The Risks Digest

The RISKS Digest

Forum on Risks to the Public in Computers and Related Systems

ACM Committee on Computers and Public Policy, Peter G. Neumann, moderator

Volume 27 Issue 34

Wednesday 12 June 2013

Contents

NSA, FBI collecting content from Google, Facebook, other services
Stephen Lawson via Gene Wirchenko
"NSA to everyone: Take your PRISM, it's good for you"
R.X.Cringely via Gene Wirchenko
On PRISM and admins
PGN
New Xbox by NSA partner Microsoft will watch you 24/7
William Green via Henry Baker
Government wants to block much cell phone, etc. use in cars
WiReD via Lauren Weinstein
Yet another Google Update affecting many sites...
Paul Wilcock
Robbing a Gas Station: The Hacker Way
Nicole Perlroth via Matthew Kruk
Peninsula woman battles DMV over alleged false conviction
Paul Saffo
"Patients' Medical Info Left on Bus"
Gene Wirchenko
Energy-efficient lighting may worsen sleep deficiencies
Johnson/Greenhouse via Monty Solomon
Risks of spreadsheets—and leap seconds
Gene Wirchenko
Re: BA plane's emergency landing at LHR caused by maintenance error
Phil Smith III
Andy Cole
Dag-Erling Smorgrav
Jim Geissman
Re: Cowlings Coming Loose from Engines - Human Factors at Work?
John C. Bauer
Re: The Hazards of Gambling
Martin Ward
Info on RISKS (comp.risks)

NSA, FBI collecting content from Google, Facebook, other services (Stephen Lawson)

Gene Wirchenko <genew@telus.net>
Fri, 07 Jun 2013 13:41:46 -0700
Stephen Lawson, InfoWorld, 07 Jun 2013
http://www.infoworld.com/d/the-industry-standard/reports-nsa-fbi-collecting-content-google-facebook-other-services-220285

Reports: US spy agencies gather data in Internet and social network usage
under a six-year-old program called PRISM.


"NSA to everyone: Take your PRISM, it's good for you" (Cringely)

Gene Wirchenko <genew@telus.net>
Mon, 10 Jun 2013 10:49:49 -0700
http://www.infoworld.com/t/cringely/nsa-everyone-take-your-prism-its-good-you-220344
Robert X. Cringely, InfoWorld, 07 Jun 2013
Is this 2013 or 1984? It's hard to tell when reports say national security
services are delving into our private Apple, Facebook, Google, and Microsoft
accounts


On PRISM and admins

"Peter G. Neumann" <neumann@csl.sri.com>
Wed, 12 Jun 2013 11:31:37 PDT
The aftermath of this seems to be causing an enormous amount of media
coverage, much of which is not relevant here.  However, I am struck
by the dichotomy between two particular positions:

  On one hand, Edward Snowden (again on NPR this morning) adamantly insists
  that he had authorized access to every call.  (As usual, we can quibble
  over what "authorized" means, especially when in some cases no
  authorization is required!)

  On the other hand, many government people and some others state that
  Snowden's claim is impossible, implying that he must be lying, bloviating,
  overendowing his abilities, or whatever.

It seems to me that in most systems in use today (typically with many
inherent security design flaws and exploitable software bugs, operational
misconfigurations, subvertible audit trails, and enormous opportunities for
insider misuse—partly because of inadequate access controls), system
administrators often have direct or indirect access to essentially
everything, and perhaps even worse, they might supervisors who do not have a
good grasp of the risks.

On the *other* other hand, because of secrecy, inadequate monitoring, and
other factors, it is often difficult to know what is really going on.

The Trusted Xenix system might have been a rare potential approach to
blocking overly powerful admins (with something like 17 different
admin-relevant privileges and mandatory access controls), but then it was
only a B2 system under the old Orange Book evaluation criteria and still had
many potential vulnerabilities.

Incidentally, reminiscing on George Orwell's novel *1984* (NPR noted this
morning that sales of the book increased by 6000 percent in the past week),
I noted that in the ubiquitous *1984* banner, Big Brother might now be
replaced with Big Data:

   Big Data Is Watching You!


New Xbox by NSA partner Microsoft will watch you 24/7 (William Green)

Henry Baker <hbaker1@pipeline.com>
Fri, 07 Jun 2013 13:34:33 -0700
FYI—Perhaps the new Tivo box can notice when I fall asleep, so that it
can pause the playback so I won't miss anything!

http://dailycaller.com/2013/06/07/new-xbox-by-nsa-partner-microsoft-will-watch-you-247/

William Green, *Daily Caller*,  7 Jun 2013

Possible privacy violations by Microsoft's upcoming Xbox One have come under
new scrutiny since it was revealed Thursday that the tech giant was a
crucial partner in an expansive Internet surveillance program conducted by
the National Security Agency and involving Silicon Valley's biggest players.

One of the console's key features is the full integration of the Kinect, a
motion sensing camera that allows users to play games, scroll through menus,
and generally operate the Xbox just using hand gestures. Microsoft has
touted the camera as the hallmark of a new era of interactivity in gaming.

What Microsoft has not promoted, however, is the fact that *you will not be
able to power on the console without first enabling the Kinect*, designed to
detect both heartbeats and eye movement. and positioning yourself in front
of it.

Disturbingly, a recently published Microsoft patent reveals the *Kinect has
the capability to determine exactly when users are viewing ads* broadcast by
the Xbox through its eye movement tracking.  Consistent ad viewers would be
granted rewards, according to the patent.

Perhaps the feature most worryisome to privacy advocates is the *requirement
that the Xbox connect to the Internet at least once every 24 hours.* Many
critics have asserted that Microsoft will follow the lead of other Silicon
Valley companies and use their console to gather data about its users,
particularly through the Kinect, and collect it through the online
connection users can't avoid.

Microsoft has promised that customers will be able to `pause' the camera's
function, but have put off questions on the precise specifics of their
privacy policies.


Government wants to block much cell phone, etc. use in cars (WiReD)

Lauren Weinstein <lauren@vortex.com>
Fri, 7 Jun 2013 12:24:33 -0700
  "Nathaniel Beuse, associate administrator for vehicle safety research at
  the National Highway Traffic Safety Administration, says government
  regulation coupled with standards set by automakers and the electronics
  industry could reduce fatalities. He says we need "a technological
  solution, some sort of innovation" in which the device or the car would
  recognize when the driver is using a mobile device and deactivate
  it. "This would be the ultimate solution," he says."
  http://j.mp/13qqY5h  (Wired via NNSquad)

In a word: impractical.


Yet another Google Update affecting many sites...

"Paul Wilcock" <paul@thebluecube.co.uk>
Fri, 07 Jun 2013 09:35:18 +0100
Another version of Google's major Penguin update hit last week! That's now
officially confirmed, so you have you been hit?

When launched last time, the update affected not only black-hat, but also
legitimate sites, with many of them unable to recover their rankings even
now.

And as the new update has been even harder, we recommend you take our free
Website Review to make sure your site goes through the new wave of ongoing
Penguin updates unaffected:

http://www.thebluecube.co.uk/link.php?M81558&N2&L=3&F=T


Robbing a Gas Station: The Hacker Way (Nicole Perlroth)

"Matthew Kruk" <mkrukg@gmail.com>
Fri, 7 Jun 2013 17:15:17 -0600
http://bits.blogs.nytimes.com/2013/06/06/robbing-a-gas-station-the-hacker-way/?nl=todaysheadlines&emc=edit_it_th_20130607

Nicole Perlroth, Bits - Business, Innovation, Technology, Society,
*The New York Times*, 6 Jun 2013

Thieves of the future will look back on today's stick-up artists and have a
good old belly laugh. Why would anyone ever rob a cashier with a gun, when
all that is needed is a smartphone?

Matt Bergin, a security consultant at Core Security, discovered he could
hack a cash register remotely, popping it open, by sending two digits from
his smartphone to the service running on the cash register's point-of-sale
system.  No gun or holdup note was required. He was able to do so through a
vulnerability in Xpient, which makes point-of-sale software that runs on
cash drawers.

"It was extremely trivial," Mr. Bergin said in an interview Wednesday. He
reverse-engineered Xpient's point-of-sale system, expecting that to interact
with it he would have to crack a password or break through a layer of
encryption. To his surprise, he encountered neither. By simply sending a
two-digit code from his phone to the point-of-sale system, he discovered he
could pop open the cash register remotely.

Christopher Sebes, the chief executive of Xpient, said in an interview
Thursday that the company had issued a patch for the vulnerability, which
Xpient customers can download to their systems. Mr. Sebes noted that
customers who had a Windows firewall switched on would be protected from the
hack, regardless of whether they had downloaded the patch. He also noted
that someone could just as easily pop open a cash register by physically
hitting the "No Sale" button on the register itself.

Increasingly, criminals are finding ways to use digital tactics for physical
theft. In February, thieves stole $45 million from thousands of New York
City A.T.M.'s in a few hours using a few keystrokes. It was one of the
largest heists in New York City history, the authorities said, on par with
the 1978 Lufthansa robbery at Kennedy Airport that inspired a scene in the
1990 film "Goodfellas."


Peninsula woman battles DMV over alleged false conviction

"Saffo.com" <paul@saffo.com>
Fri, 7 Jun 2013 18:59:40 -0700
Wanna bet this is a database glitch?  PS

http://www.mercurynews.com/san-mateo-county-times/ci_23394630/peninsula-woman-battles-dmv-over-alleged-false-conviction

Peninsula woman battles DMV over alleged false conviction

Joshua Melvin <jmelvin@bayareanewsgroup.com>, 5 Jun 2013

REDWOOD CITY—Maryann Raab says she hasn't been to Florida since 1977, yet
the DMV claims it has proof she was convicted of DUI there in 2005 and as a
result suspended her driver's license last month.

In an effort to clear things up, the Portola Valley woman's lawyer contacted
a DMV clerk May 29 and then the agency's legal affairs office. The DMV
response? Take it to court, they said, and so she did.

Raab, 63, won a temporary hold Monday on the suspension in San Mateo County
Superior Court, where she filed an appeal of the suspension the same
say. Judge George Miram ordered both sides back July 11 for a hearing, which
means the dispute is far from over.

"They can't prove it, because it wasn't me," Raab said in a phone
interview. "They'll have a pretty hard time there."

Department of Motor Vehicles spokeswoman Jan Mendoza declined to discuss any
aspect of the case.

"It's an open litigation, and we don't make comment," she said.

Raab was hit with a letter May 22 announcing the suspension, which was
effective immediately, according to her appeal. The order cited a June 6,
2005 "DUI—Drug" conviction somewhere in Florida. It didn't provide a case
number, any court documents or even a court or city name, the appeals says.

Raab says she's never had a DUI anywhere, ever. And she claims she hasn't
been to Florida since 1977 for a visit she barely recalls. She said since
she lives in a rural Portola Valley, walking or taking public transit isn't
a realistic option for her. Also, she says she's on the Mission Hospice
board and that involves numerous meetings per week in neighboring San Mateo.

Her attorney contacted a DMV clerk on May 29 to try to fix the problem. But
the clerk said Raab's file had no other information about the Florida
conviction. So that same day her lawyer talked to.

Jennifer Berry at the DMV legal affairs office, who said the agency wouldn't
set aside the stay, the appeal says.

Instead she said "Ms. Raab's only remedy would be through the courts,"
according to the appeal.

Berry also told Raab' lawyer the DMV can't remove the conviction from the
system because the conviction information came from the National Driver
Register. It's a national database that lists drivers who have had their
licenses revoked or suspended for serious violations.

It wasn't immediately clear how Raab's name ended up in the database, but
drivers' records are put in and taken out by the DMV's around the
country. As the duration and full cost of the episode remained unknown, Raab
said she was left feeling "pretty helpless."

"There needs to be some verification," she said, referring to license
suspensions. "How do they know it's actually you?"


"Patients' Medical Info Left on Bus"

Gene Wirchenko <genew@telus.net>
Fri, 07 Jun 2013 21:59:05 -0700
  [This article is not on the newspaper's Website.  I did not consider it
  worth typing it all in.  I will keep it for a week in case you want more
  data.]

Not all risks are computer-related even if the risk is data-related.  The
front-page story for today (June 7, 2013) in The Daily News (Kamloops,
British Columbia, Canada) was "Patients' Medical Info Left on Bus".

a couple of sentences:

  "The document on the bus was a list with 83 patients' names, health numbers
  and other personal data."

  "The letter said the bus driver retrieved the file within a few minutes and
  turned it over to IHA."

(The letter being referred to was a letter from IHA—Interior Health
Authority—disclosing the breach.)


Energy-efficient lighting may worsen sleep deficiencies

Monty Solomon <monty@roscom.com>
Sun, 9 Jun 2013 14:39:09 -0400
Carolyn Y. Johnson and Pat Greenhouse, Energy-efficient lights a threat to
sleep?  *The Boston Globe*, 22 May 2013

For years, Dr. Charles Czeisler has studied sleep. The chief of the division
of sleep medicine at Brigham and Women's Hospital knows that the invention
of the light bulb has profoundly altered human life and biology.

But in a perspective piece published Wednesday in the journal *Nature*,
Czeisler argues that the sleep deficiencies that have become so pervasive
among adults and children may be threatened yet again by technological
progress: LEDs (light-emitting diodes) that consume far less energy than
incandescent bulbs.

The reason Czeisler is wary of LEDs, which are already in our laptops,
televisions, and cellphones, is that they are projected to become even more
abundant as governments retire the incandescent bulb and shift toward more
energy-efficient sources of light. That matters because it isn't just all
artificial light that can reset our body clocks; the exact type of light
makes a difference. ...

http://www.bostonglobe.com/lifestyle/health-wellness/2013/05/22/tired-energy-efficient-lighting-may-worsen-sleep-deficiencies/kAXK0yBbQDrwN9A6kmy9vJ/story.html

Energy-Efficient Lightbulbs May Have Dark Side When It Comes To Health
8 Aug 2012
http://www.npr.org/blogs/health/2012/08/08/158426970/energy-efficient-lightbulbs-have-a-dark-side-when-it-comes-to-health


Risks of spreadsheets—and leap seconds (Frankston, RISKS-27.33)

Gene Wirchenko <genew@telus.net>
Fri, 07 Jun 2013 12:02:20 -0700
> ... I wonder how many financial instruments depended on the 1-2-3 bug,
> which treated 2000 as a leap year. ...

I think Frankston meant 1900, not 2000.  Mind you, there were people who
thought that 2000 would not be a leap year.

  [Quite a few other RISKS readers noted that 2000 *was* a leap year.  PGN]


Re: BA plane's emergency landing at LHR caused by maintenance error

"Phil Smith III" <lists@akphs.com>
Fri, 7 Jun 2013 14:14:25 -0400
A co-worker's dad retired from United as a Boeing maintenance guy; I thought
his comments on the incident were interesting.

  With the disclaimers that I have been retired for more than 10 years, and
  do not have any experience with Airbus manufactured aircraft, I have the
  same initial reaction as everyone else.

  How did the flight crew miss a well identified hazard of the model of
  aircraft they fly during their preflight? They most certainly received
  detailed printed bulletins outlining the risk, including the appearance in
  both the normal and hazardous configurations.

  From what I have observed, the missing variable is the motivation of the
  crewman assigned the preflight duties to kneel down and examine the
  latches for proper closure. I have seen pilots make the inconvenient extra
  effort on each and every preflight. I have also seen examples of
  preflights that would be challenged to find a missing engine. It's up to
  the individual, someone who may be having a bad day like we all do. Or, it
  may have been raining during the preflight. The pilot would have gotten
  dirty and wet looking under a dripping engine.

  The same can be said for the mechanic who was tasked with completing the
  maintenance that required opening the cowling. He did not perform his task
  properly. While there are government mandated schedules to meet, it is up
  to the individual to exercise the self discipline needed to complete all
  tasks safely. Depending on the level of maintenance performed, there may
  have been a final inspection performed by an Aircraft Inspector, a
  different pair of eyes. For all I know, this job may have been eliminated
  to keep ticket prices low.

  Next is the design of the latches. Because the cowling is heavy and
  flexible, it is unwieldy. The latches have a first level of attachment,
  typically called the safety latch, that easily catches and controls the
  cowling. It is called a safety latch because it allows a mechanic to
  safely close the cowling without injury, not because it is safe to fly the
  aircraft using it.

  Next is the second level of attachment where the latch configuration
  provides mechanical advantage for the mechanic to pull the cowling tight
  against the engine frame, thus providing physical rigidity. This rigidity
  streamlines the cowling and keeps the boundary layer of high speed air
  from ripping it off the engine. Think of a 90 mph tornado getting under a
  roof eave and lifting the roof. Imagine what a 600 mph wind can do to a
  piece of sheet metal. The rigidity also stretches the cowling over engine
  structures which provide cowling support, along with the
  compartmentalization of the immediate engine exterior for fire
  control. This all imparts structural strength to the cowling, much like
  riveting fuselage skin to the circumferentials and longerons.

  The final level of cowling attachment is typically in the form a quick
  attach screw. It functions as a safety to securely lock down the second
  level attachment lever. It would seem that the design of the cowling
  latches does not have any visible way of observing an improperly latched
  cowling with a quick glance, or by sliding one's hand over the latch.

  Long ago, the NTSB noted that it takes three independent events coming
  together at the same time to enable a crash. In large part, this is
  because of the numerous redundancies designed into aircraft. It is worth
  noting that, in this case, no crash occurred despite the loss of
  structural parts of the aircraft.

  Aircraft design and maintenance are now mature industries. The race is on
  to reduce costs and meet mandated on-time performance metrics by utilizing
  these build in redundancies. As with all human endeavors, determining what
  is adequate, and how much it costs, is subject to trial and error. This
  event seems to be on the error side of the seesaw.


Re: BA plane's emergency landing at LHR ... (Page, RISKS-27.33)

"Andy Cole" <andy.cole3@baesystems.com>
Fri, 7 Jun 2013 07:14:11 +0000
Further to Page's note, Reuters referenced the AAIB report into the incident
that said:

  "...fastening the fan cowl door latches usually required maintenance
  personnel to lie on the ground to reach the latches, and that the latches
  were difficult to see unless the person was crouching down."

This would make inspection to ensure that latches are properly secured
difficult and possibly explains the number of reported incidents.

Andy Cole, Successor Whole Boat Software Lead, BAE Systems Maritime -
Submarines Phone: +441229873334 Mobile: +44(752) 5389787


Re: BA plane's emergency landing at LHR ... (Page, RISKS.27.33)

Dag-Erling Smorgrav <des@des.no>
Fri, 07 Jun 2013 12:12:43 +0200
This could arguably be considered a user interface issue.  Both the A320
family (which, confusingly, includes the A319) and its main competitor, the
B737, have very low-slung engines.  On that particular plane, the cowl
latches are on the bottom of the engine (I believe this varies from engine
model to engine model), and you practically have to crawl under the engine
to verify that they are closed.

This reminds me of the DC-10 rear cargo door, where the handle could be
forced into the closed position even if the lock was not properly engaged.
This design error killed 346 people and injured another 11 in two separate
incidents.


Re: BA plane's emergency landing at LHR caused by maintenance error

"Jim Geissman" <jgeissman@socal.rr.com>
Sat, 8 Jun 2013 16:36:26 -0700
> "A lump of metal falling from a preceding plane is now thought to have
> caused the Concorde disaster at Paris some years ago."

A few years ago my VW Passat was subject to a recall, where the dealer put a
sheet of Kevlar in the wheel well on the side with the fuel tank, to protect
against this very risk.  Concorde should have done the same.


Re: Cowlings Coming Loose from Engines - Human Factors at Work? (RISKS-27,32,33)

"John C. Bauer" <johncbauer.xx@gmail.com>
Mon, 10 Jun 2013 21:12:49 -0400
If cowlings on a particular type of engine and nacelle come loose 33 times,
then there may be more behind it than simple carelessness.  A human factors
investigation may be indicated.  We no longer automatically ascribe
accidents to "pilot error" without delving into the complete sequence of
events and associated outside factors.  Similarly merely exhorting mechanics
to be more careful may not be the right approach for this situation.


Re: The Hazards of Gambling (Drewe, RISKS-27.32)

Martin Ward <martin@gkc.org.uk>
Fri, 7 Jun 2013 10:07:06 +0100
There are several falsehoods and fallacies in Chris Drew's response.

> My favourite quote here is "a politician who robs Peter to pay Paul can
> probably rely on Paul's vote".

This is obviously false.  The economic policies proposed by the Green Party
would benefit the vast majority of the population.  So why doesn't the Green
Party win every election with a landslide victory?

> If the Government takes money off rich people and gives it to poor people,
> this may seem to be "fairer" and reduce inequality, but it rewards people
> who rely on welfare and punishes those who provide for themselves

Throwing a lifeline to a drowning man is "rewarding" him for falling in the
river? While allowing the banker to keep only half of the 5M-pound bonus he
awarded himself this year is such a severe "punishment" that nobody wants to
be a banker? Recently the UK Government cut public services to the extent
that 500,000 jobs were lost, in order to fund a cut in the top tax rate from
50% to 45%.  So that the banker can now take home another 250,000 pounds of
his bonus.  Even The Economist, that hotbed of communist sympathies, warned
that these cuts would prolong the recession.  Which they did.

> (hence in the UK a lifetime on welfare is quite a popular career option).

Ask anyone on welfare whether they would rather be a banker earning
5-million pounds per year: few would refuse.  Yet somehow, bankers need
millions in "incentives" to get them to do the job: even when they make a
terrible hash of it!

The fallacy presented by the media is that Governments are in trouble
because they are taking more and more from the rich and giving it to the
poor.

The reality is that over the last 40 years, productivity has almost
*doubled* while the median wage has fallen, the poor are substantially worse
off, and taxes on the rich are at the lowest ever: with many of the largest
corporations paying virtually no tax.  So where has all the extra production
gone?  It has been appropriated by the wealthy: the top 1% now own more than
half of all the wealth in the country.

The biggest and most damaging fallacy is that cutting taxes creates jobs
while higher minimum wages will increase unemployment.

This "seems" obvious but it is not how companies work.  Give them a tax
break and it goes straight to the bottom line and executive bonuses: they
will not employ a single extra person.  Allow them to halve the wages of all
minimum-wage employees, and the same thing will happen.  Force them to pay
higher wages and they may have to raise prices.  BUT since labour costs are
less than 100% of the total price: the price rise will be smaller than the
wage rise.  For example, a $1 increase in the minimum wage would add only 2
cents to the cost of a hamburger.  Nobody wants prices to rise: but if the
minimum wage doubled, and at the same time all prices rose by even as much
as 50% then all the workers would be able to buy more, despite the higher
prices, thereby boosting the economy and creating more jobs.

Nick Hanauer directly addresses these fallacies in his Ted talk:
http://www.youtube.com/watch?vx2Y5HhplI

The country with one of the highest tax rates, with correspondingly high
unemployment benefits and minimum wage (Norway) also has the highest rate of
successful business startups.

As I said about gambling: for some the moral argument (in this case: paying
your workers sweatshop wages while taking millions in bonuses is stealing)
is stronger than the economic one.  But even those who believe that
economics trumps morality should be convinced by the economic argument.

To bring this post back into relevance to comp.risks: in the 70s the
potential for automation and computerisation were recognised as presenting a
challenge as to what kind of society we wanted to build.  For the first
time in history, it would be possible to keep full employment and full
productivity while greatly reducing working hours for everyone.

What instead happened was rising productivity meant that companies could get
away with employing fewer workers, which meant rising unemployment, and an
increasing pool of desperately poor people willing to work for ever smaller
wages.  Which, in turn, means that expensive robots are not needed when
sweatshop labour can do the same job for next to nothing.

In the US in 2010 a remarkable 93% of the growth in incomes accrued to the
top 1% of the population.--Emmanuel Saez, 2nd March 2012.  Striking it
Richer: the Evolution of Top Incomes in the United States (Updated with 2009
and 2010 estimates).
http://elsa.berkeley.edu/~saez/saez-UStopincomes-2010.pdf

"To end extreme poverty worldwide in 20 years, Sachs calculated that the
total cost per year would be about $175 billion.  This represents less than
one percent of the combined income of the richest countries in the world."

"In 2012, the world's 400 richest people became $241 billion richer".

http://www.visionofearth.org/economics/ending-poverty/how-much-would-it-cost-to-end-extreme-poverty-in-the-world/

STRL Reader in Software Engineering and Royal Society Industry Fellow
martin@gkc.org.uk  http://www.cse.dmu.ac.uk/~mward/

Please report problems with the web pages to the maintainer

Top