At Build, which will run April 29-May 1 in San Francisco, Microsoft must make the case for Windows 10 to developers, who have largely shunned its separate mobile and touch-based app ecosystems in favour of Google's Android and Apple's iOS.
"It will be critical how they engage developers," said Wes Miller of Directions on Microsoft, a research firm that specializes in tracking the technology company's moves. "Making it as easy to develop once for all Microsoft platforms is key."
"They have to be clear, both for consumers and developers, how they will benefit from this 'universal' app approach," added Jan Dawson, chief analyst at Jackdaw Research.
Both were referring to what Microsoft has called "universal" Windows apps, ones that allow developers to call the same APIs and leverage much of the same code to create software designed for multiple devices, from smartphones and tablets to PCs and hybrids. Universal apps -- and the associated single OS, that concept dubbed "OneCore" -- are the cornerstones of Microsoft's effort to decrease development time and expense, and thus tempt more to commit to the currently-anemic platform.
Microsoft is expected to focus on Windows 10's universal apps at Build.
Before the launch of Windows 8, then-CEO Steve Ballmer boasted that Modern app developers -- Microsoft once dubbed them "Metro" apps -- would have hundreds of millions of customers, basing that on the number of PCs sold annually, plus anticipated upgraders. "It's going to create a heck of a lot of opportunity for folks in this room to make millions," Ballmer claimed in October 2012.
That opportunity did not materialize when Windows 8 stumbled badly. Microsoft hopes to reignite developer interest and make up for the lost three years by first, streamlining app creation, and second, pushing all Windows users, on all platforms, toward 10, where they become potential app customers.
Even though the experts said Microsoft had to wow developers at Build, they were skeptical the company could pull it off.
According to Miller, Microsoft's pitch sounds familiar. "We got Metro [with Windows 8] and no one could figure that out," Miller said of Microsoft's contention that its new Metro apps would appeal to both owners of touch-enabled devices and those who stuck with keyboard and mouse. "Now, it's the same sort of thing: Microsoft says you can develop a runtime across all modern platforms with some reasonable work.
"The problem is it's really that last mile that is the hard part, that sets your app apart," Miller added, referring to the user interface (UI) and user experience (UX) design and coding that would still have to take place on a universal app destined for multiple device categories.
"We've seen this promissory note before," Miller said. "Java did this back in the '90s. But there's a lot of really custom work that will have to go into each platform."
Failing that "last mile" work, an app conceived for a smartphone will be seen as too simplistic for the desktop, while one with a desktop focus will be overwhelming and unusable on a smaller, touch-based screen.
Dawson was even more pessimistic about Microsoft's implied promise that universal apps will turn things around. In an analysis he published on his personal blog, Dawson argued that the universal app concept suffered from a fatal flow.
Microsoft's assertion, said Dawson, was that universal apps will let the millions of Windows PC developers easily port their apps to mobile, boosting the number of apps for Windows-powered smartphones and tablets and closing the so-called "app gap" between Microsoft's OS and those of Google's and Apple's.
"[But] the apps Windows Phone is missing simply don't exist as desktop apps on Windows," Dawson wrote. "Among the top 50 free iOS and Android apps, there is not one which is not on Windows Phone but exists as a desktop app on Windows."
His point? That it's the kind of mobile-first, even mobile-only apps that resonate with consumers that Microsoft most needs, and that desktop-focused developers are least likely to provide.
"I don't see universal apps helping Microsoft cross that chasm," said Dawson in an interview. "Either you already saw the logic in developing for Windows PCs, in which case you've likely already done a Windows Phone app, or you never have, in which case I don't see this changing things much, especially as it relates to increasing the number of apps on Windows Phone."
And other moves Microsoft has made seem to conflict with the universal app conceit, Dawson continued, ticking off the retreat from touch in Windows 10, as well as Microsoft's push into the low-priced part of the smartphone market, where users are less likely to pay for apps.
Windows 10's re-emphasis of traditional computing may make it even less of an attraction to app developers than was Windows 8, Dawson said. "By dialing back on some of the UI commonalties, Microsoft makes Windows 10 look more like Windows 7, but that means there's not going to be that visual commonality between apps [on different devices]. It may arguably perpetuate the core problem with Windows 8, [but] at the apps rather than the OS layer this time around," Dawson said.
So what's Microsoft to do? Or does the app issue even have a solution?
"Azure is becoming increasing important," said Miller of Microsoft's cloud-based platform. "It might not be the first choice for developers, but maybe Microsoft can own the back end for all their own platforms."
"[Universal apps] does sort of bring these three [smartphones, tablets and PCs] together," said Dawson. "It seems possible that that could lead to the kind of adoption Microsoft's lacked with Windows Phone and Windows 8. But ultimately, Windows has nothing to do with the whole world, which is Android and iOS."