Text output [was "short" Introductory Message (fwd)]

[Matt Chatterley]

On Sat, 7 Jun 1997, Chris Gray wrote:

> [Martin K:]

[lots of snipping of language stuff, etc]

> Strongly agree! I haven't done anything to make changing output, but I've
> gone to a lot of trouble to make output that is as close to correct as I
> can get. The junk that comes out of many MUDs just drives me away. My
> newsfeed has been very bad for nearly all of this year, so I think I've
> missed most of your discussions on this. Can you summarize here?
 
Oh, I should imagine he can (Hi Martin, btw ;).
 
> Here's a question for you, which is one that continues to bug me in my
> own system: How do you handle 'a' versus 'the' for NPC's? Some of my
> NPC's are unique, and so marked by a flag. When they do something, no
> article is used, e.g.:

This is one that I pondered for a while not so long ago.
 
>     Packrat picks up the pen.
> 
> However, for non-unique NPC's, I usually use:
> 
>     An orc picks up a rock.

I decided on this, with a suitable rationale:

Since the player cannot really tell the difference (not on a player level,
but on a character level), between two orcs, and Gobbler the unique orc,
although Gobbler may look different (to a point.. he is afterall, still an
orc). If he looks the same.. they definitely can't, but if he doesn't,
they still have no real reason to mark him as 'unusual' and treat him
separately. To generalise him, he is just an orc. So we have three orcs.

One orc, one rock -> The orc picks up the rock.
One orc, multiple rocks -> The orc picks up a rock.
Multiple orcs, one rock -> An orc picks up the rock.
Multiple orcs, multiple rocks -> An orc picks up a rock.
 
> but there are situations where this isn't appropriate. From the English
> language point of view, one of those is when there is only one orc present,
> but that can be a costly thing to have to determine whenever an orc
> does something. Another situation is when the orc's action is a direct
> result of an action by another character, which has in some way selected
> that orc as temporarily special. Many systems punt on this by just never
> using an article, and get:

Yes. It can get awkward - I'm adding global 'Efuns' to allow the checking
(simply and fairly efficiently) of how many things in a room match a
certain ID, x.
 
>     Orc picks up a rock.
> 
> but I really don't like that. Even worse is:

It's not as bad as some, but it's clumsy, and sounds very computer
generated, if that makes sense.
 
>     orc takes rock
> 
> Ick!

Ack, damnit ;) If I log into a mud and see something similar, I log out
almost immediately.
 
> [Of course, 'a rock' should sometimes be 'the rock', but I don't try to
> handle that one either!]

Heh. I do. I want these fairly simple bursts of text, which are very
common, to be as well set up as possible, it makes the whole thing read
cleaner, and much better.

Also pondering 'output stacking' so that any line which is repeated more
than 3 times is cocencated, for instance; It is very annoying to see:

Giant bee misses Fred.
Giant bee misses Fred.
Giant bee misses Fred.
Giant bee misses Fred.
Giant bee misses Fred.

And so forth. The dilemma is how to *present* the cocencated text. The
above is put into storage as "Giant bee misses Fred.":5, so that we know
what was outputted, and how many times. Then, when the stack is processed,
it's sent to the recepients.

But.. do we send it as:

[x5] Giant bee misses Fred.

or something else? Sending 'Five Giant bees' is not possible, because it
might just be one bee missing five times! Ditto 'Giant bee misses Fred
five times', it might be five bees!

Regards,
	-Matt Chatterley
	http://user.itl.net/~neddy/index.html
"He can't stop us, we're on a mission from Glod!" - Soul Music (Pratchett)