OK, so now we have the rumour that Facebook is going to buy Opera. That would be unexpected. And interesting.
In this entry I’m going to pretend the news is true, even though that’s not certain yet, because it is a good starting point for some serious strategic thinking.
So. What exactly is Facebook buying here?
The obvious answer is that it’s buying the largest (or second-largest) mobile browser and its user base. More importantly, the browser is already there (i.e. doesn’t need years and years of work to get it up to speed) and it already runs on lots of operating systems, including the low-end ones such as S40.
So that’s cool. Facebook will suddenly have a presence on many phones running many operating systems and spanning the entire mobile market from high-end smartphone to low-end feature phone.
Still, if this were only about adding a few Facebook features to the (second-)largest mobile browser, I’d be disappointed, and it probably wouldn’t be worth a few of Facebook’s recently-hauled-in billions.
Opera has more to offer, though. It has a fully functional browser team, a rendering engine, operator relations, influence in the standardization process, it’s ready for TV, cars, and the rest of the next wave of devices, and it has a (second-rate, but still interesting) mobile ad network.
Buying all this plus a browser is lots better than building your own browser — a process that’s universally underestimated among new mobile players — and laboriously creating your own team, proxy browser, operator and standards relations, etc.
Finally, Opera has its Opera Mini proxy browser. The point here is that all data that proxy browser users request and receive go through central servers, which makes it quite easy to “map their social graph” or whatever the hip buzzphrase is this month. All your traffic data are belong to us, where “us” is now Facebook.
Incidentally, a tiny but vocal minority of Opera Mini users wouldn’t like this at all due to privacy concerns. The problem is that there is no alternative to Opera Mini in the market. This could have been a good opportunity for Mozilla to get some market share, except that as far as I know they don’t even have plans for a proxy browser. Other than that there are only Chinese alternatives such as UC. The web-savvy crowd is not yet desperate enough to go Chinese to avoid Facebook.
Tough luck for Opera Mini users.
It’s not entirely clear to me how good or bad Facebook’s current relations with operators are. On the one hand, operators want to drive traffic through their mobile networks (and thus make money), and Facebook is maybe the world’s best traffic generator. Operators like having Facebook on their phones. On the other hand, Facebook is (until now) a service provider, and not a browser, OS, or device vendor.
In any case, Opera has good operator relations. Take a look at the press releases: a good many of them announce an alliance with operator X in country Y to bring Opera Mini (and thus data traffic) to X’s phones. If that already-good position is combined with Facebook’s traffic generating prowess Facebook won’t have to worry about operators any more.
Facebook also buys a rendering engine; in fact, the rendering engine. By now it’s clear that there are only four rendering engines that matter on mobile: WebKit, Presto, Trident, and Gecko — exactly the same four as on desktop, because the proprietary ones several mobile vendors created turn out not to be good enough.
Of those four rendering engines Trident is not for sale, Gecko is irrelevant and possibly too heavy for mobile use, and WebKit is interesting but rather crowded. Presto, Opera’s engine, is the only one that’s available.
As far as I can see the point here is not actually owning a rendering engine, but owning the influence in the web standards bodies that comes with it. Facebook could have joined the WebKit crowd, but it would be hard-pressed to make an impression there, since there are already plenty of web-savvy companies involved.
Facebook is making a dedicated turn toward HTML5, because native apps aren’t going to bring it to the mass of low-end phones around the world. It recently unveiled Ringmark, its own browser test suite. Although I’m not entirely convinced of its usefulness (that probably merits a separate blog entry), it’s a clear sign that Facebook is very, very serious.
In fact, I recently heard a rumour Facebook is pressuring operators to deliver “Ringmark-compliant” mobile browsers (whatever that may mean), with the expectation that the operators will pressure device and browser vendors in turn. Facebook wants agenda-setting power in the web standards world, and the acquisition of Opera is going to help a lot there.
Would it be bad for web standards if Facebook becomes involved? Not necessarily; it all depends on how much proprietary stuff they want to add (insert stupid joke about
<like> tag here).
In fact, I’m rather avidly watching one shelved web technology that Opera still has lying around: W3C Widgets, which are essentially installable web apps. I spent most of 2009 working on them, and although they weren’t quite ready for prime time back then and haven’t improved since, I’m still convinced that installable web apps are the way forward for web standards on mobile.
Installable web apps are an excellent way of delivering code-heavy Facebook content such as games to not-too-highly-specced devices, so it’s clearly in Facebook’s interest to get them out there. And Opera used to have a widget manager embedded in its browser, so it wouldn’t be too hard to dust it off and restart the project.
If Facebook adds its weight to getting a workable spec for W3C Widgets or a similar technology, more power to them.
On the other hand, if Facebook runs with W3C Widgets or their successor, the other players will implement their own version, because they don’t want to give Facebook too much power. That’s how it goes in web standards land.
Still, the very fact that installable web apps will once more be on the agenda trumps these problems, at least in the short run. Once we have W3C Widgets up and running we can start complaining about proprietariness. At least we’ll have something to complain about.
Just yesterday a rumour surfaced that Facebook is going to create its own phone. Ewan McLeod explains why this would be great for the mobile industry, but doesn’t give too many clues about why it would be great for Facebook.
My point here is that creating its own phone would be exactly the opposite of Facebook acquiring Opera. Opera gives Facebook instant access to every single smartphone OS except for Windows Phone, as well as most feature phone OSs. It will end up in people’s pockets pretty quickly, in other words.
A Facebook phone would not have nearly the same impact. It would have to be created first (and that’s a hell of a job, don’t underestimate it), then a hardware vendor would have to get involved, then it would have to be marketed to operators, and then sold in bulk. In other words, in two to three years Facebook would be in a few people’s pockets, far less than with Opera in far more time.
So I don’t really believe in the Facebook phone. It’s a sop thrown out there to propitiate investors, who are worried about Facebook’s lack of a mobile strategy but have no clue how a winning mobile strategy looks. In the end Opera trumps the Facebook phone — in spades.
With all that said, will Facebook actually acquire Opera? I believe that it would be in Facebook’s interest to do so, so it all comes down to money. Google is rumoured to also be in the market (why? mostly to cut off Facebook, I’d say), as is Russian search giant Yandex. (Opera Mini is big in Russia, so Yandex would make sense.)
Let’s wait and see. In any case, it’s going to be one hell of a ride. Once again.
Upcoming speaking gigs: