[31805] in RISKS Forum

home help back first fref pref prev next nref lref last post

Risks Digest 31.39

daemon@ATHENA.MIT.EDU (RISKS List Owner)
Thu Aug 29 14:29:11 2019

From: RISKS List Owner <risko@csl.sri.com>
Date: Thu, 29 Aug 2019 11:29:03 -0700
To: <risks@mit.edu>

RISKS-LIST: Risks-Forum Digest  Thursday 29 August 2019  Volume 31 : Issue 39

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

***** See last item for further information, disclaimers, caveats, etc. *****
This issue is archived at <http://www.risks.org> as
The current issue can also be found at

"Why positive train control is vulnerable to a cyber-attack"
  (D G. Rossiter)
Frequency-sensitive trains and the lack of failure-mode analysis
  (Clive Page)
Inside America's Dysfunctional Trillion-Dollar Fighter-Jet Program
  (Valerie Insinna via Richard Stein)
Sometimes simplicity is dangerous ... (Rob Slade)
A Bitter Divorce Battle on Earth Led to Claims of a Crime in Space (NYTimes)
Premier's office accidentally publishes name of secret agent (TheAge)
WeWork's Wi-Fi network is laughably easy to hack (Fast Company)
Wake Up! Your House Is Calling (NYTimes)
OpenAI releases larger GPT-2 dataset. Can it write fake news better
  than a human? (Boing Boing)
SecurityWatch: Backstabbing, Disinformation, and Bad Journalism:
  The State of the VPN Industry (PCMag)
Security Researchers Find Several Bugs in Nest Security Cameras (VICE)
Found: World-readable database used to secure buildings around the globe
  (Ars Technica)
Credit card privacy matters: Apple Card vs. Chase Amazon Prime Rewards Visa
Regis University's technology systems targeted by malicious threat
  likely from outside the country (Denver Post)
A Harvard freshman says he was denied entry to the U.S. over
  social media posts made by his friends (WashPost)
Ring, the doorbell-camera firm, has partnered with 400 police
  forces, extending surveillance reach (WashPost)
FBI seeks to monitor Facebook, oversee mass social media data
  collection (Charlie Osborne)
Facebook's big win: Will this ruling have global impact on how
  your data is used? (Cathrin Schaer)
Re: Playing God: Japan temple puts faith in robot priest (Amos Shapir)
Re: Phishing spam is getting better (Amos Shapir)
Abridged info on RISKS (comp.risks)


Date: Sun, 25 Aug 2019 20:48:40 +0000
From: "D G. Rossiter" <d.g.rossiter@cornell.edu>
Subject: "Why positive train control is vulnerable to a cyber-attack"


Positive Train Control (PTC) is a federally-mandated replacement of
traditional rail signaling on the largest railroads with a network of on-
and off-train electronics to space trains and prevent collisions or
runaways.  Railroads are installing PTC on nearly 57,848 route miles and on
19,912 locomotives.

``Unlike other critical infrastructure, such as energy or water management
systems, rail networks have avoided regulations as lawmakers have focused
recent efforts on safety due to high profile crashes,'' says Jesus Molina,
director of business development, for Waterfall Security Solutions.  ``There
is no question that a PTC rollout without managing the cybersecurity risk
will open new attack vectors due to increased connectivity and new software
added to the networks and onboard train, In these cases, PTC may actually
decrease the safety of passengers due to an unacceptable increased risk of
cyberattacks that may lead to accidents.''

``The use of IT-focused security tools, in particular, software tools such
as firewalls to protect control critical networks is a huge mistake, and
with increasingly connected rail networks, it is becoming a dangerous trend.
The focus of critical control networks is to be reliable and safe, and IT
tools meant to protect data and confidentiality are not suitable to defend
them. The most secure rail sites are not concerned with the steadily
increasing sophistication of cyber-attacks, nor with the steadily increasing
rate of disclosure of new attack vulnerabilities in control systems,
network, firewalls and other security software, This is because the most
secure sites protect their automation systems from cyber-attacks physically,
with hardware-based solutions such as unidirectional security gateways.''

  In other words, this networked solution is not being treated as one linked
  to a physical reality, i.e., moving trains.  DGR


Date: Mon, 26 Aug 2019 23:20:14 +0100
From: Clive Page <clivegpage@gmail.com>
Subject: Frequency-sensitive trains and the lack of failure-mode analysis

On 9 Aug 2019 around 4:53pm, lightning struck a transmission cable in
south-eastern England.  This had the unexpected result that a gas-fired
power station and a large wind-farm detected grid anomalies and
disconnected.  This loss of generating capacity made the frequency drop from
its nominal 50 Hz, reaching 48.8 Hz for a few seconds.  To restore it, the
grid control system cut power to about 1.1 million people for up to 50
minutes.  A report from OFGEM, the Government regulator describes the events
in more detail.

The railway system was much more badly affected, even though the traction
and signaling power had been maintained.  Most services from London to
Bedford, Cambridge, and Peterborough depend on electric trains built in
Germany by Siemens about two years ago.  It now turns out that these trains
stop if the frequency drops below 49 Hz.  About 60 of them were running at
the time: unfortunately only half of them could be restarted by the driver,
the others had to be visited by a technician which took many hours.  Many
stranded passengers had to walk along the tracks to the nearest station.
Even the inter-city services could not run as the lines were so badly
blocked by stalled suburban trains.  Practically no trains ran on these
lines until the next day and in total over 1200 train services were canceled
or delayed.

I found this a surprising failure because pretty much all domestic and
commercial equipment is designed to work on a wide range of frequencies,
especially to cope with both 50 and 60 Hz regions of the world.  The UK's
National Grid Code says that the mains frequency could be as high as 52 Hz
or as low as 47 Hz "in exceptional circumstances".  So it is unfortunate
that a train would be so sensitive to a 1.2 Hz deviation.  Indeed with
hindsight, one feels that a train that trips out at 49 Hz and then requires
a technician to reset it is a very poor design and could easily lead, as
this did, to a widespread system failure.

It seems to me that in several industries failure mode analysis is no longer
being performed adequately.  Taking the crash of AF447 in 2009: the initial
cause was that both pitot tubes froze up.  The second failure was that the
autopilots disconnected, leaving inexperienced pilots to cope unaided with
flying in the middle of the night at maximum altitude over a tropical storm
with some of their speed sensors not working.  In their panic they first
stalled and then crashed the plane, even though all they really needed to do
to the controls was absolutely nothing.  There are so many ways of measuring
the speed of a plane that the loss of two sensors should not, in my opinion,
lead to the autopilots simply giving up.  Pilots depend on them so much that
they ought to degrade more gracefully.  A thorough failure-mode analysis
might have brought up the possibility that in conditions where one pitot
tube iced up, the second one might too, and that inexperienced pilots might
then panic.

The recent crashes of the 737 Max planes show a similar inability to
consider the effects of a failure mode that is obvious to everyone in
hindsight.  Identifying all these failure modes in advance obviously takes
more expertise and foresight - but is that really too much to ask of the
relevant experts?


Date: Sun, 25 Aug 2019 17:57:03 -0700
From: Richard Stein <rmstein@ieee.org>
Subject: Inside America's Dysfunctional Trillion-Dollar Fighter-Jet Program
  (Valerie Insinna)

  [Excellent long article excerpted -- first para culled by PGN, the second
  by RS, in which `Winter' refers to Vice Admiral Mat Winter.  The
  subsequent analysis is Richard's.  (A snitch in *Times* sways Stein?)

Valerie Insinna, *The New York Times*, 21 Aug 2019

On the morning of June 23, 2014, an F-35 burst into flames just moments
before its pilot was set to take off on a routine training mission. He heard
a loud bang and felt the engine slow as warning indicators began flashing
`fire' and other alerts signaled that systems in the plane were shutting
down. Witnesses at Eglin Air Force Base near Pensacola, Fla., reported
seeing the pilot escape from the cockpit and run away from the fighter jet,
which was engulfed in thick plumes of black smoke. It was the first major
mishap involving a F-35 Joint Strike Fighter, and it couldn't have happened
at a worse time.  [...]

"Winter also made it a priority to push for drastic streamlining in the
process for testing new software in the F-35. Under the existing procedures,
the Pentagon can require test flights for more than 300 different factors or
functions when a new software load is installed.  Winter worked to cut that
down to a single validation flight, to test just the software and the
systems it affects, rather than retesting the performance of the whole
aircraft. A trial program staffed with a team of Air Force and Lockheed
coders proved that the method works and doesn't put pilots at risk, and
Winter's rapid software development strategy is now being implemented. But
moving to an agile software approach for the F-35 presents a huge challenge
for the sluggish and bureaucratic military acquisition system, and there's
no blueprint for how to integrate it alongside the traditional processes for
developing and testing hardware."

In http://catless.ncl.ac.uk/Risks/28/47#subj4, Henry Baker noted several
operational flight plan (OFP) readiness issues that could compromise F-35
system performance, mission and pilot safety.

Software stacks possess latent defects waiting discovery under appropriate
stimulus conditions. Truncated OFP qualification (regression test) limits
detection potential. The test assets may be exhausted in their capacity to
discover latent defects.

Payload exchange among the F-35 subsystems can often reveal anomalous
behavior, especially if the content is partially corrupt or inconsistent.
Subsystem test stimulus restriction is most cost effective, but at what
cost, to whom and when will the benefit be realized?

In earlier programs (~1970-1980 or so), The Air Force insisted on full,
end-to-end OFP qualification for any change. That the costs (schedule and
performance) have ballooned beyond estimates, and now preclude comprehensive
qualification coverage, is cause for concern and apparently represents a
significant operational risk.


Date: Sun, 25 Aug 2019 10:28:40 -0800
From: Rob Slade <rmslade@shaw.ca>
Subject: Sometimes simplicity is dangerous ...

We, in security, hate complexity.

Complexity is the enemy of security.

KISS, for us, isn't just an admirable principle, it's almost a way of life.
We want to keep things as simple as possible, since they are going to get
complex enough eventually anyway, and we *hate* that.

But sometimes life is just complex, and there's nothing we can do about it.

So, what has prompted this rumination on my part?

Well, suddenly everyone has become aware that the Amazon rainforest is
burning.  This isn't new, of course.  We should have been aware that the
rainforest was burning some time ago.  It's been burning for quite a while.
But, hey, so what?  There have been forest fires in other places, and we've
survived.  And most of us don't even know anyone who speaks Portuguese, so
what's the problem?

To understand that, you need to know about geology.

There are different types of soils in the world.  They have different
components, one of which is regolith.  Regolith is the breakdown product of
the underlying rock.  It contributes elements which, in turn, fix or release
nutrients that plants need to grow.  There are different soils, but they all
have regolith.

Except for tropical soil.

The soil in the Amazon rainforest has so little contribution from regolith
that it doesn't matter.  So how do things grow, without the nutrient boost?

To understand that, you need to understand biology and ecology.

Trees grow in the tropical rainforest.  Other plants grow on the trees.
Because they have no roots, they collect water in pouches and cups.  The
water, as well as watering the plant, collects and kills bugs to get
nutrients that those plants use to grow.  The insects eat fruit and leaves
up in the trees.  Other animals eat fruit and drop the husks and leaves down
to the ground.  The leaf litter gets cut up by ants who use it to farm mold.
Et cetera, et cetera until we get back to the trees.  All of the huge
complicated process has to go on to provide nutrients for the tropical soil,
without which none of it lives.

That's why ten percent of the *total* biodiversity on the planet is in the
Amazon alone.  They need it.

Stand in a hemlock forest, and all you have is the canopy above you.  Except
for the dead branches that poke you and grab your clothes, there is nothing
to impede you below that.  Tropical rainforests have five separate and
distinct layers, starting at the top canopy.

But what does this have to do with the fires?

Well, we (most of us) live in temperate rainforests.  We don't understand
the problem with forest fires.  Fires go on all the time.  Fires are
actually useful in some ways.  In the eastern forests, the First Nations
used to set fires to make the land more productive.  In the west, we know
that, even if we weren't throwing cigarette butts around with gay abandon,
the storms from the ocean (that bring the rain), also bring thunderstorms,
and therefore lightning, and therefore, even without us, forest fires are a
natural part of the forest growth, ecology, and procession.

That's not the case in tropical rainforests.

In temperate rainforests, after the fire goes through, all we have to do is
plant douglas fire, and, within a few years, the trees are taller than we
are and there are mice and salal and mule deer and blackberries and bears
are pooping in the woods fertilizing the douglas fir.

(And we have to hurry to plant the douglas fir, because, if we don't, five
minutes after the fire goes through alder starts growing.  We'll still have
a forest, just with a different economic value.)

That's not the case in tropical rainforests.

After a fire, you can't just plant some trees.  You've got this whole
complex system that means that the fact that some insect you can't even name
is missing means that *that* frog doesn't pollinate *that* bush which
doesn't feed *that* fish and the whole thing falls apart.  (Or, more likely,
doesn't start in the first place.)

In the tropical forest, after a fire, the grass (and crops, if you plant
them), grow spectacularly.  The first year.  The second year, the grass is
great.  The third year, it's pretty good.  After that, it's crap.  Because
the system isn't putting anything back into the soil.

In the temperature rainforest, the rains come from the ocean.  (Remember?)
Even if we burned down all the trees, the rains would still come.  Not in
the tropical rainforest.  Most of the rain comes from the forest itself.
The trees are lifting tons of water into the atmosphere every day.  It takes
energy.  And that's part of the reason that tropical rainforests have so
much rain, and are four or five degrees cooler than tropical savannah.

If we leave burned areas in the tropics alone, they might recover.  But,
whereas in the temperate rainforests it takes years, in the tropics it takes
an equivalent number of millennia.  The soil is dead, the land is in
drought, and isolated stands of forest will probably die, unless they are
miles in extent.

OK, now look at a map of the world.  Can you find the Amazon?  Remember that
not all of that bump is, in fact, the Amazon.  Not even all of Brazil is all

 And that part of that bump recycles 20% of all the oxygen in the
atmosphere.  And when we lose that oxygen recycling capacity, we lose that
carbon sequestration capacity, all that rain, and that biodiversity (and all
the undiscovered pharmaceuticals it contains).  And it won't grow back.

That's why a few fires in another country far away are important ...


Date: Mon, 26 Aug 2019 09:22:31 -0400
From: Monty Solomon <monty@roscom.com>
Subject: A Bitter Divorce Battle on Earth Led to Claims of a Crime in Space

NASA is examining a claim that an astronaut improperly accessed the bank
account of her estranged spouse from the Space Station.



Date: Mon, 26 Aug 2019 09:24:03 -0400
From: Monty Solomon <monty@roscom.com>
Subject: Premier's office accidentally publishes name of secret agent



Date: Mon, 26 Aug 2019 17:32:15 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: WeWork's Wi-Fi network is laughably easy to hack (Fast Company)



Date: Mon, 26 Aug 2019 17:41:17 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: Wake Up! Your House Is Calling (NYTimes)



Date: Mon, 26 Aug 2019 17:57:42 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: OpenAI releases larger GPT-2 dataset. Can it write fake news better
  than a human? (Boing Boing)



Date: Mon, 26 Aug 2019 18:11:27 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: SecurityWatch: Backstabbing, Disinformation, and Bad Journalism:
  The State of the VPN Industry (PCMag)



Date: Mon, 26 Aug 2019 19:11:38 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: Security Researchers Find Several Bugs in Nest Security Cameras



Date: Tue, 27 Aug 2019 10:59:40 -0400
From: Monty Solomon <monty@roscom.com>
Subject: Found: World-readable database used to secure buildings around the
  globe (Ars Technica)



Date: Wed, 28 Aug 2019 00:31:29 -0400
From: Gabe Goldberg <gabe@gabegold.com>
Subject: Credit card privacy matters: Apple Card vs. Chase Amazon Prime
  Rewards Visa (WashPost)

In a privacy experiment, he bought one banana with the new Apple Card -- and
another with the Amazon Prime Rewards Visa from Chase. Here's who tracked,
mined and shared our data.


Good luck following these details, let alone protecting yourself from being


Date: Wed, 28 Aug 2019 00:49:09 -0600
From: Jim Reisert AD1C <jjreisert@alum.mit.edu>
Subject: Regis University's technology systems targeted by malicious threat
  likely from outside the country (Denver Post)

Elizabeth Hernandez, *The Denver Post*, 23 Aug 2019

A forensic investigation at Denver's Regis University confirmed Friday that
the private college's technology systems were attacked by a malicious
threat, likely from outside the country.

University officials declined to say whether the situation at Regis was a
ransomware attack, saying the matter is still under investigation.
``Immediately upon discovering this issue, we quickly and intentionally took
our information technology systems offline in an effort to protect the
university and your information while we initiated an investigation and
notified law enforcement.  We are unfortunately only the latest entity to
face this kind of incident.''



Date: Wed, 28 Aug 2019 08:21:03 -0600
From: Jim Reisert AD1C <jjreisert@alum.mit.edu>
Subject: A Harvard freshman says he was denied entry to the U.S. over
  social media posts made by his friends (WashPost)


Deanna Paul and Susan Svrluga, 27 Aug 2019

  Ismail B. Ajjawi touched down at Boston Logan International Airport on
  Friday night, prepared to begin his freshman year at Harvard
  University. The 17-year-old Palestinian student never left the airport.

  The Harvard Crimson reported that U.S. officials detained Ajjawi for eight
  hours. After interrogating the minor and searching his phone and computer,
  they revoked his visa and sent him home to Lebanon.


  According to a statement by Ajjawi, an immigration officer claimed she
  ``found people posting political points of view that oppose the U.S.,'',
  though she discovered nothing Ajjawi had posted himself.


Date: Wed, 28 Aug 2019 12:20:56 -0400
From: Gabe Goldberg <gabe@gabegold.com>  DUP???
Subject: Ring, the doorbell-camera firm, has partnered with 400 police
  forces, extending surveillance reach (WashPost)

The doorbell-camera company Ring has quietly forged video-sharing
partnerships with more than 400 police forces across the United States,
granting them access to homeowners' camera footage and a powerful role in
what the company calls the nation's new neighborhood watch.

The partnerships let police automatically request the video recorded by
homeowners' cameras within a specific time and area, helping officers see
footage from the company's millions of Internet-connected cameras installed
nationwide, the company said. Officers don't receive ongoing or live-video
access, and homeowners can decline the requests, which Ring sends via email
thanking them for ``making your neighborhood a safer place.''

The number of police deals, which has not previously been reported, is
likely to fuel broader questions about privacy, surveillance and the
expanding reach of tech giants and local police. The rapid growth of the
program, which began in spring 2018, surprised some civil liberties
advocates, who thought that fewer than 300 agencies had signed on.



Date: Wed, 28 Aug 2019 10:39:09 -0700
From: Gene Wirchenko <gene@shaw.ca>
Subject: FBI seeks to monitor Facebook, oversee mass social media data
  collection (Charlie Osborne)

Charlie Osborne for Zero Day | 12 Aug 2019
Plans to track social media activity will potentially clash with existing
privacy policies.

The Federal Bureau of Investigation (FBI) is planning to aggressively
harvest information from Facebook and Twitter, a move which is likely to
cause a clash between the agency and social media platforms.

As reported by the Wall Street Journal, the FBI has recently sought
proposals from third-party vendors for technological solutions able to
harvest publicly-available information in bulk from Facebook, Twitter, and
other social media outlets.


Date: Wed, 28 Aug 2019 10:43:23 -0700
From: Gene Wirchenko <gene@shaw.ca>
Subject: Facebook's big win: Will this ruling have global impact on how
  your data is used? (Cathrin Schaer)

Cathrin Schaer for The German View, ZDNet, 27 Aug 2019
What was seen as one of the best ways to regulate social-media giants like
Facebook has just fallen apart in a Düsseldorf court.

opening text:

A decision by a regional court in Germany has derailed what many saw as the
world's best chance to regulate the behavior of data-gobbling social-media
giants like Facebook.


Date: Sun, 25 Aug 2019 17:23:14 +0300
From: Amos Shapir <amos083@gmail.com>
Subject: Re: Playing God: Japan temple puts faith in robot priest (RISKS-31.38)

I think there was a story by Isaac Asimov about an intelligent robot who
turned religious and became a Muslim.


Date: Sun, 25 Aug 2019 17:29:16 +0300
From: Amos Shapir <amos083@gmail.com>
Subject: Re: Phishing spam is getting better (RISKS-31.38)

This should be a golden rule for anyone reading email: Never click on any
link in an unsolicited incoming message, especially not one from your bank
(or any other service which may have access to your money).

If your bank needs you to click a link in their email message, it's *their*


Date: Mon, 14 Jan 2019 11:11:11 -0800
From: RISKS-request@csl.sri.com
Subject: Abridged info on RISKS (comp.risks)

 The ACM RISKS Forum is a MODERATED digest.  Its Usenet manifestation is
 comp.risks, the feed for which is donated by panix.com as of June 2011.
=> SUBSCRIPTIONS: The mailman Web interface can be used directly to
 subscribe and unsubscribe:

=> SUBMISSIONS: to risks@CSL.sri.com with meaningful SUBJECT: line that
   includes the string `notsp'.  Otherwise your message may not be read.
 *** This attention-string has never changed, but might if spammers use it.
=> SPAM challenge-responses will not be honored.  Instead, use an alternative
 address from which you never send mail where the address becomes public!
=> The complete INFO file (submissions, default disclaimers, archive sites,
 copyright policy, etc.) is online.
 *** Contributors are assumed to have read the full info file for guidelines!

=> OFFICIAL ARCHIVES:  http://www.risks.org takes you to Lindsay Marshall's
    searchable html archive at newcastle:
  http://catless.ncl.ac.uk/Risks/VL.IS --> VoLume, ISsue.
  Also,  ftp://ftp.sri.com/risks for the current volume
     or ftp://ftp.sri.com/VL/risks-VL.IS for previous VoLume
  If none of those work for you, the most recent issue is always at
     http://www.csl.sri.com/users/risko/risks.txt, and index at /risks-31.00
  Lindsay has also added to the Newcastle catless site a palmtop version
  of the most recent RISKS issue and a WAP version that works for many but
  not all telephones: http://catless.ncl.ac.uk/w/r
  ALTERNATIVE ARCHIVES: http://seclists.org/risks/ (only since mid-2001)
 *** NOTE: If a cited URL fails, we do not try to update them.  Try
  browsing on the keywords in the subject line or cited article leads.
  Apologies for what Office365 and SafeLinks may have done to URLs.
==> Special Offer to Join ACM for readers of the ACM RISKS Forum:


End of RISKS-FORUM Digest 31.39

home help back first fref pref prev next nref lref last post