Most of my programming work these days is done on a Debian laptop or an Ubuntu desktop, so I actually acquired a copy of Windows 8.1 the "old-fashioned" way--by which I mean pre-installed on a new computer. Since my first PC ran DOS 3.3 applications off two 5.25" floppies in the early 90s, I've become pretty adaptable when it comes to user interfaces. That doesn't mean that I don't hate the touch-optimised tiles. When you make your living being the person who's "good with computers," it's kind of insulting to know that that the "least common denominator" user Microsoft's designers obviously had in mind was a toddler who discovered the pretty bottles in the liquor cabinet.
In the Small Mercies Department, Microsoft bowed to backlash and brought back the "Start" button with the 8.1 version of Windows. (Historical aside: Remember when Windows 95 debuted and the Start button was the Worst. Thing. Ever.? Yeah, me too. Humans are so weird sometimes.)
So while yesterday's buzz mostly focused on Microsoft's numbering juke (intended to put perceived distance between Windows 8 and the newest new thing), the real news is how much the blow-back was apparently taken to heart. At least for full-size monitors, Windows 10 will be 7-ish enough to appease corporate users. It's a critical demographic, given how businesses would rather use the (dangerously) unsupported Windows XP than pay employees to thrash their way back to baseline productivity in Windows 8.
True, whenever the interface for a workaday piece of software is radically changed, it rarely pleases anyone. No news-flash there--if it ain't broke, don't fix it, amirite? (Although it's always fun watching the religious war between the Team Flat Design and Team Skeuomorphic Design whenever Apple toggles between the two.) And while people expect more for less all the time, too many features decreases the usability (and thus the value) of pretty much anything, including software
The number of reasons companies like Microsoft and Facebook jerk users around are probably as many as the reasons we keep coming back to them anyway. But, apart from sheer sunk costs, it's important to keep in mind that software companies labour under one disadvantage not shared with most other industries. And that is they don't have the luxury of contracting out their software development. After all, who'd trust any company that outsources its core competency?
Granted, in a Microsoft-sized company, you can shuffle people between products. (And my gut feeling is that such cross-pollination is worth the disruption in nearly all cases.) But the fact is that keeping people on staff through the slower periods is cheaper than boom-and-bust hiring and firing. People always need time to "onboard" (in HR parlance). Time to unsnarl the spaghetti code-base. Time to trip over the land-mines in the home-grown tools. Time to figure out which brain-sucking meetings can be avoided without political repercussions. In short, time to absorb and internalise the company's own special dysfunctions.
Most importantly, adding more people to an already late project makes it even later.
All of which makes a solid organisational case for staying fully-staffed for as long as it makes strategic sense. But I have to wonder whether it also incentivises people to look busy--and thus to push pixels and twiddle bits that were fine just the way they were, spank-you-very-much. It's easy to blame the pushers and the twiddlers themselves, but the fact is that any given change has to be approved by multiple hierarchies of someones before it goes on public view. Because managers and executives worry about looking busy too. Worse, some of them even worry about how they look to all the cool, edgy kids.
And, next thing you know, the monster escapes and kidnaps Fay Wray and tries to play volleyball with bi-planes from atop the Empire State Building. (Which probably doesn't get a second look from any self-respecting native New Yorker--even in 1933 when there were no cellphones. But, hey, for us hayseeds out here in the sticks...)
"Creature Feature," indeed.
