If IBM had managed to push an affordable compiler into the market just three years earlier, OS/2 would be better positioned today.
At least for me, not programming OS/2 was not due to a lack of interest, but a lack of money. Both the freeware/shareware and the commercial market could have been WAY more vital with a "Turbo Pascal for OS/2". But this problem just matches the general problems with OS/2: the best technology around, but no interest in the mass market. :-(
BTW, some consolation for depressed OS/2 users can be found on the most excellent web page "OS/2 Headquarters" by Tom Nadeau at http://www.mindspring.com/~os2headquarters/ with the two excellent, witty series "Understanding Microsoft" and "Understanding IBM", plus more.