Browser news from the Mobile World Congress

Yesterday evening I returned from my fourth foreign trip this year. This time I went to the Mobile World Congress, the annual Barcelona-based get-together of the mobile industry, and I can tell you, it’s something else.

This post gives an overview of announcements by mobile players that might be of interest to web developers. There’s an incredible lot of it, too, because every single major mobile player except Apple feels that MWC is the ultimate forum for major announcements.

If you know of more news, or have links to additional information, please leave a comment.

I was there because Vodafone had invited me to sit on a panel in a technical “embedded conference” about W3C Widgets and related technologies. The concept can use some fine-tuning; I’m hoping to do some of that in the future. I was there mainly to stress that the mobile browser situation is not as simple as it looks. THERE IS NO WEBKIT ON MOBILE!
While I was at it I also invented guerilla browser testing.

MWC is huge, and by that I don’t mean SxSW-style huge. You could drop all of SxSW smack-bang in the middle of MWC, and nobody would notice (apart from the people in the drop area). With only 50,000 delegates this was a decidedly bad year, as one Vodafone MWC veteran pointed out to me.

All major and a lot of minor mobile players were there, with the exception of Apple and (curiously) Nokia. Apple’s absence was expected; it follows its own unique trajectory and basically ignores the rest of the industry when it comes to announcements. (Apple might have a point here; it gets more attention for its announcements when they’re done entirely outside the MWC timeframe.)

Nokia’s absence created a huge buzz; it was the main MWC sponsor for many years before it suddenly withdrew this year. Unfortunately I’m not yet well enough versed in mobile political sciences to be able to explain this. (And don’t leave a “simple” explanation in the comments! This stuff isn’t simple!)

Phone-wise I was deep, deep inside BlackBerry land, as you’d expect with business executives and marketing people. There were few iPhones, I saw almost no Androids, though Samsung had a surprisingly solid presence.

Anyway, the main general question right now is when browsers will become a hot topic on MWC. This year they weren’t; the 30-minute panel for about 100 attendees I participated in was about the most browser-focused event of the entire conference. Still, I believe that browser quality is an emerging theme in mobile, so I expect a future MWC (next year? year after? 2013?) to be more browser-aware.

This will be something to watch; if even business executives start to seriously consider browsers, web technologies will have basically annexed the mobile space. That would be a cool extension to our current reach, right?

For those following the smartphone market Tomi Ahonen’s smartphone war frontline update is required reading. He currently expects Apple to be in serious trouble because Christmas sales were bad. It will be interesting to follow that storyline as it unfolds.

Announcements

Everybody and their dog considers MWC the prime venue to make exciting announcements. As a result you hear all announcements at once, and few make more than a fleeting impression.

Besides, MWC is not a technical conference. It’s a business and marketing conference where business and marketing people try to impress other business and marketing people by firing off a stream of tech-based buzzwords.

Neither the broadcasting nor the receiving party have the slightest clue what they’re talking about, and it’s this curious mechanism that gives all MWC-related press releases the unique vagueness that make them so hard to interpret by techies.

Nonetheless I’ve carefully gathered most (all? nah, probably not) announcements that are important to web developers. Here’s the list, in the approximate order in which they made a splash in web dev land.

Microsoft

Microsoft announced its new Windows Phone line which runs Windows Mobile 7, and a new version of IE that I currently expect to be based on IE8. I saw from my Twitter feed that this announcement was well-received. Unfortunately the Microsoft stand turned out to have no actual Windows Phones in it, so I couldn’t test anything and will reserve judgement until I’ve actually used one.

More in general, this is Microsoft’s last chance to matter on mobile, and they know it. Windows Mobile 6.1 is just spectacularly bad. Windows Mobile 6.5 is a lot better, but still no real competitor to iPhone, Android, or Symbian. Will Windows Mobile 7 help here? Only time will tell.

Obviously I’ll post a full test report as soon as I can get my hands on a device.

Opera

Opera was canny enough to send out its exciting announcement slightly ahead of the crowd. It will shortly submit Opera Mini to the Apple App Store. Although at first I doubted the technical wisdom of this move (Safari iPhone is far ahead of Opera Mini, which, after all, does not offer any client-side interactivity), I’m now starting to revise my opinion.

Personally, I’d welcome Opera Mini because my Safari iPhone has taken up the nasty habit of crashing on my webmail site. The more I reload it (and have to re-login), the higher this crash chance is.

More in general, Opera Mini uses far less bandwidth than Safari iPhone because it sends a highly compressed version of the site to the phone. Right now that might matter on lousy connections (such as the ones in Barcelona, where 50,000 mobile professionals clogged the networks considerably). In the future, though, it will start to matter even more because eventually the operators will put an end to the economically untenable flat-rate iPhone data plan. When that happens Opera Mini will become an interesting alternative because it saves you money.

Still, the main point of this release is political. By publicly announcing their intent, Opera has put Apple in the position of either allowing the first competing browser on the iPhone or being slammed for not allowing it. (Currently all alternative iPhone browsers use the Safari rendering engine; Opera Mini would be the first to use a different one.)

It’ll be interesting to see how this one plays out.

WAC and network APIs

Then, no less than 24 global operators announced the Wholesale Applications Community which aims at spreading applications to an installed base of not 100 million iPhone users, but 3 billion users of other phones.

This is typically something operator executives love; it’s an “answer” to the App Store (which, in my opinion, doesn’t need an answer because it’s slowly going to be transformed into a niche market mostly for games and other applications that need to be native, and not web-based. But I digress.)

Obviously, the announcement gave no technical details whatsoever. This is a business and marketing thing, after all.

Still, to anyone who spends more than half a second thought on it it should be clear that there’s only one single technology that gives the operators a fighting chance of attaining their objective: web technologies.

Unfortunately, the fact that web standards are the obvious answer does not mean that they will actually play a role in the WAC initiative. Ignoring the obvious is something that executives are very good at.

Still, Vodafone is part of the consortium, and Vodafone is investing heavily in web standards (even to the extent of paying me to do fundamental research). So I guess we can hope. As far as I know the Vodafone widget-based system is currently the only one that works on more than one platform.

Related to this is the news that operators are working on “network APIs.” I’m not yet totally sure what they are (technical info, as usual, is completely absent), but it seems likely that they’ll include APIs for mobile payments.

The point here is that operator-based payments are the most user-friendly ones possible because, contrary to every single current app store, you do not need to log in. Your identity is ascertained through your SIM card, and the payment is added to your operator bill. Mobile payments cannot be made any simpler than that. (Besides, the operators obviously love this middle-man position.)

BlackBerry WebKit

Then BlackBerry announced that a new WebKit-based browser is coming up. This is no surprise to anyone following the mobile industry. About six months back BlackBerry acquired Torch Mobile, the creator of the Iris browser. Since the Iris browser scored rather well in my tests, it was obvious that BlackBerry wanted a high-quality replacement for their proprietary browser whose Best-Before date was reached somewhere in 2008.

As far as I know there were no actual devices with the new browser in Barcelona (but by the time this news reached me I had become very tired, and I didn’t try very hard to find them).

Still, the 100/100 Acid 3 score gives us a clue as to what’s going on. I found only four WebKit-based browsers with this score: Safari desktop and iPhone, Chrome, and Iris. (Android 2 not yet tested; wouldn’t be surprised if it scored 100/100, too.)

I assume that the Torch Mobile team has been working on an actual port of the Iris browser to the BlackBerry OS instead of creating a wholly new one. This would bode well for the mobile web.

But remember: THERE IS NO WEBKIT ON MOBILE! This is no iPhone WebKit magically transmuted into a BlackBerry application, it’s a completely separate WebKit implementation. A good one, most likely, but that doesn’t mean it’s the same as the iPhone.

Ignore any post you read about the new BlackBerry browser somehow making the creation iPhone-only sites more acceptable. That’s just uninformed nonsense. However, using BlackBerry WebKit as an additional argument for serious use of progressive enhancement, as Andy proposes, is an excellent strategic idea that’s worth a try.

Obviously I’ll post a full test report as soon as I can get my hands on a device.

Adobe

The Adobe stand was all about Flash, and sported the tagline “One Web, any device.” Although I agree with the basic sentiment, I’m not sure if Flash is going to deliver on that promise. I’d like a list of mobile platforms on which it is supported, but obviously the Adobe site doesn’t give any clue of that. (Nobody ever does that ever; everybody’s very vague about platforms and just assures anyone who’s willing to listen that their mobile solution works practically everywhere. But I digress.)

This list of mobile Flash videos is the closest I came, and it mentions Android and Palm, but no other platforms. Not exactly a huge user base.

The odd thing is that Adobe’s announcement was not about Flash but about AIR. This, apparently, will be the technology to deliver content across all platforms — well, to Android and BlackBerry, at least.

We’ll see. The pro of AIR (or Flash) would be that it is the single RIA environment (except for plain old HTML, CSS, and JavaScript) that works on more than one platform. The con is that it doesn’t work on that many platforms.

Adobe AIR uses WebKit, by the way. I never tested it, but it’s safe to assume that it’s different from all other WebKits.

Samsung bada

One of my personal top stories was the unveiling of Samsung bada (Korean for ocean; not capitalised), the new smartphone OS that runs a spiffing new WebKit implementation that’s different from all other WebKit implementations.

I give kudos to Samsung for actually allowing anyone to play with their new Wave phone in their stand. I took the opportunity to invent guerilla browser testing; I fired up the browser and did some very quick tests while standing in the Samsung stand. I returned twice when I’d thought of more tests to do. I crashed two phones in the process.

The WebKit-based browser seems to be called Dolfin; in line with the ocean metaphor that Samsung has chosen. It is not to be confused with the Dolphin browser for Android — I think.

The bad news is that Dolfin is not yet very good. The good news is that Samsung clearly stressed that the software wasn’t yet final; changes would be made. I hope they change a lot on the browser side of things.

I was able to prove that Dolfin is closely related to the widget manager WebKit on the H1/M1 Vodafone devices; something I’d more or less expected. There’s one bug in my still-unpublished width research that occurs only on the Samsung bada and widget manager WebKits, and nowhere else. (There is no “WebKit on Mobile!” They’re all different!)

As a whole bada seems a reasonably good platform. The point here is that it’s not meant as a high-end smartphone platform that competes with iPhone, Android, or Maemo, but instead a mid-range one that will compete with Symbian. Some quality has been sacrificed on the altar of affordability.

Bolt and widgets

The Bolt browser, which is the main competitor to Opera Mini, now supports W3C Widgets in version 1.7. Bolt is WebKit-based, by the way, but it’s different from all other WebKits.

This news is interesting because it requires Bolt to use a different architecture than the standard widgets W3C is defining and Vodafone and Opera are implementing. Such a widget is an application stored on the device that uses a full browser that offers full JavaScript capabilities to run in.

Like Opera Mini, Bolt does not support client-side interaction, but instead requires a full page refresh every time a script changes something to the DOM. Besides, the widgets cannot be stored on the device because there’s no full browser there, just a thin client that gets its instructions from the server. Therefore the widgets will be stored server-side. The downside of this is that you need something like a Bolt account to access your widgets. (Details aren’t clear to me.)

Android

Google’s announcement was that 60,000 Android phones are sold per day. Let’s be clear: this encompasses all Android phones, whether they’re branded by Google itself, HTC, Samsung, Motorola, SonyEricsson, or minor players. Still, it’s an impressive number that might upgrade Android to third-largest smartphone operating system in 2010, after Symbian and BlackBerry, but before the iPhone.

Still, all’s not well in the Android world. The problem is the Nexus One. Google’s release of its own branded phone (created by HTC) has widely been interpreted as a stab in the back of Motorola, who’s gambled its existence as a smartphone vendor (or even as a device vendor in general) on the Droid-centred Android strategy. The Google phone is obviously a serious competitor to the Droid in the eyes of the affluent tech-savvy US smartphone purchaser. (Motorola is irrelevant outside the US.)

To partly make up for that HTC unveiled the Desire, an HTC-branded and -skinned phone that runs on exactly the same hardware as the Nexus One. HTC’s brand awareness doesn’t come close to Google’s, but still it has a decent market share in Asia and Europe, and of the generic vendors without their own OS I definitely rank HTC first when it comes to interface design and general performance. They even made Windows Mobile 6.1 work — more or less.

Right now I don’t know how much all this matters. Some people will buy an Android phone specifically; others buy a SonyEricsson, HTC, or Motorola phone without knowing anything about the OS. Motorola will likely fall out of the race in 2010, and it could full well blame Google, but it’s certain that Google won’t care. Meanwhile it’s unclear whether Google cares about the HTC Desire.

In any case, of the sub-markets the Android one will remain by far the most complicated because there are so many players involved. Every single Android player, however, has at least one alternative platform in case of emergency. Except for Google.

Android will be one of the major unfolding stories of the year.

Nokia and MeeGo

As to Nokia, despite being absent from MWC it had one announcement for each of its smartphone platforms. Symbian^3, the successor to S60, is coming this year. No more details are known, but I expect its WebKit to be significantly upgraded. (It still will be different from all other WebKits, though.)

More interesting is the Maemo news, because industry pundits generally expect Maemo to become Nokia’s high-end smartphone platform, while Symbian will serve the low end and the business market.

Point is, Maemo’s gone. It has been merged with Intel’s Moblin platform to produce MeeGo (silly name in my opinion, but nobody asked me). I have no clue what this means in terms of browsers (technical details, obviously, are absent). And no, I have no idea what Intel was doing in the mobile OS world, either. Maybe this merger was the point. (Whose chipsets will the new MeeGo devices use?)

For now I assume that the Gecko-based MicroB browser will remain the default one of the new system, and that Mozilla will create a Firefox port for it. That’s just guesswork, though.

The most interesting tidbit is that the new MeeGo platform is not restricted to Nokia; LG just announced it will release a MeeGo phone. It was already working with Moblin, anyway, so moving to the new platform wouldn’t be such a huge step.

I hope the photo in the article does not depict LG’s new phone, though. The photo shows the Obigo browser, and if there’s one browser you should definitely avoid if you want to give your users a forward-looking browsing experience it’s Obigo.

That concludes my coverage of the MWC news. If you have more mobile browser news, please leave a link in the comments.

This is the blog of Peter-Paul Koch, mobile platform strategist, consultant, and trainer. You can also follow him on Twitter.
Atom RSS

I’m speaking at the following conferences:

(Data from Lanyrd)

Categories:

Monthlies:

Comments

Comments are closed.

1 Posted by Lars Gunther on 18 February 2010 | Permalink

Just to complete the news on Flash 10.1. It will be available on the n900 as well.

http://maemocentral.com/2009/10/08/video-flash-10-1-on-the-n900/

2 Posted by Jason Grigsby on 18 February 2010 | Permalink

Thanks for the run down on MWC. I hope to get over there next year.

Regarding the Nexus One vs. Motorola phones, I think the answer is the opposite. I expect Motorola phones, particularly the Droid line that Verizon promotes to vastly outsell the Nexus One.

What the media in the U.S. is missing and that you can't see unless you are in the U.S. is the massive amount of money and promotion that Verizon has put behind the Droid.

There are billboards everywhere. I see it ads at sporting events. Ads on TV.

Yes, Google has prime real estate on its home page to promote Nexus One, but after the first couple of weeks promoting it there, it no longer does so. Currently that spot is held by "Doodle 4 Google: Your student's design could be our logo - sign up now." So they're no longer promoting the Nexus One a little over a month since it was announced.

Andy Rubin told GigaOm that Google hoped it "can sell, at the very least, 150,000 Nexus One devices." That's what they are on record as saying they hope they sell for the first year of the device http://gigaom.com/2010/01/05/andy-rubin-on-google-phone-android/

I don't know what the future holds for Motorola, but the Nexus One will have little impact.

3 Posted by Codesquid on 18 February 2010 | Permalink

Good to see that some people realise that mobile development is about more than just the iPhone. We definitely need to do more to push web standards on mobile devices.

4 Posted by Michael on 20 February 2010 | Permalink

At a huge conference, where Apple was not present and did not make any concurrent announcements, your recap mentions them or their products 24 times. Just sayin'...

5 Posted by John Dowdell on 21 February 2010 | Permalink

Hi, you're right that there isn't a full list of Flash-enabled devices... here's an older list from October 2008:
http://www.adobe.com/mobile/supported_devices/handsets.html

Since then Flash Lite 3.x licenses picked up dramatically, despite the announced move to license-free Player 10.1. Once partners start shipping it will be easier to make such a list.

jd/adobe

6 Posted by Bartosz on 25 February 2010 | Permalink

Thanks PPK, nice coverage. BTW, I'm a bit confused about the whole MeeGo thing. So far I haven't heard about any Moblin device on the market, while Maemo is already out there and seems to be a more mature platform... I'm not sure if merging those two will do any better for Nokia. Still, MeeGo as a future replacement for Symbian may be a huge win for Linux.

7 Posted by Dan on 2 March 2010 | Permalink

Would it be too far fetched that if Google "accidentally" begins snuffing out Motorola, that Google would actually just eat them up (you know acquire them)?

8 Posted by Tom Malcolm on 9 March 2010 | Permalink

Hello, as a hard-headed newbie to this mobile world (my experience is in the fixed web)I am being pulled hither and fro (more like getting my ass jerked around)on trying to get my mind around what kind of code I should use to accommodate most phones. I am in the camp of NOT forsaking the entire rest of the mobile market in favor of the iphone. So toward that end I need some direction on how to code a dedicated mobile web site that will accommodate most smartphones as well as most feature phones. I have jumped in with both feet but am getting a constant stream of conflicting opinions on what to do and how to do it.(even conflicting definitions of what constitutes, say, a WAP site as opposed to an XHTML site!)
So I am asking YOU, where the hell does one find a good document or group of documents to understand it all? And how to test the sites like with PC and mobile emulators, etc.
Thanks for any advice.
FYI - I'm here for the long haul, NO MATTER how frickin' convoluted this all seems now, I'm stayin'!

Thank you,
Tom Malcolm