Newsgroups: rec.arts.int-fiction
Path: news.duke.edu!newsgate.duke.edu!nntp-out.monmouth.com!newspeer.monmouth.com!europa.netcrusader.net!152.163.239.129!portc01.blue.aol.com!feed.newsreader.com!uunet!ash.uu.net!world!buzzard
From: buzzard@world.std.com (Sean T Barrett)
Subject: Re: How do you work?
Message-ID: <G8xL51.MxM@world.std.com>
Date: Sun, 18 Feb 2001 02:19:49 GMT
References: <3A8DE985@MailAndNews.com>
Organization: The World Public Access UNIX, Brookline, MA
Lines: 58
Xref: news.duke.edu rec.arts.int-fiction:83615

Kathleen M. Fischer <greenGargoyle@MailAndNews.com> wrote:
>For the most part it avoids it, because such low level routines are only
>written to support higher level text. You need it when coding, you don't
>need it when deciding what the table should look like. It's nitpicky code
>like that which quickly make an object a fixture in the game, whether you
>need it or not,

Of my three works in progress:
   1. very unfinished, writing the transcript a few rooms
      ahead of the implementation at a time
   2. in final beta, wrote a transcript of the opening sequence,
      the first room description, and then just implemented the
      whole thing
   3. very unfinished, wrote a transcript without doing any
      coding at all, missed the smoochiecomp deadline and
      abandoned it

Despite the fact that the only successful one is the one where I didn't
go transcript based, I believe in the transcript approach as well (there
are other factors not described above that influence the failure/success
of each project).  I believe in it both for the reasons you give, and
because I've found that there's a different mind-frame necessary to
programming versus writing English text; I find it more effective not
to have to swap constantly back and forth between them.  This was even
more true with my item #1, which is written in a language that is
not my primary one, and thus is very slow to even just write as
a transcript.

(I suppose I should argue there are really three separate modes:
designing, writing, and coding, and debugging, FOUR modes...)

>"You see a child/man/boy/woman/space alien
>creeping/crawling/walking/slithering across the grass."
>
>It's quick and easy to write, easy for me to proof read, and later I'll turn
>it into the appropriate code... something roughly like:
>
>print "You see a ", random ("child", "man", "boy", "woman", "space alien");
>print " ";
>print random ("creeping", "crawling", "walking", "slithering");
>print " across the grass.";

Of course, if you used my varying strings compiler extension,
you could just code:

  "You see a {%child|man|boy|woman|space alien}
  {%creeping|crawling|walking|slithering} across the grass."

or perhaps even better

  "You see a {!child|man|boy|woman|space alien}
  {!creeping|crawling|walking|slithering} across the grass."

which prevents the random numbers from generating the same
text twice in a row.

SeanB
but I'm not sure why I'm plugging a compiler mod I can't release
