Re: Giving an application a window icon in a sensible way

Bent C Dalager wrote:
In article <1164532424.537227.125330@xxxxxxxxxxxxxxxxxxxxxxxxxxx>,
<nebulous99@xxxxxxxxx> wrote:
Bent C Dalager wrote:
As for game theory, if that is what you are trying to apply here then
it seems to me you are entirely ignoring the shadow of the future in
your analysis, and you may be stuck in a trivial model where you think
there are only two distinct players in the game.

(Explanation)

Very well, then, it seems that [snip insult]

1) It completely disregards the shadow of the future.

If you'd care to speak English, I might be able to debate you further
on this matter. If you refuse to, then I will have to just blanket
disclaim whatever negative things about me you are saying or implying
and have done with it.

But an attempt to parse what you might be meaning suggests that you've
made a significant error. Far from ignoring the future, I'm concerned
with it. If I didn't care about the future I wouldn't care about the
fact that someone had claimed negative things about me in public. But
thinking about the chain of consequences if they do so unopposed leads
me to act differently. Consider a person who hears the insult, but does
not hear anything that provides a counterbalancing viewpoint -- for
example, a hostile claim about my IQ with no corresponding positive
claim, nor any pointing out of a flaw in the argument used to "prove"
my supposed low IQ. True, this person might discover the flaw on their
own, or simply be skeptical. On the other hand, they might be
credulous, and their belief in my inferiority will then influence how
they treat me in the future.

Now consider the potential scope of the internet, and multiply that one
person by, oh, say, 6 billion or so...

Fortunately, the very people most influenced by these insults, the
credulous, are equally susceptible to be influenced in *both*
directions, from which it isn't difficult to conclude how to prevent
the nightmare scenario of some jerk online convincing the world to have
nothing to do with me (or worse, to actively abuse me). And that method
is simple: for each insult, there must be an equal and opposite
rebuttal. The two influences will exactly cancel. Some people are
credulous, and they will swing widely both ways before settling at the
center again. Others are less susceptible to persuasion, and may budge
slightly or stay put.

Of course, this is with regard to claims that are false, or that are
matters of opinion with no real empirical basis at all. Unfortunately,
if someone were to discover a true but unpleasant claim about me and
start broadcasting it, I don't see any recourse other than to forcibly
shut them up (assuming they marshaled something resembling actual
evidence to support their claim). On the other hand, that has
fortunately not actually happened yet, and there may not be any such
claim anyway. Maybe the real worst-case is that such a claim is found
that, while false, is in some ways particularly convincing anyway, in
much the way it can be difficult to persuade a tribal people that the
earth isn't flat because the alternative is counter to their intuition
and they are unequipped to understand the evidence against its
flatness. Physically escorting them up to orbit and back would likely
be necessary, but that is rather expensive. I hope that a similarly
false-but-plausible-sounding hostile notion regarding me doesn't crop
up, the disproving of which might be similarly involved or expensive...

More generally, this seems to indicate a possibly serious problem with
the mere existence of the Internet, for all its benefits. I'm not a
luddite by any means, but this one particular area is a recurring
concern -- it seems to be possible for any kind of libel, slander, or
mudslinging directed against an individual to be broadcast unreasonably
and undeservedly far and wide now, impossible to recall it (and so to
force its recall using e.g. legal action to obtain a court injunction),
and difficult to counter it in any other way than by a corresponding
broadcast of the negation of what was claimed (in the simplest case,
"what the preceding message said is false", but producing arguments and
evidence against it is surely more effective). And of course every
opinion anyone has ever expressed about anyone else is now susceptible
to a Google search...there seem to be only two defenses. Besides the
counteropinion defense, there's the nymshifting defense -- everyone
could create a new pseudonymous identity for every separate online
interaction, for example one for each separate question or conversation
in comp.lang.java.programmer. This strikes me as a lot of work, until
our communication tools make establishing new pseudonyms easy and
automatic. Also, it needs to be harder to tie together anyone's
pseudonyms. Much harder. As a rule, the originating IP address of every
posting tends to be viewable by everybody else (not just those at
providers who deal with abuse complaints), so we're at minimum talking
about automating something equivalent to a) setting up a new GG account
for every new thread a person participates in, or at least for every
newsgroup or equivalent context, and b) Tor-routing everything
transparently with a minimum of fuss. As far as I can tell, the bare
*minimum* to avoid the rebuttal-requirement scenario is Tor-routing
combined with the creation of a new GG account after each insult,
thereby effectively erasing it from history.

And it is obviously critically important that none of those identities
ever be traceable to your name offline, lest you have to actually have
your name legally changed just to escape an undeserved reputation
created by some loser on the Internet with too much free time. The last
time I checked, that is not only a massive inconvenience, it actually
*costs money* too...and that means that a penetration enables anyone
who wants to to cost you money at any time, at the push of a button. In
actual fact, it's even worse; at least when past insults that are
believed accumulate to the point of interfering with activities like
job-finding or getting loans, the victim will have to change their name
legally, relocate, find a new job, and about six thousand other things.
To top it off, so will their whole family, and most of their friends
will have to fall by the wayside, leaving only those who can definitely
be trusted with the ability to connect the old and new identities! The
estimated expense of all this is upwards of $6000, *if* an online
problem develops into a problem of large enough magnitude. And this can
potentially be caused at any time by anyone who simply decides they
have a beef with you... I don't see any solution other than strong
online pseudonymity to preventing anyone from being able to cost anyone
else $6000 and months of inconvenience at the drop of a hat. It should
be noted that in less affluent countries and in the US with its
hole-riddled safety net, the consequences can even be death by
starvation or exposure, if someone starts spreading internet rumours
and nonsense that gets you fired, gets you unhireable, and thus you
don't eat ever again. The only insurance would be to sock a few tens of
thousands away as an emergency relocation-and-name-change fund, which
you might not have saved up yet by the time the bomb gets dropped...

If anything, there's worse. After all, people offline will know your
legal name and place of business and things like that. If one of
*those* decides to ruin you with online rumour-spreading, then to put
things bluntly, you're fucked even if you *do* use strong pseudonymity.
So the internet may well force us to limit our offline transactions to
only that which we can't do online. Human social interaction becomes a
largely online-only affair, and offline is limited to highly ritualized
formal transactions that can't be done online and in which everyone is
nervous and walking on eggshells because any other participant that
gets a mind to can blow their whole life sky-high. It's nearly as bad
as a Wild West type setting where everyone packs a gun and there's very
little accountability regarding their use of it. At least you can start
over after the attacks depicted in this scenario, unlike after getting
fatally shot. On the flip side, it's easy to set up justice systems to
catch and punish the majority of people who wantonly shoot people, and
in the process prevent by deterrence a lot of the same. There's also
the mutually assured destruction factor that you might *be* shot when
you pull the gun, if your target is a quicker draw. Whereas the same
strong pseudonymity necessary to enable someone to quickly amputate an
online "limb" that becomes "gangrenous" before it threatens the whole
organism also enables someone who penetrates your real name to attack
it online untraceably. There's no hope for justice and no MAD threat
when the victim can't identify who nuked him to target their own
missiles and the police also can't identify the perp. And that means
there's no deterrence.

I can't see any solutions to the final dilemma at all. On the one hand,
we could try to somehow *prevent* strong pseudonymity, but this opens
up a Pandora's box of surveillance, police state behavior, chilling of
speech, and so forth, and doesn't stop someone "nuking" someone else
with strongly negative belief-spreading about them. It only ensures the
MAD option and possibly enables some kind of legal action, which is
cold comfort since in that world you also can't change your identity
and start over after being nuked.

On the other hand, we could push a magic button and make everyone in
the world disregard anything bad they hear (however plausible) about
anyone else over the internet, and while we're at it make gullibility
itself a crime, banish war and poverty, and institute peace on earth by
fiat. I can tell you, though, that I already tried one of those
miraculous easy buttons from Staples and it definitely didn't perform
as advertised, so that option seems to be right out. ;)

Where does that leave us? We can't make everyone else not be gullible,
and we can't make everyone else like us or even just not say nasty
things about us on the net. We can only respond with a change of
identity if our current one ends up in undeserved disrepute (and
enabling that, mind you, enables people to get rid of identities in
*deserved* disrepute too...)

If you have any brilliant suggestions, I'd like to hear them.
Currently, my two rays of hope here are things like Tor and Freenet, on
the one hand, and (paradoxically) the proliferation of surveillance on
the other. The latter may *force* society to become really, really
tolerant in a big hurry once it reaches the knee of the curve, and that
will take the potentially serious consequences out of the equation, and
reduce insults again to what they once were before the Internet:
something you could just shrug off with impunity.

.