We had a mobile browsers focused Mobile Monday earlier this week where we heard from Skyfire, Mozilla, and Opera. The message was the same across the board, they’re all striving to be able to take the same experience available on the desktop and make it available on your mobile. It’s the area where the iPhone has really reversed a lot of people’s thinking (though for me personally that really happened with the Nokia Internet Tablets, the N800 and N810). It’s a point that not everyone agrees with, but that I’m firmly in support of. The phone in your hand should be able to display the same pages, forms of markup, and objects as your desktop. And this whole discussion of “design for the mobile web” should shift from the technical details to interaction design and optimization.
First off was Jay Sullivan from Mozilla. We went through an overview of what the Firefox mobile effort will be, there are a lot of details on the Mozilla wiki. He was showing off a demo running on the N810, so of course I had to setup the Maemo SDK on my system and build my own. I’m having some trouble getting the network working under the SDK, and it doesn’t seem to be the nslookup problem that used to exist for previous scratchbox versions. But it is a Firefox browser running under Maemo and I can load up pages once I mirror them locally. I’m going to have to poke around with it some.
The port is going to be a full FIrefox port, so everything including XUL and extensions should “just work” from the desktop side. The version he had on his N810 had Adblock Plus and Weave running. He was also talking about the different potential “users” of Mozilla. Obviously there’s the end user, but he also spoke some about what impact this would have on web developers. There’s a lot of functionality that a phone has which it would be great to expose to a browser, but currently there are no bindings for. They’re looking at using the extension mechanism to allow folks to start experimenting with the right way to solve those problems before industry wide acceptance of any given technique. I’m personally just really excited at the potential to use arbitrary extensions on a mobile browser.
All the development is being done out in the open. Info is on the wiki, weekly company calls are open to anyone who wants to join in, IRC channels and newsgroups are open to outside input. I was really excited by what I heard. Excited enough that I’ve downloaded the code and started poking at it.
Next up was Nitin Bhandari, one of the founders of Skyfire. Disclaimer here that I’ve been helping out the team at Skyfire over the last few weeks so I have a somewhat biased opinion. They have the same overall message, delivering a desktop quality web experience to a mobile device. They do it by running an instance of Firefox on a server however and keeping the install on the handset relatively minimal. Because they are handling all the rendering and interpretation using a desktop version of Firefox all the rendering and supported media styles are the same. And because the server is doing a lot of the heavy lifting the client can run without too large of a footprint, generally consumes less bandwidth, and is kinder to the battery than running a full fledged browser. Right now their limited beta is for Windows Mobile devices only, but they have a version for Symbian in the works as well.
Nitin demoed pulling up a few different sites and showing how the browser works. One of the things that keeps getting picked up in the press is that flash is supported. You can pull up the normal desktop version of YouTube and watch and hear the videos. That’s something that even the iPhone can’t do yet, and was a source of some discussion when talking about Firefox mobile. Unfortunately Flash has become one of the defacto standards for rich web content. It’s really the thorn in the side of an otherwise open and expansive environment (with all the talk about open protocols and data portability and user control in the web 2.0 environment I always wonder why no one complains about flash being a closed proprietary system). The browsers that I have on my N800 and N810 play flash as well, but I can honestly say that the proxy based system that Skyfire is built on beats that particular set of on-device implementations hands down.
Last up was Tatsuki Tomita, SVP of Consumer Products at Opera. Opera itself restructured relatively recently into enterprise and consumer divisions. The folks on the consumer side spend a lot of time talking about the Opera Mobile and Opera Mini browsers. I have no idea what the folks on the enterprise side talk about. Opera Mobile and Opera Mini are two different flavors of mobile browser. Opera Mobile is the thick client version. All of the smarts is down on the client and it’s available for Windows Mobile and Symbian. I’m sure it’s a great product still, but I haven’t used it in quite a while. I actually had stopped using Opera pretty much all together, and was just about to just write them off completely when Opera Mini came out and forced me to reevaluate.
Opera Mini is hands down the best browser available for most phones out in the market today. It runs on Java, it’s free, and it’s proxy based. Like Skyfire, most of the smarts are up on a server somewhere and the client is a relatively thin display engine. Opera of course has Opera desktop renderers running on the servers instead of Firefox, but the general concept is the same. The server in this case also speeds things up quite a bit. I have an N95 with 3G, and I still use Oper Mini for most of my browsing instead of the built in WebKit browser because pages load faster and I like the navigation style better. And like both Mozilla and Skyfire, Opera Mini is aimed at bringing the desktop browsing experience to the handset.
They demoed the browser itself, but also spent a lot of time talking about Opera Link, which is a service for keeping your bookmarks in sync between your device and the desktop. It’s an interesting service, but for my usage it falls a bit flat. First off, what is it? It’s an online storage system that Opera Desktop, Opera Mobile, and Opera Mini can all use. It’s not just on demand syncing either, once you connect an instance to the online service it automatically keeps all of your bookmarks in sync. You can also view and use the bookmarks online at link.opera.com. It’s a great idea, but for me it really smacks of driving lockin through vertical integration at a time when I don’t think it’s justified. I would love to keep my bookmarks in sync with Opera Mini, but me personally I use del.icio.us to store my bookmarks and would love to plug that into the Links system. Using the Links website itself doesn’t really give the feel of a direct consumer site inline with other properties. No bookmarklets or extensions that I can take advantage of in Firefox.
The system itself seems pretty powerful though and particularly well engineered. To get a feel for how it works give this a try:
- sign up for the service on the web
- put your username and password in Opera Mini
- go into manage bookmarks in Opera Mini and manually sync
- leave opera mini open to the bookmarks and start deleting the builtin bookmarks from the web
As you delete bookmarks using the website the bookmarks will disappear from the bookmark page in Opera Mini without you having to touch anything. Just leave your phone sitting on your desk and the actions from your desktop browser immediately update your mobile. That’s a pretty sweet bit of functionality. If I were Opera I would be thinking about how to open that up as a generic platform for other folks to plug new functions into instead of working just the links bit. Links itself is just ill positioned given the overall environment (if I were an Opera Desktop user I’m sure I would disagree however).
Overall it’s a pretty odd world we live in right now with respect to mobile browsers. There’s still plenty of talk about WML and XHTML-MP and MobileOK floating around. But there’s also this vocal and active (but small in comparison to overall market size) community around the iPhone and other advanced feature browsers who insist that the whole idea of NEEDING different technical requirements for mobile browsing is absurd.
So how is a potential web property owner supposed to deal with it? Nothing plasters over the complexity completely. But it’s certainly true that you can ignore most of the legacy mobile web oddness and just focus on a growing number of high capability device owners and you’ll be seeing your potential user base grow pretty rapidly over the next few years. You do however ignore a whole set of users who are out there right now with devices in their hands and a hunger for good mobile content and services. But at least it’s a choice, to either do nothing and restrain your user base or to attempt to service the users further down market to pick up volume. Over time hopefully that decision will shift even more and become justification for tweeking the design for different layouts based on expected value to the end user rather than required changed needed to function on different capability browsers. There are a lot of folks working to make that the case.
This was a fantastic MoMo, I learned a ton from the presentations and had a lot to follow up on and poke at after the event. After almost 4 years of running the events I’m amazed at how often I still run across stuff that I didn’t know about (or had a mistaken understanding of). Plus it’s nice to get fired up about something like Firefox and find I can just grab it and start playing around with it on my own. Guess I could never really get rid of that hacker attitude even if I wanted to.