The faster they push this along, the sooner
this whole debacle will be over.
Regardless of the lack of merit in SCO's central claims,
in CIO-land uncertainty is anathema.
It would have been worse for SCO to have simply left the
situation as FUD.
-Ed
Re:Business Models or "Developers, developers!"
on
Minitel Hits Twenty
·
· Score: 1, Informative
I don't know quite where you'd find the news or weather, but dialing into the ARPANET 30 years ago was just a matter of knowing the phone number and having a terminal and modem. There was no security on the TAC (terminal access controller, what might be called a "modem pool"). There were a couple systems at MIT that allowed guest access, and if you managed to connect there you could request an account and most likely get it.
Open TACs and systems with guest accounts existed into the early 80s, when a few people started knocking on the wrong doors and DARPA (and eventually MIT) cracked down. But they're how I first got on the ARPANET back in 1977
I know people who have been programming in C for over a
decade who think that a null pointer is always
a zero. Can you name a common computer architecture where null isn't zero?
There are some really obscure architectures where null
pointers aren't zero, but even in those cases well-written
code will usually work just fine; the standard
specifies that a zero value assigned to any pointer
or cast to any pointer type
takes on the value of the appropriate null pointer.
Furthermore, a null pointer value converted to a
_Bool is always
false,
so the expected thing happens when a null pointer value
appears in a conditional context.
Finally, when a function prototype exists, an implicit
conversion "as if by assignment" (the words are from the standard) occurs.
So you might say that the standard goes out of its way to
make ignorance of null pointers relatively harmless in
otherwise well-written code.
Capitalism is like Communism in an important respect:
both theories, in their pure form, fail to scale.
(A corollary is that supporters of either can
claim that their system "has never really been tried.")
Information moves too slowly and
is far too easily subverted for a large
free market to function, even if you assume that
buyers are perfectly rational.
This is why third parties (private or governmental)
need to be introduced for large markets to work.
Private entities like Consumers Union can help to
improve the information pool (insofar as they aren't
subverted by special interests) but there are cases where
government's ability to compel disclosure
is required.
Whether further regulation (beyond reasonable
disclosure) improves
the functioning of markets is more questionable.
But a market where sellers control all the information
isn't going to function well at all outside of an open-air bazaar.
What the world could have used was some rolling-up of sleeves and efforts to do better, either by bringing those fabulous other systems to workstation-class hardware, or by at least porting over bits and pieces of them (shells, programming languages, etc.).
Dennis Ritchie did just that: witness Plan 9 and
Inferno.
Or even those versions of Research Unix after V7.
Unfortunately, the focus of AT&T and later Lucent
was on commercialization, so the innovations found in these
products were essentially inaccessible to the communities who could most appreciate and further them.
It's a sad fact that the original creators of Unix had very little to do with its evolution after 1980 or so.
It's an uncomfortable truth that complete suppression
of terrorism requires complete suppression of
freedom.
If we want to maintain our freedom, we'll have to
combat the fear of terrorism every bit as strongly as
we fight terrorism itself.
We'll have to risk that our promotion of freedom will
at some points allow terrorism to operate.
In a word, we need courage.
But if we depend entirely upon our government and military to
be courageous for us, we're already far along the
road to losing our liberty.
Your lack of familiarity in these groups is telling.
One of the most common efforts that the women-in-engineering groups I've seen make is in developing mentoring and tutoring programs.
More general community outreach, like classroom visits, are another common
activities.
Also, networking is very important -- for many group members
that's all they're interested in.
Some male CS types can make it very difficult for
women to network with them; they tend to interpret
professional interest from a woman as an opportunity to
hit on them.
And this is just one of the issues women have that men typically don't; they need a place to discuss how to
handle such situations.
Although networking only with other women would be unecessarily limiting, it can be a safer and more comfortable starting point.
Btw, NetBSD has had x86-64 some time back. Free is just catching up, it seems.
FreeBSD has focused on Itanium up until now, given that
production hardware has been available for a while.
There are people who already have them in their
datacenters and who want to run FreeBSD on them.
It's part of the pragmatism that is at the base of
FreeBSD's philosophy.
NetBSD has more of a research-focused, "climb the mountain
because it's there" philosophy.
Compare mottos:
"FreeBSD -- The Power To Serve" vs. "Of Course It Runs NetBSD."
That's not to say NetBSD doesn't make a fine, practical
embedded platform (say), or that FreeBSD is useless in
research (especially when that research can result
in improved performance in the datacenter, FreeBSD's
home turf).
It's just a difference in emphasis, and the BSD community
is richer for it.
Actually, LA takes as many seismic precautions than
any other metropolis in the world (probably equalled
only by Tokyo).
Evidence of this is that the only apartment building
to actually collapse in the Northridge 'quake
had serious code problems.
Scores more were rendered uninhabitable, but
among apartment buildings only
the Northridge Meadows complex pancaked; it was later
discovered that it had been built without some of the
required reinforcing between floors.
Most of the rest of the country has no such requirements,
and under the building codes in, say, St. Louis, you
would have had dozens of Northridge Meadows.
The problem with the Puente Hills fault is that it goes
right under downtown.
Although the buildings there are designed to
withstand a large earthquake on more distant faults,
they weren't designed for a Northridge-style event
right under them.
(Northridge was 6.8 Richter, more or less the size of
event expected on the Puente Hills fault.)
I lived around a mile from the Northridge epicenter,
in West Van Nuys.
Real Estate values in our neighborhood went down
briefly, and didn't start climbing until a year and a
half later.
But the market hardly colapsed.
(It helped that the houses in our area were well-built,
with only those with second-story add-ons suffering much,
if any, structural damage.)
Nobody I know moved to Phoenix.
When we sold our house four years later, we got what
we expected for it, around 20% more than its valuation
shortly before the Northridge quake occured.
Bjarne Stroustrup wrote
a paper
a few years back proposing overloading the C++
whitespace operator, and claimed that the next C++
standard would include this feature.
This allows for intuitive pratices such as using a
space to indicate multiplication:
double x = 1.4; double y = 2.5 x;
or string concatenation:
string s("first"); string s2 = s "post";
Other uses, such as making whichspace equivalent to the
-> operator for particular classes can go
far to make C++ syntax less obscure.
It's a cool paper; check it out.
If you have problems finding it, just Google for
"B Stroustrup: Generalizing Overloading for C++2000. Overload, Issue 25. April 1, 1998."
Your first paragraph doesn't make much sense to me.
The US Constitution says:
Section 8. The Congress shall have power...
To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries...
Now, how does this expressed intent relate to "expanding
the public domain?"
I suspect that it has more to do with facilitating commerce
by limiting
the public domain.
But benefit to the public domain was certainly the intent for making the exclusive right for a limited time.
And that is why your argument that DRM is fundamentally
different from copyright is spot-on.
I don't think your other claim holds up, however;
although copyright hardly requires that a creator benefit
financially from exclusivity, such rewards are certainly
one of its benefits.
Thus the business interests involved are hardly irrelevent.
It's a PITA once, for one person.
Then everyone else uses the same setup,
since they have the exact same card.
I agree that there are a lot of variables involved
in analog recording, and a lot of complex decisions to be made; a good recording engineer is
as much artist as engineer.
But virtually none of those variables exist in a simple
analog out -> analog in loop.
Once properly adjusted, full-scale on the D/A will
result in full scale on the A/D -- maximum
safe S/N, no clipping, and so on.
This has always puzzled me:
why is there such an obsession with preventing bit-for-bit
copies?
Properly done, digital re-recording from the analog output is likely
to cause considerably less distortion than MP3 encoding,
even at 256Kb/s.
And the RIAA will be the first to tell you that MP3's
are by far the biggest "piracy" threat represented by the
internet.
Preventing digital output will only be a minor impediment to
copying, while the inability to use digital interconnects
is a major blow to functionality.
DARPA funding for BSD goes way back -- long before
OpenBSD, FreeBSD, and NetBSD existed.
One of the most important instances was DARPA's funding of
the development of BSD's TCP/IP network stack back in
the mid-1980's.
This made BSD the first system in wide deployment
that supported TCP/IP.
It's hard to overestimate the affect that this
has had on Unix and the Internet since then.
Sun can play both sides of the street.
Since they paid some ungodly sum to buy
full rights to Unix SVR4, they're out of
SCO's reach.
Even if SCO gets its way (which I doubt, but assume
it for the sake of argument), Sun's Linux
offering would be unaffected by SCO's suit
(but see next paragraph).
I know who I'd bet on in a matchup between Sun and SCO...
Of course, SCO's claims are completely incompatible
with the GPL, so if they win, Linux would be stopped
dead in its tracks until it can be somehow be
made "untainted" by SCO's IP.
And that's the real danger here.
BSD was stopped dead in its tracks by a similar legal
situation in the early 1990's; it almost
certainly would have been stronger competition
against commercial Unixes if that hadn't happened.
SCO might even be attempting to duplicate
this, and slow Linux down under a legal cloud for
a few years even if they ultimately fail in court.
Of course, I hope IBM hits them like a load of bricks
(and I expect they will).
But don't expect it to blow over very quickly.
Yup, y'all are right.
Teach me to post while working on
other things...
But the point remains: if transfering this one data source
to a single destination takes up ~10% of the total
bandwidth available, we're still not quite there yet...
If SLAC is generating a terabyte of information a day,
900+ bits/second isn't nearly fast enough to transfer it.
(Yes, I know that only parts of that data are likely to be
useful enough to transfer, but it does suggest that there
is still quite a ways to go in the quest for bandwidth.)
So you'd argue that making someone liable for things
likely to be beyond their control is a way to encourage
those who have the control, but no liability, to reform?
That's novel.
Remember, the "malicious users" responsible
for the incoming bandwidth
aren't at all likely to be on his system,
and as many postings here
have pointed out, the customer could be up-to-the-minute
with patches and even have shut down his server at the
first sign of trouble, but that would have no effect
in an inbound UDP flood (like the slammer worm)
or a SYN flood (like a DDOS attack), and so forth.
The malicious or unpatched originators of that bandwidth
aren't likely to be motivated to prevent the flood
by the fact that the victim pays (and perhaps the contrary
if DDOS was the objective).
Now, whether the victim pays or not is a separate issue,
and could be decided based on contract law and business
considerations.
I frankly don't think the original article gives enough
information to argue one way or the other.
But I don't see anything on which to base an opinion
that the customers are actually at fault.
The terms "big labor," "gun-control lobby," etc., are just as common as their political opposites.
There is a greater tendency to refer to business-sponsered
groups as "lobbies" and grassroots organizations as
"activists" but I don't see that as exactly being prejudicial.
(Members of pro-life and pro-choice groups are both typically
called "activists," for example.)
"Big Government" has been tossed around as part of
anti-liberal rhetoric even more frequently than
"Big [insert Business of choice]" has in anti-conservative
circles.
The arguments for liberal media bias are typically anecdotal.
Liberals have been soundly (and accurately, IMHO)
criticized for this sort of thing -- there are few matters
so clear-cut that an anecdote supporting either side can't
be found.
So it's amusing and a bit distressing to see conservatives
engage so heartily in similar tactics.
They certainly can pick enough examples of "liberal bias"
given the tens of millions of words of "news" broadcast
and printed each day.
Liberals have been doing this to "prove" how conservative
the media is for some time, so I suppose turnabout is fair
play.
Here's something to chew on:
far from failing in the marketplace,
people tend to be fairly accepting of liberal
points of view in local news -- much more so than
in national or world news.
(Fox was the first to figure this out, and thus made
the smart move of reporting their national and world news
accordingly.)
This is human nature.
People are much more likely to be generous and forgiving
to someone from their own community.
They are much happier to see their money spent on
fixing their own infrastructure and given as charity to their
own down-and-out than on some freeway or "welfare queen"
in a far-off city.
It's the same with talk radio.
Local programs can take a more liberal perspective, at
least when focused on their own community, but that
sort of thing doesn't travel well.
"Compassion" is a local phenomenon.
Away from home, people want restraints, they want
things kept under control.
(Many a conservative has scored points with the
electorate by pointing out that "liberal programs"
usualy mean government spending somewhere else.)
I frankly don't think the news media have gotten any
more or less biased, but I think they have gotten much
more cynical, with Fox leading the way.
The country is in a conservative mood, and they play
to that, but I'm quite certain that they'd be the first
to push a liberal line if the national mood shifted
that way.
So your idea that concervatism is winning in the
"marketplace" is dead on -- but I, for one, am a little
uncomfortable in putting the marketplace in charge
of how the important issues of the day are reported.
-Ed
Re:Disaster could have been averted
on
A 1974 Review of D&D
·
· Score: 4, Insightful
Actually, back then (yes, I know I'm old) I knew several
women who were D&Ders, and at least two male players
who met their future spouses playing the game.
I'm not sure when role-playing games became a guy thing,
but they didn't start out that way.
In some cases security has to take into account not only
current threats, but future threats as well.
Magnetic technology has been advancing quickly.
A technology which can pack a terabytes in a square inch
is also likely to be able to find and separate
the remnants of multiple writes at today's gigabyte densities.
If you have something you want to keep secret for the
next decade or two, it's prudent to take extreme
measures when you wnat to destroy it.
You're engaging in the same fallacy as the original poster.
You cannot change the license of the original work.
It's as simple as that.
No amount of handwaving will change this simple fact.
However, you can GPL your patch file, and thus prevent
your changes from being incorporated back into the
original and distributed under
the original license.
And you can claim compilation
copyright on the distribution of the original plus
your patch file.
But you cannot claim copyright on the entire result of
applying that patch file and so discard the claim
of the original copyright holder.
It may seem like a fine distinction given that the end
result is the same, but it is the form of the
distribution that is significant, not the form of
what eventually gets compiled.
Slashdot Mirror
The faster they push this along, the sooner this whole debacle will be over. Regardless of the lack of merit in SCO's central claims, in CIO-land uncertainty is anathema. It would have been worse for SCO to have simply left the situation as FUD.
I don't know quite where you'd find the news or weather, but dialing into the ARPANET 30 years ago was just a matter of knowing the phone number and having a terminal and modem. There was no security on the TAC (terminal access controller, what might be called a "modem pool"). There were a couple systems at MIT that allowed guest access, and if you managed to connect there you could request an account and most likely get it.
Open TACs and systems with guest accounts existed into the early 80s, when a few people started knocking on the wrong doors and DARPA (and eventually MIT) cracked down. But they're how I first got on the ARPANET back in 1977
I know people who have been programming in C for over a decade who think that a null pointer is always a zero. Can you name a common computer architecture where null isn't zero?
There are some really obscure architectures where null pointers aren't zero, but even in those cases well-written code will usually work just fine; the standard specifies that a zero value assigned to any pointer or cast to any pointer type takes on the value of the appropriate null pointer. Furthermore, a null pointer value converted to a _Bool is always false, so the expected thing happens when a null pointer value appears in a conditional context. Finally, when a function prototype exists, an implicit conversion "as if by assignment" (the words are from the standard) occurs. So you might say that the standard goes out of its way to make ignorance of null pointers relatively harmless in otherwise well-written code.
You don't know your hacker's lore. This article is about as on-topic as Slashdot gets.
Capitalism is like Communism in an important respect: both theories, in their pure form, fail to scale. (A corollary is that supporters of either can claim that their system "has never really been tried.") Information moves too slowly and is far too easily subverted for a large free market to function, even if you assume that buyers are perfectly rational. This is why third parties (private or governmental) need to be introduced for large markets to work. Private entities like Consumers Union can help to improve the information pool (insofar as they aren't subverted by special interests) but there are cases where government's ability to compel disclosure is required. Whether further regulation (beyond reasonable disclosure) improves the functioning of markets is more questionable. But a market where sellers control all the information isn't going to function well at all outside of an open-air bazaar.
Dennis Ritchie did just that: witness Plan 9 and Inferno. Or even those versions of Research Unix after V7. Unfortunately, the focus of AT&T and later Lucent was on commercialization, so the innovations found in these products were essentially inaccessible to the communities who could most appreciate and further them.
It's a sad fact that the original creators of Unix had very little to do with its evolution after 1980 or so.
It's an uncomfortable truth that complete suppression of terrorism requires complete suppression of freedom. If we want to maintain our freedom, we'll have to combat the fear of terrorism every bit as strongly as we fight terrorism itself. We'll have to risk that our promotion of freedom will at some points allow terrorism to operate. In a word, we need courage. But if we depend entirely upon our government and military to be courageous for us, we're already far along the road to losing our liberty.
Your lack of familiarity in these groups is telling. One of the most common efforts that the women-in-engineering groups I've seen make is in developing mentoring and tutoring programs. More general community outreach, like classroom visits, are another common activities. Also, networking is very important -- for many group members that's all they're interested in. Some male CS types can make it very difficult for women to network with them; they tend to interpret professional interest from a woman as an opportunity to hit on them. And this is just one of the issues women have that men typically don't; they need a place to discuss how to handle such situations. Although networking only with other women would be unecessarily limiting, it can be a safer and more comfortable starting point.
FreeBSD has focused on Itanium up until now, given that production hardware has been available for a while. There are people who already have them in their datacenters and who want to run FreeBSD on them. It's part of the pragmatism that is at the base of FreeBSD's philosophy.
NetBSD has more of a research-focused, "climb the mountain because it's there" philosophy. Compare mottos: "FreeBSD -- The Power To Serve" vs. "Of Course It Runs NetBSD." That's not to say NetBSD doesn't make a fine, practical embedded platform (say), or that FreeBSD is useless in research (especially when that research can result in improved performance in the datacenter, FreeBSD's home turf). It's just a difference in emphasis, and the BSD community is richer for it.
Actually, LA takes as many seismic precautions than any other metropolis in the world (probably equalled only by Tokyo). Evidence of this is that the only apartment building to actually collapse in the Northridge 'quake had serious code problems. Scores more were rendered uninhabitable, but among apartment buildings only the Northridge Meadows complex pancaked; it was later discovered that it had been built without some of the required reinforcing between floors. Most of the rest of the country has no such requirements, and under the building codes in, say, St. Louis, you would have had dozens of Northridge Meadows.
The problem with the Puente Hills fault is that it goes right under downtown. Although the buildings there are designed to withstand a large earthquake on more distant faults, they weren't designed for a Northridge-style event right under them. (Northridge was 6.8 Richter, more or less the size of event expected on the Puente Hills fault.)
I lived around a mile from the Northridge epicenter, in West Van Nuys. Real Estate values in our neighborhood went down briefly, and didn't start climbing until a year and a half later. But the market hardly colapsed. (It helped that the houses in our area were well-built, with only those with second-story add-ons suffering much, if any, structural damage.) Nobody I know moved to Phoenix. When we sold our house four years later, we got what we expected for it, around 20% more than its valuation shortly before the Northridge quake occured.
How do you (or he) explain this? When did Bjarne change his mind (and forget to update his website)?
It's a cool paper; check it out. If you have problems finding it, just Google for "B Stroustrup: Generalizing Overloading for C++2000. Overload, Issue 25. April 1, 1998."
Your first paragraph doesn't make much sense to me. The US Constitution says:
Now, how does this expressed intent relate to "expanding the public domain?" I suspect that it has more to do with facilitating commerce by limiting the public domain. But benefit to the public domain was certainly the intent for making the exclusive right for a limited time. And that is why your argument that DRM is fundamentally different from copyright is spot-on. I don't think your other claim holds up, however; although copyright hardly requires that a creator benefit financially from exclusivity, such rewards are certainly one of its benefits. Thus the business interests involved are hardly irrelevent.
It's a PITA once, for one person. Then everyone else uses the same setup, since they have the exact same card. I agree that there are a lot of variables involved in analog recording, and a lot of complex decisions to be made; a good recording engineer is as much artist as engineer. But virtually none of those variables exist in a simple analog out -> analog in loop. Once properly adjusted, full-scale on the D/A will result in full scale on the A/D -- maximum safe S/N, no clipping, and so on.
This has always puzzled me: why is there such an obsession with preventing bit-for-bit copies? Properly done, digital re-recording from the analog output is likely to cause considerably less distortion than MP3 encoding, even at 256Kb/s. And the RIAA will be the first to tell you that MP3's are by far the biggest "piracy" threat represented by the internet. Preventing digital output will only be a minor impediment to copying, while the inability to use digital interconnects is a major blow to functionality.
DARPA funding for BSD goes way back -- long before OpenBSD, FreeBSD, and NetBSD existed. One of the most important instances was DARPA's funding of the development of BSD's TCP/IP network stack back in the mid-1980's. This made BSD the first system in wide deployment that supported TCP/IP. It's hard to overestimate the affect that this has had on Unix and the Internet since then.
Sun can play both sides of the street. Since they paid some ungodly sum to buy full rights to Unix SVR4, they're out of SCO's reach. Even if SCO gets its way (which I doubt, but assume it for the sake of argument), Sun's Linux offering would be unaffected by SCO's suit (but see next paragraph). I know who I'd bet on in a matchup between Sun and SCO...
Of course, SCO's claims are completely incompatible with the GPL, so if they win, Linux would be stopped dead in its tracks until it can be somehow be made "untainted" by SCO's IP. And that's the real danger here. BSD was stopped dead in its tracks by a similar legal situation in the early 1990's; it almost certainly would have been stronger competition against commercial Unixes if that hadn't happened. SCO might even be attempting to duplicate this, and slow Linux down under a legal cloud for a few years even if they ultimately fail in court.
Of course, I hope IBM hits them like a load of bricks (and I expect they will). But don't expect it to blow over very quickly.
Yup, y'all are right. Teach me to post while working on other things... But the point remains: if transfering this one data source to a single destination takes up ~10% of the total bandwidth available, we're still not quite there yet...
I meant, of course, 900+ megabits/sec.
If SLAC is generating a terabyte of information a day, 900+ bits/second isn't nearly fast enough to transfer it.
(Yes, I know that only parts of that data are likely to be useful enough to transfer, but it does suggest that there is still quite a ways to go in the quest for bandwidth.)
So you'd argue that making someone liable for things likely to be beyond their control is a way to encourage those who have the control, but no liability, to reform?
That's novel.
Remember, the "malicious users" responsible for the incoming bandwidth aren't at all likely to be on his system, and as many postings here have pointed out, the customer could be up-to-the-minute with patches and even have shut down his server at the first sign of trouble, but that would have no effect in an inbound UDP flood (like the slammer worm) or a SYN flood (like a DDOS attack), and so forth. The malicious or unpatched originators of that bandwidth aren't likely to be motivated to prevent the flood by the fact that the victim pays (and perhaps the contrary if DDOS was the objective).
Now, whether the victim pays or not is a separate issue, and could be decided based on contract law and business considerations. I frankly don't think the original article gives enough information to argue one way or the other. But I don't see anything on which to base an opinion that the customers are actually at fault.
The terms "big labor," "gun-control lobby," etc., are just as common as their political opposites. There is a greater tendency to refer to business-sponsered groups as "lobbies" and grassroots organizations as "activists" but I don't see that as exactly being prejudicial. (Members of pro-life and pro-choice groups are both typically called "activists," for example.) "Big Government" has been tossed around as part of anti-liberal rhetoric even more frequently than "Big [insert Business of choice]" has in anti-conservative circles.
The arguments for liberal media bias are typically anecdotal. Liberals have been soundly (and accurately, IMHO) criticized for this sort of thing -- there are few matters so clear-cut that an anecdote supporting either side can't be found. So it's amusing and a bit distressing to see conservatives engage so heartily in similar tactics. They certainly can pick enough examples of "liberal bias" given the tens of millions of words of "news" broadcast and printed each day. Liberals have been doing this to "prove" how conservative the media is for some time, so I suppose turnabout is fair play.
Here's something to chew on: far from failing in the marketplace, people tend to be fairly accepting of liberal points of view in local news -- much more so than in national or world news. (Fox was the first to figure this out, and thus made the smart move of reporting their national and world news accordingly.) This is human nature. People are much more likely to be generous and forgiving to someone from their own community. They are much happier to see their money spent on fixing their own infrastructure and given as charity to their own down-and-out than on some freeway or "welfare queen" in a far-off city.
It's the same with talk radio. Local programs can take a more liberal perspective, at least when focused on their own community, but that sort of thing doesn't travel well. "Compassion" is a local phenomenon. Away from home, people want restraints, they want things kept under control. (Many a conservative has scored points with the electorate by pointing out that "liberal programs" usualy mean government spending somewhere else.)
I frankly don't think the news media have gotten any more or less biased, but I think they have gotten much more cynical, with Fox leading the way. The country is in a conservative mood, and they play to that, but I'm quite certain that they'd be the first to push a liberal line if the national mood shifted that way. So your idea that concervatism is winning in the "marketplace" is dead on -- but I, for one, am a little uncomfortable in putting the marketplace in charge of how the important issues of the day are reported.
-EdActually, back then (yes, I know I'm old) I knew several women who were D&Ders, and at least two male players who met their future spouses playing the game. I'm not sure when role-playing games became a guy thing, but they didn't start out that way.
In some cases security has to take into account not only current threats, but future threats as well. Magnetic technology has been advancing quickly. A technology which can pack a terabytes in a square inch is also likely to be able to find and separate the remnants of multiple writes at today's gigabyte densities. If you have something you want to keep secret for the next decade or two, it's prudent to take extreme measures when you wnat to destroy it.
You're engaging in the same fallacy as the original poster. You cannot change the license of the original work. It's as simple as that. No amount of handwaving will change this simple fact.
However, you can GPL your patch file, and thus prevent your changes from being incorporated back into the original and distributed under the original license. And you can claim compilation copyright on the distribution of the original plus your patch file. But you cannot claim copyright on the entire result of applying that patch file and so discard the claim of the original copyright holder. It may seem like a fine distinction given that the end result is the same, but it is the form of the distribution that is significant, not the form of what eventually gets compiled.