Password Security: What Users Know and What They Actually Do posted on Bruce Schneier’s blog.
The finding that participants in the current study use such simplistic practices to develop passwords is supported by similar research by Bishop and Klein (1995) and Vu, Bhargav & Proctor (2003) who found that even with the application of password guidelines, users would tend to revert to the simplest possible strategies (Proctor et al., 2002). In the current study, nearly 60% of the respondents reported that they do not vary the complexity of their passwords depending on the nature of the site and 53% indicated that they never change their password if they are not required to do so. These practices are most likely encouraged by the fact that users maintain multiple accounts (average = 8.5) and have difficulty recalling too many unique passwords.
Privacy
GOP Phone Home
From the everyone-has-an-angle dept: Minnesota GOP’s CD Raises Privacy Concerns
ThinkProgress blog is reporting that the Minnesota Republican Party has been distributing a new CD about a recent proposed amendment. The CD poses questions about some of the hot-button issues like abortion, gun control, and illegal immigration. The problem with this CD, however, is that it “phones home” to the Minnesota GOP, without making it clear that your name is attached. So, if you take a look at the CD and take time to answer the questions, beware. Once you are finished they will know not only who you are, but where you stand on the issues at hand.
Bruce Schneier: Why Security Has So Little To Do With Security
Sorry for the delay this is Bruce Schneiers talk at RSA on Why Security has so Little to do with Security.
There is no such thing as absolute security!
Security involves trade-offs:
- If no airplanes flew – 9/11 couldn’t have happened.
- Air port security has trade offs – tax dollars, time, “calmness”)
- Gated communities offere more security but less privacy vs. buying home alarm system costs money and convenience.
- We are not wearing bullet proof vests… we don’t want to spend the money or fashion tradeoff.
We make decisions every day about these trade-offs
Additional security depends on the risk and the effectiveness of the counter measure. People are security consumers the right question is ‘it worthit.’
- People RARELY perform this analysis explicitly
- People succumb to fear and uncertainty
- People beleive falus promieces
- peopel regularly do things that compromise their security
- people don’t do what they say
Security trade offs can be financial, social and non-security
The Barrier model doen’t quite work…There are legitimate users and attachers are bounced. Failures in this model are that the good guys get bounced or the- bad guys in. When system fails most likely fail against legitimate users!!!
Threat = Attacker -> Goals -> Attacks
There are multiples of all these criminals and hobbiest hackers
Attackers – can also be legitimate users (they can get a lot of information about systems to find out how they work – 9/11 they flew airlines).
Attacks divert to other targets…(go attack someone else)
Assets are owned by someone else. -> Trade offs are made by the asset owner. They don’t make the decision.
Security system (access control tarmack, passengers)
ASSETS (airplane)
Perceived Risks + Other considerations “everything else” [ social/moral | technological | legal | economic ]
The owner of the asset around which risks can be mitigated.
Owner -> Policy -> Trusted people -> Trusted systems – security systems to protect assets.
There are things we can not do to protect assets:
- Banks – strip search everyone who comes into bank.
- House – put landminds in lawn
Legitimate users influence owners:
- We can’t put cameras in dressing rooms.
- Government – wanted to ban laptops on airplanes (legitimate users forced decision)
Trusted people influence: –
Urin test for federal employees in Regan years. They said no.
The effectiveness of security system…minor component in complex decision graph.
Look at the sum of the stuff pointing at the owner. Every security decision affects multiple players and the party who gets to make the decision will make one thats’ benificial to him.
Every security decision affects multiple players… Look at Guns on airplanes .. pilots, flight attendants, citizens legislators
If they don’t want to buy it they are not ‘irrational’ it is rational within their world view…
Every player has his own unique perspective, his own trade-offs and his own risk analysis. You have to evaluate security options based on the positions of the players. Often security decisions are made for non-security reasons. The major security issues have nothing to do with security technology.
Detecting counterfeit money there is no incentive to detect it. I as a customer if we find it in our wallet just want to spend it the merchant if they find it will not report it as long as they can deposit it.
Look at KAL 007 the last western flight to get shot down over Russian.
- The reason was that the prior time a western airline flew over Russian airspace general in charge lost his charge. So this general did not want that to happen to him. The agenda general who decides to shoot plane down (I need to save my career.)
Look at salesclerks and credit card verification.
- They never check – they don’t care it is not their agenda. Make transaction go through with minimal stress. Owner of store more likely to check.
Look at Tylenol poisonings and random stupid crimes
- Security is the science of tamper resistant packaging. They are silly security measures but they address the real problem (public no longer trusts over the counter drugs).
Look at banning things on airplanes.
- Didn’t ban matches and lighters – cigaret lobby got to congress.. better the knitting needle lobby
SECURITY PROXIES are a way we address risk.
Proxies are players who acts in the interest of other players
(airline security:TSA…airlines not alowd to compete on everyone – more secure airlines, background checks – less secure airlines – no lines)
Proxies are a necessary aspect of the security because people arn’t proficient at risk management. Proxies are not going to make the same risk management decisions that the people they represent would have…
Home building inspectors
They ultimately work for the real estate business if deny to many timse they won’t have business…They are Mercenaries. If you hire them make sure you pay them….turn around and attack people who hire them.
DVD region encoding
- There are secret features to defeat…manufactures as proxies…manufactures want the cheat codes piss of media companies.
Government regulatory bodies
- yes safety but also their jobs (my comment: embedded autonomy)
Banks’ verification of signature on checks
- They don’t do this – why check? No security problem until customer complains. Push security onto the consumer. Bank to centralize security – checks signatures? distirubuting it is cheaper then…
Banks’ verification of balance in account
Compare – bank security of overdrafts – they will spend a lot to deal with that kind of fraud.
Cell phone security against eavesdropping
- Ridiculous that they don’t have encription. Phone companies don’t care – voice privacy not affect bottom line. In who’s agenda is voice privacy? the customers.
Cell phone security against third-party batteries
- Third party batteries they care – security device ‘security of their revenue’
Security is never an isolated good. It always makes sence in the context of a greater system. That system is more important then security. Security is a secondary consideration ( it is an after thought). Understanding the context is just as important understanding the security.
The context of the interent today is a lot of spying stuff. Fighting the context won’t help…must work with the context.
Subscribe to Crypto-Gram: Free Monthly Security Newsletter
Questions:
About Proxie diversion: NSA protects us by snooping on us. NSA – protect ours and attack theres. Equities issue If you are in the NSA and tell Microsoft about major flaws…If you don’t then do you tell them — you can spy on chinese etc. defend good guys or attack bad guys. NSA eaves dropping on Americans. Police start making these decisions you get security that looks like a police state – powers of investigation and who. That is the way they think… This is a good example of proxie divergence. Important that the POLICE don’t run the countyr. somone above the police making the decisions above them civilian control of intelligence – pull the proxie back a bit.
Someone in government today [elected official]…their agenda in getting re-elected is measures to incent proxies properly understanding it we might be able to correct for it. Elections are about fundraising. Clever electoral reform systems…try to deal with proxie problem recognize them and then figuring out risk…I act as a proxie for corporation and vendors. What are the METRICS for measureing risk.
Regulations not most of my data is not controled by me company X (Choicepoint) the have controle over it but I am not a costomer…
The delta value is extreme. underprotected….
Regulation to choice point to take security more serioiusly. Does it work? is it effective? Increasing the cost to the company – make it more expensive to ignore security – vendors SOX is good for us there is more money to spend more money on security.
Bruce Schnider is brilliant!!!!
I saw Bruce Schnier twice at RSA and missed the third time he was on stage for a debate on Real ID. If you don’t read his blog you should. I was excited to see him at RSA and he surpassed my expectations. He is a very clear communicator and full of common sense. He handed out Individual-I stickers and his book of San Jose Restaurant Reviews.
Individual-i stands for:
* Freedom from surveillance
* Personal privacy
* Anonymity
* Equal protection
* Due process
* Freedom to read, write, think, speak, associate, and travel
* The right to make your own choices about sex, reproduction, marriage, and death
* The right to dissent
The next few posts are what I transcribed as best I could while he talked.
What kind of e-mail is it?
I recently got an e-mail with this at the bottom – a low tech way to assert privacy and ownership. This email is: [ ] blogable [ ] ask first [X] private
TSA data cloud searching – Flights today, Subways tomorrow?
This article was slashdotted today.
TSA had promised it would only use the limited information about passengers that it had obtained from airlines. Instead, the agency and its contractors compiled files on people using data from commercial brokers and then compared those files with the lists.
The GAO reported that about 100 million records were collected.
The 1974 Privacy Act requires the government to notify the public when it collects information about people. It must say who it’s gathering information about, what kinds of information, why it’s being collected and how the information is stored.
And to protect people from having misinformation about them in their files, the government must also disclose how they can access and correct the data it has collected.
Before it began testing Secure Flight, the TSA published notices in September and November saying that it would collect from airlines information about people who flew commercially in June 2004.
Instead, the agency actually took 43,000 names of passengers and used about 200,000 variations of those names – who turned out to be real people who may not have flown that month, the GAO said. A TSA contractor collected 100 million records on those names.
It brings up some serious concerns about how information collection and validation is done by the TSA for airline passengers. How can we trust governments to collect this much information about us just because we travel.
This week I wonder why care about airlines passengers because security is so tight that airlines do not seem to be a place where the next round of attacks will be. If London is any indication it will be on mass transit. Given the level of police/security presence on the transit systems in the Bay Area this week is certainly seems like there is some concern that mass transit will be attacked. They have started random searching of bags to get on the NYC subway. One wonders if they will start issuing ‘identity passes’ to get on such systems.
On the city subways, which are used by 4.5 million people on the average workday, the inspections started on a small scale Thursday afternoon and were expanded Friday.
The New York Civil Liberties Union opposed the searches, saying they violated the Fourth Amendment. Mayor Michael Bloomberg said he hoped the NYCLU would recognize that the city had struck the right balance between security and protecting constitutional rights. He said the bag-checking program is part of a policy to “constantly change tactics” and “may, or may not, be there tomorrow.”
Index Finger Scanning at Disney World + FastTrack Scanning
This article was Slashdotted…
Tourists visiting Disney theme parks in Central Florida must now provide their index and middle fingers to be scanned before entering the front gates.
The scans were formerly for season pass holders but now everyone must provide their fingers, Local 6 News reported. They have reportedly been phased in for all ticket holders during the past six months, according to a report.
I think it’s a step in the wrong direction,” Civil Liberties Union spokesman George Crossley said. “I think it is a step toward collection personal information on people regardless of what Disney says.
I think this is self explanatory in terms of why it is concerning. It seems to goes along with what is now happening with FastTrack passes (automatic toll readers) that I heard about last night at the Hillside Club CyberSalon where Esther Dyson was speaking. I googled the phenomena and here are some excerpts of what I found.
In New York State, readers have been multiplying ever since September 1997, when the New York Police Department (NYPD) used E-Z Pass toll records to locate and track the movements of a car owned by Nelson G. Gross, a New Jersey millionaire who had been abducted and murdered. The NYPD had neither a subpoena nor a warrant to obtain those records; the police simply asked the Metropolitan Transportation Authority (MTA), and the MTA complied. This set a very bad precedent. Though Gross wasn’t alive to complain about it, his privacy had been violated. Access to those toll records also permitted access to all sorts of sensitive information, including his billing address, his credit card number, his license plate number and his Social Security number.
In February 1998, the MTA announced that — near the Tappan Zee Bridge (the site of the first reader in New York State, installed in 1993) — it had just concluded a successful “experiment” with readers that could detect and extract information from transponders even though the cars to which they were attached didn’t slow down. These “high-speed readers” were only three-feet tall and could be placed just about anywhere. As a result, they permitted the ETC system to do something it was never intended to do: namely, collect truly huge amounts of information about such non-toll related phenomena as traffic flows, speeds, densities and delays (all of which, incidentally, can be videotaped by either flow monitoring or security cameras that have been automatically activated by the readers).
Since then, high-speed readers have been installed along a great many State-owned roads and highways; they’ve also been installed atop many residential buildings in New York City.
Catalyst: Logic of Identity – Bob Blakley Chief Scientist IBM
This is a summary of Bob Blakley’s talk at Burton Catalyst:
Opening – Sermon on Laws
Laws of Planetary Motion
Kim’s Laws what happens to Identity if you make stupid or subtle mistakes
Newtons Law – gravity
Why things happen
Introduction – Looking Back Digital Signatures
A while back we decided we needed non-repudiation and did digital signatures by issuing certificates.
We forgot to figure out why do signatures work in the real world.
So, we got how they worked wrong in the technical world.
Having signatures not work is bad looking forward having privacy not work is bad.
Body of Talk
Definition:
Identity is a collection of attributes by which a person or thing is generally recognized or known
Identity Relativity
The Identity of X according to Y: The set of attributes believed by Y to be true of X.
Axiom: Utility
An identity attribute has value if and only if knowing that attribute reduces risk for some party
Reducing one party’s risk often creates risks for other parties.
Consequence: Identification is Power
Identity allocates risk.The ability to create or eliminate a risk for another confers power over the other.
Axiom: Contention
Because identity claims allocate risks, they will be disputed.
Identity Attributes
- Commercial Interest – Convenience
- Government Interest – Security
- Individual interest – Privacy
Definition
Privacy: is the ability to lie about yourself and get away with it.
Axiom: Subjectivity
People disagree about one anothers identity attributes
In general, there’s now easy way to tell who’s right and who’s wrong
Axiom: Temporality
The name that can be named is not enduring and unchanging name. All identity attributes change over time.
- Prince -> symbol
- Michael Jackson Black -> Plastified
Axiom: Obscurity
Identity attributes can be
- what you know – you can lie
- what you have – loose / leave
- what you are – alter disguise
Axiom: Publicity
Identity attributes cannot be secret
By definition attributes aren’t observable can’t be used to use attributes
Axiom: Contextually
Identity is inherently subject to effect of scale.
Brandon Mayfield – guy who did not blow up trains
His finger print matched one at Madrid Bombing (it was not an accurate assertion)
Large databases -> not completely reliable
To scale identity information one needs to collect — more information
Consequence: Powerlessness
Identity is in they eye of the beholder – subjectivity.
- You can’t control what other people think or say about you.
- You can’t even know who knows what about you.
- Can control what you tell people but not what people find out
Consequence: Privacy Erosion
Scale requires distinguishing between lots of individuals which requires lots of information.
In a sufficiently large population the commonly agreed to be public attributes will not distinguishing individuals well enough.
So information about sensitive attributes will be collected.
In the UK they are look at putting in scanners (QinetiQ) while entering the subway to detect knives but what about creep in the use of other things identifying tatoos?
People push back against government identification.
Consequence: Due Process
Because identity is subjective, contextually, contention and obscurity and temporality.
IDENTIFICATION REQUIRES DUE PROCESS
But due process undermines the business case for identity. Due process requires transparency. Transparency reveals how identity attributes are collected and synthesized to make judgment. Collection and Synthesis are the only sources of completive value.
They do it because they like costumer intimacy.
Supply and Demand mismatch between favorable and unfavorable information.
Favorable information is easy to get.
The subject is happy to give it to you and the subject is happy to help you authenticate it. Therefore the supply is large and the value is low. But it’s worse: Demand is also low! Because favorable information is less likely to reduce another party’s risk. Especially the case when the other party has lots of potential customers.
The business case fore identity service provider infringes privacy.
The business of identity service providers is risk reduction withholding adverse information decrease the value of business.
Collecting more adverse information makes more.
Identity and Privacy are Incompatible.
Adverse information has positive identity value but negative privacy value.
Favorable information has zero identity value and zero privacy value.
Fable about MARIA
Recent guatemalan immigration
she has AIDS and she doesn’t want anyone to know. The health insurance company wants to know this information because it is a $180,000 not to know this.
Meta-data, collaborative filtering and identity
Collaborative Filtering and Identity from John Udell. A use case we should be thinking about.
What if one would create a crawler application that, using all of these sources, could compine a complete “view” of my digital online self?
The problem though, is that “social” internet tools, that effectively visualize this stuff, are not all that common yet. Bloglines, Del.icio.us, and Flickr, and a few others, are still few and far between.
And then there is privacy. Now personally, I have no problem sharing all this metadata about myself. I would not even mind my browsing habbits being monitored to service the “social” metaspace like I do with the above-mentioned services, provided I had complete and transparent control over when what was monitored.
Wouldnt it be kinda creepy, once we indeed had a centralized match-and-compare system for all of this data, if you where to run into another person online that mirrored you and your interestes in every way? And online doppelganger, so to speak.
Canada has some crazy laws too.
Canada has some crazy laws too. I kind of was thinking of Canada where I was born as the ‘friendly’ nation to the north but it seems not to be true :(.
Before privacy laws or the Charter, there was little if anything to stop police or national security operatives from cajoling or coercing information from private sector organizations. A civic-minded government department or company could blab all it wanted about its customers or employees.
Our privacy laws changed this, although they didn’t really try to put a stop to it. In BC, our public sector privacy law gives public bodies discretion to disclose personal information for law enforcement purposes, without warrant, but there are (some would argue, weak) constraints on this. The same can be said for our private sector privacy law. Still, these laws, together with the Charter, have until recently insulated against over-enthusiastic private sector co-operation with all and sundry state inquiries. Is this still true? If it is, how long will this last?
After the 9/11 attacks, governments everywhere felt compelled to act, and to be seen to act. This was in an important sense responsible of government. It was also mandated by political Darwinism. But a profoundly important aspect of the post-9/11 changes is the blurring of lines between collection and use of personal information for law enforcement purposes under criminal and other penal laws and use for national security purposes. A defining characteristic of police states is the blurring of distinctions between law enforcement and national security functions, the danger being that the rule of law eventually gives way to arbitrary decision-making by law enforcement authorities and the rights of ordinary citizens lose meaning. Democracies depend on clear and effective rules suited to the state activities that the rules are intended to govern and that reflect the essential values of a free society.
In Canada, post-9/11 amendments to the Customs Act and regulations authorize officials to require private sector organizations to provide border officials with extensive advance information about arriving passengers. These changes expanded the federal government’s ability to use and share that information, not only for national security purposes, but also for ordinary law enforcement and other purposes, including (according to government statements in 2002) public health surveillance. The information-sharing authority includes a broad ability to share personal information about Canadians and others with foreign governments. The amendments don’t restrict information-sharing arrangements to national security uses they could easily include ordinary law enforcement or other purposes defined on a case-by-case basis or in an agreement with another nation.
Also, Public Safety Act amendments to the Aeronautics Act allow the RCMP Commissioner to require any air carrier or operator of an air reservation system to, for the purposes of transportation security, disclose specified information in its control to any person the Commissioner designates. Despite the Public Safety Act reference to transportation security, the amendments allow this data to be matched with other data and to be disclosed to assist in executing certain outstanding arrest warrants. This effectively compels the private sector to assist the state, in the absence of a warrant or court order, in surveillance of all air travellers for the broader general purposes of both national security and ordinary law enforcement.
Consistent with these powers to conscript the private sector into both national security and law enforcement activities, Public Safety Act amendments to the Personal Information Protection and Electronic Documents Act (PIPEDA) permit private sector organizations to collect personal information without an individual’s knowledge or consent in circumstances that amount to an invitation to, and in some cases compulsion of, the private sector to assist the state in surveillance for both general national security and ordinary law enforcement purposes.
The Public Safety Act also amended the Proceeds of Crime (Money Laundering) and Terrorist Financing Act to authorize the Financial Transactions and Reports Analysis Centre of Canada to collect information it considers relevant to money laundering or financing of terrorist activities from publicly available information, including commercially available databases. FINTRAC is also authorized to obtain, under information-sharing agreements, information maintained by federal or provincial governments for law enforcement or national security reasons.
FINTRAC expanded powers point to the fact that, when it comes to co-opting the private sector, 9/11 can’t be blamed for everything. Laundering of dirty money was of sufficient concern before 9/11 to lead to extensive transaction-reporting requirements for banks and others. You can easily find other examples of legislative responses to individually pressing policy challenges that draft private sector organizations into state service in the name of public safety or order. One example is the current federal government lawful access proposals, some of which would apparently require ISPs to hand over at least identifying customer information and perhaps more on simple request by state officials, and for a pretty broad range of uses.
Also, at the local level, at least in BC, we see more and more local government bylaws compelling businesses to hand customer information over to police for a variety of reasons. Pawnshop reporting requirements have been around for a long time, but now were seeing bylaws requiring businesses to regularly give police information, without request, in a variety of situations (such as information who’s been buying pepper spray, hydroponic supplies or chemicals that could be used to make drugs and who’s been renting mailboxes at commercial mailbox centres).
And governments are now large purchasers of personal information from the private sector. So far this is being seen mostly in the US think of Total Information Awareness, MATRIX, Secure Flight and so on but to think that our own governments will ignore the expanding private sector trove of electronic personal information much longer.
As databases proliferate, become more comprehensive and become lifelong, it’ll be harder and harder to resist those who say that, since the information is out there, the state should be able to use it. Time and time again over the last six years I’ve been told by middle-aged, middle class Caucasian males that they have nothing to hide, so why should anyone else feel differently? Let the government have the information it needs to protect us, they say.
Now, I don’t doubt the good faith of BC’s police agencies not for a minute. But, thinking thirty or fifty years down the road to a time when the lines between national security and law enforcement have blurred to vanishing, will there be any meaningful rules? If not, will our belief in the good faith of state officials, set adrift without guiding rules, be enough to sustain our privacy and other rights?
Belgium Identity Cards
From ID Corner comes this story about Belgium Identity Cards.
The card provides strong security against traditional outsider attacks, but unfortunately has not been designed with privacy in mind. In fact, it features one of the worst privacy designs imaginable. Two glaring problems:
The citizen certificates on each ID chipcard contain the cardholder’s name and RRN (the œrijksregistratienummer,” a single government-wide identification number for each natural person). The name and RRN are disclosed whenever a card is used at a relying party. The RRN (which has a simple structure based on the citizen’s birthday) serves as the key to numerous databases containing citizen information; on the basis of this number, all cardholder actions and movements with the eID chipcard can be electronically traced and linked (not merely by the government itself!).
The eID card specifies the following information, both visibly on the card itself and stored within the card’s chip: cardholder’s photo, surname and first names, gender, nationality, place and date of birth, signature, RRN, and the validity period of the card. In addition, the chip also stores the cardholder’s current address. Some of this information is privacy-sensitive, yet the cardholder has no control over its disclosure. (Historically, this is the same information as has always been on Belgium identity cards, and so arguably this does not constitute a reduction in privacy; however, in most countries around the world an information-rich national identity card would not pass in the first place.)
The privacy problems do not stop here. Each eID chip contains two X.509v3 identity certificates (each specifying the citizen’s name and RRN number, one for authentication and one for digital signing), as well as a basic signature key to authenticate the card with respect to the RRN. The certificates and public keys, which are assigned by the central issuing authority, by themselves serve as “omni-directional” identifiers that are globally unique. For a detailed account on the various privacy problems caused by this use of PKI, see, for instance, here.