Re: AppleScript a miserable and utter failure
Re: AppleScript a miserable and utter failure
- Subject: Re: AppleScript a miserable and utter failure
- From: has <email@hidden>
- Date: Wed, 28 Sep 2005 23:47:37 +0100
Jon Pugh wrote:
>In their vein of throwing gasoline on a fire, I'll present a critique
>of something we've all struggled with for years. Terminology
>conflicts.
It's an inherent weakness of the language's syntax design - a reminder that there's no such thing in this world as a free lunch. You can't take dumb character-based source code, throw out a bunch of essential semantic information, perform crude, premature mapping of identifiers to AE codes, pour in an arbitrary amount of global namespace pollution, and then be too surprised if it throws a wobbly now and again. Clever compiler tricks only get you so far; push too hard, throw in too many variables, and you'll soon start discovering the weak points in their brilliant plan.
>A colleague, in discussing this, posits that programmers look at
>AppleScript and see something that falls into "the uncanny valley".
>A monster in human clothing. I think he's onto something.
>
><http://en.wikipedia.org/wiki/Uncanny_Valley>
There's some truth in this naive newcomers assume that because AS code looks like English, it will work like it too. It doesn't, of course: it's a very conventional programming language at heart, with very strict rules about what is and isn't legal code - as these folks quickly and rudely find out. (I'm quite confident this problem is solvable, BTW - just not in the way that AS tries to do it.) That's a completely separate issue to terminology woes, however, which are due to AS's penchant for sloppy sleights-of-hand.
>My take on the fuss is simply that there needs to be a language czar
>who ensures that terminology conflicts are fixed.
The Department of Information Adjustments? I'm sure that will work just swimmingly. Anyway, it's nonsense to blame application and osax authors for something that's entirely AppleScript's fault. (Allowing osaxen to pollute the global namespace was a really, really dumb decision; AS probably would've gotten away with all its clever trickery if it hadn't been for that.) Application/extension authors should be free to use whatever identifiers they want, not made to fight bloody combat to the death over who gets the last non-crappy identifier on some massive, lumbering "Officially Approved" list. I can't think of any other environment that would impose such bloated, ridiculous bureaucracy on contributors and get away with it, and I don't see AppleScript doing so either. It's hard enough getting developers to follow completely sensible design guidelines, never mind bow to the mad whims of some Great Terminology Dictator as well.
The problem is AppleScript's: it should suck it up and stop expecting everyone else to cover for it. It's perfectly normal for two parties to use common names; it's the language's responsibility to provide an environment where this doesn't cause collision problems. For AppleScript users, that means knowing these problems exist and how to minimize/avoid them, or switching to another language that's immune to such troubles if they really can't stand it. Moaning about it just gets terribly tiresome after a while though; yes, we all know the language isn't perfect, get over it.
/2c
has
--
http://freespace.virgin.net/hamish.sanderson/
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Applescript-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden