In 1985 I attended a conference in Palo Alto, California, of the Modula 2 Users' Association. There were about 150 of us there, possibly including some of you. It was perhaps the most high-powered intellectual group (of that size) I've ever been in. More than half of those present were compiler writers. Borland had a couple of guys there showing off a pre-beta version of their Turbo Modula 2 compiler (running in a 64K DOS environment) that I later beta tested, but that never got released.
On the last day of the conference, I asked the group to take a straw vote on whether the underscore should be permitted in Modula 2 indentifiers. (Wirth excluded underscores from the language specification, and none of the compilers had them. Modula 2 is case sensitive, and Wirth modeled identifiers InMixedCaseFormatLikeThis, a format that many people, including me, can't stand to look at.)
To my utter astonishment, the conference deteriorated from worthy intellectual discussions into "religion". The group divided roughly in half between pro-underscorist and anti-underscorist. The two Borland guys were in the pro-underscorist group.
The reasons in favor of the underscore are several and obvious, and today, most languages support them.
There is only *ONE* reason that was offered by the anti-underscorist group to exclude underscores from the language... and it was an emotionally-charged reason. Guess what it was? Here it is:
"Underscores are not Wirthian."
These otherwise brilliant people were worshipping Wirth to the point of absurdity. I talked with Wirth on the phone once, and he's not a die-hard anti-underscorist. He's a humble professor who NEVER had any idea Pascal would become so hugely important in the world. Had I been in the right place at the right time, I'm pretty sure I could have convinced him to include the underscore in the Modula 2 spec.
Mark Emerson
_______________________________________________ fpc-pascal maillist - fpc-pascal@lists.freepascal.org http://lists.freepascal.org/mailman/listinfo/fpc-pascal
