To add my two cents to this debate. I have always thought the ideal form of standards (or de facto thereof) came from benevolent dictators. Borland did that job well for quite a long time, a single vision developed well beats the design-by-committee half-compromise path any day. The problem only really arose when the benevolent dictator stops being benevolent. (circa .net). The next best thing is what Free Pascal provides. An open architecture is lower to develop due to the additional discussion required, but it is also immune to taking the wrong path leaving people in limbo. Such events just inspire forks. In the article I wrote at http://screamingduck.com/Article.php?ArticleID=43&Show=ABCE I touched upon Free Pascal as an alternative to a standard. Being open means that people will never be denied access to the compiler, as long as it remains compatible with itself, that works for me. There are a number of languages out there for which the only implementation is the open source one. The other problem that I encounter is purely an evangelical one, People assume pascal is the pascal of old. I would suggest (perhaps at the time of a major release) creating a language syntax mode with a unique name.

Ah yes, the solution to the world's problem is Qomp. Getting rid of "Pascal" name and turning the language into a more practical and elegant syntax while we are at it. b Ln('Qomp Rules'); e.