[GENERAL] Styles of Asking NPC About Objects

8 views
Skip to first unread message

Eric Mayer

unread,
Jan 14, 2001, 2:52:54 PM1/14/01
to
Hi,

I've been pondering how to handle asking an NPC about objects in an IF
game. Since I don't have a really extensive background in IF I'm not
sure what approach is most used/preferred/disliked etc so I thought
I'd. . .er. . .ask about it.

I can think of three major approaches:

1. The simplest would seem to be that you ask the NPC about an object
and the NPC knows one thing about the object and keeps telling you the
same thing no matter how many times you ask.

This would seem to have the advantage that if the NPC is telling you
something you need to know, but don't pick up on the first time, you
can keep asking until you understand what's going on.

There is a "problem" in that the player can't be sure that the NPC has
exhausted its store of knowledge, since it never says so, but I would
think that most players would assume, safely, that once an NPC starts
repeating itself it has run out of information. (Is that usually the
case?)

Also, I suppose such one fact NPCs might seem a little shallow, or is
this an acceptible convention?


2. The NPC might know more than one thing about an object, run through
its known facts, in order, with each new "ask about" and then tell
the player it has nothing more to say. (or just start repeating the
last bit of information it gave)

This seems more realistic. However, is it usual for players to try
asking more than once to see if the NPC has more information. I tend
to but can an author assume this?

Also, it would seem to present a problem if the NPC says something the
player needs to know but then does not repeat it.

3. The NPC has more than one thing to say about an object but doesn't
necessarily offer the information in any given order but rather
according to the state of the game, what's been asked already, its own
mood etc etc

This is most realistic but aside from presenting the writer with
difficulties (my including this variation in my question is purely
theoretical !) it does present the player with obvious difficulties in
getting at information, knowing when the NPC has dried up etc.

I realize there's endless variations one can work and my feeling is
that for a puzzle oriented game NPCs with one thing to say are OK
(because you can always review what they've said to find any clues in
it) while in a story oriented game, even though you can't review
what's been said, somewhat more knowledgeable NPCs might be better.

But I'd appreciate any thoughts.

Best,

Eric
--
Eric Mayer
Web Site: <http://home.epix.net/~maywrite>

"The map is not the territory." -- Alfred Korzybski

Sean T Barrett

unread,
Jan 14, 2001, 3:38:48 PM1/14/01
to
Eric Mayer <emay...@epix.net> wrote:
>1. The simplest would seem to be that you ask the NPC about an object
>and the NPC knows one thing about the object and keeps telling you the
>same thing no matter how many times you ask.

I'd be cautious about drawing a story/puzzle-game dichotomy
when answering this question, but it does seem a bit natural:
if the NPC responses are needed to advance (the game or the
plot or whatever), then making it apparent that you've gotten
all the information is very important; if the NPC response is
just supplying information or atmosphere that isn't needed to
advance, things are more flexible.

>Also, I suppose such one fact NPCs might seem a little shallow, or is
>this an acceptible convention?

Characters that are only willing to tell you one fact about an
object don't need to seem shallow; they simply think that's the
only thing worth discussing about the object. They may seem
artificially repetitive if they say the same exact words to
reveal the one fact; this can either be accepted as a convention,
or you can write the character to say the same thing different
ways if you want them to sound more alive and less like
console RPG signposts; for example, I wrote this to demonstrate
my "serial printing" modification for Inform:

>ASK Q ABOUT WATCH
Q says, "That 'watch' is an exquisitely compact timed explosive,
007. Turn the minute hand to the number of seconds you want, the
hour hand to the number of minutes, shake it, push the alarm
button, and it's activated. Push the button twice rapidly to
deactivate it. Once it's activated, changing the time won't affect
the actual delay before it goes off."

>ASK Q ABOUT WATCH
Q says, "Do pay attention, 007. The minute hand is seconds, the
hour hand is minutes; shake the watch and push the button to arm
it; push the button twice to disarm it."

>ASK Q ABOUT WATCH
Q explains again how the watch is a timed explosive, that minutes
and hours on the watch represent seconds and minutes until it goes
off, that shaking it and pushing the button activates it, and that
pushing the button twice in a row deactivates it.

>ASK Q ABOUT WATCH
An exasperated Q explains yet again how the watch is a timed
explosive, that minutes and hours on the watch represent seconds and
minutes until it goes off, that shaking it and pushing the button
activates it, and that pushing the button twice in a row deactivates it.

>ASK Q ABOUT WATCH
...same thing...

I believe that shifting out of quoted text and into narration makes
the repetition more forgiveable, since one can imagine the character
is actually saying different things; it's the "adventure game narrator"
who's becoming repetitive, but the narrator's always repetitive with
e.g. room descriptions (except in My Angel).

[saying different things on each query]


>This seems more realistic. However, is it usual for players to try
>asking more than once to see if the NPC has more information. I tend
>to but can an author assume this?

Probably not; but then again, perhaps it's possible to train players
in this; for example, the first NPC encountered can say something
on the order of "If you wish to know more, ask again".

>I realize there's endless variations one can work and my feeling is
>that for a puzzle oriented game NPCs with one thing to say are OK
>(because you can always review what they've said to find any clues in
>it) while in a story oriented game, even though you can't review
>what's been said, somewhat more knowledgeable NPCs might be better.

I don't know; I kind of feel like drawing that line is a cop out; to
me, the interesting side of IF is when there's a good story, but it's
significantly interactive; i.e. a story-oriented puzzle game... in
which case what's the compromise?

SeanB

ems...@my-deja.com

unread,
Jan 14, 2001, 6:58:00 PM1/14/01
to
In article <3a61f829...@newsserver.epix.net>,

emay...@epix.net (Eric Mayer) wrote:
> I can think of three major approaches:
>
> 1. The simplest would seem to be that you ask the NPC about an object
> and the NPC knows one thing about the object and keeps telling you the
> same thing no matter how many times you ask.

I tend to be of the opinion that if you are putting in an NPC *only* to
provide a single piece of information, that information could be
provided in some other way. For one thing, such an NPC will either a)
seem very shallow or b) have a lot of resource-draining code attached to
making him interesting that doesn't go into telling the player anything.

> This would seem to have the advantage that if the NPC is telling you
> something you need to know, but don't pick up on the first time, you
> can keep asking until you understand what's going on.

Another option for review might be to allow the player to remember what
the NPC said without having to re-ask the character. (As someone on
this group suggested in another context.) This would have the advantage
that the knowledge would be portable: to branch off Sean's example you
could then

>THINK ABOUT WATCH

at the critical moment, even if Q is long departed from the scene, and
get

You recollect that to activate the watch you must... (blah blah).

> There is a "problem" in that the player can't be sure that the NPC has
> exhausted its store of knowledge, since it never says so, but I would
> think that most players would assume, safely, that once an NPC starts
> repeating itself it has run out of information. (Is that usually the
> case?)

See, in my dream world, the NPCs *never* obviously exhaust their full
store of knowledge.

I hear the howls of rage. But okay; here's my point. What we're
basically talking about here is how to write a game in which a puzzle
(or the main puzzle) is knowledge-based: the player must obtain enough
information to do something.

In classic-model IF, the player just wanders around asking every NPC he
sees (of which there are, however, probably not that many) until he gets
an answer. This is a simple model from a writing perspective, and has a
kind of pleasant obviousness from the player's, as well. There might be
subpuzzles where it becomes clear that one of the NPCs has necessary
information but that you must first bribe/trick/compel him to tell it to
you.

Arguably in a more realistic world model, especially in situations where
one is investigating a complex situation or concept, there would not be
a one-to-one correspondence of informers and information, but a group of
people with a bunch of knowledge some of which is held in common and
some of which can be acquired through non-NPC means. This leads to a
non-linear, multiple-solution-path kind of thing: you can either spend a
while looking up and consulting the leather-bound book in the library OR
sweettalk the crotchety old professor into sitting you down for a little
chat OR piece the answer together by getting pieces of it from several
of his less-well-informed but more friendly proteges.

The point here is to discourage the player from regarding the NPC as a
kind of quirky puzzle box that must be "worked" until it has revealed
all its information. Preferable would be for the player to play by
doing things that make sense within the logic of the game world (ie,
finding out that crotchety professor knows stuff, asking other NPCs
about how to get to him, etc), returning to people when he has reason to
believe that they have more information.

Combined with an adaptive system where the player is directed towards
information he hasn't discovered yet and the potential to recall past
info, it seems as though this needn't be too difficult from the player's
perspective.

Then: give NPCs as many answers to the same question as you like and
after that point disable repetitive questioning, either by a menu system
that loses that question at the relevant point, or by having the NPC
react by saying some variation of "I've told you all I'm going to say
about that," or whatever.

ES

--


Sent via Deja.com
http://www.deja.com/

Sean T Barrett

unread,
Jan 15, 2001, 12:12:35 AM1/15/01
to
In article <93tee5$7b3$1...@nnrp1.deja.com>, <ems...@my-deja.com> wrote:
>Arguably in a more realistic world model, especially in situations where
>one is investigating a complex situation or concept, there would not be
>a one-to-one correspondence of informers and information, but a group of
>people with a bunch of knowledge

Ok, I can't wait to see the comp01 game where you play an IF author
investigating the complex situation of how best to handle NPC responses,
and one of the possible ways of acquiring the info to solve the problem
involves posting about it to Usenet...

SeanB

Eric Mayer

unread,
Jan 15, 2001, 3:35:44 PM1/15/01
to
On Sun, 14 Jan 2001 23:58:00 GMT, ems...@my-deja.com wrote:

<snip>


>Another option for review might be to allow the player to remember what
>the NPC said without having to re-ask the character.

As I said in my reply to Sean, thanks very much. This is yet another
idea I hadn't thought about.
>
<snip>

>
>The point here is to discourage the player from regarding the NPC as a
>kind of quirky puzzle box that must be "worked" until it has revealed
>all its information. Preferable would be for the player to play by
>doing things that make sense within the logic of the game world (ie,
>finding out that crotchety professor knows stuff, asking other NPCs
>about how to get to him, etc), returning to people when he has reason to
>believe that they have more information.
>

Very useful discussion. I guess it might be safely said, that the same
rule applies to IF as to non-IF writing, that the easiest or most
superficial way to write something usually isn't the best!
(unfortunately...)

>Combined with an adaptive system where the player is directed towards
>information he hasn't discovered yet and the potential to recall past
>info, it seems as though this needn't be too difficult from the player's
>perspective.
>
>Then: give NPCs as many answers to the same question as you like and
>after that point disable repetitive questioning, either by a menu system
>that loses that question at the relevant point, or by having the NPC
>react by saying some variation of "I've told you all I'm going to say
>about that," or whatever.
>
>

This is what I'll have to go away and mull. Although I do want to put
in a caveat -- I hope that not every IF author feels that they can't
release a game until the characters are all as fully realized as
Galatea otherwise there's going to be very little new IF :)

Eric Mayer

unread,
Jan 15, 2001, 3:35:43 PM1/15/01
to
On Sun, 14 Jan 2001 20:38:48 GMT, buz...@world.std.com (Sean T
Barrett) wrote:

<snip>

> I wrote this to demonstrate
>my "serial printing" modification for Inform:
>
> >ASK Q ABOUT WATCH
> Q says, "That 'watch' is an exquisitely compact timed explosive,
> 007. Turn the minute hand to the number of seconds you want, the
> hour hand to the number of minutes, shake it, push the alarm
> button, and it's activated. Push the button twice rapidly to
> deactivate it. Once it's activated, changing the time won't affect
> the actual delay before it goes off."
>

<snip>


>
> >ASK Q ABOUT WATCH
> ...same thing...

I had not thought of that, thanks. I really appreciate you and Emily
giving me some stuff to chew on. I had been turning some
possibilities over and realized I needed some nudges.


>
> I kind of feel like drawing that line is a cop out; to
>me, the interesting side of IF is when there's a good story, but it's
>significantly interactive; i.e. a story-oriented puzzle game... in
>which case what's the compromise?
>

Yes, you're right. Just thought I'd see if I could find an excuse to
make it easy on myself :)

J.D. Berry

unread,
Jan 15, 2001, 9:57:52 PM1/15/01
to
In article <3a635ccf...@newsserver.epix.net>,
emay...@epix.net (Eric Mayer) wrote:

> This is what I'll have to go away and mull. Although I do want to put
> in a caveat -- I hope that not every IF author feels that they can't
> release a game until the characters are all as fully realized as
> Galatea otherwise there's going to be very little new IF :)

Especially keep this in mind for Smoochie Comp. :-) I plan to submit a
game that does keep track of what was discussed (and sometimes shown),
but sheesh, with Christmas holidays and what not, it's not like I've
given it the love and attention it deserves.

"Feared" review start: "I know thes games were supposed to be short,
and I know the authors didn't have a LOT of time..." (wait for it)
"BUT..." ("No! No buts, not the BUT Anything but the BUT." "Yes,
the but.")

Jim "Be Gentle" Berry
;-)

Giesbrecht, Joshua [CAR:9F52:EXCH]

unread,
Jan 16, 2001, 9:42:37 AM1/16/01
to
ems...@my-deja.com wrote:

> Another option for review might be to allow the player to remember what
> the NPC said without having to re-ask the character. (As someone on
> this group suggested in another context.) This would have the advantage
> that the knowledge would be portable: to branch off Sean's example you
> could then
>
> >THINK ABOUT WATCH
>
> at the critical moment, even if Q is long departed from the scene, and
> get
>
> You recollect that to activate the watch you must... (blah blah).
>

This is a verb you'd have to train the player to use, but it's a great
idea. It brought to mind that something similar could be used when
actually asking the NPC for the bazillionth time:

> ASK Q ABOUT WATCH

"You start to ask Q about this watch you're wearing, but recall that he
already explained it to you. The minutes and seconds on the watch count
down how long you have before the explosive detonates, shaking it and
pushing the button arms it, and pushing the button twice disarms it. No
sense bugging Q about it again when all you have to do is think about it."

It's sort of cheating, in that it forces the player to be in character
(Bond would never pester Q about a device repeatedly, he barely listens the
first time!), but it does keep the NPC from being repetitive.

- josh g.


John Colagioia

unread,
Jan 17, 2001, 9:11:24 AM1/17/01
to
Again, for those who don't know, I'll be your idiot for the evening,
having posted this to the wrong thread, and now trying to put this in
the right spot. Whoops...

ems...@my-deja.com wrote:
> In article <3a61f829...@newsserver.epix.net>,
> emay...@epix.net (Eric Mayer) wrote:
> > I can think of three major approaches:

[...]


> > There is a "problem" in that the player can't be sure that the NPC
has
> > exhausted its store of knowledge, since it never says so, but I
would
> > think that most players would assume, safely, that once an NPC
starts
> > repeating itself it has run out of information. (Is that usually
the
> > case?)
> See, in my dream world, the NPCs *never* obviously exhaust their full

> store of knowledge.
> I hear the howls of rage.

Maybe it's because I'm the new guy, but I don't really see how such a
thing could be *bad*. Hard to program, sure. Hard to cram into a
Z-Machine, probably. But not a bad thing at all.

> But okay; here's my point. What we're
> basically talking about here is how to write a game in which a puzzle

> (or the main puzzle) is knowledge-based: the player must obtain
enough
> information to do something.

[...]

Hmmm...This is actually a very interesting idea. Programming-wise, I
would imagine that this could be done with a "table" of
information--basically, interconnected phrases, or something (which
wouldn't be more than a mild nuisance in Inform, I think). The
professor, of course, would have nearly all the information at hand
(maybe missing some bits). Each book may contain pointers to a handful

of these notes. And the "grad students" would similarly have pointers
to an assortment.

Which is interesting, in a way, because this means that the students
could possibly be generated on-the-fly (by generating a new sub-table)
from the same game object, so that the player can always find someone
new--specifically, someone who reacts slightly differently to everyone
who has come before--to fill in that final bit of knowledge, or even to

replace a previous student who has been horribly offended by something
you've said or done to him.

I suppose that using such a scheme *might* even lead to NPCs that could

do some of the research for you, by allowing them to select random
books and accumulate those bits of knowledge until you come pick them
up.

I really like this!

Ms. Short, between Smoochie Comp and this idea, you're now my official
hero...at least for a couple more days.


ems...@my-deja.com

unread,
Jan 17, 2001, 2:52:12 PM1/17/01
to
In article <3A645DDC...@pretendaddress.com>,

"Giesbrecht, Joshua [CAR:9F52:EXCH]" <jo...@pretendaddress.com> wrote:

> It's sort of cheating, in that it forces the player to be in character
> (Bond would never pester Q about a device repeatedly, he barely
listens the
> first time!), but it does keep the NPC from being repetitive.

No, it is not at all cheating. There's nothing wrong with hijacking
the player's input on the way to being carried out, if you don't feel
like dealing with all the implications. Many games also refuse to
carry out instructions like >JUMP OFF CLIFF and >BURN BRIDGES.
Basically I approach it thus:

a) Have the NPC provide as many varied reponses as is reasonable. The
number may be (1).
b) Then prevent the player from asking again [adaptive menus that lose
the relevant option, and/or some kind of refusal on the part of the
parser] or
c) switch to meta-conversation: have the NPC react to the repetition
rather than to the topic. This latter is perhaps not productive in a
game where the NPCs exist primarily to provide clues to puzzles. YMMV.

ES

ems...@my-deja.com

unread,
Jan 17, 2001, 3:00:39 PM1/17/01
to
In article <3a635ccf...@newsserver.epix.net>,

emay...@epix.net (Eric Mayer) wrote:
> This is what I'll have to go away and mull. Although I do want to put
> in a caveat -- I hope that not every IF author feels that they can't
> release a game until the characters are all as fully realized as
> Galatea otherwise there's going to be very little new IF :)

I'd be the first to admit that it is not, in fact, desirable that they
should be so. There seems to be a trade-off between depth of
characterization and complexity/sturdiness of plot: the more
complicated a story you have to tell and the more you want to make a
given outcome inevitable, the less room there presumably is for erratic
NPC behavior.

That said:

1) It is reasonable to limit what the player is allowed to do and say,
for instance by making the PC one who wouldn't knowingly be
offensive/crass/forward/whatever.
2) It is reasonable to make NPCs who are by social situation unlikely
to have strong emotional responses or to burden the player with their
life stories (eg., checkout people at the supermarket).

Total mimesis is impossible. Select the degree thereof that suits your
abilities as a designer and the nature of the subject at hand.

Martin Julian DeMello

unread,
Jan 18, 2001, 1:38:06 AM1/18/01
to
ems...@my-deja.com wrote:

> c) switch to meta-conversation: have the NPC react to the repetition
> rather than to the topic. This latter is perhaps not productive in a
> game where the NPCs exist primarily to provide clues to puzzles. YMMV.

This might be less obstructive if you add a RECALL command.

> ASK MAN ABOUT PSMITH

"That's all I know about him," he says.

> ASK MAN ABOUT PSMITH

The fat man gives you an annoyed look and returns to his newspaper.

> RECALL ABOUT PSMITH

Recalling conversations about Psmith

[Mike]
.
.
.

[Fat Man]
.
.
.

Sort of a filtered log summary. Of course, you could try wrapping it in text
of the 'you recall Mike saying' variety so as not to break mimesis too
badly, but the way I look at it it'd be more a convenient tool to avoid
having to manually scroll through a game log than an actual part of the
game.

--
Martin DeMello

Magnus Olsson

unread,
Jan 18, 2001, 9:28:20 AM1/18/01
to
In article <944t57$esd$1...@nnrp1.deja.com>, <ems...@my-deja.com> wrote:
>a) Have the NPC provide as many varied reponses as is reasonable. The
>number may be (1).
>b) Then prevent the player from asking again [adaptive menus that lose
>the relevant option, and/or some kind of refusal on the part of the
>parser]

This is of course mimetic, but it's not very player-friendly, since it
essentially forces the player to keep notes of all the potentially
important things the NPC has said earlier but refuses to repeat. (or,
even worse, which the game refuses you to even ask about one more
time).

Perhaps the ideal is having the NPC react to the repetition, but
repeat the information anyway (as in the example with Q and the
explosive watch). Or the game could in some way keep track of all
information you've received.

--
Magnus Olsson (m...@df.lth.se, m...@pobox.com)
------ http://www.pobox.com/~mol ------

ems...@my-deja.com

unread,
Jan 18, 2001, 10:14:22 AM1/18/01
to
In article <94630e$fp6$1...@joe.rice.edu>,

Martin Julian DeMello <mdem...@kennel.ruf.rice.edu> wrote:
> ems...@my-deja.com wrote:
>
> > c) switch to meta-conversation: have the NPC react to the repetition
> > rather than to the topic. This latter is perhaps not productive in
a
> > game where the NPCs exist primarily to provide clues to puzzles.
YMMV.
>
> This might be less obstructive if you add a RECALL command.

and Magnus Olsson wrote:

>This is of course mimetic, but it's not very player-friendly, since it
>essentially forces the player to keep notes of all the potentially
>important things the NPC has said earlier but refuses to repeat. (or,
>even worse, which the game refuses you to even ask about one more
>time).

Right. cf. my comments on a THINK ABOUT command (which would be
essentially the same thing) elsewhere in this thread:

***

Another option for review might be to allow the player to remember what
the NPC said without having to re-ask the character.  (As someone on
this group suggested in another context.)  This would have the advantage
that the knowledge would be portable: to branch off Sean's example you
could then
 
>THINK ABOUT WATCH
 
at the critical moment, even if Q is long departed from the scene, and
get
 
You recollect that to activate the watch you must... (blah blah).

***

One potential advantage here is that you can also use this to provide
backstory to the player about things that his character already knows
(cf. Kathleen Fischer's _The Cove_ -- I think the technique might have
been better telegraphed to the player here, but the basic idea was an
interesting one.)

Ultimately, though, I think there's no single right answer about how to
do this. There's nothing really wrong with having the NPC repeat
information (other than that having her rephrase the same information in
multiple ways makes a lot of work for the author); it's more realistic,
character-wise, if the PC is a little kid, for example, and the NPC her
patient grandma, than if they're both highly trained operatives on a
secret mission where they need to talk as little as possible. (Note to
self: write game about being inside the Trojan Horse.) Anyway, I
certainly don't want to make life harder for the player just for the
sake of obfuscation.

My general goal is to prevent the player from regarding all the NPCs as
glorified parrots. (Unless, of course, they are supposed to BE parrots.
[Here's an idea for solving the recapitulating-information problem: put
a {non-detachable} parrot on the shoulder of the player character, who,
whenever the player asks about a keyword the parrot already heard,
interrupts with a repetition of the earlier statement. This requires
that all future IF be set on pirate ships, but hey, that's not such a
big loss.

Daryl McCullough

unread,
Jan 18, 2001, 3:56:56 PM1/18/01
to
m...@pobox.com (Magnus Olsson) says...

>
>In article <944t57$esd$1...@nnrp1.deja.com>, <ems...@my-deja.com> wrote:
>>a) Have the NPC provide as many varied reponses as is reasonable. The
>>number may be (1).
>>b) Then prevent the player from asking again [adaptive menus that lose
>>the relevant option, and/or some kind of refusal on the part of the
>>parser]
>
>This is of course mimetic, but it's not very player-friendly, since it
>essentially forces the player to keep notes of all the potentially
>important things the NPC has said earlier but refuses to repeat. (or,
>even worse, which the game refuses you to even ask about one more
>time).

A nice feature of "Return to Zork" (which I liked a lot, in contrast
to many people) was that it allowed the player to take photographs
of scenes and to record conversations. For text games, there would
be no point in either of these, but you could easily have a notebook
that records previous room descriptions and NPC conversations.

--
Daryl McCullough
CoGenTex, Inc.
Ithaca, NY

Daryl McCullough

unread,
Jan 18, 2001, 4:52:34 PM1/18/01
to
ems...@my-deja.com says...

>Right. cf. my comments on a THINK ABOUT command (which would be
>essentially the same thing) elsewhere in this thread:
>
>***
>
>Another option for review might be to allow the player to remember what
>the NPC said without having to re-ask the character.  (As someone on
>this group suggested in another context.)  This would have the advantage
>that the knowledge would be portable: to branch off Sean's example you
>could then

>>THINK ABOUT WATCH

>at the critical moment, even if Q is long departed from the scene, and
>get

>You recollect that to activate the watch you must... (blah blah).

Another possibility would be to have a "notebook" object into which
the player could record room descriptions and NPC conversations.

OKB -- not okblacke

unread,
Jan 19, 2001, 10:32:37 AM1/19/01
to
ems...@my-deja.com wrote:
>[Here's an idea for solving the recapitulating-information problem: put
>a {non-detachable} parrot on the shoulder of the player character, who,
>whenever the player asks about a keyword the parrot already heard,
>interrupts with a repetition of the earlier statement. This requires
>that all future IF be set on pirate ships, but hey, that's not such a
>big loss.

For variety, they could also be set in the ifMUD lounge. :-)

--OKB (Bren...@aol.com) -- no relation to okblacke

"Do not follow where the path may lead;
go, instead, where there is no path, and leave a trail."
--Author Unknown

Reply all
Reply to author
Forward
0 new messages