months later the flowers bloom and from above you can see the "page". Talk about latency...

Web designers generally accept the fact that their page designs are more like guidelines, and even celebrate it. It's certainly different, and in many cases exciting (empowering?) to create things in a dynamic medium. You can go the other way, too. As a designer or artist, you might decide that a work you want to convey depends not just on the words "this is not a pipe", but on those words being presented in a particular typeface in a particular size, color, and position. You can do that, but when you do, "the web" doesn't work quite as well. To control those things completely you might resort to one big jpg file instead of any text at all. You can certainly do that, of course, but there are costs. Searchability is one, and page size (and thus loading time, particularly on a mobile) is another. Art and communication are nothing if not sets of combined compromises, after all. And yet...

A big thing about the web is deconstruction. Let's call it "dismantling" to get away from any connotations of the term. A web page is called a "page" because it resembles a page in a book or newspaper. (That is why it's called a page, isn't it? Seems obvious, but I've been wrong before.) The characteristics of a printed page include words, typeface, size, absolute position, position relative to other items on the page, and sometimes various kinds of pictures. A printed page includes all those things together; they have fixed interrelationships. The page has its own context. A web page dismantles all those interrelationships. Each characteristic becomes a standalone thing that can be presented all by itself in different contexts. The text could become part of an RSS feed. An image can be used in any number of ways. As for the layout, typeface and the like, well, those things have meaning only in the context of their interrelationships. In other words, the meaning inherent in the context is lost in the process of dismantling.

There's always a context, of course. Just because text from a web page is presented via RSS doesn't mean it lacks context; it just acquires a new context. In other words, after deconstruction there's reconstruction with some new materials.

When the meaning of a page is “just the text”, that’s okay. It’s even an advantage, as the new context may work out better, changing the meaning in a useful way. An RSS feed may increase the importance of a news story, for example. In my own case, I’ve noticed some news items in feeds that I suspect I would have missed on the original page.

But sometimes pages are more than "just the text". Contextual meaning involves subtleties and nuance. Here's an example of a page that I think suffers when you start to dismantle it. You could reproduce the "factual" content of the page, I suppose, without much of the context. But something would be lost, and I think on a page with elegant, subtle design the loss is significant.

What I'm really arguing is that in some cases the natural web-centric process of dismantling all the aspects of content and reassembling them in other ways can be destructive. Creative distruction, you might say, but sometimes it makes me uneasy.

For much more depth on some of these ideas, have a look here.

Sorry about the title; I just get tired of "blogosphere". Anyway, did you know Forum Nokia also hosts blogs? Being Forum Nokia, many of the folks there are master-level developers, and better able than I am to talk about things like Symbian development. Gabor Torok is a Forum Nokia champion who has some informed things to say about Symbian vs. OS X in a mobile context.

This, of course, differs from me not in having things to say, but in those things being informed! It's all Gordon Moore's fault, I think. Without that blasted law of his my Lattice C on MS DOS skills would still be relevant!

Before I start, I should just say for the record that I'm not a Symbian developer. In fact, I'm not a developer at all. Other than writing C code back in the day, and writing some scripts when I can't invent any plausible excuse not to, my coding skills are pretty much nonexistent. Don't ask me about Objective C, although I did once have some comments about Objectionable C!

That said, here's an interesting set of comments about Symbian development. I don't know how realistic any of it is. Doing UI design isn't particularly constrained by Symbian -- or if it is, we're so constrained by the display and input capabilities of the device that we haven't really gotten as far as being constrained by Symbian!

Maybe some of the developers will have some more substantive contributions.

There are a lot more mobile phones in the world than there are desktop or laptop computers. Many of these phones have web browsers of one sort or another. But web pages and other web content is designed for desktop/laptop use, for the most part.

Some people think sites should be designed for use on mobile phones (or mobile devices, to be politically correct). On the other hand, maybe mobile devices should be able to (or aspire to) display web content designed for full-size computers. In this way of thinking, mobile users shouldn't have to operate in an internet "ghetto" containing only "mobile content".

This argument is still raging. Mobile devices are slow (compared to full-size computers), have tiny displays, limited interaction, and far less storage space to work with. Given these limitations, maybe it makes sense to have a special version of your website for mobile use.

But mobile devices are constantly improving in speed and storage, and lots of developers at Nokia (and elsewhere, believe it or not! :-) are working on better ways to display and interact with information. So maybe it makes sense to have a mobile device that works with the main (or only) version of your website.

I don't have the answer to this, but I do have a suspicion. Smaller, less capable, limited versions of things generally related to computer and communication technology don't seem to have held up very well over the past few decades. Hardware always gets more capable and less expensive. User Interfaces have been progressing more slowly recently, but still they progress.

I think it's also a human tendency to prefer the "real" version of something over a simplified, abridged, or reduced version. I could be wrong about this, of course; people do prefer easy/simple/clear to difficult/complex/confusing. I think most people would simply prefer one version of a website, book, movie, database, or what-have-you, but that single version should be useful.

The S60 browser is an attempt to take that way of thinking into use. There are other ways of thinking, and so there are other browsers. If you're interested, here are some links:
The .mobi top-level domain
The WAP Forum, now subsumed under OMA.
Shakespeare made, um, shorter.
Condensed Books -- maybe this could only come from the US!
W3C's thoughts on Mobile Web Best Practices.

Have you ever noticed that the S60 browser does sort of a 'double-take' when it loads a page? It seems to load, then it seems to start over. That's by design! In case you don't read blog comments, here's what Bradley Morrison of the Browser team has to say about it (this is in the comments section; I'm just pasting it here for convenience):

The browser is not actually loading the page twice; what you're seeing is deliberate and is known as our "double-tree" implementation.

Studies have shown on slow networks that users may come unsure about the status of the browser if content is not rendered in a reasonable timeframe. Like the progress bar, the purpose of the double-tree is to visually inform the user that the browser is well on the way to rendering the content.

So while fetching all of the required subresources (such as external style sheets, images or even other web pages) to fully display a web page, the browser engine temporarily shows an approximation of how the page will eventually look [*]. As you note, the effect is minimal with a quicker network connection (such as Wifi), and more noticeable with a slower connection speed.

An option to turn off double-tree could indeed be possible, thanks for the feedback! Hope that makes things a little clearer, please rest assured the browser doesn't download the content twice.

-Bradley Morrison