A new Windows? Please supplant the old one.

✎ Published: 2010-03-10 10:30:30

John Gruber discusses potential futures for Windows, highlighting the idea of launching two parallel operating systems: a new, streamlined version and the existing Classic version. He critiques the adoption challenges posed by varying user demographics, arguing that the new version may lack sufficient developer interest and user transition unless Microsoft implements strategies like virtual machine support for legacy applications.

A new Windows? Please supplant the old one.

John Gruber discussed a Technologizer article on the Future of Windows yesterday. In said article some of the "big brains" of the software industry were asked for their opinion on what Microsoft ought to do to keep Windows relevant in the years to come. Gruber quotes a scenario that imagines a tabula rasa, a new Windows without cruft and legacy code (much like Apple did when they switched from Mac OS 9 to Mac OS X), but then takes this idea even further, wondering whether it is actually necessary to replace the current Windows.

Or just maintain Windows in parallel. Point is, there’s no reason why Microsoft should have one and only one PC desktop operating system. Why not two: the new cool no-cruft one; and Windows, the established, familiar, chock-full-of-baggage-and-legacy-compatibility one.
I disagree, however. As splendid, as the idea actually feels, the characteristics of the two Windows user bases (Developers and Users) won't resonate with such a setting. Let's call the two versions "WIN" (New, no legacy code, cool) and "Classic" (Current Windows) Developers only invest their time and money up front if there is a viable market available (or to come), so that the investment is at least likely to generate a positive outcome. They won't start developing for a new system until the user base is of a certain size (or a huge growth is expected, see iPad). Windows users, on the other hand, are not really known for their progressiveness. I usually tend to divide them into three groups:
  1. Business users, who have no choice anyway since IT installs their machines. As businesses are actually the main force behind Microsofts keen upholding of legacy code, there is little reason for them to ever leave Classic. So here we have a very low adoption rate.
  2. Non-Computer users: The kind of people who almost accidentally bought their PC at Walmart, still run Windows XP (maybe even 98SE), and mostly use the system to print ugly birthday flyers or play Solitaire. They would probably enjoy WIN, but they won't ever notice it exists until they go and buy a new PC at Walmart. Here, we have another low adoption rate.
  3. The enthusiast Computer user: These are people that grew up with Windows, know (and sometimes love) all of it's kinks and quirks, and use it for a variety of tasks. They're its avid defenders in Engadget comment threads and tend to be a tad regressive, since Classic has always worked for them quite well and as they fear loosing the high investment in Windows knowledge that they accumulated over the years. The less regressive, the higher the chance that this particular person has already switched to Mac or Linux. The more regressive, the higher the chance that he/she would actually reject or even despise WIN as it deviates too much from her/his well known Classic. The adoption rate should be pretty solid, but this market segment is also heavily courted by the alternatives.

The problem with the above scenario is that the actual market for a brand new cruft free Windows deems me not big enough to attract enough developer traction to get a larger user base to switch over. One possibility, of course, would be to leverage .NET so that new applications would natively work on Classic and WIN sans problems (much like Apple did with Carbon). This, however, would disable any chances for Microsoft to radically change the user interface, accommodating usability research and results of the past decade. Instead, I think it would make more sense for Microsoft to create a brand new OS that still runs old apps by means of a virtual machine or something resembling the classic environment in Mac OS X. Microsoft needs to enforce adoption as much as they can, otherwise most of their user base simply won't switch. (Even the almost 10 years old Internet Explorer 6 is still being used way too much. This simple fact is telling about the Windows demographic)