Archive of UserLand's first discussion group, started October 5, 1998.

Re: Outliners, Unix, and Scripting

Author:Jim Moy
Posted:3/5/1999; 11:40:12 AM
Topic:Frontier and PIM's
Msg #:3660 (In response to 3641)
Prev/Next:3659 / 3661

Expanding on Dave's comments, I think a standalone outliner is focused on the nodes of the outline, rather than on the characters and formatting like in a WP. I've never been comfortable in embedded outliners because I'm always doing something like backspacing over some formatting boundary which tells the code that's faking the outlining that it's a node, and suddenly my concentration is gone because my outline is messed up.

Anyways, you're right, when Unix folk talk to Mac folk about scripting, they're not talking about the same thing. I'm not sure if this helps, but when Mac folk are talking about scripting, they mean having the app present an object model externally, to be manipulated by the scripting language, usually in some object or pseudo-object oriented way. Unix scripting has always seemed to be about munging text in some way or another, though that seems to be changing with the object features of Perl & Python.

You write a script on Unix that invokes utilities which typically go away after they've done a single-purpose job. On Mac you launch apps which stick around over many invocations of verbs on nouns, and then you tell it to quit. The typical AppleScripter or Frontierite can easily manage this process. On Unix, it's hard to do something other than just chain commands together in a shell-style pipe, and they all quit when they've processed all their stdin/stdout. Even in Perl you've gotta manage pipes and forks and all that, so the whole idea of having something sitting out there waiting for object requests is foreign.

Rambling, my apologies...

There are responses to this message:

This page was archived on 6/13/2001; 4:48:26 PM.

© Copyright 1998-2001 UserLand Software, Inc.