Come back from vacation and find a half-dozen credit cards opened in our name, all with $1K-$2K already charged on them. They were opened over a period of 48 hours, across a swath of states over 1K miles from our residence. Mostly MC/Visa; a few store cards.
One or two banks called our home number, and since we weren't there to confirm, refused to open the account. The rest all opened the accounts and allowed the first-day charges.
We filed a police report and contested every charge. There were some forms to fill out and a few phone calls, but every bank "resolved the dispute in our favor", they all notified the credit reporting agencies that it was fraud, and the whole thing just went away over the next few months.
A few of the banks subsequently tried to sell us credit monitoring services (meant to keep this kind of thing from happening) which I thought was a bit slimy.
Raising prices lowers JoinRate and raises LeaveRate, and therefore shrinks the user base. Lock-in reduces the effect of Price on LeaveRate. But (most likely)
NonUsers >> Users
so the effect of Price on JoinRate may swamp the effect of lock-in on LeaveRate.
They do something like this in Florida. They put a sign up on one of the interstates saying "Drug Checkpoint Ahead". There is no checkpoint, and if there was one, it would probably be unconstitutional. But they don't need one. They just pull over everyone who suddenly pulls a U-turn across the median (which is a genuine traffic violation).
I saw the same thing back in the mid-1990s. Sequencing technology was ramping up hyper-exponentially. That means that it curves up on semi-log paper. It was outstripping Moore's Law, and crushing our data systems.
Finished DNA sequence only needs 2 bits/base pair, but the raw data behind those 2 bits can be much bigger; in our case, the raw data was scanned images of radiograms.
In the early '90s, a typical sequencing project was a few hundred DNA fragments. Each fragment is a few hundred base pairs. You put each fragment in a file, and put all the files in a subdirectory on disk. The file system becomes your de facto database, and it works OK, because there are only ~10K base pairs in the whole project.
When I got there, a typical sequencing project had grown to a few thousand fragments. They were still keeping everything in the file system, and it took a dozen networked DECstations with big (1GB!) drives to manage all the projects.
Then the biologists went 40x in one generation. (Remember, RAM only goes 4x per generation.) That meant that a sequencing project now had 40K fragments. They bought HUGE (9GB!) drives, and had a TB of storage online, which was unprecedented at the time.
They were still keeping all the fragments in files in subdirectories, because who has time to rewrite your data systems when you have all that DNA to sequence. The system was slowly grinding to a halt; the weekly (tape) backups were taking longer than a week to complete; they fell behind on their OS updates. Eventually there was file system corruption, and then the whole thing came crashing down around our ears.
It was an exciting time.
FWIW, the graphs in the article show don't show continuous hyper-exponential growth. They show two discrete jumps (doubtless due to introduction of new sequencing technologies), one in 2003 and one in 2008. After each jump, the growth rate returns to exponential (straight line). And the last three data points show the cost bottoming out at $0.30/MBase. Of course, that may just be the plateau before it falls off the next cliff.
Today's NY Times gives the first account I have seen of the actual circumstances and allegations
The two women who accused him were volunteers who had offered to assist WikiLeaks and met him in his first days in Sweden.
According to accounts the women gave to the police and friends, Swedish officials have said, they had consensual sexual encounters with Mr. Assange that became nonconsensual. One woman said that Mr. Assange had ignored her appeals to stop after a condom broke. The other woman said that she and Mr. Assange had begun a sexual encounter using a condom, but that Mr. Assange did not comply with her appeals to stop when it was no longer in use.
The existence of the allegations is a fact.
We have two competing hypotheses to account for that fact
Assange committed sexual assault
Assange was set up by someone who doesn't like WikiLeaks
Having read the allegations, I find it very difficult to discount the second hypothesis.
The United States Constitution, amendment 6, provides that
In all criminal prosecutions, the accused shall enjoy the right to a speedy and public trial...
however,
In Sheppard v. Maxwell, 384 U.S. 333 (1966), the Supreme Court ruled that the right to a public trial is not absolute. According to Press-Enterprise Co. v. Superior Court, 478 U.S. 1 (1986), trials can be closed at the behest of the government on account of there being "an overriding interest based on findings that closure is essential to preserve higher values and is narrowly tailored to serve that interest."
I guess Goldman Sachs' trade secrets are a higher value...
The headline says "crashes". The article says "interrupted", but gives no details. The article has two pictures (#18 and #19). #19 looks like Stallman posing after the event for the benefit of the camera. #18 is probably the interruption. All you can see from the picture is that Stallman (and friend) stood at the front of a conference room holding poster-board signs. It looks like Stallman has a sheaf of papers in his hand, so maybe he said something.
Schneier on Security
A blog covering security and security technology.
April 1, 2006
Announcing: Movie-Plot Threat Contest
For a while now, I have been writing about our penchant for "movie-plot threats": terrorist fears based on very specific attack scenarios. Terrorists with crop dusters, terrorists exploding baby carriages in subways, terrorists filling school buses with explosives -- these are all movie-plot threats. They're good for scaring people, but it's just silly to build national security policy around them.
But if we're going to worry about unlikely attacks, why can't they be exciting and innovative ones? If Americans are going to be scared, shouldn't they be scared of things that are really scary? "Blowing up the Super Bowl" is a movie plot to be sure, but it's not a very good movie. Let's kick this up a notch.
It is in this spirit I announce the (possibly First) Movie-Plot Threat Contest. Entrants are invited to submit the most unlikely, yet still plausible, terrorist attack scenarios they can come up with.
Your goal: cause terror. Make the American people notice. Inflict lasting damage on the U.S. economy. Change the political landscape, or the culture. The more grandiose the goal, the better.
Assume an attacker profile on the order of 9/11: 20 to 30 unskilled people, and about $500,000 with which to buy skills, equipment, etc.
I knew a guy who had the fallout symbol tattooed across his left shoulder blade, maybe 4 inches across, in all its black and yellow glory. I asked him why, but I don't recall his response.
He did allow that reading physics textbooks in coffee shops was a good way to pick up girls.
What we're seeing here are the results of reality shear (props to Neal Stephenson).
Historically, people had separate legal and ethical frameworks for managing tangible objects and for managing speech.
The basic rule for objects--respected by almost everyone--is don't take other people's objects.
The basic rule for speech--generally respected by democratic governments--is you can say what you want and you can hear what you want. You also have some privacy rights in your speech.
Now the internet has inextricably and irreversibly enmeshed these two very different frameworks. Things that used to be objects (CDs, DVDs, etc) can now be moved around by acts of speech (FTP, BitTorrent, etc.).
Copying infringes the content owners property rights, and they are enraged. They have responded in three ways. Social : convince people that copying is theft, and hope that people's natural moral aversion to theft will dissuade them from copying things. Technical: DRM Legal : copyright enforcement; ISP regulation; 3-strikes, etc.
Socal doesn't work. People don't think that copying is theft (because it doesn't deprive the owner of a tangible object), and you can't rewrite people's ethical systems with a PR campaign, no matter how slick or how insistent.
Technical doesn't work. DRM doesn't stop pirates, it just annoys your paying customers.
Legal responses necessarily infringe people's conceptions of their own speech rights. What used to be a free and private act of sending and receiving signals over the internet is how subject to review, judgement, and punishment by the the government and corporations.
Just as you can't convince ordinary people that copying is theft, you can't convince ordinary people that speech acts are morally wrong. Not the kind of wrong that really guides people's actions. The kind they learned as children: don't hit, don't steal, don't lie.
So people see the legal responses of the content owners as grave infringements of their own legitimate speech rights. And they get enraged.
So we have two groups of people, each enraged, each convinced of their own right, and working from incompatible premises.
I seem to recall there was an organization that used those tactics in Britan. Irish Republican something-or-other. Some dispute over a piece of Ireland.
I haven't heard much about them lately, but they're probably still around somewhere. You might want to check on how those tactics worked out for them.
Microsoft operating systems can never patch without rebooting, because in Microsoft file systems, a file cannot exist without a name.
So you can't replace the running copy of a.exe or.lib with a new version, because you have to delete the old version first to free up its name, and you can't delete a running executable.
Instead, the new version is staged in a temp area, the computer reboots, and the OS replaces the old copy with the new copy early in the boot sequence, before the old copy starts running.
Microsoft can never fix this problem, because if they fixed this problem, (and others like it) then Windows would become Unix, and then you might as well run Unix, and then there wouldn't be any Microsoft.
Developers invest their own time and (sometimes) money learning to develop for a platform.
Some of the value of that investment accrues not to the developer, but to the platform owner.
There is constant struggling and squabbling among all parties to capture more of this value.
Typically, vendors create walled gardens (e.g. with restrictive licensing or non-portable features) to try to contain the value created by developers, while developers try to break out of those gardens (e.g. by writing compatibility layers).
The case with Apple and Flash is a bit unusual, in that Apple is walling the developers out, not in. But the underlying struggle is the same: Apple and the developers are both trying to get a bigger piece of the pie.
My wife's machine got hit last week. No idea where it came from. Been running for years with no problem. (NetGear router seems to keep the baddies out.)
All of a sudden there's a dozen dialogs flashing dire warnings about viruses and trojans and keyloggers and malware and insisting that we "register" our copy of XP security.
Pulled the network cable and started googling (from a linux box). The thing is pretty nasty. It scatters pieces of itself around the file system with random names. Then it hooks the.exe registry keys so that it gets control each time any program is run, and takes the opportunity to spawn a new copy of itself, with new dialog boxes and systray icons.
After you delete the program files, nothing runs at all, because the.exe keys are still trying to redirect through the files you just deleted. (Hint: right click -> run as). Then I fixed all the.exe (and related) keys by hand. There's quite a lot of them, because it is really important for each user on a windows box to have their own semantics for running a program. (Removal instructions on the web don't generally find them all.)
Finally (should have done this long ago) created an admin account and knocked all the user accounts down to user privilege level.
Tabs 8 is a standard. If everyone uses tabs 8, then it is easy to move code (and other text files) around. If everything is portable, and everything is interoperable, then the OS is just infrastructure, and you can use whichever one you like.
That would be bad for MSFT.
Indent is not a standard: everyone uses whatever they like. By tying tabs to indent, they scatter tabs all over the map, which effectively breaks the tabs 8 standard.
Breaking standards promotes lock-in and monopoly. Microsoft is on the record as doing this to defend against competitive threats. (http://catb.org/~esr/halloween/).
The gods had condemned Sisyphus to ceaselessly rolling a rock to the top of a mountain, whence the stone would fall back of its own weight.
They had thought with some reason that there is no more dreadful punishment than futile and hopeless labor.
EITHER
you monitory your pages every day
reverting vandalism
patiently explaining to every newbie who wanders by why their edit is wrong, or inappropriate
all the while remembering that they aren't "your" pages,
and that all you can do is make your best evidence-based case
and hope that other agree with it...
OR
you don't, and you watch as bitrot and entropy slowly but relentlessly degrade the pages to something you can't bear to look at any more.
I maintained some pages for about a year, and then after one particularly nasty edit war I gave up. Not in a petulant "they won't have me to kick around any more" way. I just stopped caring so much. Wikipedia dropped off my mental list of sites that I check every day.
I still use Wikipedia—it's near the top of every SERP. But I haven't tried to edit anything there in years.
From TFA, it doesn't sound like they simulated the cerebral cortex of a cat.
It sounds like they simulated a neural net with a comparable number of neurons.
Not the same thing.
Monkeying with the equation that generates the Mandelbrot set seems misguided.
The true definition of the Mandelbrot set is the set of points for which the corresponding Julia set is connected. This is the original motivation for the equation. If you want to get an interesting 3D object, start by searching for an interesting collection of sets that are parameterized by three coordinates.
Slashdot Mirror
We got hit with this--or something like it.
Come back from vacation and find a half-dozen credit cards opened in
our name, all with $1K-$2K already charged on them.
They were opened over a period of 48 hours,
across a swath of states over 1K miles from our residence.
Mostly MC/Visa; a few store cards.
One or two banks called our home number, and since we weren't there to confirm, refused to open the account.
The rest all opened the accounts and allowed the first-day charges.
We filed a police report and contested every charge.
There were some forms to fill out and a few phone calls,
but every bank "resolved the dispute in our favor",
they all notified the credit reporting agencies that it was fraud,
and the whole thing just went away over the next few months.
A few of the banks subsequently tried to sell us credit monitoring services
(meant to keep this kind of thing from happening) which I thought was
a bit slimy.
Revenue = Price * Users
The user base grows until
NonUsers * JoinRate = Users * LeaveRate
Raising prices lowers JoinRate and raises LeaveRate,
and therefore shrinks the user base.
Lock-in reduces the effect of Price on LeaveRate.
But (most likely)
NonUsers >> Users
so the effect of Price on JoinRate may swamp the effect of lock-in on LeaveRate.
They do something like this in Florida.
They put a sign up on one of the interstates saying "Drug Checkpoint Ahead".
There is no checkpoint, and if there was one, it would probably be unconstitutional.
But they don't need one.
They just pull over everyone who suddenly pulls a U-turn across the median
(which is a genuine traffic violation).
Jump kits (Go bags)
You put 'em by the door for when you have to rock'n'roll.
http://www.sff.net/people/doylemacdonald/emerg_kit.htm
I saw the same thing back in the mid-1990s.
Sequencing technology was ramping up hyper-exponentially.
That means that it curves up on semi-log paper.
It was outstripping Moore's Law, and crushing our data systems.
Finished DNA sequence only needs 2 bits/base pair,
but the raw data behind those 2 bits can be much bigger;
in our case, the raw data was scanned images of radiograms.
In the early '90s, a typical sequencing project was a few hundred DNA fragments.
Each fragment is a few hundred base pairs.
You put each fragment in a file, and put all the files in a subdirectory on disk.
The file system becomes your de facto database, and it works OK,
because there are only ~10K base pairs in the whole project.
When I got there, a typical sequencing project had grown to a few thousand fragments.
They were still keeping everything in the file system,
and it took a dozen networked DECstations with big (1GB!) drives to manage all the projects.
Then the biologists went 40x in one generation.
(Remember, RAM only goes 4x per generation.)
That meant that a sequencing project now had 40K fragments.
They bought HUGE (9GB!) drives, and had a TB of storage online,
which was unprecedented at the time.
They were still keeping all the fragments in files in subdirectories,
because who has time to rewrite your data systems when you have all that DNA to sequence.
The system was slowly grinding to a halt;
the weekly (tape) backups were taking longer than a week to complete;
they fell behind on their OS updates.
Eventually there was file system corruption,
and then the whole thing came crashing down around our ears.
It was an exciting time.
FWIW, the graphs in the article show don't show continuous hyper-exponential growth.
They show two discrete jumps (doubtless due to introduction of new sequencing technologies),
one in 2003 and one in 2008.
After each jump, the growth rate returns to exponential (straight line).
And the last three data points show the cost bottoming out at $0.30/MBase.
Of course, that may just be the plateau before it falls off the next cliff.
The existence of the allegations is a fact.
We have two competing hypotheses to account for that fact
Having read the allegations, I find it very difficult to discount the second hypothesis.
however,
I guess Goldman Sachs' trade secrets are a higher value...
A Girl's Guide to Geek Guys
http://www.neystadt.org/john/humor/Girls-Guide-To-Geek-Guys.htm
The headline says "crashes".
The article says "interrupted", but gives no details.
The article has two pictures (#18 and #19).
#19 looks like Stallman posing after the event for the benefit of the camera.
#18 is probably the interruption.
All you can see from the picture is that Stallman (and friend) stood at the front of a conference room holding poster-board signs.
It looks like Stallman has a sheaf of papers in his hand, so maybe he said something.
http://www.schneier.com/blog/archives/2006/04/announcing_movi.html
http://www.schneier.com/blog/archives/2006/06/movieplot_threa_1.html
Instead of symlinking to directories,
create directories of hard links to the files.
Then you can move files around whenever you like,
and you never have any dangling links.
I knew a guy who had the fallout symbol tattooed across his left shoulder blade,
maybe 4 inches across, in all its black and yellow glory.
I asked him why, but I don't recall his response.
He did allow that reading physics textbooks in coffee shops was a good way to pick up girls.
What we're seeing here are the results of reality shear (props to Neal Stephenson).
Historically, people had separate legal and ethical frameworks for managing tangible objects and for managing speech.
The basic rule for objects--respected by almost everyone--is don't take other people's objects.
The basic rule for speech--generally respected by democratic governments--is you can say what you want and you can hear what you want. You also have some privacy rights in your speech.
Now the internet has inextricably and irreversibly enmeshed these two very different frameworks. Things that used to be objects (CDs, DVDs, etc) can now be moved around by acts of speech (FTP, BitTorrent, etc.).
Copying infringes the content owners property rights, and they are enraged. They have responded in three ways.
Social : convince people that copying is theft, and hope that people's natural moral aversion to theft will dissuade them from copying things.
Technical: DRM
Legal : copyright enforcement; ISP regulation; 3-strikes, etc.
Socal doesn't work. People don't think that copying is theft (because it doesn't deprive the owner of a tangible object), and you can't rewrite people's ethical systems with a PR campaign, no matter how slick or how insistent.
Technical doesn't work. DRM doesn't stop pirates, it just annoys your paying customers.
Legal responses necessarily infringe people's conceptions of their own speech rights. What used to be a free and private act of sending and receiving signals over the internet is how subject to review, judgement, and punishment by the the government and corporations.
Just as you can't convince ordinary people that copying is theft, you can't convince ordinary people that speech acts are morally wrong. Not the kind of wrong that really guides people's actions. The kind they learned as children: don't hit, don't steal, don't lie.
So people see the legal responses of the content owners as grave infringements of their own legitimate speech rights. And they get enraged.
So we have two groups of people, each enraged, each convinced of their own right, and working from incompatible premises.
I don't know how we get past this.
I seem to recall there was an organization that used those tactics in Britan.
Irish Republican something-or-other.
Some dispute over a piece of Ireland.
I haven't heard much about them lately,
but they're probably still around somewhere.
You might want to check on how those tactics worked out for them.
Microsoft operating systems can never patch without rebooting,
because in Microsoft file systems, a file cannot exist without a name.
So you can't replace the running copy of a .exe or .lib with a new version,
because you have to delete the old version first to free up its name,
and you can't delete a running executable.
Instead, the new version is staged in a temp area,
the computer reboots,
and the OS replaces the old copy with the new copy early in the boot sequence,
before the old copy starts running.
Microsoft can never fix this problem,
because if they fixed this problem,
(and others like it)
then Windows would become Unix,
and then you might as well run Unix,
and then there wouldn't be any Microsoft.
I periodically try to buy media from some service that is trying to sell it to me. Invariably, their DRM doesn't run on my platform, and I give up.
Developers invest their own time and (sometimes) money learning to develop for a platform.
Some of the value of that investment accrues not to the developer, but to the platform owner.
There is constant struggling and squabbling among all parties to capture more of this value.
Typically, vendors create walled gardens (e.g. with restrictive licensing or non-portable features) to try to contain the value created by developers, while developers try to break out of those gardens (e.g. by writing compatibility layers).
The case with Apple and Flash is a bit unusual, in that Apple is walling the developers out, not in. But the underlying struggle is the same: Apple and the developers are both trying to get a bigger piece of the pie.
My wife's machine got hit last week.
No idea where it came from.
Been running for years with no problem.
(NetGear router seems to keep the baddies out.)
All of a sudden there's a dozen dialogs flashing dire warnings about viruses and trojans and keyloggers and malware and insisting that we "register" our copy of XP security.
Pulled the network cable and started googling (from a linux box). .exe registry keys so that it gets control each time any program is run, and takes the opportunity to spawn a new copy of itself, with new dialog boxes and systray icons.
The thing is pretty nasty.
It scatters pieces of itself around the file system with random names.
Then it hooks the
After you delete the program files, nothing runs at all, because the .exe keys are still trying to redirect through the files you just deleted. .exe (and related) keys by hand.
(Hint: right click -> run as).
Then I fixed all the
There's quite a lot of them, because it is really important for each user on a windows box to have their own semantics for running a program.
(Removal instructions on the web don't generally find them all.)
Finally (should have done this long ago) created an admin account and knocked all the user accounts down to user privilege level.
IT staffing firms don't fall into the first category, and web developers don't fall into the second.
I want web pages to stand still unless I type or click.
GUIs that respond to mouse position alone, with
- pop ups
- hover text
- raise/lower windows
- flashes or color changes
make me mental.
A GUI that responded to my eye movements...<shudder>...
Tabs 8 is a standard.
If everyone uses tabs 8,
then it is easy to move code (and other text files) around.
If everything is portable, and everything is interoperable,
then the OS is just infrastructure, and you can use whichever one you like.
That would be bad for MSFT.
Indent is not a standard: everyone uses whatever they like.
By tying tabs to indent, they scatter tabs all over the map,
which effectively breaks the tabs 8 standard.
Breaking standards promotes lock-in and monopoly.
Microsoft is on the record as doing this to defend against competitive threats.
(http://catb.org/~esr/halloween/).
EITHER
you monitory your pages every day
all the while remembering that they aren't "your" pages, and that all you can do is make your best evidence-based case and hope that other agree with it...
OR
you don't, and you watch as bitrot and entropy slowly but relentlessly degrade the pages to something you can't bear to look at any more.
I maintained some pages for about a year, and then after one particularly nasty edit war I gave up. Not in a petulant "they won't have me to kick around any more" way. I just stopped caring so much. Wikipedia dropped off my mental list of sites that I check every day.
I still use Wikipedia—it's near the top of every SERP. But I haven't tried to edit anything there in years.
It sounds like they simulated a neural net with a comparable number of neurons.
Not the same thing.
A few days ago, Slashdot ran The Math of a Fly's Eye May Prove Useful.
Those guys
and they still don't understand how the equations actually work.
That's where we are with brain simulation.
Monkeying with the equation that generates the Mandelbrot set seems misguided.
The true definition of the Mandelbrot set is the set of points for which the corresponding Julia set is connected. This is the original motivation for the equation. If you want to get an interesting 3D object, start by searching for an interesting collection of sets that are parameterized by three coordinates.
Some cautionary tales
http://world.std.com/~swmcd/steven/crypt/recruiters.html