J C Lawrence claw at under.engr.sgi.com
Thu Apr 9 17:24:49 New Zealand Standard Time 1998

On Wed, 8 Apr 1998 23:03:19 PST8PDT 
maddog <maddog at best.com> wrote:
> J C Lawrence wrote:

>> The first problem is that any such system is based on a core set of
>> mechanical assumptions -- and of course they do not translate to
>> systems which don't use those assumptions, or which base on
>> (entirely) different sets of principles.
>> Next up, given a set of systems which allow conversions between
>> them, say, A, B, C, and D, it is going to be difficult (if not
>> impossible) to do such convserions without data loss (A->B->C->D->X
>> where X is _not_ A, and worse when X translates to B etc he is
>> different yet again in a new series of oddities) without requiring
>> the base models (if not the systems themselves) of each of the
>> systems to be identical.

> In my previous post I talked about the universal system (X) and two
> systems A,B with the ability to do A->X->B.  According to WotC, they
> didn't recommend this method for their Envoy.  They felt it was
> better to do A->B and eliminate X.  I think this also has
> problems. If someone invents a low volume system C, no one is likely
> to convert to their system, so they have to write converters for
> A->C and so on. In a multiverse you have the problem of having to
> account for every possible system now and in the future.

Good point.  The problem however is simpler than that and has to do
with data loss, not permutations. 

If a character definition is translated from GUMPS to a foreign system
and back again, will the resultant character definition be identical
to the one before the translation?  Always?

Now consider the point of a character QQQ with attributes peculiar to
non-GUMPs system XXX which *don't* map to equivalents under GUMPS.  If 
that character definition is translated into GUMPS and then back
again, is it unchanged?  

Next, given that same character QQQ under system XXX, if it is
translated into GUMPS and then into system YYY which has no equivalent
for XXX's peculiarities (much as GUMPS doesn't), and that YYY
character is then translated back into GUMPs and thence to XXX, will
it remain the same character?

And lastly, Given the same scenario except that YYY now also has
different attributes not represented nateively under GUMPS or XXX,
during the translation XXX->GUMPS->YYY->GUMPS->XXX will the resultant
character preserve or lose the unique characteristics of YYY's

>> All of this can of course be worked around -- but it will be at the
>> cost of a large degree of homogeneity in all the supporting
>> systems.  Is that cost worth it?  Is it worth it this early in the
>> field's development?  I would argue that the field and the
>> technology (tightly coupled) are not rich enough to support such
>> standardisation moves yet.

> Maybe this is the time.  I think that most hypertext people feel
> that HTML is a really weak form of hypertext. Yet the simpler HTML
> has changed the way we all deal with computer systems.  MUDs are
> very good as islands of entertainment, but they not good at creating
> a vast, seamless universe.  The Web browsers with their kludgey URLs
> are many times smoother than a mud client.

Could be.  HTML and XML solve a a significant and common problem which
span multiple disciplines.  I have trouble seeing virtual character
translators similarly.  Yet.  Given a more mature and invasive
technology, yes the problem will exist in sufficient strength to be
resolvable.   No, I don't think so.

Think:  Would the Web have taken off back in 1985?  Why not?

>> There are obvious problems with attempting to translate physical
>> objects between systems:
>> Bubba has a gold coin.  It is not worth much in world/game A.  He
>> portals to world/game B.  Gold is incredibly rare and valuable
>> there.  Bubba is instantly as rich as Croecus.  Bubba does the
>> standard arbitrage trick and repeats the trip, this time with a
>> wagon full of gold...
>> Okay, you could diminish the gold repsective to its value in the
>> relative worlds.  But his is a mock solution as the value may be
>> multiplexed or otherwise complex (the valued items are the steel
>> nails holding the ship together (ignore the possibility of wooden
>> pegs)).
> Agreed.  I don't have an answer for this.

Understood.  I don't think there is an answer which isn't riddled with
special cases.  Any rote rule will attract exceptions faster than
pickles to a pregnant woman.  <kof> (Interestingly enough my wife
while pregnant only ever had cravings for eggs and bananas, both of
which she normally doesn't like)

>> Bubba has the same gold coin, and this time travels to world/game C
>> which takes place undersea with fish-like bodies.  Gold there is a
>> powerful toxin.  The gold is rapidly eroded by the acidic water and
>> everybody dies...
> This is not good. 



<<Visions of a vast sea of bobbing white bloated player corpses
outside the OOC lounge window while hordes of irate non-virtual
players lynch Bubba...>>

>But didn't the early explorers on our own planet
> end up spreading a toxin to other cultures?  Gold or microbe-- same
> problem.

>> Thus your virtual characters -- BUT the same problem exists if the
>> translations are done for mechanical factors:
>> Bubba originates from a high gravity world, wanders off to a
>> low-grav world and suddenly finds that he can whup Tiamat using
>> only his left nostril.
> Looks like bad news for Tiamat, but it certainly is possible.

Okay, but what about arbitrage loops?

  Bubba leaves the difficult HeavyWorld, goes to WimpyWorld, sneezes a
couple times, slaughters the entire workd population, and returns to
HeavyWorld a massively buffed Dude.

Yes, you can attempt pro-rate stats across the translation.  The
problem is that attempting to pro-rate stat's cross-effects between
multiple translations is a nightmare.

>> Okay, so you scale strength and the rest of it per the standards of
>> the old and new world:
>> Bubba is a rougueish over-muscled barbarian old fart.  He's not the
>> fastest or strongest on the block, but his supplemental stats
>> largely ensure that he always comes out of the cauldron alive.
>> (Translation: extremely high experience based stats).  Bubba moves
>> to a new world.  His various stats are scaled appropriately so that
>> he is still an over-muscled old-fart.  However the new world is
>> entirely based on puzzle solving, not brute force so his old
>> experience stats count for naught.  Or do they?  Should they be
>> scaled appropriately, or should he be the babe-in-the-woods?

> I asked maddog about this.  He said, "I see the monster.  I wack the
> monster with my sword.".  So essentially he wasn't much help with
> this question.


>> Whichever side you pick of the last one you're going to be wrong
>> some of the time.

This is the critical bit.  No matter which side of the fence you
decide to adopt, its trivially easy to argue strongly that its the
wrong side.  This isn't to say that either side is "wrong" -- just
that you'd better be prepared to support it against T'ed off players
who found the rule operating against them.

J C Lawrence                               Internet: claw at null.net
(Contractor)                               Internet: coder at ibm.net
---------(*)                     Internet: claw at under.engr.sgi.com
...Honourary Member of Clan McFud -- Teamer's Avenging Monolith...

More information about the MUD-Dev mailing list