S60 Open to new features

Modern browsers always have the ability to download. Downloading functionality is sometimes provided as a "secondary" UI; there might be a download manager, a download list, or something similar. Our browser takes that approach; we have a download list and some limited choices you can make about what to do with downloads that are in process or finished.

I'm not really very happy with this particular area of our UI. You can download various kinds of files to your S60, but it's not as seamless as it should be, and we often contend with differing opinions (both from within and without the team) about whether to proactively inform users or simply make information available about download status, completion, and whatnot.

I wonder how much longer "downloading" will be recognizable function. Increasing network speed, more reasonable funding methods, and better software design might -- might -- mean that the boundaries between "what's on your mobile" and "what's on the network" blur and eventually disappear. The software will still keep track of the difference, but only in the sense that today, one thing is stored on your C: drive and another is stored on your E: drive. And "downloading" will be marked in the dictionary as "obsolete".

There are some interesting trends starting to converge, I think. First of all, a truly mind-boggling (at least to me) number of S60 devices are streaming out into the world. Second, the notion that mobile devices are all about the Internet is (finally) starting to pervade upper levels of corporate decision-making. The third one is not particularly visible yet, but it makes so much sense that it's probably there, just hasn't yet emerged. The third thing is the proliferation of more S60 development projects, particularly open-source ones. Now, there are different opinions about how many third-party software choices you might want or need, but I'm really just interested in one category: browsers.

I think the convergence of these three things is going to lead to more choice in the S60 browser world, which seems like it should be one of those "inevitable" things. We're already in the tens of millions of users, and at NokiaWorld Olli-Pekka Kallasvuo talked about bringing the Internet to "the next billion users." S60 isn't Nokia, of course, but at the moment Nokia is selling the lion's share of S60 devices. In any case, there are a lot of S60s.

With such a huge population of users, it stands to reason that they'll have different needs and wants for browsing. There's plenty of room for a wider variety of browsing applications. There's also plenty of room for a wider variety of variations of our browser.

I think there's a good chance we'll see specialty browsers in S60. Given enough memory space, you might want browser tailored to your own uses. Rather than a set of bookmarks, for example, why not a Sports Browser? A Music Browser? A Financial Browser? All of these would use the same Internet, but they'd deliver a UI and content tailored to a particular set of use cases.

All of this is already possible, of course, and on desktop systems takes a slightly different form: take a "basic" browser (say, Firefox) and add plugins and skins. On mobile devices it might go a different way: different applications. Or different varieties of the same application, of course. When storage is no longer a problem, why not?

Just a brief note: have a look at the new Mobile Security Blog! I'm assuming it offers no sinecure, at least.

...just an attention-getting device! Here are some places where our browser has recently attracted attention.

Nokia World in Amsterdam is over now, but the From the Floor blog is still available. Accessing various Internet resources from mobile phones has become the thing, it seems!

Ivan Kuznetsov compared our browser to Opera.

Roland Tanglao talks about how to upload to blip TV from the WBfS60. (And thanks to Roland for reminding me to include this!)

Speaking of Opera, it's another very good browser. Too bad ours isn't available on multiple platforms as well.

If nothing else, we're doing pretty well supporting UAProf.

This is more of a hardware review, but they like the browser too.

Not recent, but a good article at OS News.

Here's a report from Phone Scoop that touches on the browser.

Wikipedia will post entries about almost anything, apparently!

And there are always interesting things going on over at Google. This isn't really about our browser, but worth a look anyway.

The term satisficing comes from the work of Herbert Simon. It's a product of the area of bounded rationality; the observation that humans are not completely rational in all areas. Actions are partly based on rationality, but also affected by emotions and other internal factors. Moreover, it's rare to have enough objective information to make a completely rational, informed decision. In most cases it's just not worth the effort to become fully informed, even if that were possible.

We might have a need, desire, or goal, but instead of completely satisfying the need, we come as close as we can within the constraints of our current situation. That's satisficing. If you need a watch, for example, a precision timepiece milled out of solid unobtanium might appeal to you, but an inexpensive model from eBay might be good enough. Good enough is the hallmark of satisficing.

In designing a software user interface we do a good deal of satisficing. Even UI designs from Apple Computer, usually touted as the best, coolest, easiest-to-use interfaces available, are at some level "good enough" rather than "perfect" or "ideal".

The trick, of course, is that as a UI designer you know this but you don't often admit it. You try to make your designs as close to perfect as possible, even when you understand that "perfect" in this arena isn't even a reasonable notion, all things considered.

Generally the more a UI designer knows about the users of a piece of software, the better the design will be. By "better" I mean "providing more of the things users want to do, in ways users discover and understand." There are three aspects to that: features, discoverability, and understandability. Sometimes these aspects work against each other.

In the case of the browser, we generally want to create more features, but some users reach an impasse in discovering those features. This is partly because the S60 user environment is more constrained than, say, Linux, Windows or MacOS. We don't have the display area to offer status bars, toolbars, docks, menu bars, and scroll bars all at the same time. We have a more "vertical" menu control structure -- that is, rather than starting with a set of categories like "File", "Edit", "View", and the like, we start with just one: "Options". That means that some features are going to be "buried" in the vertical control structure.

Features can certainly be buried in a desktop windowing UI as well (did you ever find the screen saver that used to be built into Microsoft Word?) But desktop UI designs often offer more than one path through the UI; you might find a particular feature in different ways. In S60's more constrained UI environment, there's less of that. So some users get stuck and don't discover some feature that might be useful.

We're still working on this problem. We serve a vast array of kinds of users, and generally try to strike a balance between features that would be useful to some number of them versus the additional UI complication a new feature might add. Then once we do decide to add something, we try to predict the situation that might lead a user to an impasse, and redesign in order to avoid that situation.

Often we have to simply choose what we think is the best design approach even though we know some users might get stuck. If you use a mobile device to open a web page with embedded sound, for example, it can lead to some potentially confusing situations because the user's situation may very well make the sound problematic. Or a feature that "power users" really want might be so confusing to nontechnical users that it's hard to decide to include it. If you know what a user agent string is, and how it works, you might want the ability to modify it in your browser. But if you don't know, and you end up in that part of the UI anyway, it would be bad if your browser no longer worked the way you expect.

Longer term, I think the fact that the WBfS60 is an open-source project offers and answer to this. I think eventually you'll start seeing different kinds of browsers, still based on WebKit, but serving more specific audiences. We don't have enough developers to create a special browser for a specific group of users, but just because those developers aren't here doesn't mean they aren't somewhere else. I think the future may well hold some special-purpose browsers to choose from.