[MUD-Dev] strong encryption for authentication

[Caliban Tiresias Darklock]

On Wed, 18 Jul 2001 10:54:00 -0400, Edward Glowacki
<glowack2 at msu.edu> wrote:
> Quoted from Caliban Tiresias Darklock on Mon, Jul 16, 2001 at
> 10:26:19AM -0700:
>> On Fri, 13 Jul 2001 10:12:07 -0400, Edward Glowacki
>> <glowack2 at msu.edu> wrote:
>>> Quoted from Caliban Tiresias Darklock on Wed, Jul 11, 2001 at
>>> 08:29:12PM -0700:
>>>> On Wed, 11 Jul 2001 09:35:39 -0400, Edward Glowacki
>>>> <glowack2 at msu.edu> wrote:

>>>>>  1. Cheating 2. Spying

>>>> So... you don't consider these perfectly legitimate
>>>> applications of player ingenuity?

>>> Cheating is not a legitimate application of player ingenuity,
>>> IMHO.

>> But if it's legitimate, it's not cheating. It's all in your
>> perception.  If you say "this is not OK", it's cheating; if you
>> say "this is OK", it's not. Cheating and fair play are arbitrary
>> distinctions.

> Yes, often the boundary is arbitrary.  So are boundaries in any
> game.  Think about football, soccer, basketball, volleyball,
> tennis, etc.  where the boundary is just a painted line on the
> playing surface.

Right. The rules of the game are essentially lines not to cross. You
can have a line that is okay to cross (the "key" in basketball), and
a line that is not okay to cross (the court boundaries in
basketball).  Sometimes a line is just for reference, and sometimes
it actually means something.

I think most MUD server developers want to have complete control
over the game and how it's played. We might call this "GM syndrome";
since many of the people building these games are coming from a
pencil-and-paper background, they have this notion in their minds
that some individual (or group of individuals) has absolute power
over the game and the purpose of the server and the client is to
help them *exercise* that absolute power -- by applying it to
tremendous numbers of players, characters, mobs, areas, etc. In
short, the computer is merely a tool to scale their game upwards.

But simultaneously, these people recognise a need to give the
players some leeway. They want to operate the game "fairly",
implementing player suggestions and respecting player desires,
because they know this will make the game more popular. But the
computer doesn't really help much with that, and the farther up you
scale the harder the task becomes.  It's easy to take a vote among
six players you see once a week, but it's really hard to take one
among six thousand players you see whenever they happen to feel like
it -- from "all day every day" to "five minutes once a month so my
character doesn't get deleted", and everywhere in between.

>>> As for spying, if you want to support it in a game, build it
>>> into the game itself.

>> I did. :)

> =) I think that's definately a "Good Thing".

I wish everything was that obvious. There's one legacy item I'm
eyeing suspiciously, which was called a "press pass". Essentially,
this was a nice and reasonably cheap item which players could use to
post messages on system logs -- so when you logged into the game and
looked at the system's bulletins, someone with that item might have
put "Joe was here!" in the middle of it.

I like this on the one hand, because it allows you to falsify
reports.  The system might report "Bob conquers the planet in sector
521", and Bob might then tell the system to report "Bob destroys the
planet in sector 521". This could mislead people into thinking there
is no longer a planet in sector 521. That's devious and ingenious,
and I like that.  Many people would call it "cheating".

On the other hand, I have this nightmare vision of a system log that
says things like "MAKE MONEY ON YOUR PC CALL 1-800-UP-MY-ASS!" and
"FREEE NEKKID PITCHERS AT WWW.I-SUCK.COM!" for hundreds of
lines. That's just plain annoying, and it makes the *intended* task
of system bulletins -- information -- virtually impossible.

But I honestly don't know what I'm going to do about it. Like most
of the more unusual things I want to give players, it has the
capacity to completely mess up the game. I'm not sure if it's worth
the risk. I considered allowing it to DELETE items instead of add
them, but I'm not sure if that's a good idea... perhaps if "Deleted
by [whomever]" goes in the place of the deleted line. Hmmm,
actually, I have an idea that could work. I'll think some more about
it.

>> Quake assumes people will play fair. I don't. I know that people
>> cheat.  I expect them to cheat. I account for that in the
>> design. It is then no longer cheating.

> People will always come up with new ways of cheating that you've
> never thought of yet.  If you allow some levels of cheating to
> become part of the game, there will be new ones that fall outside
> that domain, and you still have the problem.

But it will be much smaller. ;)

> Cheating will always unbalance the game.  Even if you eliminate
> all out-of-game cheating (packet sniffing, etc.), there can be
> loopholes inside the game to exploit, such as buying and selling
> items for ridiculous profits.

Only if you call that "cheating". Most games do; but if I mess up
and leave a loophole in the game, I tend to regard using that
loophole for personal gain as simple opportunism. I'll plug it up
when I find it, but there's nothing wrong with using it in the
meantime. Allowing people to use these loopholes also makes it less
likely that they will *conceal* using those loopholes, which makes
it that much easier for me to see what they're doing and go fix it.

> One difference though is that in a sport with lines painted and
> referees watching, stepping out of bounds can be seen, whereas
> something like packet sniffing is for all intents and purposes
> invisible.

I know, which is one of the reasons I don't try to stop it. There
are limits to the things over which one can and cannot hold
authority, and one of the "hard" limits to that is that you cannot
hold authority over anything you cannot *detect*.

> Quake is an extreme example of the player and the character being
> exactly the same.  All avatars in Quake are created equal, and all
> differences in them are purely the result of the player playing
> them.

This is *also* the case in my system, which has a similarly extreme
"the player is the character" model. Your "character" is
theoretically a guy typing commands on a terminal which are relayed
to a remote location that returns reports on what happened... and
that's exactly what *you* are. The only difference is that the
theoretical remote location has real objects, while the real remote
location just has a database.

> However, we're talking more about MUD-style games here where
> characters have skill sets that can be nearly as complete as
> real-life.  If the game has a skill for swordfighting, then within
> the game, the primary factor involved when two characters duel
> with swords should be that swordfighting skill.  If one of the
> players is in real life an olympic fencer and the other a 14 year
> old kid who has never picked up a sword before, whichever
> *character* has the better skill should win.

There's also a drive to make these skills more realistic. Consider,
for example, a game which has armor modifiers. The player who knows
that chainmail is only an effective defense against slashing
attacks, not bludgeoning and piercing attacks, can select a weapon
which is more effective. These efforts make real-world knowledge
more important, by allowing it to provide an advantage. The
14-year-old kid probably thinks a sword is a sword, without
realising that a cutlass and a short sword may do the same damage
but the cutlass is a slashing weapon while the short sword is a
piercing weapon. The olympic fencer, however, probably knows this --
and can therefore reason that with identical in-game skill and both
of them wearing chainmail, the short sword is the more effective
weapon.

> I guess I should have stated more clearly, I meant that the
> *characters* (or avatars, or in-game personae, or whatever) all
> start the same.  Based on a players background, a player may have
> knowledge in different areas, and that may apply to the game, but
> for the most part the characters would be equal.

But I don't like that! ;)

Seriously, when I log onto a game and play it for a year, I don't
want to start over from the BEGINNING if something happens. I think
I've earned the right to be trusted with a certain amount of initial
capability. If the admins hand me that sort of capability, people
cry "favoritism". If I go out and use my knowledge of the system to
acquire it, they cry "cheating". If anybody has the gall to suggest
that maybe these people just haven't earned the same privilege, they
cry "discrimination". And because there are a lot more of them than
there are of me, the game's admins have to go "sorry, but it's what
the players want".

Never underestimate the power of stupid people in large groups.

The problem, I think, is a misinterpretation of what the game's
administrators are supposed to be doing. Most people seem to see the
admins as a sort of "protector" caste, which is supposed to stop bad
things from happening. I don't think that's the right thing for the
admins to do. I think they're supposed to stop bad things from
happening TO THE GAME, but not to the players. Their entire judicial
capacity should be to look at a situation and say either "that needs
to be fixed" or "that works just fine". If it needs to be fixed,
they should fix it.  If it doesn't, they should say "tough!" and
kick the player in the butt.  (Well, okay, the butt-kick is
optional.) It is not possible for an admin to be everybody's friend,
but most of them still try. They want to keep their playerbase
happy, so they try to do the things that the most people want done.

The major problem with this policy is that the people who are doing
well at the game obviously *like* the game and are investing time
and energy in playing it. The people who *aren't* doing well, of
course, are the ones with all the complaints and gripes and ideas
for policy changes.  When a player comes up and says "your game
sucks because it doesn't have this", what he really means is "I
don't like your game". Why you would step-and-fetch-it for someone
who doesn't like your game is beyond me; I'd be more likely to tell
him to piss up a rope. If you don't like the game, you don't have to
play it.

I don't know why nobody has instituted a policy of "ban unhappy
players". It seems to me that this is the most efficient and
workable solution, since the players will not hesitate to run up and
bitch when they aren't happy, which makes it exceptionally easy to
identify them.  ("This isn't fair!" "The computer wants you to be
happy. Failure to be happy is grounds for summary execution. Are you
happy, citizen?" "Oh, uh, yeah! Ecstatic! See ya...") Once the
unhappy players have been banned, everyone on the MUD will be happy,
and no compromise of the admin's vision is necessary. You may have a
smaller playerbase, but you'll have the highest player satisfaction
rating in the world.

I don't really need to wave the sarcasm flag for that, do I? I
didn't think so.

I can already hear the people saying "a commercial MUD doesn't have
that option!", but that's not true. A commercial MUD most certainly
*does* have that option, if they are committed to QUALITY and not
QUANTITY.  People can and do position their commercial offerings as
"luxury" and "elite" products, and other people can and do buy them
because of it.  How many times have you heard "we may cost a little
more, but we're worth it"? An Oldsmobile luxury sedan is every bit
as comfortable as a Lexus, but the Lexus costs twice as much. Why? 
Because the Lexus is... a Lexus. (Which, strangely enough, is
synonymous with "a Toyota" -- but when you say "I have a Toyota",
nobody cares. Oh, what a feeling.) If you want to be the Lexus of
MUDs, you can be. Lexus didn't revolutionise the industry. Lexus
didn't innovate all over the place. Lexus didn't even do anything
particularly special. Check their history: all they ever did was say
"we are the best", and people agreed. Their technology was not
ground-breaking. They simply made a quality product and charged a
premium price.

There is something seriously wrong when the *important* part of
"massive multiplayer gaming" is the "massive" part. The
"multiplayer" part is being downplayed to satisfy people's desire
for privacy. The "gaming" part is being downplayed because... hell,
I don't know, maybe because nobody wants to admit they play games
all the time or write games for a living; probably because they're
sick of being asked when they're going to get a "real" job or write
some "real" software. And everyone's talking more players, more
connections, more servers, more MONEY, more, more, more.

Screw more. I want BETTER.

>> This is a naive statement. All players exist on a continuum. Some
>> players are at the top, and others are at the bottom. Those on
>> the top are reasonably safe. Those at the bottom are in trouble.

> For a specific skill, yes, but not necessarily for *every* skill.

Yes, for every skill. If you add up the skills of every person on
the game, there will be a group of people with higher scores than
all the other people, and a group of people with lower scores. No
reasonable unilateral criteria will result in any other
result. (There are some mathematical oddities you could use, like
say "a player's highest skill is worth one point and the rest of his
skills are worth nothing", which results in every player having a
final rating of 1 regardless of his skills. You might also say "each
player's skill total is scaled by the reciprocal of that total",
which would also give every player a final rating of 1. These are,
however, not reasonable representations of a ranking system.)

Unless... hmmm, this is an idea.

Imagine a system in which you are given, say, five hundred skill
points.  At each "level break", you're allowed to move some of them
around. As your level gets higher, you can move more points. You
always have only five hundred points, but over time you can build
yourself into a "better" character by putting them in areas you
personally find interesting. In such a system, the quality of your
character is defined by how accurately it supports the activities
*you* enjoy. You can build your ideal character right out of the
gate, or shuffle things around as you go to refine it.

An alternate system might allow you to "use" points from specific
skills to achieve tasks, and after using those points you get the
same number of points back for reallocation. Only allocated points
are usable, and you can only allocate points under specific criteria
-- like "be in this place" or "carry this item" or "wait X number of
minutes". As you rise in level, you are able to use larger numbers
of points at once and/or can reallocate them under less restrictive
conditions.

Both systems have your character beginning and ending the game with
the same point totals, so for all intents and purposes your
character never "advances" in the normal sense -- he just acquires
an ability to shuffle things around more. If the delta from lowest
to highest level is small enough, the player never really loses a
character, just a small amount of game ability which doesn't
preclude enjoyment of the game.

> Some people will rise to the top of many categories, but very
> rarely would someone be the best at *everything* about a game.

You seem to be confusing "best" with "perfect". In any group, no
matter how small or large, SOMEONE is the best at whatever task you
might choose. Define your playing field and what "best" means, and
someone will fit that definition on that field. Assume you have
someone who's great at math but a bad runner, and someone else who's
a fast runner but bad at math. You ask both of them to solve a
complex equation, run a mile, and then solve another complex
equation in the shortest possible time. The guy who's great at math
will win, because he can WALK the mile -- he'll solve the equation
in under a minute, while the other guy takes at least ten
minutes. Since the average walking speed of a human being is three
to five miles per hour, the math whiz can walk the mile and do the
other equation easily. The other guy will be a good five minutes
behind on solving the second problem. Now, stretch it to three
miles, and the runner wins: he runs the three miles in fifteen
minutes, and can take ten minutes on each of the equations for a 35
minute total time while the math whiz takes 45 minutes just to cover
the distance. Same situation, same skills, but alter the playing
field slightly and the outcome is reversed. In any game, among any
group of players, for any definition of "best", someone will be
closer to that definition than anyone else.

>>> If any single skill is one-way and there is no defense against
>>> it, it will be abused and unbalance the game.

>> Any skill that can be abused, will be abused. Assume players will
>> abuse it, and design accordingly.

> That seems to me to be a paradox.  If you build it with abuse in
> mind, then it won't be balanced (it will be sub-standard most
> likely), and therefore people won't have a reason to abuse it.

Not if they aren't abuse-minded people, no. But those who *are*
abuse minded will come up with all kinds of interesting ideas.

One of the items available in my system is called a "GoTo". It's a
transport device; the first time you use it, you "home" it to
whatever sector you're in. At any later point in time, you can use
it to "GoTo" that sector. Minor capability, right? Right.

I also have a flea market. You can put items up for sale in the flea
market. An immediate abusive thought is to home a GoTo on some
sector full of nasty things, and then sell it at the flea
market. But this is, well, nasty... so I have a security measure:
when you put it up for sale, your name is shown as the seller, and
people would know it was your fault. Furthermore, the guy who gets
it will probably check the sector to see what's in there, if he's
smart.

So to avoid the sector check, you can get an item called a "Sector
Reflector". This will display some *other* sector whenever someone
scans it. (An alert player will notice that he scanned sector 521
and got a report on sector 43, but there's also an "Advanced Sector
Reflector" at a much higher price that avoids this problem.) But the
most easily accessible long-range scanner -- the ship's computer --
will fail if there are fighters in the *true* sector, as a security
measure. So you have to put nothing but mines in that sector, so the
new owner can scan it and ensure that it's safe. (Mines are a
passive defense, and don't interfere with long-range scanners.)

To avoid the name problem, you exploit the system: the flea market
also automatically picks up discarded devices and sells them at
reduced prices. The way this works is that every "discarded" device
migrates toward the flea market, and when it ends up in the same
sector with the flea market they collect it and put it up for sale
at the next maintenance sweep. (Since items have limited
availability in my system, anything that gets thrown away *must*
return to circulation to preserve overall game balance. You can
still abuse this system, of course, as I'm about to demonstrate.)

So the obvious solution is to home it on some sector full of nasty
things, then throw it away in the flea market's sector. The flea
market will get it, and your name won't be on it. But if those nasty
things belong to *you*, people will know you did it anyway --
because they'll be informed whose fighters and mines killed them.

So you need to convince someone *else* to put a significant
contingent of mines -- but not fighters -- in the sector. So you go
off to some weak guy's home planet and conquer it. The catch is not
to CLAIM the planet, just conquer it; so you can pillage the planet
now, but you'd need to conquer it again if you came back. (Not that
this is normally *hard*, since you will no doubt have destroyed most
of its defenses.)

Now you exploit the system again. At the maintenance sweep, fighters
and mines on a planet are automatically shuttled out to defend the
sector.  Those fighters and mines belong to the planet's owner. So
you just grab all the remaining fighters and put a big bunch of
mines on the planet.  Since you didn't claim the planet, they'll
belong to the other guy...  not you.

So you just leave the planet, fire up the sector reflector, home
your GoTo on the sector, then go to the flea market's sector and
drop the GoTo. At the next maintenance sweep, it will be picked up
and placed on sale at greatly reduced price, pointing at a sector
that looks safe -- and surely you took the time to drop something
*interesting* in that sector, didn't you? -- but is actually filled
with mines that belong to someone else. Even if the trap doesn't
work, the GoTo's buyer will think it was set by whoever owns the
planet.

It would be easy to prevent these things. I could clear out the
GoTo's settings before putting it on sale. I could put your name on
the device being sold even if it was collected automatically. I
could make the planet attack you if you left without claiming it. I
could stop requiring you to claim the planet at all, and make you
claim it automatically. I could do away with the maintenance sweep
and handle all operations on a continuous basis. I'm sure many
people would really like some of these modifications. But *none* of
them make the game more FUN.  They make it more *secure*, but they
don't do anything fun.

That's one of my main rules. If it is not fun, it does not belong in
the game. Encryption is not fun. Abuse, however, is! It's unethical! 
It's immoral (or at least amoral)! Sometimes it's actually illegal! 
Isn't that great?! ;)

A lot of this is a question of what the *developer* considers "fun",
too. I've got features in my game that aren't done yet, but that
aren't fun to write either. It's *boring* to write code for a
transmat platform. If the player owns the platform, or is a member
of the team that owns the platform, or the platform is public, and
he has a sufficient amount of anticloaking energy to see the
platform, and the platform is set to a valid sector, and the sector
does not contain a transmat inhibitor, put the player in the target
sector. Pfffft. I could do that whenever. I'm much more interested
in writing code for enemy fighter propagation, so there's something
worth fighting out there. The actual network server code is boring,
too; I'd much rather write a nice efficient mechanism for searching
tens of thousands of player data files. Of course, if I don't write
the server code, I'll never HAVE tens of thousands of player data
files... but that's not important! This is a MUCH more interesting
feature!

I think a lot of people are looking at encryption that way. Making
the game interesting and fun for the player isn't anywhere near as
important as the *developer* having fun writing robust encryption
code that works fast enough to handle hundreds of simultaneous
encrypted connections.  And you know, I kind of have to agree that
the encryption problem is probably a lot more fun.

>> But *other* people have been so thrown. That's why seatbelts are
>> no longer an OPTION, as they used to be.

> Proving my point.  =) Other people have been sniffed, not
> necessarily for games but for information in general.  Therefore,
> encryption is my seatbelt.

Proving *my* point. Is there a seatbelt on your desk chair? No! Why
not?  Because while people *have* been thrown from their seats, it
just plain doesn't happen when you're sitting at your desk, and even
if it *did* you probably wouldn't suffer any significant damage. :P

>> The vast majority of players are not using UNIX and do not have
>> SSL readily available, so a MUD server that supported SSL
>> wouldn't help them at all.

> I believe the vast majority of computers run MS Windows.  Most
> recent versions of MS Windows come with MS IE.  MS IE supports SSL
> connections.  Therefore MS Windows has SSL.  There's also OpenSSL,
> which I believe is available for Windows.

My bad, you said SSL and I read (and wrote) SSH. Never mind. ;)

> There is plenty of encryption available today that is sufficiently
> strong and technically sound enough to work.  SSH and SSL are
> "secure" for almost all applications.  Eventually, increased
> computing power will render them vulnerable in a trivial amount of
> time, but by then larger keys and new algorithms will be
> available.

I'd go so far as to say they're essentially secure ad infinitum in
the absence of a quantum leap in mathematical theory. Computing
power is not just a limitation on key *compromise*, but also on key
*generation*.  When we can compromise a 512 bit key trivially, we
can also generate a 2048 bit key trivially. As computing power
rises, both capabilities will increase, and as long as SSH and SSL
can be implemented with *support* for the increased key size they
will remain just as secure as ever...  provided the factoring
problem remains essentially NP-complete.

> Actually using the encryption is another matter entirely.  Web
> browsers seem to do a fairly good job of implementing encryption
> transparently, but many other ways of using encryption aren't so
> easy.  PGP is one example that I think you mentioned that requires
> extra effort.  The *implementation* of using the encryption is at
> fault, not the technology behind the encryption.

The implementation, however, often *reflects* inconveniences in the
underlying technology. Sometimes you just plain can't implement the
technology with any more convenience than you already do. In those
cases, the implementation CAN'T be improved without improving the
underlying technology.

>>> Security isn't about making your system invincible, it's about
>>> making your system more difficult to break in than it's worth.

>> Or making your system worth less to break. Removing a space from
>> the previous sentence would make it even better. ;)

> People will still try the door to see if its unlocked, and if they
> can get in, they'll come and hang out for a while.  If there's
> nothing there, they might leave or they might make it a base of
> operations for doing other things.

But "suitability as a base of operations" is the same as having
something there. If they can't do anything with it, they'll just
leave.

It's also worth noting that if your door isn't locked, people who
like to break into things will generally not waste their time on
you. ;)

>> Most MUDs are *too* secure. People can smile and nod and be nice
>> and friendly to you while they PRIVATELY plan to assassinate you
>> via the MUD's person-to-person channels. You have no ability to
>> eavesdrop. You have no indication that they are even talking. And
>> they take advantage of that, don't they? Doesn't that damage game
>> balance, when a player can lie to me without the slightest
>> indication that he's lying or even doing something suspicious? 
>> Doesn't that destroy the game's immersion?

> Unfortunately, yes.  =( But I still don't think packet sniffing or
> other forms of cheating are a suitable answer to the dilemma.

They're *not* an answer to the dilemma. This dilemma is simply an
example of things MOST games offer which damage game balance and
destroy the game's immersion. Just because it damages balance and
destroys immersion doesn't mean it shouldn't be in the
game. Sometimes you have to *accept* a certain amount of compromise
to game balance and immersion from some specific feature. For
example, when you let players post on a bulletin board, you must
accept that some of them will talk about non-game matters, and some
of them will post solutions to in-game puzzles that were intended to
be kept secret. When you give them internal email capability via
this facility, you have to accept as well that some of them will
choose to hoard this information rather than share it with everyone.

Essentially, all information on a MUD *may* be public knowledge in
any given subsegment of the MUD's population. If any of that
information MUST NOT be public knowledge, then IMHO it does not
belong on the MUD.  It belongs somewhere else. The old saying
"information wants to be free" comes to mind; all information can be
learned, but cannot be unlearned.  It may become MORE widely known,
but so long as it is relevant it will not become LESS widely known.

Another saying that comes to mind is "the only way two people can
keep a secret is if one of them is dead". ;)

> I understand what you're saying, that MUDs are designed with the
> serious computer types in mind.  However, for online games in
> general, specifically larger ones that cater to a more general
> audience, I think having an established set of rules and
> boundaries for the game is important.

And generally speaking, the smaller the set, the better the game.

>> In short, people will be dishonest if and only if it is
>> advantageous to do so. You should take away the ADVANTAGE, not
>> the ability. What

> It's not always possible to take away the advantage.  

A game is certainly fair if *nobody* has an advantage, but it is
also fair if *everyone* has an advantage.

> In NPK or CPK areas, knowing an opponents location is an
> advantage, and that advantage can not be taken away.

Sure it can!

Add a homing mirror to the game which is created magically to work
for two people: a target (under some set of criteria; perhaps you
need something that belonged to him) and an owner. Whoever carries
the mirror is told where these two people are whenever he looks at
it. Initially, the owner will have it, so he'll see the target. If
the target gets it, he'll see the owner (can you say "revenge"?). If
someone else gets it, he'll see both of them. The mirror can also be
broken, which kills both the person carrying it and the person
breaking it (if it's broken by a sword blow, for example). The
mirror will break all by itself if you try to look at a player that
no longer exists, killing whoever carries it.

Now knowing the location is an advantage, all right, but you can get
it in-game too. It has risks of its own, and advantages of its
own. And rather than attempting to prevent cheating, which only
affects cheaters and simply annoys them into doing something else,
you've rewarded the players with an object all of them can use to
enhance gameplay. This object can be used in all kinds of diabolical
ways.

Of course, you can certainly just fire up a packet sniffer. Then you
don't have to buy the object. And the other guy can't take it from
you.  And you won't get it broken in a fight and die. And the guy
won't just delete his character so you die next time you look in the
mirror.

But you also can't see whomever you want with it; only those people
who happen to be on your network. And you have to be logged on at
the same time they are. Sure, you can't kill someone who's not
logged on...  usually... but you can certainly set up a
trap. Perhaps you could charm a series of monsters and lead them
into his location, so when he DOES log on he gets attacked and
mauled by an army of monsters. Perhaps you could buy a whole slew of
them and shove them in a bag of holding, then sell information to
other players. Perhaps an enterprising thief could set up a side
business in "items that belonged to other characters". An
adventuring trio could purchase one for each pair of characters, and
have the unincluded member carry each of them so he always knew
where his two friends were.

And that's FUN. Stopping people from cheating is not fun. It makes
people who do not play by the rules angry. Since they do not play by
the rules, they will not say "oooohh, I am so mad, I think I will
have a sandwich!", they will say "I'm gonna get that
sonnamabitch". (I don't even know the man, and he call me a
sonnamabitch.) But now that you can get the homing mirror, you don't
*have* to stop them! Who cares if he knows because he sniffed or
because he bought a mirror? The point is that *everybody* knows that
*anybody* might know. The advantage of sniffing a packet for
locations is twofold: you know where they are, and they don't know
you know. But now that advantage is gone with the wind, and you can
sniff all you want... because nobody cares.

And the cheater cannot identify someone or something to attack in
retaliation, so he has no choice but to go off and have a sandwich.

>> One failure of most MUDs, IMHO, is to make cheating attractive by
>> default. If you cheat, you will have a definite advantage. Remove

> That sounds like a definition of cheating.

Essentially, it is. When cheating doesn't give you an advantage,
it's not usually considered "cheating". It's considered "stupid". ;)

_______________________________________________
MUD-Dev mailing list
MUD-Dev at kanga.nu
https://www.kanga.nu/lists/listinfo/mud-dev