For most new app development, these users are not a very appealing target market -- if they don't buy computers nor upgrades, do they buy software either?
But I'm sure there are business cases where it makes sense to target such users (even GeoWorks and OS/2 are still around, after all, so there's someone interested in software that runs on 486 boxes).
Many of our customers (mainly large companies and government departments) have pathetic computers, never upgrade (and in fact downgrade any new computer purchased to XP) and buy the cheapest hardware you can imagine, yet spend huge amounts on software. I don't have any stats on hand, but at a guess I'd say at least 50% still use XP.
Or they work for a corporation who can't upgrade to a decent OS because they invested heavily in IE6-only web applications which their business runs on.
Or, they assemble computers from scratch and don't want to shell out extra cash for an upgrade, especially considering that Windows 7 is slower, has compatibility issues with older software and has made questionable "improvements" to UI.
The upgrade effort (never mind the expense) is huge, and there has to be a compelling case to move on from XP... which there certainly wasn't for Vista. Win 7 is gradually being adopted but it is taking a long time.
So while I can appreciate Microsoft's desire to move on from supporting apps on XP, it means that there is at least one large customer who won't be able to adopt VS11 for development.
See also:
http://social.msdn.microsoft.com/Forums/en-SG/vcgeneral/thre...
"Windows XP Embedded or Windows Embedded Standard 2009, both based on Windows XP Pro binaries will get security updates until Jan 8, 2019"
If 50% of all PCs run Windows XP as we speak, do you really consider it "ancient?" Do you think they will really stop delivering updates, even if they wish?