Introductions and

[s001gmu at nova.wright.edu]

On Tue, 9 Dec 1997 coder at ibm.net wrote:

> 
> On 18/11/97 at 07:47 PM, Mike Sellers <mike at online-alchemy.com> said:
> 
> is that this sort of problem descends to the minor style.  All name
> collisions are local -- which makes name space viruses something to fear
> and cultivate.
> 
> >player name internally.  The only problem comes up in areas where, for
> >example, I want to talk to "Joe" or send game-mail to "Joe."  If there
> >are a dozen Joes out there, how do I know which one I want?  Even in the
> >scheme described above, you could have several "Joe Doe" characters.  I
> >suppose you could list further, publicly known information for each, such
> >as occupation, general area where they live ("Joe Doe of Tos, who is a
> >merchant"), or in a graphical environment, a picture of them -- anything
> >to help other players differentiate their Joe from all other Joes.
> 
> >Thoughts on this?
> 
> Because I expect namespace viruses to become very common, I've been

can you give a definition for 'namespace virus' please?  I'm having 
problems pickup one up from the context clues. :}

> thinking about this one.  I currently have two types of mappings:  a name
> string that maps to an internal unique ID value, and a name string that
> maps to a list of such ID values (ie individuals and groups).  Creating a
> group is a special process as compared to creating an individual name
> assignment.  In the general case I have newer name assignments override
> ones, thus:
> 
>   > l
>   There is a troll here.
>   > name troll Bubba.
>   > l
>   Bubba is here.
>   An elf walks in.
>   > name elf Bubba
>     // I'm in debate over whether to warn the character over the prior 
>     // existance of the Bubba name assignment, or to prompt directly into 
>     // a Bubba group name assignment.  Currently I do the latter, but 
>     // allow replacement for:
>   > l
>   Bubba is here.
>   There is a troll here.

mmm.. but what if they really are both named Bubba?  I personally know 
several Mikes.  I don't concieve of them as a collective group called 
Mike, and I don't ignore all but the most recent one's name.  They are 
all different ppl with the same name, Mike.  Granted, they have different 
last names.  Do you allow for that, or just have one name ppl can assign 
during an introduction?
 
> What I'm thinking of doing is creating name stacks, such that there would
> be no warnining or alert of the name collision above.  Thus "Bubba" would
> end up refering to the elf, but could be prompted for earlier assignments:
> 
>   > l
>   Bubba is here
>   There is a troll here.
>   > l at Bubba?  // The '?' request assignment stack expansion
>   Do you want:
>     1) Bubba the elf (current default).
>     2) Bubba the troll.
>     0) Cancel command?
>   >> ...etc...

this seems to be an inconcistancy within the system.  I see Bubba and a 
troll.  Yet, if that troll was the one I named Bubba, when I get to the 
list and type 2, I look at the troll that I previously didn't recognize 
as Bubba the troll?  I don't know him, and then I do? 

> Ideally the list would be keyed by most recent reference, most significant
> interaction with character etc rathe than simple race.

I think this idea solves the problem... just show (for the above case):

>l
You see:
Bubba the elf (or other defining characteristic)
Bubba the troll (...)

This ties in nicely with another thread which appears to have died off, 
about determining what characteristics of a character are most striking, 
to generate a decent description when you don't know the person's name 
(IE: a man with striking blue eyes is here).  you can apply the same 
algorithms to the subset of Bubbas the character knows and tag them by 
their striking characteristics (IE: Bubba w/ blue eyes, Bubba the Troll, 
etc).

I'd be tempted to take it a step further and allow the player to override 
the later descirption with a surname of sorts... so instead of getting 
Bubba the troll with a large nose, the player could just assign Bubba 
Bignose or somesuch.

-Greg