Bittersweet Security

All the way north on the Island of Madagascar is a city named Ambanja. The E. Guittard company claims to produce a 65% cacao bittersweet with flavors from the region. If you believe their website, the bars are a product of Criollo beans from the fertile Sambirano Valley.

Personally, all I can say is that I found the Ambanja Bittersweet very dry and light in taste, and a stark contrast to Guittard’s Chucuri Bittersweet. The latter is apparently a Columbian bean, which I think has a far more smooth and spicy flavor with a rich and familiar aftertaste.

This all makes me wonder if the “unknown” method of distributing food will come under pressure from newer and better distribution methods for old-world and boutique-type brands.

Take for example the unpleasant situation when a restaurant tells you that ground beef can not be prepared “rare” because of a law meant to protect you from disease — bad beef. Someone should alert the big beef that automation can be counter-productive when it becomes overly efficient at promoting one value in spite of all the others. In fact I usually say I would pay more if I could get a hamburger that came right from the “trusted” local butcher because I know my body is happier when I eat better food. I guess I should find out if you can even have a local butcher, baker…

So although I truly appreciate the security control model provided by the US government to reign in the mass-automation meal industry I would much rather know that the origins of my meal could be traced and therefore controlled right at the root-causes. Come to think of it, how do I find out whether the beef industry has the same or better tolerance for risk that I do? Is their idea of “safe” one in 1,000,000 deaths or is it the big fat 0?

Consider for a second the BSE website, which was prominently advertised on the front page of the National Cattlemen’s Beef Association. It provides the following assurance:

U.S. beef producers have worked with federal authorities for more than 15 years to set up the system of science- based firewalls that is working today to keep the food supply safe.

Hmmm, last time I checked firewalls are a single control and thus widely considered insufficient on their own to provide adequate security. Not such a great marketing campaign, if you ask me. Alas, nothing else is mentioned although I found it interesting that the Cattlemen’s website also links to some anti-vegetarian propaganda.

I suspect that if a proper set of consumer-based controls were in place, they might be able to preserve “single-origin” (e.g. quality) values on a large scale, such that we would still have excellent flavor and texture along with desireable price. But until that happens, wise consumers seek out the small-batch and single-origin brands that are a healthier choice and more in tune with their real needs (better cost-benefit ratio).

Back to chocolate, I have to wonder, are you safer trying to stay on top of the additives in the giant brand chocolate bars, or are controls more likely to be present and effective with small-batch real cacao, cane sugar, lecithin and vanilla? And does fair-trade mean less chance of sabotage? Mmmm, chocolate.

When was the last time you looked at your Padlock?

The little gold SSL padlock, that is.

VeriSign is reported to be saying some interesting things about changes they would like to see to increase user trust in SSL certificates. Most would agree that the level of protection from SSL encryption has made a huge improvement to online commerce for a very minimal investment (even “official” intermediary-signed SSL certs can be purchased for as little as $30/each). However the ubiquity of SSL, and lack of a unified standard root authority, has included a trade-off in terms of validity of the certificates. In other words, as the old adage goes, the lower the barrier to adoption the higher the rate of fraud.

So, if you are a certificate-selling company, you are probably debating how to introduce new controls to (re)establish the trustworthiness of the padlock (and raise prices). The browser companies are thusly also considering how to upgrade the padlock to represent the upcoming upgrade in “assurance”. Well, actually, to be fair they are considering how to represent the assurance that was supposed happen in the first place, now that the current icon has been watered-down to represent “RC4128” and not much more:

When the padlock was first invented by Netscape in the early days of the Web, it stood for a secured connection with an identified Web site. That changed when some certification authorities started lowering their verification standards and discounting certificates, said Judy Shapiro, vice president of marketing at Comodo. “Browsers did an end-run around this. Nobody expected anyone to delete what was a key part of the certificate issuance process, which was the business verification,” she said. “Browsers were unprepared to display high assurance and low assurance certificates in a different way.”

Kudos to Comodo for saying so…I guess if you have lost control of a currency’s value, you have to print new currency to restablish control.

Balzana Olive Oil

I wasn’t going to say anything about the 2005 batches of Balzana California Extra Virgin Olive Oil (more for me, you know) but some friends convinced me that I should be generous and share the news. It’s an oil derived from several varieties of olives chosen by Mr. Edmunds himself with an unbelievably smooth and rich hint of pepper. The “extra virgin” means the olives are grown very near the place that the final oil is produced. Seriously good stuff. The last thing I can remember from Santa Cruz that tasted this good was a 2000 Bonnie Doon Big House Red. Give ol’ Merritt a call and get some for yourself:

Merritt Edmunds, Balzana
2655 Warren St., Santa Cruz, CA, 95062
831.475.7873 or 800.815.9775

Actually, is it single barrel oil or small batch…? I’m going to have to buy a case of the stuff just to be sure I don’t run out.

It seems to go well with anything, but I’ve been tasting it with the Explorateur triple crème, a super buttery and almost grotesquely moldy cheese from Ile de France.

Sober Day, 2006

F-secure has an excellent write-up on their blog that details an impending Sober attack, scheduled for January 06, 2006:

Sober.Y was the biggest email outbreak of the year. It still is responsbile for around 40% of all the infections we see. This variant is programmed to activate on January 6th, 2006. After this date all the infected machines will regularily try to download and run a file from a website, forever. The virus even synchronizes the machines via atom clocks so the activation will not happen before January 6th, even if the clock of the computer is incorrect.

Scan early, scan often. But the more interesting part of their log entry is this:

The Sober virus author can precalculate the URLs. We wanted to be able to do the same thing. So we cracked the algorithm. This enabled us to calculate the download URLs for any future date. In fact, we did this already in May 2005, and we informed the local police in Germany as well as the affected ISPs. But we didn’t want to talk about it publically then – we didn’t want to fill in the virus writer on this. But he must know this by now.

And then they give examples of the URLs. Nice work.

Roaring Forties Australian Blue

The cheese of the day was the King Island Dairy Roaring Forties Blue. Another discovery at a local grocery store, the Blue had the appearance (and name) of a typical American blue cheese, but had far less bite and an awesomely smooth texture that is hard to find in domestic varities that tend to be dry and crumbly. After I polished off the last bit this evening I searched for the King Island website and found this helpful description:

A full flavoured blue with a sweet, slightly nutty character and good aftertaste. A rindless cheese matured in wax thus retaining its moisture and creating a smooth and creamy texture. A Roquefort style mould is used to create this unique and exciting cheese style.

Mmmm. A really great cheese. A bit of googling uncovered a recent newspaper review in the San Francisco Chronicle, which might be related somehow to the appearance of the cheese at a local grocery that boasts of a selection of over 3,000 wines:

My favorite among those I’ve tasted is the Roaring Forties Blue, a creamy, mild, blue-veined cheese from pasteurized cow’s milk. Local retailers tell me it is a customer favorite, too. […] Under the wax, you’ll find a moist, smooth and creamy blue with a mellow, almost sweet taste. It has neither the saltiness nor the pungency that characterizes many blues, which probably accounts for its popularity. Its lush, velvety texture calls for an equally luscious wine. Lustau’s Rare Cream Sherry, Solera Superior, accompanies it beautifully.

I couldn’t (and didn’t) say it any better myself, especially since I’ve never heard of those wines. I can just imagine that groceries in the future will have “hyper-linked” food. For example, when you pick up a cheese and put it in your cart, the cart’s interface will alert you to the appropriate selection of crackers and wine. Talk about a powerful and ubiquitous commerce model for information…

In the meantime, does anyone ever taste cheese with bourbon?

ID breach risk debated

ID Analytics has released a report that suggests ID theft from credit cards needs to be re-evaluated in terms of actual risk to consumers. Reuters picked up the report in today’s news and suggests that the report shows “where thieves access social security numbers and other sensitive information on consumers they have deliberately targeted — only about 1 in 1,000 victims had their identities stolen.” Reuters goes on to say:

After six months of study, comparing compromised information against credit applications, ID Analytics said it discovered something counterintuitive: The smaller the breach, the greater the likelihood the information was subsequently used by fraudsters to hijack the identity of victims.

“If you’re in a breach of 100, 200 or 250 names, there’s a pretty high probability that you’re identity is going to be used,” said Mike Cook, ID Analytics’ co-founder.

“The reason for that is if you look at how long it takes a fraudster to use an identity, they can roughly use 100 to 250 in a year. But as the size of the breach grows, it drops off pretty drastically.”

I do not think that is counter-intuitive at all. Small breaches are likely to be easily explained as directed attacks, as opposed to the more complicated investigation of a story that a box of tapes that have been misplaced. The Ford Motor Credit breach is an example of a massive breach that is both highly directed and that can continue giving consumers grief for many years after their data is stolen. So it is plausable that every “loss” will in fact be discovered to be a succesful attack or “breach” after the fact.

That being said, it is hard not to notice that Reuters claims first that “ID Analytics said it discovered that identity thieves have a hard time using a stolen credit cards to hijack the identity of cardholders because the cards are usually quickly canceled” and then they go on to conclude with a rather contradictory statement from the ID Analytics co-founder:

“As far as notifications, we think there are certain instances where businesses might want to notify consumers and certain instances where they might not to inform them,” said Cook. “For instance, if they lose data, and they don’t know where it is, we think too many notices may not be a good thing. They should probably monitor that and spend dollars on consumers who are actually harmed, rather than spending dollars on 10 million consumers” most of whom won’t be affected.

Where does the certainty come from? If you have lost the data, then presumably you have lost control of its use in the future. Who should be allowed to decide when it is safe to give up on an investigation and declare a “loss” as opposed to a “breach”? In addition, if notification and cancelling card numbers has made ID theft less easy, then why should you not notify all consumers when you have lost their data to ensure the maxium reduction of risk? Monitor to detect and catch a criminal I can understand, but I don’t follow the logic of “notification reduces risk when IDs are stolen, but you might not need to notify”. This reminds me of the theory that if trace amounts of a substance kills less than one in 1,000 customers than large companies might find it more profitable to just pay off the families who suffer rather than prevent the harm.

Perhaps the lack of clarity is because Reuters did not mention the “significant finding”, which can be found at the start of the report from ID Analytics:

A significant finding from the research is that different breaches pose different degrees of risk. In the research, ID Analytics distinguishes between “identity-levelâ€? breaches, where names and Social Security numbers were stolen and “account-levelâ€? breaches, where only account numbers—sometimes associated with names—were stolen. ID Analytics also discovered that the degree of risk varies based on the nature of the data breach, for example, whether the breach was the result of a deliberate hacking into a database or a seemingly unintentional loss of data, such as tapes or disks being lost in transit. […] Another key finding indicates that in certain targeted data breaches, notices may have a deterrent effect. In one large-scale identity-level breach, thieves slowed their use of the data to commit identity theft after public notification. The research also showed how the criminals who stole the data in the breaches used identity data manipulation, or “tumbling” to avoid detection and to prolong the scam.

Precisely. That makes perfect sense to me, as everyone wants a spectrum of risk to review not a black-or-white approach. And yet we should not forget that a vast majority of companies that house ID information still look at breaches from a “cup is half full” perspective and might not be in a suitable (expert) position to make intelligent decisions about the risks. Look at CardSystems, for example. The question is not whether people are trying to classify more levels of risk, but what is the quality of the data and their analysis — how qualified are today’s executives to make information security risk decisions on behalf of hundreds of thousands of consumers (assuming larger breaches will now automatically be determined to be lower risk)? Moreover, if you publicize reports that says huge breaches are lower risk and therefore exempt from breach disclosure, it seems obvious that clever criminals will simply shift to using huge breaches, no? That makes economic sense as well since a huge breach can be diversified to many criminals who will be able to commit ID theft more efficiently. They do not call it a black market for nothing, eh?

A SanDiego newspaper article from 2003 mentions that ID Analytics is a startup with “Citibank, Dell Financial Services, Sprint, Diners Club, Discover Financial Services and First North American National Bank as clients”. The company definitely seems to be a reputable source on the subject (“30 employees, including seven Ph.D.s”) with a timely business strategy that, according to the newspaper, almost could be an extension of the Payment Card Industry itself:

“I’m convinced they have a good product,” said Beth Givens of the Privacy Rights Clearinghouse in San Diego. “It’s just a shame that the financial industry didn’t take steps to fix these problems themselves.”

Givens contends that “credit issuers should have been spending more time on each credit application all along. And by more time, I mean, by two minutes.”

Hansen grimaced a bit when the gist of Givens’ criticism was recited.

He acknowledged that a financial institution might process as many as 10,000 credit applications an hour. But he says the industry also contends with a variety of challenges, including regulatory requirements to process each credit application within 30 days.

“What we’re trying to do is bring a technology solution to bear within the context of an automatic processing environment,” Hansen said.

In other news, Visa is asking assessors to re-certify, due to recent changes to the PCI data security standards, to the tune of over tens of thousands of dollars in training fees. That is a hefty chunk of money, even for veterans of PCI security, and believe it or not the standards are expected to change again in January 2006. Contrast this money-maker with the fact that Visa is giving away compliance scans for free outside the US and that the amount of credit card fraud has dropped dramatically over the past ten years due mainly to additional card verification measures (from billions of dollars to the low hundreds of millions). Fascinating stuff and a very exciting time to be in information security.

Dec 12, 2005 Update: Written Testimony from ID Analytics that was submitted on Nov 9th, 2005 to the Subcommittee on Financial Institutions and Consumer Credit (Hearing on H.R. 3997, the “Financial Data Protection Act of 2005”) can be found here:

However, misuse rates could continue to increase drastically over time if the vibrant black market for “identitiesâ€? remains unimpeded. … By selling any amount of the remaining identities (those not able to be used because of the ‘feasible limit’), fraud rings could maximize the proceeds from their efforts and exact a far greater degree of harm to consumers, industry and government over time.

Yet another DRM malware alert

If this keeps up I may need a dedicated DRM category to keep up with the flow of malware released under the guise of protecting big-media profits at the expense of consumer rights. The Register reported today:

According to the EFF, the vulnerability centres on a file folder installed by the MediaMax software shipped on some Sony CDs, “that could allow malicious third parties who have localized, lower-privilege access to gain control over a consumer’s computer running the Windows operating system.â€? …other severe problems with MediaMax discs, including: undisclosed communications with servers Sony controls… undisclosed installation of over 18 MB of software regardless of whether the user agrees to the End User License Agreement; and failure to include an uninstaller with the CD.â€?

There is definitely a balance and a right way to do things in “digital copyright management” (DCM), which is what DRM should be called, but the fact that the EFF claims 30 other labels use this software means the big labels either do not realize the harm they are causing by distributing malware or they do not believe they are liable. A healthy market would find neither acceptable.

Diebold insider issues warnings

RawStory posted an “exclusive interview” yesterday. There are some harsh allegations that altogether appear to be a stern warning to stay away from Diebold systems until an independent and open validation is available:

Previous revelations from the whistleblower have included evidence that Diebold’s upper management and top government officials knew of backdoor software in Diebold’s central tabulator before the 2004 election, but ignored urgent warnings—such as a Homeland Security alert posted on the Internet.


The 2002 gubernatorial election in Georgia raised serious red flags, the source said.

“Shortly before the election, ten days to two weeks, we were told that the date in the machine was malfunctioning,”? the source recalled. “So we were told ‘Apply this patch in a big rush.’”? Later, the Diebold insider learned that the patches were never certified by the state of Georgia, as required by law.


Responding to public demand for paper trails, Diebold has devised a means of retrofitting its paperless TSX system with printers and paper rolls. But in Ohio’s November 2005 election, some machines produced blank paper.

The whistleblower is not surprised. “The software is again the culprit here. It’s not completely developed. I saw the exact same thing in Chicago during a demonstration held in Cook County for a committee of people who were looking at various election machines… They rejected it for other reasons.”?

Asked if Ohio officials were made aware of that failure prior to the recent election, the source said, “No way. Anything goes wrong inside Diebold, it’s hush-hush.”?

Most officials are not notified of failed demonstrations like the one in Cook County, the insider said, adding that most system tests, particularly those exhibited for sale are not conducted with a typical model.

California, which recently conducted a test of the system without public scrutiny that found only a three percent failure rate—far lower than earlier tests that found a 30 percent combined failure due to software crashes and printer jams.

Asked if the outcomes of the newest test should be trusted, the whistleblower, who does not know the protocols used in the California test, warned, “There’s a practice in testing where you get a pumped-up machine and pumped-up servers, and that’s what you allow them to test. Diebold does it and so do other manufacturers. It’s extremely common.”?


The Diebold insider noted that the initial GEMS system used to tabulate votes for the Diebold Opti-scan systems was designed by Jeffrey Dean, who was convicted in the early 1990s of computer-aided embezzlement. Dean was hired by Global Election Systems, which Diebold acquired in 2000. Global also had John Elder, a convicted cocaine trafficker, on its payroll.

Someone convicted of computer-aided embezzlement designed the system? Security clearance is mandatory for many government jobs related to handling sensitive information, one would think that election systems should be treated in a similar fashion. Diebold should be held to a strict burden of proof that their systems are safe, at this point, and not allowed to release any product for public consumption until all uncertainty has been thoroughly clarified.

Alternatively, perhaps Diebold management should ask their staff to use their own systems to vote on future direction for the company and swear that they will abide by the outcome. Live by the sword…

Gates wrong about spam

Apparently as many as 80% of people surveyed did not trust Gates in 2004 when he announced that spam would be gone by 2006. An article in today’s ZDNet suggests that within 30 days that number might jump to as high as 100%:

Bill Gates’ prediction of January 2004 that spam would be “a thing of the past” within two years has virtually no chance of coming true, according to security company Sophos this week.

Beware those who say “security will happen by x date”. True security is far more complex and subject to uncertainties than a short-term objective such as a functionality enhancement. Moreover, there are usually so many influential factors that it is better to say “security will have x control in place by y date” and predict a resultant soft “decline” rather than any “absolute” or “total” eradication.

ZDNet put it slightly differently when they covered the original announcement.

John Cheney, chief executive of email security firm BlackSpider Technologies, which conducted the survey, said the results show that the industry doesn’t perceive Microsoft as a security authority, despite its chairman’s enthusiasm for the task

To his credit, at least Gates did not land on the roof of Symantec in 2004 for a photo-op in front of a “Mission Accomplished” banner.

the poetry of information security