At 11:35 PM -0500 2000-02-28, Chris Nandor wrote: > >These modules can be dangerous. People SHOULD be scared to the point >where, at least, they won't experiment too much without thinking things >through first. >Well, as John Carmack said (paraphrased), the Zen of Mac programming is in >waiting for your machine to reboot each time you do something wrong. Mac >programming with the toolbox CAN be a lot of non-fun. MacPerl itself >protects you from that, for the most part, until you start accessing the >toolbox. >I don't think it is elitist. I think that Matthias basically saw the real >danger in using these toolbox modules, and would rather warn people about >using them rather than fend off the flames of people who corrupted their >system and lost a week's worth of work. What if instead of providing (essentially) direct access to the Toolbox calls, we had modules that implemented objects, like, say, a Window? Window->new() would call [Get]New[C]Window as appropriate, and when it went out of scope, (it automatically calls DisposeWindow, and) the window goes away. If your script dies, the window goes away. If MacPerl runs out of memory and aborts your script, the window goes away. I suspect most MacPerl users would appreciate the added safety and have no need of finer control (or desire for added responsibility) from calling the Toolbox directly. The primary issue that I see is that a new API is called for, which means someone has to design (as well as implement) it. This is on my list of projects to take on if someone doesn't beat me to it, but there are many projects ahead of it. Is anyone interested in using or developing such a beast? Josh -- Joshua Juran Metamage Software Creations =) Tools for Wizards wanderer@metamage.com <http://www.metamage.com/> * Creation at the highest state of the art * # ===== Want to unsubscribe from this list? # ===== Send mail with body "unsubscribe" to macperl-request@macperl.org