if you guys allow me to squeeze a few words in this (hot) thread, I'd like to say that the initial post by Dan Shafer contains a couple of sentences that I would qualify as, ahem "embarrassing", especially due to Dan's huge contribution to the xTalk world for many years...
actually it reminds me of another "embarrassing" discussion that took place on this very list a few months ago, in which Dan wrote that goal-oriented interfaces were a huge step forward in UI design, no matter if the rules were set by Microsoft... Here's an example of what I find "embarrassing" : "I hope it *does* in fact adopt dot notation so that all of us who have trained our brains to think in those terms when we create and program with objects will be comfortable doing so" the embarrasing part being, IMHO : "all of us who have trained our brains to think in those terms" I have a background in ergonomics and psychology, and it is proven fact for decades that newbies and experienced programers approach and memorize algorithms in completely different ways (I have already discussed this a few times on this list during the past years, so please check the archives for more info). But at the same time, AFAIR, xTalk (and specifically HC) were designed to allow newbies, but also ppl with no background as programers but with enough intellectual skills and rigour (and mostly no time and/or no desire to learn a cumbersome language & notation) to develop sophisticated projects by themselves. IOW, the design of xTalk & HC was to "train" computers to think in the human way... And it seems to me that now Dan is promoting the other way around : to adopt dot notation because so far all OO languages use dot notation (mainly because they have been created by prof. programers for prof. programers, except perhaps for SmallTalk)... It really sounds like a big step backwards, back to a pre-HC and pre-xTalk era... I've learned assembler in 1976-78 and believe me I do know what it means to be "forced" to think in the same way as the computer... Almost any algo can be described in a natural (verbose) language. If one says that OO concepts benefit from dot notation (non natural language), does that imply that the associated OO concepts are non natural by nature ? Frankly I don't think so : AFAIR OO languages like SmallTalk have been created to promote more "natural" concepts (sorry guys if I use such approximative words & concepts, but I really lack time to develop this)... Anyway, I guess you see the pojt I'm trying to make, and the paradox I'm pointing at... Last but not least : as for the fact that dot notation means better implementation and better performances, with computing power doubling every 18 months, I really wonder how long this will remain an issue... JB _______________________________________________ use-revolution mailing list use-revolution@lists.runrev.com Please visit this url to subscribe, unsubscribe and manage your subscription preferences: http://lists.runrev.com/mailman/listinfo/use-revolution
