Dan Shafer wrote:
At the end of the day, speaking as an old Smalltalker who thinks objects rule, I've just had give up on the OO dream when using Rev. As you've learned, it's not implemented and although there are some worthy work-alikes, they're not really satisfactory to an OO thinker.
There has been a lot of back-channel talk about either creating an OO version of Transcript or forking a new IDE/Language combo for OO folks but my guess is that's a pretty distant dream at this point.
Yes, if the goal is OOP. But OOP is a means to an end, not an end in itself.
If the goal is to implement OOP as a merely intellectual exercise, Rev will be a disappointing experiment.
But if the goal is to get results that benefit development and maintenance of software systems, there's likely a highly profitable way to achieve those ends in Rev as it is today.
This is not to suggest that OOP is without value; on the contrary, I've seen otherwise. But I've also seen the Mac Toolbox and other non-OOP systems that still get great benefit from well-structured interfaces that achieve similar results. And with a million fewer JSRs. ;)
The original poster presented a practical problem, and a practical solution was offered that was simple and incurs only 0.006ms of overhead on a modest computer (using a frontScript to allow well-factored code that applies only to specific objects). It's not OOP, but it gets the job done and lets the developer move on to the next challenge....
-- Richard Gaskin Fourth World Media Corporation ___________________________________________________________ [EMAIL PROTECTED] http://www.FourthWorld.com _______________________________________________ use-revolution mailing list [EMAIL PROTECTED] http://lists.runrev.com/mailman/listinfo/use-revolution
