Weblog

The End of Mowser is Not the End of the Mobile Web

In the last few days there has been a certain amount of rather sensationalistic and poorly informed commentary floating around on tech sites and blogs, predicting the immediate death of the mobile web. For example, this piece on CNET, and The Register's dramatically titled A Requiem for the Mobile Web.

And what is the basis for this doom-and-gloom mongering? Well, it turns out that a poorly-marketed twelve-month-old startup, named Mowser, which has never been able to attract VC, and seemingly staked its future entirely on its ability to attract VC, has called it a day.

That's it.

The End of Mowser

Mowser's offering, in a nutshell, was a wholly ad-funded service which adapts web sites, allowing them comfortably to be viewed on the small screens of mobile handsets. This process is known as transcoding, and is nothing new or original, as we'll soon see.

There are many reasons why I don't believe that the failure of Mowser allows us to make any predictions, positive or negative, for the future of the mobile web. The most obvious of these is the simple fact that whilst Mowser's technology was very promising, their business model was flawed.

Now, you'll need to bear in mind that there are several companies out there, perhaps most noticeably OpenWave and Novarra, who are making a not-so-small fortune selling web transcoding services, and that this fact was conveniently overlooked by both CNET and The Register. The difference between all of these companies is that whilst Openwave/Novarra are selling their services to mobile operators, in exchange for real money, Mowser were giving away their service to consumers, in exchange for showing them ads.

Could it ever have worked? It's anyone's guess, but let's have a look at the criteria that a consumer would have to meet before they would be sufficiently motivated to actually use Mowser's service. The user must:

1. Wish to view non-mobile websites on their mobile handset; and yet
2. Own a mobile handset/browser combination which does not do a good job of displaying non-mobile web pages
3. Have heard of Mowser
4. Be willing to open their mobile browser, type in Mowser's URL, wait for Mowser to load, and then type in the URL of the site which they wish to visit
5. Be willing to be shown advertisements, as a kind of payment in return for the ability to browse a stripped-down version of the public web on a handset which they paid for, on a data plan which they pay for, no doubt handsomely

Furthermore, they need to do all of that in sufficiently enormous numbers that they provide a viable target market for an ad-funded, direct-to-consumer proposition.

I'm told by those in the know that the magic number of users, viewers or readers that it takes to make advertisers sit up and take notice is one million. I don't know how many users Mowser had, or claimed to have. However, bearing in mind that I've worked on the mobile web for longer than Mowser even existed, and had never heard of them before this week, I find it hard to believe that one million consumers even knew what Mowser did, let alone could have been tempted to use the service regularly.

The medium is irrelevant: Mowser failed as a business.

A Different Angle

There's another angle here too, which has been overlooked by commentators. I don't believe that this was the case here, but let's try it on for size: one could suggest that the success of a transcoding company such as Mowser should be directly inversely proportional to the success of the mobile web. That is, as more and more mobile websites come online, and more and more full websites offer mobile alternatives, or adapt themselves gracefully to mobile devices, the value of a third party transcoding service diminishes. Thus the demise of Mowser could be presented as evidence that the mobile web is absolutely flying right now.

Conjecture aside, the simple fact is that the mobile web is going through an enormous period of growth: at OverTheAir 2008, BBC mobile head Matthew Postgate pointed out that over the three and a half years up to mid 2007, the BBC's mobile site picked up 1.5 million users. In the following six months, it gained another million. That's an incredible acceleration in growth, and it's a big pointer as to how exciting 2008 and beyond are going to be.

Of course, the most successful sites and applications are going to be those whose mobile and desktop versions are seamlessly integrated, and provide the best possible experience available on the device from which they are being accessed. That much is a no-brainer. Facebook is an obvious example of a site that is getting that right already.

A Storm in a Teacup

All in all, I think that this doom mongering is no more than lazy journalism on a slow news day; a storm in a teacup.

Personally, I feel that the column inches would be better devoted to raising awareness of some of the real challenges facing the mobile web, such as the shocking liberties taken by a number of operators which are suffocating innovation at every step. I hope to cover a few of these matters here in due course, so stay tuned.

Many thanks to Ciaran for helping to inspire this rant.

Why is mail.police.gov.bd Attempting to Hack Me?

I keep a distracted eye on failed logins to the mighty Pointbeing.net datacentre, courtesy of logwatch [1]. Here's the list of attempts from yesterday:

Authentication Failures:
root (mail.police.gov.bd): 452 Time(s)
unknown (mail.police.gov.bd): 403 Time(s)
root (58.241.84.193): 17 Time(s)
adm (mail.police.gov.bd): 6 Time(s)
lp (mail.police.gov.bd): 4 Time(s)
mysql (mail.police.gov.bd): 4 Time(s)
root (securityscan.xtraordinary.net.uk): 4 Time(s)
unknown (securityscan.xtraordinary.net.uk): 4 Time(s)
apache (mail.police.gov.bd): 3 Time(s)
ftp (mail.police.gov.bd): 3 Time(s)
bin (mail.police.gov.bd): 2 Time(s)
bin (securityscan.xtraordinary.net.uk): 2 Time(s)
daemon (mail.police.gov.bd): 2 Time(s)
games (mail.police.gov.bd): 2 Time(s)
gopher (mail.police.gov.bd): 2 Time(s)
ftpsecure (mail.police.gov.bd): 1 Time(s)

Now, securityscan.xtraordinary.net.uk is a monitoring service provided by my hosting company, Xtraordinary Hosting, who I strongly recommend. I don't know who or what 58.241.84.193 is, but a quick visit to DNSstuff suggests that it's probably a run of the mill hacking attempt orignating from China.

Of particular interest to me are the many hundreds of attempts originating from mail.police.gov.bd. That appears to be the Bangladeshi police's own webmail server, running an alpha of SquirrelMail. The pattern of the login attempts corresponds strongly with that of a compromised server.

Has anyone else seen this in their logs? Is there any point in emailing the Bangladeshi police to let them know?

[1] Seriously guys, who runs their website on port 81?

New Pitch Website

I'm pleased to learn that Pitch's corporate website has today been relaunched.

Sure, it's a corporate website with the traditional combination of slick agency design, press releases and mugshots of the usual suspects. But have a dig around and I think you'll see that we're up to some pretty cool stuff.

I find the work section particularly exciting, of course. It mentions a few of the things which we're doing right now, including the recent work with Red Dwarf, Warner Brothers and a new project with legendary Thin Lizzy guitarist Scott Gorham.

Update 01/06/2008: This post has quickly become entirely irrelevant! We've now joined the PlayPhone family and will be rebranding accordingly. Watch this space.

An Introduction to Fire Eagle

A definite highlight of Over the Air 2008 was London-based Yahoo Steve Marshall's introduction to Fire Eagle. For those not in the loop (which, to be fair, is most people: Fire Eagle is currently only open to a limited number of invited developers) Fire Eagle is Yahoo!'s brand new API for location-based services.

The genius of Fire Eagle, and the reason why it will be an enormous success, is its sheer simplicity. It does absolutely nothing beyond storing your current location, and disseminating it to your choice of sites and applications. Sure there's an API, wrappers for a few languages and some relatively fine-grained user privacy controls, but that's about it. No, actually, that is it.

By way of a simple use case: you, the consumer, log into Fire Eagle with your Yahoo! id, manually enter your location on the web page (you can enter this in countless formats - for example, geographical coordinates, a street address, a postcode or town name), and all your envious Facebook, Twitter or MSN friends get a notification that you're on the beach in Hawaii.

That's no more effort than Twitter requires, but the possibilities are way, way more interesting.

As one delegate pointed out, the killer app for Fire Eagle will be mobile, and will be one which automatically detects and uploads the user's location to Fire Eagle without user intervention. (Let's face it, who has time to constantly update it manually? [1]). I'm certain that those kinds of apps will be around in short order for GPS-enabled S60 smartphones or Windows Mobile devices such as the ubiqitous N95 or the XDA, but I won't hold my breath waiting for this functionality for my LG KU990.

Once that's in place, along with other tools such as the ability to SMS your location into the system, Fire Eagle will be a goldmine for application development. There's already a Facebook app and the rather nifty wikinear.

And no doubt countless further applications are on the way. Because, at the risk of repeating myself, the genius of Fire Eagle is its simplicity: that the intelligence is at the edge of the network [2]. Fire Eagle - the network - itself makes absolutely no assumptions about how it will be used, and thereby places no limitations on its use. The intelligence is you, the developer or entrepreneur, sat at home or in your office dreaming up incredible ways of using the technology.

You can request your invite to Fire Eagle here, but don't hold your breath. A nice touch was that Steve brought along handfuls of developer invite codes, so I made a point of snagging a couple. Fortunately, I don't think it will be long until Fire Eagle is opened up to the masses (presumably in perpetual Beta, as is de rigeur these days).

Footnotes

[1] Judging by the massive and inexplicable success of Twitter, perhaps quite a lot of people have this much time on their hands.

[2] That's a loose quote from financial boffin Andy Kessler, and is one of his criteria for what constitutes a good technology investment. The principle can be used to explain both the unmitigated success of TCP/IP and HTTP, and the drab featureless world of fixed telecoms.

Hint to Employers

Here's a handy hint to the employers and facilities managers out there, which shouldn't really need stating:

Do not expect your developers to be happy and productive in an office heated to 28° or 29°: they won't be.

A corollary to this rule is:

Don't rent offices from Regus.

Luckily, we're moving in a few weeks' time.