3 ms·
I'm not sure. If you do this the right way, you could build a Smalltalk the way it should have been. And that's not really an over-engineered enterprisey thing
by wlievens 18y ago
I'm not sure. If you do this the right way, you could build a Smalltalk the way it should have been. And that's not really an over-engineered enterprisey thing by itself is it?
- plinkplonk 18y ago"If you do this the right way you could build a Smalltalk the way it should have been" it all depends on the details of "the way it should have been" and "the right way". which I believe was the grandparents original point - the devil is in the details. So far all we have is lots of hype and a video or two. I personally believe the whole "multitide of dsls" angle, which is where Mr.Fowler is coming from, is a load of fluff wrapped around a genuine engineering principle (linguistic abstraction - see SICP). The fluff is the "enterprisey" part.
- vorador 18y agoWhat do you mean by "the way it should have been" ?
- wlievens 18y agoSmalltalk (at leas the flavor I used) stores code as a data structure rather than plain text. That same idea is at the core of the language workbench concept Fowler is championing. What the language workbench concept adds to that is that you can render and edit your "code" data structure in a multitude of languages. That includes graphical flow languages, for instance. One could take Smalltalk and take it to this step, too. I agree that right now it all looks like slideware, but I do think the concept has merit.