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 30 Issue 13

Tuesday 7 February 2017


Russians Engineer A Brilliant Slot Machine Cheat - And Casinos Have No Fix
TLS vulnerability in popular iOS apps allows user data to be intercepted in man-in-the-middle attack
Malcolm Owen
Popular apps with 18 million combined downloads in the Apple App Store found vulnerable to silent data interception
Greg Barbosa
"Dozens of iOS apps fail to secure users' data, researcher says"
Michael Kan
Security flaws in Pentagon systems "easily" exploited by hackers
Zack Whittaker
Data from man's pacemaker led to arson charges
Lauren Pack
Vizio to Pay $2.2M to Settle Charges it Illegally Collected Data from TV Owners
Gabe Goldberg
The Truth about Unix—my version, anyway—for comic relief
Don Norman
"Vulnerability in Microsoft SMBv3 protocol crashes Windows PCs"
Woody Leonhard
"GitLab database goes out after spam attack"
Paul Krill
Cisco: Spam is making a big-time comeback
Tom Greene
How WhatsApp is fighting spam after its encryption rollout
Trump's Vote Fraud Guru is Registered in Three States
Re: Hackers Use New Tactic at Austrian Hotel: Locking the Doors
Amos Shapir
Re: Network-enabled ICBMs for the USAF?
Amos Shapir
Re: site
Re: Nim Language Draws From Best of Python, Rust, Go, and Lisp
Dimitri Maziuk
Arthur Flatau
Alexander Klimov
Re: Quantum Computers Versus Hackers, Round One
Rob Slade
Werner U
Info on RISKS (comp.risks)

Russians Engineer A Brilliant Slot Machine Cheat - And Casinos Have No Fix

"Peter G. Neumann" <>
Mon, 6 Feb 2017 22:53:51 PST

  "At the same time, most casinos can't afford to invest in the newest slot
  machines, whose PRNGs use encryption to protect mathematical secrets".
  These folks did what others have done in the past—bought one of the
  machines and reverse engineered it.  In this case the PRNG.

Also see
Two-armed Bandits, 7 Feb 2017 (thanks to Tom Lambert)
This item also has an interesting sequence of following comments.

  ...the operatives use their phones to record about two dozen spins on a
  [slot machine] they aim to cheat. They upload that footage to a technical
  staff in St. Petersburg, who analyze the video and calculate the machine's
  pattern based on what they know about the model's pseudorandom number
  generator. Finally, the St. Petersburg team transmits a list of timing
  markers to a custom app on the operative's phone; those markers cause the
  handset to vibrate roughly 0.25 seconds before the operative should press
  the spin button.

No surprise to RISKS readers.  The first interesting case of this kind that
I reported was the Harrah's Tahoe $1.7 Million payoff internal fraud.  This
was a progressive payoff on 16 adjacent machines.  The casino seems to have
hidden this story, but it appears to have been an inserted Trojan horse
chip.  See the ACM SIGSOFT Software Engineering Notes, 8, 5, Oct 1983, pages
7-8.  There are several other similar stories in our RISKS archives.  PGN

TLS vulnerability in popular iOS apps allows user data to be intercepted in man-in-the-middle attack (Malcolm Owen)

"Peter G. Neumann" <>
Tue, 7 Feb 2017 14:21:48 PST
  Malcolm Owen, Apple Insider, 07 Feb 2017

  A number of popular apps are vulnerable to a 'man-in-the-middle' attack
  due to poorly implemented TLS protection, an examination of apps in the
  iOS App Store has revealed, with a security researcher claiming it is
  possible to read data sent back to the app developer's servers for 76
  apps. [...]

This should be no surprise to those of you who read the best paper at the

  Benjamin Beurdouche, Karthikeyan Bhargavan, Antoine Delignat-Lavaud,
  Cedric Fournet, Markulf Kohlweiss, Alfredo Pironti, Pierre-Yves Strub, and
  Jean Karim Zinzindohoue, A Messy State of the Union: Taming the Composite
  State Machines of TLS, Proceedings of the 36th IEEE Symposium on Security
  and Privacy, San Jose, CA, May 18-20, 2015.

As I probably noted once before, this paper analyzes the composition of
client-side and server-side TLS implementations, and finds flaws
(`unexpected behaviors') including the FREAK vulnerability, in popular TLS
implementations, and in OpenSSL and JSSE.  It is a remarkable paper, and
well worth reading.  PGN

Popular apps with 18 million combined downloads in the Apple App Store found vulnerable to silent data interception (Greg Barbosa)

geoff goodfellow <>
Tue, 7 Feb 2017 10:36:39 -1000
Greg Barbosa., 6 Feb 2017

After scanning through the binary codes of applications in the iOS App
Store, Will Strafach's verify.lyservice has detected that 76 popular apps in
the store are currently vulnerable to data interception. The interception is
possible regardless if App Store developers are using App Transport Security
or not.  A few months ago, similar vulnerabilities were discovered with
Experian and myFICO Mobile's iOS apps.

"Dozens of iOS apps fail to secure users' data, researcher says" (Michael Kan)

Gene Wirchenko <>
Tue, 07 Feb 2017 09:06:01 -0800
The developers have misconfigured the apps to accept invalid TLS
certificates, says the security researcher who detected the app vulnerabilities
Michael Kan, InfoWorld, 7 Feb, 2017

selected text:

Dozens of iOS apps that are supposed to be encrypting their users' data
don't do it properly, according to a security researcher.

The developers of the apps have accidentally misconfigured the
networking-related code so it will accept an invalid Transport Layer
Security (TLS) certificate, ...

In all, the 76 apps have 18 million downloads, ...

Security flaws in Pentagon systems "easily" exploited by hackers (Zack Whittaker)

geoff goodfellow <>
Thu, 2 Feb 2017 09:25:18 -1000
Zack Whittaker for Zero Day, 1 Feb 2017
Hackers are likely exploiting the easy-to-find vulnerabilities, according
to the security researcher who warned the Pentagon of the flaws months ago.

Several misconfigured servers run by the US Dept. of Defense could allow
hackers easy access to internal government systems, a security researcher
has warned.

The vulnerable systems could allow hackers or foreign actors to launch
cyberattacks through the department's systems to make it look as though it
originated from US networks.

Dan Tentler, founder of cybersecurity firm Phobos Group, who discovered the
vulnerable hosts, warned that the flaws are so easy to find that he
believes he was probably not the first person to find them.

"It's very likely that these servers are being exploited in the wild," he
told me on the phone.

While the Pentagon is said to be aware of the vulnerable servers, it has
yet to implement any fixes—more than eight months after the department
was alerted.

It's a unique case that casts doubts on the effectiveness of the Trump
administration's anticipated executive order on cybersecurity, which aims
to review all federal systems of security issues and vulnerabilities over a
60-day period.

The draft order was leaked last week, but it was abruptly pulled minutes
before it was expected to be signed on Tuesday.

Tentler, a critic of the plans, argued that the draft plans are "just not

"It's laughable that an order like this was drafted in the first place
because it demonstrates a complete lack of understanding what the existing
problems are," he said.

"The order will effectively demand a vulnerability assessment on the entire
government, and they want it in 60 days? Just that one vulnerability
finding from me... it's been months—and they still haven't fixed it," he

In the past year, the Pentagon became the first government department to
ease up on computer hacking laws by allowing researchers to find and report
bugs and flaws in systems in exchange for financial rewards.

But security researchers like Tentler are still limited in how much they
can poke around the military's public-facing systems.

The department's official bug bounty governs the scope of what networks
researchers can access. Researchers must limit their testing to two domains
-- "" and its subdomains, and any ".mil" subdomain.

In an effort to pare down the list of hosts from "all public Department of
Defense hosts" to "only the ones in scope," Tentler was able to identify
several hosts which answered to the domain names in scope.

"There were hosts that were discovered that had serious technical
misconfiguration problems that could be easily abused by an attacker inside
or outside of the country, who could want to implicate the US as culprits
in hacking attacks if they so desire," he told me.

"The flaw could allow politically motivated attacks that could implicate
the US," he added.  [...]

Data from man's pacemaker led to arson charges (Lauren Pack)

Jim Reisert AD1C <>
Tue, 7 Feb 2017 19:51:55 -0700
Lauren Pack, Staff Writer, *Journal News*, 27 Jan 2017

  Investigators used the data from a Middletown man's pacemaker to help get
  an indictment in a fire that caused about $400,000 in damages.  Ross
  Compton, 59, has been indicted on felony charges of aggravated arson and
  insurance fraud for allegedly starting the fire on Sept. 19 at his Court
  Donegal house.

  Police said Compton gave statements that were inconsistent with the
  evidence at the fire.

  Compton, who has extensive medical problems, including an artificial heart
  implant that uses an eternal pump, told police that when he saw the fire,
  he packed some belongings in a suitcase and bags. He told police that he
  then broke out the glass of his bedroom window with a cane and threw the
  bags and suitcase outside before taking them to his car.

  So police got a search warrant for all electronic data stored in Compton's
  cardiac pacing device, according to court records obtained by this news

  The data taken from Compton's pacemaker included his heart rate, pacer
  demand and cardiac rhythms prior to, during and after the fire.

  A cardiologist who reviewed that data determined “it is highly improbable
  Mr. Compton would have been able to collect, pack and remove the number of
  items from the house, exit his bedroom window and carry numerous large and
  heavy items to the front of his residence during the short period of time
  he has indicated due to his medical conditions,''

  The fire caused about $400,000 in damages to the structure and contents of
  the 2,000-square-foot home on Court Donegal, according to Deputy Fire
  Chief Jeff Spaulding.

Vizio to Pay $2.2M to Settle Charges it Illegally Collected Data from TV Owners

Gabe Goldberg <>
Tue, 7 Feb 2017 13:52:21 -0500
  [Wow, $2.2 million penalty for spying on 11 million TVs. That's 20 cents
  per offense.  That'll sure teach them and the industry not to trifle with
  peoples' information.]  [And it's fives cents less than your two-bits
  worth of information.]

Washington, DC—*Vizio*, a California-based manufacturer of
Internet-connected "smart" televisions, has agreed to pay $2.2 million to
settle charges by the *Federal Trade Commission* (FTC) and the New Jersey
attorney general that it installed software on its TVs to collect viewing
data on 11 million TVs without consumers' knowledge or consent.  The payment
includes $1.5 million to the FTC and $1 million to the *New Jersey Division
of Consumer Affairs*, with $300,000 of that amount suspended. The federal
court order also requires Vizio to prominently disclose and obtain express
consent for its data collection and sharing practices, and requires the
company to delete data collected before March 1, 2016. According to the
complaint, since Feb. 2014, Vizio and an affiliated company have
manufactured smart TVs that capture second-by-second information about video
displayed, including from cable, broadband, set-top box, DVD, over-the-air
broadcasts and streaming devices. In addition, the agencies allege that the
company added specific demographic information to the viewing data, such as
sex, age, income, marital status, household size, education level, home
ownership and household value, then sold this information to third parties.

The Truth about Unix—my version, anyway—for comic relief

Don Norman <>
Fri, 3 Feb 2017 22:45:47 -0800
> I'm reminded of a famous article Don Norman wrote in 1981 about how awful
> the UNIX shell language (which at that time was the user interface)
> was. One of the UNIX guys pointed out that commands he complained weren't
> "natural" were because they weren't like the PDP-10 he was used to.

That statement is cute, clever, and false (or maybe it is an alternative

Don Norman (that is, I) never said Unix Shell (and APIs) were bad because
they were unfamiliar.  I thought the underlying philosophy brilliant
allowing people to string together lots of modules gto do powerful actions.
I was a wizard at writing shell scripts.

I said Unix was bad (actually, I said horrible) because of its lack of
consistency.  Argument handling and specifications were different from
routine to routine.  You had to keep looking up the syntax for all except
the most frequently used calls in the manual, which kept getting larger and
larger and larger. There were a zillion other examples of really
unthoughtful design that not only had inconsistencies.

There was also a distinct lack of interest in error checking, so that simple
slips of the finger could lead to erasure of data, files, and entire
directories (to remove all files starting with the word temp, say temp1,
temp2,and temp3, simply type rm temp*. An accidental space character would
transform that into rm temp * which would remove ALL files in that
directory). Early text editors lost everything that had been done if the
person quit the file without first saving it. ("Even experienced users have
been known to lose their work" said the manual. Isues like these gave rise
to the joke that "if it is documented, it isn't a bug, it is a feature.")

There are lots of reasons for this, but I don't want to repeat my paper
here. Let me simply say that after the initial shock of my paper wore off,
the Unix creators came to agree with me and even became friends. Today's
various flavors of Unix have overcome most of the problems. Alas, MacOS (and
Windows) which runs on a Unix kernel, hides great power of Unix except to
techies who can find the terminal program and pipe and redirect to their
hearts content.

The design rules that I started to develop in that paper in the 1980s are
still true and important today. Consistency, feedback, good mappings, and a
good conceptual model are the hallmark of good, useful design. Unix did have
a powerful underlying conceptual model: it failed at the others. But hey,
that was over 35 years ago!

More facts: I never used a DEC (Digital) PDP-10, although I did use (and
own) many every other DEC machine: PDP 1, 4, 7, 8, 9, 11 and Vax. I managed
to skip the 10, which was replaced by the Vax. More importantly, it never
occurred to me to contrast the Assembly language of these machines with
Unix's shell script language. I fell in love with shell scripts: I never
fell in love with assembly languages.

Don Norman, Prof. and Director, DesignLab, UC San Diego  <>

"Vulnerability in Microsoft SMBv3 protocol crashes Windows PCs" (Woody Leonhard)

Gene Wirchenko <>
Mon, 06 Feb 2017 10:21:37 -0800
Woody Leonhard, InfoWorld, 3 Feb 2017
Computers running fully patched Windows 10, 8.1, Server 2012, and 2016 are
hit by Blue Screens when trying to connect to an infected server

opening text:

Security experts warn that it may be possible to exploit a vulnerability in
a protocol widely used to connect Windows clients and servers to inject and
execute malicious code on Windows computers.

Computers running fully patched Windows 10, 8.1, Server 2012, or 2016 that
try to access an infected server will crash with a Blue Screen triggered in
mrxsmb20.sys, according to a post by Günter Born on today's Born's Tech
and Windows World blog.

"GitLab database goes out after spam attack" (Paul Krill)

Gene Wirchenko <>
Thu, 02 Feb 2017 09:26:40 -0800
Paul Krill, InfoWorld, 1 Feb 2017
Approximately six hours of data, including issues, merge requests,
users, comments, and snippets, will be lost as GitLab restores from a backup

opening text:

Code-hosting site GitLab has suffered an outage after sustaining a "serious"
incident on Tuesday with one of its databases that has required emergency

The company today said it lost six hours of database data, including issues,
merge requests, users, comments, and snippets, for and was in the
process restoring data from a backup. Data was accidentally deleted,
according to a Twitter message.

Cisco: Spam is making a big-time comeback (Tim Greene)

geoff goodfellow <>
Thu, 2 Feb 2017 11:38:09 -1000
*Adware is also on the rise, Cisco's Annual Cybersecurity Report says*
*By Tim Greene Senior Editor, Network World*

Spam is making a surprising resurgence as a threat to corporate security
and becoming a more significant carrier of attacks as varied as spear
phishing, ransomware and bots, according to Cisco's 2017 Annual
Cybersecurity Report.

The company's 10th such report says spam is way up. It accounts for 65% of
all corporate email among customers who opted in to let the company gather
data via telemetry in Cisco gear...

... Adware and other threats

Another growing problem is adware, whose primary purpose is to display ads
on Web pages or pop-ups to the benefit of advertisers. In the hands of
malicious actors, though, they can carry malicious payloads that change
settings in browsers and operating systems, undermine security products and
even gain full control of the host. So rather than being an annoyance,
adware is a threat.  “Which means the focus is going to have to come onto
adware from the corporate side to defend whereas historically it was more of
a nuisance,'' Antes says.

The report looked at adware in 130 organizations distributed across vertical
industries for a year and found that 75% had adware infections...


How WhatsApp is fighting spam after its encryption rollout (Techcrunch)

Lauren Weinstein <>
Thu, 2 Feb 2017 17:02:08 -0800

  Rolling out end-to-end encryption raised not just political concerns, but
  practical ones. If WhatsApp couldn't read the contents of its users'
  messages anymore, how would it detect and fight spam on the platform?
  WhatsApp could have become a haven for scammers pushing pills and
  get-rich-quick schemes, which would have driven users off the platform and
  harmed its business even more than short-term court-ordered shutdowns.

Trump's Vote Fraud Guru is Registered in Three States (AP)

"Peter G. Neumann" <>
Thu, 2 Feb 2017 9:01:49 PST
Gregg Phillips, whose unsubstantiated claim that the election was marred by
3-million illegal votes was tweeted by the president, was listed on the
rolls in Alabama, Texas and Mississippi, according to voting records and
election officials in those states. He voted only in Alabama in November,
records show.  [AP 30 Jan2017]

Re: Hackers Use New Tactic at Austrian Hotel: Locking the Doors (RISKS-30.12)

Amos Shapir <>
Thu, 2 Feb 2017 12:19:09 +0200
I find it hard to believe there is no mechanical workaround for such locks.
What happens when the power is off?  Since hotel staff did not cut power to
the affected rooms, does this means that in case of power cut the locks
would stay in locked position?  (Or maybe they just never thought of that).

Re: Network-enabled ICBMs for the USAF? (RISKS-30.12)

Amos Shapir <>
Thu, 2 Feb 2017 12:11:50 +0200
The initial goal of DARPA when initiating the Ethernet was to establish a
robust and secure network, so that communication to ICBMs could be
maintained even in the event of a nuclear attack.

Considering that this means that almost nothing but ICBMs was intended to be
connected to the net, it seems we have made a full about-face...

Re: site (LW, Risks-30.11)

"Arthur T." <>
Tue, 07 Feb 2017 00:39:05 -0500
I don't have the tools or the expertise to answer a question I've had for a
long time. Perhaps you do. What percentage of browser exploits require the
user to have active scripting enabled? Pretty much every serious threat I've
read about requires it. If the percentage is as high as I think it is, I
don't think my caution could reasonably be considered paranoia. It seems to
me that running without active scripting can protect you against most
zero-day exploits.

I believe that the trade-off between privacy and functionality is something
that we should each be weighing and deciding for ourselves. I've made my
choice. I'm willing to believe that I'm mistaken about the choice of the
average RISKS reader.

Re: site (AT, RISKS-30.13)

Lauren Weinstein <>
Tue, 7 Feb 2017 07:34:36 -0800
In reality, only a vanishingly small percentage of exploits depend on
scripting these days. The reason why is simple—the scripting systems have
been massively hardened by the browser makers, and most modern Web sites
depend on JavaScript and other scripting methodologies to provide primary
functionality now without difficulty.  The only people mainly at risk now
for scripting attacks are people who voluntarily run ancient, non-updated
browsers. Most successful attacks today are based on credential thefts,
which themselves are based on sophisticated phishing techniques that don't
depend on scripting attacks at all. From a practical standpoint, the age of
scripting exploits is past history now.

Re: Nim Language Draws From Best of Python, Rust, Go, and Lisp (Levine, RISKS-30.11)

Dimitri Maziuk <>
Wed, 1 Feb 2017 20:57:38 -0600
I have to respectfully disagree: it is just as hard to match nested if/elses
on the indentation level as it is to match the braces. It may be somewhat
easier if it all fits on one screen—as python scripts were originally
intended to—but any half decent brace-based-language editor will
highlight matching the brace, so not really even then.

What you can't safely do is pretty-print python code. With braces there's
any number of pretty-printers that will reformat your code so that indent
matches the braces. With python I can't even manually reformat anything
longer than a two screenfuls without messing up the un-indent somewhere down
below. Or at least without being very very careful not to.

And of course inserting a closing brace for scope that was never opened will
generate a compiler error just like a wrong un-indent; what you can't do is
un-indent past the left edge of the page, so there's that.

Re: Nim Language Draws From Best of Python, Rust, Go, and Lisp (Levine, RISKS-30.12)

Arthur Flatau <>
Thu, 2 Feb 2017 13:44:42 -0600
I like Python, but the fact that it uses indentation for grouping is one of
the things I dislike, for two reasons.  First it is easy to re-indent C or
Perl code with braces, for instance with emacs (my text editor of choice).
If there are spacing typos, they can easily be removed.  This does not work
with Python.  If I refactor some code in Python, so it needs to re-indented
this is a largely manual process.  Not so with C or Perl.  Second, I find it
hard to visually line up different groups in a "large" piece of code.  It
has been my coding guideline for many years to have functions fit on a
single screen (I saw a talk by Bjarne Stroustrup a little while ago where he
mentioned this as well).  Even following that guideline I sometimes find it
hard to figure out the groupings with only indentation.  This may well be,
in part, due to my poor eyesight, but I still prefer the braces

Re: Nim Language Draws From Best of Python, Rust, Go, and Lisp (RISKS-30.12)

Alexander Klimov <>
Tue, 7 Feb 2017 19:12:30 +0200
> While it took a little while to get used to it, now I find the
> python way works at least as well. Compilers remember the open
> levels of indentation so they can diagnose spacing typos where you
> return to an indentation level that was never opened, something C
> and perl can't do since all braces look the same.
> It also avoids a whole category of hard to find bugs in C programs
> where the indentation suggests one thing but the braces say
> something else.

To find the discrepancy, it is required to have redundancy. In languages
with braces, there are two ways to encode the block structure, so GCC can
notice the problem:

Wmisleading-indentation warns about places where the indentation of
the code gives a misleading idea of the block structure of the code
to a human reader. For example, given CVE-2014-1266:

    sslKeyExchange.c: In function 'SSLVerifySignedServerKeyExchange':
    sslKeyExchange.c:629:3: warning: this 'if' clause does not
     guard... [-Wmisleading-indentation]
        if ((err = SSLHashSHA1.update(&hashCtx, &signedParams)) != 0)
    sslKeyExchange.c:631:5: note: ...this statement, but the latter
     is misleadingly indented as if it is guarded by the 'if'
            goto fail;

Re: Quantum Computers Versus Hackers, Round One (Werner, RISKS-30.12)

Rob Slade <>
Thu, 2 Feb 2017 02:05:48 -0800
I'm a bit ambivalent myself.

I've worked on this topic for about a dozen years now.  I wish someone would
take it seriously, but this article doesn't give any indication that anyone
is.  (Of course, that may simply be a reporter covering something he doesn't
understand ...)

I've posted a little bit on it recently:

Re: Quantum Cryptography (Feb 2 in WiReD)

Werner U <>
Mon, 6 Feb 2017 22:17:37 +0100
[ reads like a prequel for "Quantum Cryptography *for Dummies*" ]
[ worth reading, for those tracking the media coverage. ]
[ needs some pruning, I'm afraid.]

*Physicists, Lasers, and an Airplane: Taking Aim at Quantum Cryptography*


( #cryptography #Hacking #quantum computing #quantum mechanics )

...pivoted their telescope to catch the photons, one by one. On their best
run, they caught over 800,000 photons in just a few minutes, but it wasn't
easy. “Out of every 10,000 photons they sent, we'd get one,'' says Pugh, who
studies at the University of Waterloo. “One to a hundred of them.''

The point of this high-altitude game was to test a technology known as
quantum cryptography. For decades, experts have claimed that if executed
properly, quantum cryptography will be more secure than any encryption
technique used today. They also say it will be one of the lines of defense
when quantum computers crack every existing algorithm. But it's hard to
pull off; quantum cryptography requires precise control of individual
photons over a long distance. Pugh's group was the first
<> to successfully test the technology from
ground to airplane.

It works like this: The sender transmits carefully prepared photons, over
optical fiber or through the air, to a recipient. The recipient reads the
photons like Morse code, with physical signals corresponding to a letter or
a number. Instead of listening for long and short beeps, Pugh and his
colleagues measured how the photons are oriented—what physicists call
polarization. In their setup, photons could be polarized in four directions,
and the team translated that polarization into 1's and 0's: a binary message
known as a cryptographic key. Using that key, a sender can encrypt their
information, and only a recipient with the key can unscramble the message.

Quantum cryptography is so powerful because it's physically *impossible* for
a hacker to steal a key encoded using quantum particles. In the quantum
world, when you measure or observe a particle, you change it. It's like
Schrodinger's cat, which is both dead and alive when you're not looking, but
immediately becomes one or the other when you look. If you try to measure a
quantum key, you immediately change it—and by design, the sender will
know and throw the key out.  “It's secure by the laws of nature,'' says
physicist Thomas Jennewein, who led the work at the University of Waterloo.

Commercial quantum cryptography products have been around for over 15 years,
but they have limited range.  “You can guarantee security between the White
House and the Pentagon, or from the corner of one military base to
another,'' says Caleb Christensen, the chief scientist at MagiQ
Technologies, a Boston-area company that makes commercial quantum
cryptography systems.  “In the telecom business, that's way too short.''
So far people have been able to send quantum keys just 250 miles.

This tech will be important when computers become too powerful for current
encryption algorithms. It takes today's computers far longer than the age of
the universe to decode an encrypted message, but it'll be a cinch for
quantum computers.  “It might take hours or days as opposed to age of the
universe,'' says Pugh.

Still, quantum cryptography won't be tech's security savior. Most hacks
today are due to simple human error.  “Most times when a corporation gets
hacked, it's not necessarily because someone went in and spliced into their
telephone line,'' says Christensen.  “If you lose all your secrets because
someone phishes the e-mail of your middle management, you're not going to
spend millions of dollars installing a quantum cryptography backbone.''

For those with higher security standards, the eventual goal is to deliver
quantum keys to a satellite, which could make it possible to send
quantum-secured messages across the globe. Last August, the Chinese Academy
of Sciences, collaborating with Austrian physicists, launched a satellite
called Quantum Experiments at Space Scale, although they haven't
successfully sent it a key.

Jennewein's team has been rehearsing for a satellite mission for over three
years. In 2013, they started by sending quantum keys to a moving truck. Now
that they've shown they can transmit enough quantum signal through a mile of
Earth's atmosphere, Jennewein wants to beam a key 300 miles into the air, to
a satellite in low-Earth orbit. With proper funding, Jennewein thinks his
team could do it in two or three years. He's optimistic: “The airplane
experiment is, in some respects, harder than an actual satellite,'' he says.
“A satellite has much smoother and more predictable motion than an
aircraft.''  Just ask Pugh.

Please report problems with the web pages to the maintainer