Showing posts with label pci. Show all posts
Showing posts with label pci. Show all posts

Monday, July 20, 2009

If you only read one article on Information Security...

[... this is it]

Actually, this is a bit unfair because after reading this one article, you'll be compelled to read more.

Richard Bejtlich's article sums this up nicely. He links to another blog post by Verizon Business.

I have some issues with Verizon Business's annual report but it is probably the most important document on Information Security to be published.

My one criticism of the Verizon Business Breach Report is that it shows credit card data to be more at risk than anything else. I was never sure if this is because it is easier to abuse than other data (such as Intellectual Property) or is just easier to detect when it is abused. According to the article, it is the latter. IP is leaving our companies, we just don't know it.

When a whole bunch of credit card information is stolen then the banks track which credit cards are abused. They are good at this and they slowly work out where all the credit cards were used together. So, if 5 credit cards were all used at a specific shop and then end up being abused that points to that shop having had an information breach. In the case of IP, there is no bank tracking abuse so you have to track it yourself... and companies are really bad at that.

The other point which I found quite amazing is that very few times when a PC is lost, is it used for fraud. End point encryption is cheap and easy to apply so it should be done, but most information is lost, not through assets being lost but through network attacks.

Thursday, March 20, 2008

Information Security, Governance, Compliance and Safety Belts

The state of Victoria in Australia made wearing safety belts compulsory in 1970. This is now almost universal practice.

I don't know the exact statistics but a study done in South Africa found that more people used safety belts after it was made illegal to not use them than when it was left up to the driver.

The conclusion really is that people are more likely to obey a rule because it is law than because it may just save their life.

I think that the same is true with Information Security. It won't (necessarily) save your life but it is good practice. And yet companies are only doing it because it is now law.

The problem with this is that it is not accepted by people in their hearts. I know of people who drive around without their belts on and put them half on when they see a traffic cop.

The Information Security equivalent is jacking up your InfoSec program when the auditors come to visit and letting it slide when they are not around. Or making sure that they don't see some issues that you are well aware of.

I have seen a lot of complaints about PCI and SOX etc etc in the same way that people complain about "self protection" laws like safety belt laws. The thing is that the government is stepping in only because people are very bad at self regulation. Really, what a number of InfoSec experts are trying to promote is - understand why you need to protect yourself, understand how and abide by it. Do it for your company, not because the government demands it.

That way, not only will you be "compliant" and full of "good governance" but more importantly - your company will be safe.

Friday, October 26, 2007

TJX - Who suffers?

Just a quick break from the 7 habits. They take awhile to think out and I need to post something..

All the signs are pointing that TJX has suffered a text book case hack attempt and so all the Security Chicken Littles were salivating because this would be the "I told you so" opportunity of a lifetime.

And it didn't happen. I blogged about it here and here.

So, what happened? My personal feeling is that this was just the first punch in the fight. Consumers have taken the knock and have felt a bit upset by it but they can deal with it.

In the back of their minds though they have decreased the amount that they like both TJX and credit cards and maybe their bank ever so slightly depending on how much this breach has impacted them.

TJX is lucky in that if their service levels are up to scratch and if they have no more major breaches then over time their image will be improved and their customers will be happy once more.

For the credit card companies it will be a bit harder. If someone now suffers a breach at another store it won't impact TJX but the consumer may feel a bit less trusting of the whole credit card process.

This is problematic in the same way my swimming pool theory is bad for networks. Every store only suffers a bit of the problem but the whole credit card process suffers the most. Perhaps this is why the PCI members (Visa, Mastercard, etc) are working hard to get the stores to implement the PCI DSS security standard. They may find consumers start to give up using credit cards as much or at all ever.

Maybe the answer is actually for the whole process to be scrapped and redone.

Monday, August 27, 2007

The TJX stock is just not co-operating!

According to wikipedia :
The TJX Companies, Incorporated is the largest international apparel and home fashions off-price department store chain, based in Framingham, Massachusetts in the United States.

[...]

On January 17, 2007, TJX announced that it was the victim of an unauthorized computer systems intrusion.

TJX ended up as the default PCI black sheep. PCI, for those not in the know, is an industry standard created by the credit card companies telling stores how to protect their customer's information, specifically credit card information.

Basically TJX did everything wrong including storing information they should never have stored in the first place. 45 million credit card numbers are now being traded on the black market because of this breach.

Net income for their 2nd quarter dropped 57% due to information security costs related to the breach.

Bad news for the company, right? Wrong. Maybe someone can explain this to me but on January 16th, 2007 the share price was $29.94. Friday's closing price was $30.75. Man, did the market get them (not)! To be fair they have underperformed the S&P500 until recently but the company does not seem to be very hurt by the breach.

Monday, August 13, 2007

3rd Party Security - The big question

As happens in the "Blogworld" I read a blurb in the Daily Incite which then linked to a good Blog entry by Andy It Guy which in turn linked to a really good PDF document by Rebecca Herold who has more letters after her name than in her name.

While we are so busy concentrating on our own security structures (You are, aren't you?) how do we make sure that our partners are protecting our data?

There are several places where this is important

  1. The obvious first one: you give your credit card information to someone. What they can and can't do with it is governed by a standard made by the credit card companies. It is called PCI compliance. It seems most companies don't abide by the rules but the fact is that the rules have been very well designed and slowly, hopefully, companies will abide by them. The nice thing is that PCI complience is worked out already. You don't have to worry. You should as a matter of principal make sure that a company is PCI compliant. I think it would be a good idea for the credit card guys (visa, mastercard, etc) to actually promote PCI compliance as a marketing tool for companies to diplay proudly on their websites and in their stores.
  2. You fill in a form, any form, anywhere, online or offline. This is your personal, private information and you should be aware exactly what happens with it. If you have to give the information across for some law such as the ones preventing money laundering, you don't want that form going to the company's marketing department. ("You are a treasured customer of ours, do you want to be the first to use our new services...?") You also don't want it put into a dustbin and used by anyone who finds it in the street. '
  3. You are trusted with someone's details and have to send them to a 3rd party. If something happens to the details - its you to blame.
Basically, wherever someone has some of your personal data, your company's confidential data or data that has been given to you by some entity that trusts you with it, you should be able to make demands on how they treat it. No security is 100% but you should be able to at least, without getting into all the details, know enough about how your data is treated to make an informed decision on whether or not you trust the person you are giving it to.

The PCI standard came out of a need to protect data but there should be a broader standard for all types of data allowing us to make spot decisions on who to trust and who not to trust with our data.

And, taking an observation from Andy but broadening it: the specification of how data is looked after should be more specific than a framework. A framework is fine for protecting your own data, but other people should be able to judge exactly how you treat their data.

But, on the other hand, you don't exactly want to go around to every company that you deal with (perhaps all over the world) investigating in minute detail exactly what methods they use to protect their network and data. You can't be expected to watch that none of their staff take their laptops home etc.

You shouldn't even be expected to take a look at their policies.

You should just want to be able to see a logo that says "we are secure up to the level 3 of the "3rd party information control standard (3pics)". This should be good enough for a bank but a video shop may be able to get away with level 2 and a doctor should have level 4.

By the way, I made up 3pics because, as far as I can see, there is no widely accepted standard with clearly defined levels that the man in the street can trust and be used to (except PCI and that is for credit card information only). But shouldn't there be? Wouldn't it be nice to be able to trust that a company you are about to deal with is going to treat your information the same way you do?

Rebecca's PDF document (linked to above) goes into great detail about how one can manage personal information that is given to 3rd parties but it is a lot of work and is fine for companies who have few partners but when there are many partners it would be nice to be able to just check their "3pics" compliance level and start dealing with them.


In case you argue that it is possible already using ISO, SOX etc, then read what Andy said in his article about how they are just frameworks and not generally accepted standards.

What we need is someone (who me? I'm too busy ;) to create a(n auditable) standard with a few levels that are easy to understand and implement. And for companies to use the standard and brag about their level of security.

I think part of my thinking comes from discovering this week (but not being rich enough to follow through with actually buying and reading) a book by Stephen Covey (Jnr) about how once trust is established, business can proceed quickly. It is up to us as the public to demand that companies show how they can be trusted with our private information. It is up to us Information Security specialists to make it easy for them to do it.

Wednesday, July 4, 2007

PCI Auditors selling stuff?!


In a really good blog entry, Mike Rothman talks about how PCI assessors (auditors) are pitching products and other solutions once the audit is done. He goes on further to talk about separation of duty and how the client should make it clear from the beginning that there will be no further business to be made after the audit.

I agree with Mike but I don't think he took it far enough. In an earlier blog entry of mine I discuss this very issue. Once the auditor has been too visit, it is too late. Have a good strategy and see it through long before you call in auditors. Then once they have arrived and start to sell you products and solutions that you don't need - you'll know that you don't need them.

Never use auditors to tell you what should be done.. use your security experts... use auditors to do the checking.