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 24 Issue 77

Friday 3 August 2007


Structural problems with the I-35W bridge span
Driver follows GPS when he should not
Erwan David
"Meteorology Police -- you're BUSTED!"
Annie Johnson via Paul Saffo
Hacked passport crashes RFID readers
Jeff Jonas
IRS computer security/privacy problems
User-hostile behavior
Steve Summit
Location-Based Dictionary Attacks
Diomidis Spinellis
Amazon chasing 2-cent Web services bill
Martin Redington
Windows Live Messenger blocking even more completely innocuous text
Cody Boisclair
Re: Accuracy of Hawkeye at Wimbledon
Paul Wallich
Fraudproof voting protocols from scientists
Warren Smith
REVIEW: "Implementing ITIL", Randy A. Steinberg
Rob Slade
Info on RISKS (comp.risks)

Structural problems with the I-35W bridge span

<"Peter G. Neumann" <>>
Thu, 2 Aug 2007 10:18:35 PDT

  Two reports published since 2001 pointed to structural problems with the
  Interstate 35W. ...  The bridge's deck truss system has not experienced
  fatigue cracking, but it has many poor fatigue details on the main truss
  and the floor truss system. ...  In another report two years ago, the
  U.S. Department of Transportation's National Bridge Inventory database
  concluded the bridge was "structurally deficient."
    [noted by Mike Hogsett]

  Bridges are generally built with a high level of redundancy, so that if
  one part fails the load is distributed through the structure.  The I-35W
  bridge did not have a high level of redundancy, and the failure of a
  single significant component could have led to the collapse of the entire
  structure.  [Annotation in A Deadly Collapse, a half-page set of graphics,
  *The New York Times*, 3 Aug 2007, National Edition A14]

The propagating bridge structure collapse on 1 Aug 2007 in Minneapolis
exposes just one more tip of an iceberg among a large collection of
icebergs.  Many of our infrastructures such as roads (some with sink-holes
lying in wait), bridges, railroad track beds, pipelines, storage tanks
(including fuel and nuclear waste), and so on are in serious need of repair,
decommissioning, or replacement.  For example, some of road infrastructures
have endured loads far in excess of what was expected in their original
designs and operating environments, and have been steadily declining.  This
is just another example of the old adage, "an ounce of prevention is worth a
pound of cure".  In this case, the scales are unbalanced by deaths that
cannot be cured and collateral losses.

There is of course a lesson here for information system infrastructures,
Removing information security vulnerabilities seems to be a nonstarter in
the eyes of government and system developers that might otherwise stimulate
remediation.  The short-term costs of preventive maintenance always seem to
blind folks to the long-term costs of inaction.  This situation reminds me
once again of the importance of farsighted design and continual oversight.
See my two-page note on holistic systems in the November 2006 issue of the
ACM SIGSOFT Software Engineering Notes, in case you have not looked at it

I seem to have gone all these years of moderating RISKS without citing one
of my favorite multipurpose mixed metaphors: Pandora's cat is out of the
barn and the genie won't go back in the closet.  It certainly seems
applicable here.

Driver follows GPS when he should not

<Erwan David <>>
Wed, 1 Aug 2007 14:21:43 +0200

On 22 Jul 2007, a Polish bus had a grave accident in Vizille (France). The
bus used a road with a 14% (1/7) descending slope, it seems its brakes went
too hot and could not stop the bus at the end of the slope.

The inquiry made appear that the driver blindly followed the indications of
is GPS receiver, ignoring the 11 signs forbidding him to use this route.

Risk: always relying on technology, even if used a little bit out of spec.

"Meteorology Police -- you're BUSTED!"

<Paul Saffo <>>
Tue, 31 Jul 2007 16:40:45 -0700

A suspicious looking box found near Lewis-Gale Medical Center [on 19 Jul
2007] was, in fact, a remote weather station that had been affixed to a tree
by an employee, and not an explosive device.  Constructed with putty and
wires, it was probed by the Virginia State Police Bomb Squad -- which blew
it up before realizing it was a weather station.  An employee had placed a
putty-like substance around the box to make it weather proof.  ...
  [Source: Annie Johnson, *The Roanoake Times*, 20 Jul 2007; PGN-ed]

Hacked passport crashes RFID readers

<Jeff Jonas <>>
Wed, 1 Aug 2007 03:36:39 -0400 (EDT)

The report that voting machines are not trustworthy is joined by this: RFID
passports and readers are vulnerable:

Hacked passport crashes readers

A hacker has demonstrated an exploit against the RFID tags in the new US
passports that allows him to clone a passport and modify the RFID with bad
code that will crash the passport readers.  Lukas Grunwald, an RFID expert
who has served as an e-passport consultant to the German parliament, says
the security flaws allow someone to seize and clone the fingerprint image
stored on the biometric e-passport, and to create a specially coded chip
that attacks e-passport readers that attempt to scan it.

Grunwald says he's succeeded in sabotaging two passport readers made by
different vendors by cloning a passport chip, then modifying the JPEG2000
image file containing the passport photo.  Reading the modified image
crashed the readers, which suggests they could be vulnerable to a
code-injection exploit that might, for example, reprogram a reader to
approve expired or forged passports.

"If you're able to crash something you are most likely able to exploit it,"
says Grunwald, who's scheduled to discuss the vulnerabilities this weekend
at the annual DefCon hacker conference in Las Vegas.

IRS computer security/privacy problems

<"Peter G. Neumann" <>>
Fri, 3 Aug 2007 10:47:55 PDT

The Treasury Inspector General for Tax Administration reports that in a
recent test of 102 people with direct access to internal IRS data (employees
and contractors), 62 of them complied with a request from a caller posing as
a technical support person to provide their user name and temporarily change
their password.  Only eight called the IG's office or IRS security personnel
to verify the identity of the caller.  Similar tests in 2001 and 2004 were
intended to improve security practices, but apparently were not effective.
[Source: AP item in *The New York Times*, 3 Aug 2007; PGN-ed]

User-hostile behavior

<Steve Summit <>>
Tue, 31 Jul 2007 23:31:11 -0400

As an infrequent user of Microsoft Windows, I'm often belatedly surprised by
its various foibles -- new ones seemingly every time I use it -- that
everybody else saw long ago or is used to by now.  Today was no exception.

Out of the blue, a dialog box popped up, saying (from memory):

  Upgrade of your system is almost complete.  A restart is required to
  complete this upgrade.  Windows will automatically restart your computer
  in 3:47 minutes.

  Restart now                                 Restart later

The time 3:47 was continually counting down, once per second, and there
was a progress bar showing about 25% complete.

There were several interesting things about this message.  I had initiated
no upgrade, and Windows had not even asked me (as it so often does) if I
wanted it to start an upgrade, nor even notified me that one was available.
And there was no indication whatsoever (nor any obvious way of investigating
to try to find out) what this imposed upgrade actually was.

Most significantly, if I had done nothing, my computer evidently would have
rebooted, all by itself, in less than five minutes.  But of course I had
several windows open, containing all sorts of context relevant to the
problem I was working on and which I most certainly did not want to log off
and lose just then.  And I was turning back and forth between the Windows
computer and another one; I could easily have turned away for more than five
minutes, and missed this charming little dialog box entirely.

Naturally I clicked "Restart later", and the dialog box went away.
But about five minutes later it reappeared, exactly as before.
I clicked "Restart later" again.  About five minutes later it reappeared.
I clicked "Restart later" again.  This went on for the next hour or two.

In what universe is this acceptable behavior?  I've got work to do; I don't
have time for unprovoked restarts; I'd really rather not have to keep a
weather eye on a machine so as to be able to repeatedly click "Restart
later" just to keep the damn thing up and my work intact.  I can't help but
wonder what might happen with a machine being used for vital real-time work,
or as an unattended server.

I do know the answer to the "In what universe?" question, of course: in
Microsoft's universe.  And I suspect that the update they were so insistent
on applying was one for their benefit, not mine.

I further suspect I know (although it wouldn't say so) what the
whether-I-liked-it-or-not upgrade specifically was.  A couple of weeks ago
the same machine had been asking me if I wanted to (voluntarily) install and
activate another update, namely a more-fully-functional version of its
"Windows Genuine Advantage" component.  But I had declined that upgrade,
because I know that the machine's software is genuine, and I don't want the
machine "phoning home" all the time or complaining if it can't, and I
certainly don't want it locking me out some day if it ever makes a mistake.

I hadn't noticed that it had stopped asking about that earlier upgrade.
Perhaps I ought to have been suspicious.  Like the back-alley con man who is
perfectly happy to rob you if you decline the proffered game of three-card
monte, I suspect Windows simply decided to fall back on "Plan B" after I
declined the "voluntary" upgrade too many times.  That machine is probably
now assimilated, and I can feel secure that it is WGA-safe from non-genuine
software.  Yippee.

Location-Based Dictionary Attacks

<Diomidis Spinellis <>>
Thu, 02 Aug 2007 10:08:29 +0300

I get daily security reports from the hosts I manage. Typically these
contain invalid user attempts for users like guest, www, and root.
(Although FreeBSD doesn't allow remote logins for root, I was surprised to
find out that many Linux distributions allow them.)

Today's log surprised me, because it contained only Greek names. Here is an
excerpt from the log.

Aug  1 00:19:42 istlab sshd[22137]: Invalid user achaikos from
Aug  1 00:19:45 istlab sshd[22191]: Invalid user achilleus from
Aug  1 00:19:48 istlab sshd[22218]: Invalid user actaeon from
Aug  1 00:19:51 istlab sshd[22244]: Invalid user acteon from
Aug  1 00:19:55 istlab sshd[22279]: Invalid user adelpha from
Aug  1 00:19:58 istlab sshd[22302]: Invalid user adelphe from
Aug  1 00:20:01 istlab sshd[22321]: Invalid user adelphie from
Aug  1 00:20:04 istlab sshd[22353]: Invalid user adonia from
Aug  1 00:20:08 istlab sshd[22387]: Invalid user adonis from
Aug  1 00:20:11 istlab sshd[22400]: Invalid user adrasteia from
Aug  1 00:20:14 istlab sshd[22417]: Invalid user adrastos from

The attack to this host (which is based in Athens, Greece) came from a
Hong-Kong-based machine, and the list contained many exotic Greek names
while also missing many common ones. Therefore, I doubt that this was a
local attack. A Google search revealed that the name list was obtained by
merging male Greek names and female Greek names from Most probably an attack tool contains lists of
names for specific countries (the same site also provides, African, Chinese,
English, French, German, Hebrew, Irish, Italian, Japanese, Polish, Spanish,
and Welsh names). The tool also maps the IP address of the host it attacks
to a specific country, for instance, through the geolocation data of the
IP-to-Country databases Finally, the
attack tool uses the country-specific list for trying to log in to those
accounts. Attackers seem to be getting more sophisticated with every passing

Diomidis Spinellis -

Amazon chasing 2-cent Web services bill

<"Martin Redington" <>>
Fri, 3 Aug 2007 02:19:36 +0100

I recently signed up for an Amazon Web services account, to try out their S3
service, supplying my credit card number for them to bill me.  I played with
the service very briefly, enough to incur $0.02 of charges, which appeared
in the statement they sent me on Wednesday.

Today I received a notification from Amazon that their attempt to charge my
credit card had failed (presumably because the amount was too low), and
asking if I could amend my account with valid payment method.

Hopefully sanity checking will prevail before they start seriously chasing
me for the money.

Windows Live Messenger blocking even more completely innocuous text

<Cody Boisclair <>>
Wed, 1 Aug 2007 16:42:25 -0400

In RISKS-24.35, there was an entry I submitted detailing how Microsoft's
Windows Live Messenger service silently filtered out any message containing
".scr" or ".pif", in a very ham-handed attempt to prevent links to trojans
from coming through.

Even more recently, Microsoft has decided that any IM containing the
substring ".info" should be silently discarded.

Yes, that's right. In an attempt to combat links to malicious executables
hosted on a few random .info domains, they've blocked every reference to an
entire top-level domain... and even *that*, as heinous as it may be, isn't
the full extent of the block. Sharing a link to an article on via Messenger will be a futile effort indeed, for
instance. And good luck trying to ask other .NET developers whether
MessageBoxIcon.Information is the best icon for a given dialog.

The RISKS here are enough to leave one speechless-- in more ways than one!

Cody "codeman38" Boisclair

Re: Accuracy of Hawkeye at Wimbledon (PGN, RISKS-24.76)

<Paul Wallich <>>
Tue, 31 Jul 2007 20:26:37 -0400

Under the most obvious assumptions about the distribution of "in" vs "out"
in disputed calls, that would mean the system was performing worse than
chance on its four worst days. That's really not good at all.  (There are
other plausible distributions, of course. If players object to any call they
think has a reasonable possibility of being reverse to their advantage,
including some where their objective judgment agrees with the call, that
would mean the system was performing seriously worse than chance. If players
only object to calls that they think were utterly bogus, the system might
well be doing better than chance on disputed calls. Of course, in that case,
it might still be doing worse than chance on close shots in general.)

Fraudproof voting protocols from scientists

<"Warren Smith" <>>
Thu, 2 Aug 2007 09:44:42 -0400

Simple New Voting Protocols provide Ballot Secrecy AND Fraud Resistance

Conventional wisdom says elections with "secret ballots" are protected
against vote-buying and coercion, while elections publicizing the list of
all voters with their votes are immune to fraud -- but you can't have it
both ways.  In a paper at EVT 07 (Boston, 6 August) mathematicians Ronald
L. Rivest and Warren D. Smith refute that conventional wisdom, potentially
enabling a new level of voting integrity.

"You can have your cake and eat it too with some very simple new voting
protocols," said Professor Daniel Sleator of Carnegie-Mellon's computer
science department.  "These are explainable to children.  It's surprising
this wasn't thought of 50 years ago."

Previous attempts to create such protocols have "succeeded" in mathematical
senses, but only by employing very complicated cryptographic algorithms,
challenging even for math PhDs.  Humans can't vote in those systems without
computer aid, which means that each voter would have to own a small computer
"helper" they trusted to be running correct, unhacked, voting software.

Rivest & Smith's new protocols, called "VAV," "Twin," and "ThreeBallot,"
don't require computers or cryptography, and need only low-tech mechanical
voting devices.  In each, voters get take-home "receipts" they can use later
to check their vote was correctly counted -- or prove fraud -- but which
nevertheless bear absolutely no relation to that voter's vote, hence aren't
helpful for vote-selling.

How can that be?  Your take-home receipt in Twin is a copy of a random
_other_ person's vote.  In VAV, each voter casts two votes and one matching
"antivote" and gets a copy of one of these three (she chooses which) as her
receipt.  Either way, the receipt has no logical relation to that voter's

All three Rivest-Smith protocols allow "mixing in" old-style unsafe ballots
with the new safe ones.  That not only permits happy coexistence with voters
who don't want to use the new system, but also "contagiously protects" even
the unsafe ballots against fraud.  "I really love this 'easy upgrade'
feature," said Doug Jones, former chair of Iowa voting systems examiners and
computer science professor at University of Iowa.

The Rivest-Smith protocols work with a wide variety of vote-totaling
systems, not just the "plurality" system most familiar in the USA.
"Plurality is a very poor voting system," said Guy Ottewell, an astronomer
and author regarded as the inventor of Approval Voting in 1968.  "We've
known better ones for 200 years."  "In plurality voting, it's 'name one
candidate then shut up'," said Ottewell.  "With Approval, you name _all_ the
candidates you 'approve.'  It's actually simpler because there is no special
rule outlawing 'overvoting,' and it both delivers more information in each
vote and allows voters to approve their true favorite without being
strategically foolish, so it's also more honest information."

But why would voters want dishonestly to vote for someone other than their
true favorite?  "Two words," said Ottewell. "Ralph Nader."  "With approval
voting, Nader voters aren't a problem, they're beneficial."

But Ottewell and Smith now instead advocate "Range voting," essentially the
system used in the Olympics: as their vote, voters score all the candidates
they want to within some fixed score-range (say 0 to 9); highest average
wins.  (Range becomes the same as Approval if the range is 0 and 1.)
"Honeybees have been using range voting for millions of years, and my
computer simulations indicate it outperforms every other common
vote-totaling proposal," said Smith.  ###

Fuller Story (including how VAV & Twin actually work):
Rivest-Smith actual paper:
    also in html:  
Addenda to the paper:
Follow-up stories: 
EVT 07 Conference: 
Center for Range Voting:

  Dr. Warren D. Smith   631-675-6128    warren.wds AT  (prefer email)
*Approval & Range voting (AV & RV):
  Guy Ottewell  +1297-442247   guy AT
*(AV, RV, and also most other vote-totaling systems too)
  Prof. Steven Brams,  NYU politics dept. 212-998-8510  steven.brams AT
  (co-author of book "Approval Voting")    FAX: 212-995-4184
*Computer Science:
  Prof. Daniel Sleator, CMU CS dept. Office ph 412-268-7563, fax: 412-268-5576,
     home ph: 412-HACKERS

REVIEW: "Implementing ITIL", Randy A. Steinberg

<Rob Slade <>>
Wed, 01 Aug 2007 08:29:51 -0800

BKIMITIL.RVW   20070228

"Implementing ITIL", Randy A. Steinberg, 2005, 141206618-2
%A   Randy A. Steinberg
%C   Suite 6E, 2333 Government Street, Victoria, BC   V8T 4P4
%D   2005
%G   141206618-2
%I   Trafford Publishing
%O   888-232-4444 FAX 250-383-6804 sales@trafford.Com
%O   Audience i- Tech 1 Writing 1 (see revfaq.htm for explanation)
%P   489 p.
%T   "Implementing ITIL"

Chapter one notes that there are problems in how information technology (IT)
works in supporting the enterprise.  Steinberg does mention that there
should be better integration of the various parts and functions of IT
service, that IT service management (ITSM) should be performed better, and
that the Information Technology Infrastructure Library (ITIL) is a framework
for improving ITSM, but does not, at this point, define either ITIL (and
never does explain ITSM).  Nine general principles for success are listed in
chapter two.  The precepts are sound (such as targeting the "Pareto"
processes that are going to give you the best results for least effort), but
vague: there are almost no details on how to accomplish this wonderful
state.  Chapter three provides a generic and rather terse outline of a
general project management cycle, under the heading of a process for
implementing ITSM over a period of a year.  Modification of the culture of a
corporation is a massive and difficult task: the suggestions in chapter four
have some interesting and useful detail in regard to communications, but
disregard the challenges involved.  A catalogue of roles for large teams and
projects is given in chapter five: this is probably too large for most ITSM

Chapters six through eleven outline the general stages in a project cycle,
albeit with idiosyncratic names for most phases (and missing a few steps,
such as requirements definition, testing, post- implementation assessment,
and maintenance).  The material is reasonable, although quite terse and
vague.  A great deal of space is devoted to forms, checklists, and
questionnaires.  These would probably be quite useful as templates for those
involved in an ITSM improvement project, but would have to be refined for a
specific situation.  "Vision," in chapter six, is basically the project
concept or initiation phase.  "Assessment" is given a separate chapter
(seven), but seems to be part of the concept definition.  Planning is in
eight, and implementation in nine.  "Initial wins" are described, in chapter
ten, as small, quick projects that provide some early "high" returns on the
efforts.  The text outlines a management cycle for small projects and so
duplicates a good deal of material that was presented earlier.  There is
also a list of initial win projects, although the value of most is
questionable and they would have to be carefully reviewed for a specific
environment.  "Control work," in chapter eleven, is partly implementation of
small projects, partly overall project documentation and management, and
lots of workflow model charts: the content is rather a mixed bag.

Chapter twelve finally gets around to some details of ITIL: the text does,
rather briefly, present the topical areas (known, in ITIL parlance, as
processes) of the management of incidents, problems, change, release (of
software), configuration, service levels, availability, capacity,
continuity, finance, the service desk, and security.  A poorly explained and
formatted two-dimensional chart of the information flow between processes
makes up chapter thirteen.  Various software utilities and their bare-bones
functions are listed in fourteen, while fifteen mentions miscellaneous
documents related to the ITIL processes.  Chapter sixteen has a terse
catalogue of roles and job descriptions for the processes.  Guiding
principles are defined, in chapter seventeen, in a way that is very similar
to vision or mission statements, albeit with somewhat more detail.

(ITIL is a decent overview of the provision of IT services, but note
that it has gaps.  For example, incident response is seen only in
terms of customer service, without any relation to security.  Security
management has solid and important directives on management, a
holistic approach, policies, and audit, but when it comes to the
actual provision of controls, the advice is to have proper ones,
without much detail on what those might be.)

The title of the work is somewhat misleading.  The largest part of the book
has to do with generic project management.  ITIL does get some presentation,
but not until the book is more than half over.  In addition, the work is
poorly structured and written.  The end of chapter sixteen, as one example,
talks about roles for "ICT," but ICT is not defined until the end of chapter
seventeen (and then only as "Infrastructure Control").  The material is not
complicated, but the writing is frequently unclear, and it is only the
simplicity of the basic concepts that prevents the reader from getting lost.
(Sometimes the writing is completely off the wall.  "Fix just one IT service
problem per day and within 90 days you will have made 107 service
improvements" is clearly self-contradictory.)

For those who have not done much in the way of project management, there are
some helpful guides that will get you going (although you will need to check
in other references such as Scott Berkun's "The Art of Project Management"
[cf. BKARPRMA.RVW] or "Applied Software Project Management" by Stellman and
Greene [cf. BKAPSWPM.RVW] in order to deal with the missing bits).  For
those not familiar with ITIL, chapter twelve is a reasonable introduction.
For those working to improve ITSM within their enterprises you will probably
need a bit more help than is provided herein.

copyright Robert M. Slade, 2007   BKIMITIL.RVW   20070228

Please report problems with the web pages to the maintainer