As some of you may have noticed, Google has received some heat the last couple of weeks due to claims that they intercepted private data from open wifi-networks when driving around to complete Google Street View coverage. One of the many articles on this subject can be found here: http://www.computerworld.com/s/article/9176810/Google_stops_sniffing_Wi_Fi_data_after_privacy_gaffe

First off: I am very much against any form of privacy infringement and believe quite strongly that most forms of proactive surveillance against non-criminals are futile at best and damaging for national security at worst. However this whole case is just somewhat ridiculous.

Yes, Google made a mistake in not disabling that specific piece of software, but calling the data they gathered private is a bit of a joke. What they gathered was data sent unencrypted over a public network. If you’re sending confidential information over a public network unencrypted, Google stealing your deep-dark secrets is the least of your worries. They did it by mistake – many others do it intentionally!

In fact where I’m sitting right now, I can see no less than 7 open wifi-networks. Most are private homes and most of them have, according to Kismet, traffic flowing over them right now. This means that if I wanted to, I could activate software like Kismet or Wireshark and use this to steal every single bit of unencrypted data sent over this network. In fact, I would be able to do this with almost no chance of ever being detected in doing so. Even if the network owners tried to catch me, they most likely would not be able to. That’s simply how easy and risk-free it is.

The reason why I can do this, is because wifi-networks work by transmitting data outward on a given frequency and then let all clients in that network receive all data. It’s then the client’s computer that needs to filter out what was meant for it and what was meant for everyone else. If a computer behaves “nicely” it’ll discard anything not meant for it, but if it’s been put up to intentionally receive everything, you’ve created a so called “sniffer” and all unencrypted data is up for graps.

While software like Wireshark allows you to only “sniff” data sent over the network you’re connected to, Kismet let’s you “sniff” from any network without ever connecting to that network. This effectively makes you completely invisible to the network owners, so they have no way of knowing, that you’re stealing everything they send.

Sadly, most users are completely oblivious to these facts and use open networks as if they we’re their home networks. And sadly in some cases they even are (as was the case with most of the 7 networks here). So effectively, when Google was driving around gathering private data from open wifi-networks, they weren’t really “sniffing” because they had no intention of gathering that data. The users on those networks were however shouting every single bit of so called “private” information in all directions, forcing Google wifi-analysis software to capture and save it.

Now, to be fair: Google weren’t really being smart here and should not have captured data sent over unencrypted networks. It was a bad move and while they didn’t intend to do so, it probably still didn’t give them a boost in their reputation!

That being said, I must however still say, that the real problem here is the user and the open networks. If you don’t want your data to be scooped up by Google, don’t send it unencrypted over an open network. Chances are someone far worse than Google is listening in – especially if it’s a public network near train stations or the like. Sending data over a open wifi-network is, for all intents and purposes, the equivalent of shouting the same information out your office window.

Back in April 2010 we published a blog post describing the secure way of working from open wifi-networks – We recommend you read up on that and use the techniques mentioned there in order to keep private data private in the future.

I recently came across this article which talks about an ongoing trial concerning a “hack” in 2008.

Now this “hack” is made especially interesting by the fact that the victim was the, at the time, 2008 Republican vice presidential candidate Sarah Palin. Now you may have noticed I write “hack” instead of hack or security break-in, and the reason for this is actually quite simple: It really wasn’t a hacker attack!

What this kid (David C. Kernell) did, was that he simply used the “Forgot your password?” feature on Yahoo Mail to guess his way in to change Sarah Palins password and gain access to her mails. In the end, he actually gained access by using publicly available information and subsequently reacted by bragging on a discussion board while posting pictures of Sarah Palins e-mails. That’s not hacking! What he did was, at best, correctly estimating Sarah Palin’s knowledge of proper password policy.

Now what would an appropriate reaction be to such an incident? He’s clearly a reckless idiot, so some action should probably be made. Yet, at the same time, he showed Sarah Palin knows very little about basic security, thereby making a quite powerful (and perhaps needed) political point. If she can’t even secure her e-mail against amateurs, how is she going to secure the nation against ruthless psychopaths?

But basically, no matter what the appropriate reaction was, the prosecutors and Sarah Palin went with, for all intents and purposes, ending this person’s life! They went to court and tried to get him convicted for crimes with a combined punishment of up to 21 years and 250000$ in fines. All for guessing a password!

Until now he’s been convicted of felony destruction of records to hamper a federal investigation and of a misdemeanor charge that he unlawfully accessed a protected computer. He was however acquitted of a federal wire fraud charge. What level of punishment he’ll end up getting is hard to guess at, at the moment. But almost no matter what kind of punishment he gets, there’s a high likelihood it’ll be grave overkill.

Let’s face it: This 22 year old man is a moron – he did something enormously stupid. Not so much because he showed a grave lack of security understanding from the possible future vice president, but because he didn’t inform her, or her people, and didn’t give them sufficient time to correct the issue before sending it to the proper media channels. All in all he should have been a lot smarter! But that being said: What harm did this person really do?

Now, I’m all for punishing criminals and putting dangerous people behind bars. However a person like this isn’t really dangerous, he’s just not well-mannered. Had there been a proper, legal and well-documented process for reporting security issues in systems or procedures, then he would most likely have used these to get what he wanted: To show Sarah Palin knew little about security! Bare in mind, almost all IT-security professionals have learned primarily by doing – as in, they’ve tested their methods in more or less moral ways. Personally, I’ve always tried to keep to the moral part, however many others have been somewhat morally challenged – yet are now enormously talented and hard working. In fact, some of the best security professionals I’ve met are former “criminal” hackers to some degree.

So what’s my point with all this? Well basically: A young man/woman who “hacks” into a system and flaunts about it, is a person who lacks a place to be. We have full-fledged university degrees for biochemists, computer scientists, lawyers, politicians, engineers etc., yet we don’t have one for a hacker? We even educate police and military in the tactics of their enemies and how the enemy operates – even to the degree that certain soldiers have to act like the enemy in training in order to simulate combat. Yet we still don’t have any equivalent program to educate military hackers or security experts, even though we know for a fact that hacking has been used to attack a country’s infrastructure.

My five cents here is that David C. Kernell shouldn’t be prosecuted to the fullest extent of the law. He should get a slap on the wrist for handling it stupidly and for publishing/reading the content of Sarah Palin’s e-mails, and should then be thanked for showing the problem and put into a training program for IT and Security somewhere in the US. Even though this “hack” was enormously simple, he might still have some talent that could be used for so much good, instead of just throwing a 22 year old kid in jail and wasting his life.

We’re in a very problematic place in our society if showing the government aren’t doing some task well enough, results in oneself being imprisoned for the majority of one’s adult life.

Such a society is quite surely insecure!

As we have stated several times before (New OWASP guide: Secure Application Development on Facebook and Ruby on Rails Security Guide) OWASP, The Open Web Application Security Project, is a great organization tasked with providing comprehensive security knowledge for companies, individuals, organizations and developers.

This week they came out with a new finished OWASP Project: The Top 10 Security Threats of 2010.

The project website is located here and the full 22 page report can be found here: OWASP Top 10 for 2010 (pdf)

Basically what this is, is a break down of the most severe security issues in web applications for the year 2010. What’s especially scary about it is however, that these 10 security issues have stayed largely unchanged since the Top 10 of 2007. In fact only two issues have been replaced on the list, making the OWASP top 10 security threats of 2010 (the new ones are bold):

1. Injection
2. Cross-Site Scripting (XSS)
3. Broken Authentication and Session Management
4. Insecure Direct Object References
5. Cross-Site Request Forgery (CSRF)
6. Security Misconfiguration
7. Insecure Cryptographic Storage
8. Failure to Restrict URL Access
9. Insufficient Transport Layer Protection
10. Unvalidated Redirects and Forwards

What this shows us is that despite the efforts of OWASP, Aconiac and similar organizations, the security field has stayed largely unchanged and developers are still making the same mistakes in their designs and code. It might very well not be entirely possible to change this fact in general, even given 10 years from now.

But while companies in general may be making these mistakes, you don’t have to! The OWASP report includes several pages describing the security issues in detail, including an analysis of the risk it imposes on your business and what impact a breach might result in. We encourage you to download and read the entire 22 page PDF and make it mandatory reading for every developer and designer in your organization.

Have you ever been on the road towards a meeting or a vacation, and then just suddenly stumbled upon an open network while waiting for a plane or drinking a cup of coffee? Most people probably have..

And have you even been a bit too tempted and logged onto this open network? Again, most probably have.

Now, have you then started working while on this network and directly sent corporate information or handled information on your corporate systems? Sadly, many have and if you’re one of them: Read on! Using open networks directly for sensitive data (like corporate data) is a big security no-no!

So why would this be a problem? Isn’t it just free internet for the masses? Well, yes and no. Yes it’s probably a network you are completely free to use. It might even be a network owned by the office building, hotel, airport or which ever company you’re at. But due to the way wifi-networks are designed, everything you send over this network is completely public. Every person, on the network or simply in the vicinity, can easily set up a simple network scanner like Wireshark or Kismet and directly save all the information you send over this network, including all e-mails, websites you visit, data you send to websites, data you receive – plain and simply everything! And you have no way of detecting this! None what so ever! There is absolutely no way to check for eavesdroppers on an open unencrypted network.

To add insult to injury, eavesdropping on a network is extremely easy to do and there are several easy to use tools out there that hordes of 15 year old script kiddies love to use to steal as much information as they possibly can – for no other reason than: They can!

So are we advocating not using public open wifi-networks? No, not at all – you just need to use them correctly!

You can look at it like this: A public open wifi-network gives you a gateway on which you can build a connection to your workplace and work from there. How do you do this? Well basically there are several solutions here:

• Make the network encrypted.
  Well normally you won’t have the option of doing this, but in most cases it is simply better to keep smaller networks encrypted and then only use encrypted networks. Preferably using WPA2-PSK or WPA-Enterprise as encryption schemes. This is however most likely not a possible solution!
• Use a VPN connection
  A VPN (Virtual Private Network) is a technology with which you can remotely connect to your organization’s network in a completely encrypted manner. It is by far the most transparently secure solution available and is generally the one we would suggest to companies wanting their employees to be mobile always.
  There are several VPN solutions available out there, including big corporate solutions from companies like Cisco and open source solutions like OpenVPN.
• Access resources with SSL/TLS
  While VPN applies to all network traffic sent from your computer, there is also the other option of encrypting critical parts of your work like e-mail, FTP access, critical websites etc. There are protocols to support this for almost all the different kinds of traffic including: POP3S and IMAPS for email, SFTP for FTP and HTTPS for websites.
  Using this solution may in many ways be simpler, but it assumes you know beforehand every place from which you will be needing critical information. It also puts a considerable extra security concern onto the individual employee, since this person now has to deduce whether or not the given communication he/she is doing at the moment is secure or not. Using VPN, these concerns go away in most cases.
• Remote desktop solutions
  Another option, that’s somewhat similar to the VPN option, is to have the employee make a secure connection to a server at the workplace and from there open up a terminal service running another computer remotely. Solutions like this are available in many forms like VNC, RDP and proprietary solutions from companies like Citrix. This gives the employee a remote view of his/her workstation desktop even though he/she is no way near the actual office and, most importantly, it makes it possible for him/her to work securely from any network.

So you can look at it like this: If you’re not doing any of the above, you have a problem and should take it up with your company in order to get a security policy on the matter and making it safe for the company to work from anywhere! Mobility is one of the top priorities in business these days, and you really want to use the opportunities laid before you well, without screwing yourself because of bad security.

So remember: Public open networks aren’t bad, but you need to keep your assets safe while using them!

As is sadly often the case, well-meaning newcomers to programming take on the newest and/or most popular programming language/framework available. And as you might suspect, they usually get it wrong and make every design and security mistake possible along the way!

This is a trend that we’ve seen with pretty much every popular programming language out there – like for example PHP, which sadly still holds the record for most insecure websites written in the language.

One very popular web framework these days is Ruby on Rails, created by the very talented 29 year old Danish guy David Heinemeier Hansson. It uses a Model-View-Controller construction and emphasizes good design by such principles as DRY (Don’t Repeat Yourself). By all means it’s a very good web framework! But as with PHP, a lot of newcomers get it wrong. Either by not following the Ruby on Rails conventions or by ignoring security!

Now, we’re not here to teach you about design. If you want to learn more about proper software design, a good place to start is simply your local library. Look for books on topics like Design Patterns, Software Engineering, Extreme Programming, Test Driven Development and Agile Software Development.

However what we do want to teach you about, is proper security in Ruby on Rails! Luckily, we don’t have to take out extreme amounts of time from the work we need to do, to get you trained in RoR security – instead we can simply refer to the work already done by the OWASP organization. OWASP is an organization working to improve web application security in the entire world, by means of a whole bunch of different projects for developers, security professionals and end users. One of these projects is the Ruby on Rails Security Guide V2 project which includes a PDF file detailing the different security concerns and solutions concerning Ruby on Rails development.

If you are going to develop Ruby on Rails applications (or if you’re simply curious) please download the Ruby on Rails Security Guide from OWASP and read it before doing any production deployment of applications.

Note: If you’re too busy to go to the project page on OWASP and find the download link, then here’s a direct download link instead: Download the Security Guide

NOTE: This news item was originally posted on January 4, 2009

Since many of our clients have turned out to be fully capable of correcting their security issues themselves or really just wanted to get their own security corrections checked, we have now launched a service that can fulfil this need.

This service is Vulnerability Testing and is basically like Security Testing, however without Aconiac correcting the security issues and without Aconiac needing to have access to specifications about the system. A vulnerability test is therefore a real simulated attack from a hacker, so that companies can find whatever security issues a hacker would have found.

The service is sold at a fixed price of 540€ excl. VAT, with the sole exception that if your system is abnormally large or complex, Aconiac may deny to do the service at the fixed price and will instead suggest other solutions, like e.g. a real security test.

Further description of the service can be found here.