January 30th, 2012

full steam ahead

My tweets

snark maiden

Just the facts: downgrading ratings for Russian tech rumours

There's a Russian blogger and consultant by the name of Eldar Murtazin who has in the past had surprisingly good success at getting advanced access to Windows Mobile, Nokia and other smartphone releases; way back when, he used to leak details that later turned out, from time to time, to be fairly accurate. In fact he had enough access that at one point, Nokia asked the Russian police to retrieve all the 'unauthorised' Nokia phones Murtazin had been writing about. Other Russian leaks gave us an early build of Office 2012 and several Windows 8 builds, along with a running commentary of internal build numbers. But recently, Murtazin and similar tipsters have had a far poorer record and their rumours range from dodgy to plain laughable.

Claims that Microsoft would buy Nokia or RIM don't make sense as a business model, and neither do rumours that Microsoft will sell its own PCs or tablets, because Microsoft doesn't compete with its OEM partners in the phone or PC market. At the time I demolished the rumour that a Windows 8 CTP would be unveiled at last year's Partner Conference. Now Windows Phone evangelist Brandon Watson has busted Murtazin's latest rumour, that Windows Phone 8, AKA Apollo, won't run Windows Phone 7 apps.

That's an even clearer version of the message Aaron Woodman gave me at CES: "We are absolutely dependent on developers falling in love with the product. It's so important to the value we ultimately deliver to the customer that they're always first." Murtazin is probably thinking of the discussion around what 'common core' means for Apollo and what the new kernel will be; I have my own theory but at this point no-one outside Redmond actually knows. Although I did also wonder if he'd read too much into something else Woodman said to me: "We will always have to evolve - this an incredibly scary competitive space. There will always be challenges with backwards compatibility, but that's the price of innovation."

Rumours are fun (unless they're so far off base they're infuriating) and so is the detective work of digging through job postings and technical documentation and conference presentations to spot connections and make deductions. One of the pieces I'm proudest of writing last year explains how the Metro interface in Windows 8 relies on HTML5 standards that Microsoft built into IE9 and IE10. At the time we thought Microsoft was just catching up to the open Web but when I saw the first Metro demo, I recognised some principles and techniques and put two and two together. Simon and I have just thrown our hat into the ring over on Ed Bott's blog with predictions for the dates when we'll see the consumer preview and other milestones for Windows 8, though that's more like informed guesswork.

Of course, crazy rumours get clicks - though there's an issue if your crazy rumours are always wrong. There's one mobile news site I used to respect that has come out with so many batshit rumours recently that I now dismiss everything they say as more wolf-crying. That's where the Russian rumours seem to be ending up these days.

The problem may be that they've been too good. Microsoft and Nokia have been far more protective of what's in development than in previous years and they've been cracking down on leaks; Microsoft has resorted to near-Apple levels of misdirection and internal need-to-know about some things (while maintaining a great deal of transparency about technical details). From what we hear, Russia is where a good deal of that cracking down has taken place. If all the sources of leaks have dried up, it's not surprising that the rumours are turning out to be so wrong, so often.

What does surprise me is how many people carry on listening to rumours from pundits who have a less than stellar track record. I'm happy to put my hand up when I've got it wrong (I was expecting a CES Windows 8 beta right up until the Building Windows blog said it would be the end of February, for example). Maybe we need a rumour-scoring service for tech pundits that keeps track of what we predict and what we get right? Now that would be a Klout score worth tweeting about...

(I'm currently unimpressed with Klout because it will not allow me to claim that I have expertise in either Windows 8 or snark; clearly Klout has not met me...)