Original URL: http://www.theregister.co.uk/2007/04/23/online_impersonations_validation/

Online impersonations: no validation required

How do you know what's real and what's not?

By Neal Krawetz

Posted in Security, 23rd April 2007 10:40 GMT

Back when I lived in the Silicon Valley, there was an ongoing employment scam. Prospective employees would show up with perfect resumes and immediately get hired. It would not take long before it was clear that these people did not have the experience stated on their resumes. Within six months they would be fired.

However, now they had six months of legitimate experience with real companies that they could reference. Their next jobs might not be as good or glamorous, but it would be much better than if they started with their real resumes.

In order to combat false credentials and gain insight into potential candidates, hiring managers have turned to web search engines and social networking sites to augment their screening process. According to a survey last October, hiring managers claimed that up to a third of candidates lied about their qualifications.

In addition, about one in ten candidates had posted inappropriate messages or images in public forums, making them less desirable than candidates who always appear professional. The problems with having a less than stellar online persona were covered in a recent NPR story.

However, while resume padding has been around for years, online impersonation has only begun playing a role. What if your online profile included an inappropriate statement that you never made? In most online forums, anonymity and impersonations are trivially accomplished since you are never asked for verifiable information during enrolment. This becomes a two-stage problem. First, you must be able to identify that an impersonation is taking place. And second, you need to know how to take corrective actions.

Bad impersonations

When people think of impersonations, they usually think of identity theft and financial crimes. However, the impacts from online impersonations can be significant.

For example, one of my associates had a free email account similar to "john.doe@gmail.com". Someone with a grudge registered "john_doe@gmail.com" (underscore instead of a period) and began to send out emails that impersonated the individual. The emails were intended to undermine his credibility. Unless you looked very closely at the email sender's address, you would not realise that it was an imposter. As another example, in 2006 a police officer impersonated his ex-girlfriend and used the account to solicit dates. Men actually appeared at her house expecting a romantic interlude.

While financial impersonations may take years to rectify, information and false postings on the web may circulate indefinitely. Distinguishing fact from fiction and true actions from an imposter's can be virtually impossible. The requirements for removing an imposter vary from simple web forms to nightmarish runarounds.

Hunting imposters

It's fun playing hide-and-seek with my four-year-old niece. She will shout from behind the curtains, "I'm over here!" Unfortunately, most online imposters will not tell you where they are hiding, or even when they are going to start. You have to find them and you must constantly be looking. If you maintain a low profile online - never posting to forums and not maintaining a public image - then you might only need to look around once every few months. However, if you are more widely known, then consider looking more often, such as weekly or even every few days.

The first step in the search process is to perform an ego-search. Start with search engines such as Google, Alta Vista, and Yahoo!. Perform searches for yourself. Look for your name, company, anything that you are well-known for, and anything you recently did publicly that may have been noteworthy. Use a variety of different search engines since different tools return different results (While Google may have the largest index, other search engines find things that Google misses.)

Strictly searching for web pages will only return web pages. However, impersonators may use a variety of forums. Consider tools such as Google Groups to search newsgroups and MARC to scan technical mailing lists.

Technorati is an excellent place for searching blogs - particularly since Google does not index MySpace pages.

There are two main things to look for during your ego-search. First, look for things posted by you. Make sure they are things that you actually posted. Second, look for things attributed to you. Even if you don't initially find an impersonator, you will probably find references to an impersonator.

Rather than doing the searches yourself, companies such as Reputation Defender and Naymz have been established to help manage your online profile. For a fee, these companies will identify potentially damaging online information about you - whether it was created by you or by an imposter (however, I have no direct experience with either of these services).

Hopefully you won't find an impersonator. However, if you do then the next step is to mitigate damage. The mitigation process depends on the type and degree of impersonation.

You have mail?

Configuring a mailer to send email as someone else is beyond trivial. When you configure your Microsoft Outlook mailer (or Gnome Evolution or Firefox Thunderbird or Eudora, etc.), you simply need to enter in a fake email address. While laws such as the US CAN-SPAM act makes email impersonation illegal, who is going to catch you? Fortunately, uni-directional emails can usually be identified as forgeries fairly quickly. Any kind of return correspondence will not reach the imposter.

Unfortunately, free mail and web service, such as Yahoo! Mail and Google's Gmail, make it trivial to create "similar enough" addresses that can fool many people. As a recipient, you should check the sender's email address to make sure it really is from the person you expected. Do not assume that the sender is using a new email address unless they told you about it ahead of time.

Refuting email

In sensitive environments, PGP signatures or similar key-based systems can be used to authenticate the sender. These cryptographic systems validate your identity because only the sender has the right keys. However, few people use PGP for general email communications and it does not authenticate you with new acquaintances.

With free email services, disabling an imposter becomes more difficult. In most cases, you must authenticate yourself before the hosting provider will take action, but not to open the account. For example, anyone can register a Yahoo! Mail account using your name. If Yahoo! Mail has an email account that is impersonating you, then you can fill out their abuse form. This form requires you to describe the incident and identify yourself.

In contrast to Yahoo!, Gmail requires you to print out the imposter's email and mail it to them through the postal service. Gmail offers no method for submitting an online complaint about an imposter.

Although sites such as Yahoo! and Gmail do provide options for refuting and disabling an imposter's account, other public mailing systems are not as responsive. For example, Hushmail allows people to report abuse. However, they will not investigate abuse complaints (other than spam complaints) and refuse to disable any imposter's account without a court order.

Since Hushmail does not authenticate during enrolment and does not remove abusive accounts, you should personally verify the sender's identity before responding. Any email from "hushmail.com", "hush.com", "hush.ai", etc., could be an imposter. While Hushmail does use PGP for email authentication, this only validates that the email was sent using Hushmail; it does not validate the person who opened the Hushmail account.

External authentication is one option to mitigate the impact from general email impersonations. Build a reputation around a known email address and maintain a website that can quickly be used to identify you. If your emails always come from "mydomain.org", emails from some other domain should be suspect.

Mitigating mailing lists

While an imposter's email account can be disabled, the damage from posted messages is usually permanent. While some mailing lists will remove messages from their archives, most will not. And even if the message is removed, many mailing lists are mirrored; an imposter's posting could live indefinitely on hundreds of archive sites. Exposing the imposter by posting to the list or informing the list manager, is usually enough to mitigate any damage.

Unfortunately, some unmoderated forums have a large number of imposters. Contesting an imposter may not be worthwhile; if many people are impersonated, then malicious postings are usually suspect without any feedback from you. In addition, any identification of an imposter could be overlooked in high-volume forums.

Web of lies

Along with impersonating email addresses, imposters can create fictitious websites. Yahoo! Mail includes web space at Geocities, Gmail includes Googlepages, and there are thousands of other hosting providers. The inability to authenticate an owner opens the door to impersonations such as phishing; anyone can create a web page that looks like Citibank and anyone can register a domain name similar to "citibank.com". No validation required.

Refuting a fake website really depends on the type of impersonation and the hosting location. Companies such as Yahoo! and Google are extremely responsive to phishing reports. Due to well-organised efforts by groups such as the Anti-Phishing Working Group, reported phishing sites are usually taken down within hours, and sometimes within minutes. If the hosting site is non-responsive, then network routers can restrict access to these sites until the phishing site is removed.

While web impersonations usually refer to companies, they can also apply to individual people. Unfortunately, if the site is impersonating a person - and not a company - then refuting the site becomes more complicated. Yahoo! Geocities links web pages to email accounts. If you can refute the Yahoo! email account, then you can disable the imposter's web page. In general, your ability to refute a fake web page really depends on the hosting provider. Look at their site for a method to report abuse and be prepared to validate that you really are you.

Invading my space

While email allows you to communicate and web pages provide a narrow view of your world, blog services such as MySpace allow you to create an entire online persona. Visitors can see up-to-date content as well as friendship relationships. Unfortunately, as with email, anyone can create a MySpace profile and impersonate anyone else. Politicians, sports figures, and celebrities are frequently impersonated. Even law enforcement and kidnap victims are not immune. Teachers are another common target for MySpace impersonations.

Refuting a MySpace profile

Disabling a MySpace imposter is much more difficult than other online services. While Yahoo!, Google, and even Hushmail have well-defined, single-step processes for reporting abuse, MySpace seems intentionally complicated. I have assisted a half-dozen people with the removal of false MySpace profiles, and none have been painless.

MySpace provides a specific FAQ and request form for reporting abuse and impersonation. They want you to (and I'm not kidding) send them a "salute". The salute is "an image of yourself holding a handwritten sign with the word 'MySpace.com' and your Friend ID". If this sounds familiar, it is because it is the same type of embarrassment used at 419eater.com against advanced fee fraud scammers. There are a lot of problems with this type of authentication:

If you are the victim of a MySpace impersonation, then you have two options for reporting abuse:

In all cases, you will receive no acknowledgment that the complaint has been received, and no case number for tracking the complaint.

Addressing other blogs

Most bloggers are friendly and want to provide accurate information. If you find that an imposter is posting on someone else's blog, then consider contacting the blog owner. A simple, polite email explaining the situation is usually enough to have an impersonator's comments corrected or removed.

Can you picture it?

Online impersonation does not always need a specific profile or account. For example, YouTube allows anyone to upload virtually anything. All you need is a YouTube account, and that only requires a username, password, and email address - none of which are validated.

With YouTube, it is easy to upload videos - whether you have permission or not. And while you can remove a video that you uploaded, it can be difficult to remove a video that someone else uploaded. According to YouTube, you need to file a DCMA "counter-notice". However, this is only for the removal of materials where you claim copyright. What about harmful or malicious videos? For example, what if someone has a video of you in the bathroom that was taken with a hidden camera? What if someone says that the video is you when it really isn't? Well, these are not covered by the DMCA and YouTube has no documented procedure for this type of removal request.

However, even if you have a valid DMCA counter-notice claim for YouTube, you might not see any response. Recently Viacom filed suit against YouTube for $1bn, claiming that there are almost 160,000 unauthorised videos that have been viewed more than 1.5 billion times.

These are not the only problems with YouTube's repudiation system. Recently, a 15-year-old impersonated Australia's ABC Television and sent a DMCA counter-notice to YouTube. YouTube responded by sending infringement notices to users and many video clips were removed. (This begs the question, why couldn't Viacom get this kind of response?)

Virtual credibility

While email, web, and MySpace require imposters to register accounts, impersonations can happen in any forum where the identity is not authenticated and validated. For example, Wikipedia has long struggled with impersonators. In some cases, vandals have entered fake information in Wikipedia. For example, the comedian Sinbad is not dead, and John Seigenthaler was not behind the Kennedy assassinations.

Following the Seigenthaler incident, Wikipedia changed their system to require registration prior to creating pages or uploading images. Creating an account simply requires a user-supplied name and password. There is no authentication and no validation. You can create an account with a one-time username and password, and then immediately upload files, create pages, or edit existing ones to your heart's content.

Currently, Wikipedia is considering methods to validate contributors' credentials. This comes on the heels of another Wikipedia scandal, where a user falsely claimed to be an expert in sociology.

Fortunately, misinformation on Wikipedia can be painlessly resolved. Ironically, the one thing you should not do on Wikipedia is correct the information yourself (or have a friend make the corrections). Instead, you should follow Wikipedia's very detailed process for handling disputes and abuse. In the worst case, pages can be quickly corrected and locked from future editing.

Conclusion

The real problem with online identities is not that anyone can impersonate you. The problem is that the impersonations are enabled by public technologies and services. Gmail, Yahoo!, and MySpace all have methods to refute an identity and authenticate yourself after an account is created. With each service, you must validate yourself in order to cancel an imposter's account or fake profile. However, these services do not use any form of validation for opening the account in the first place.

In addition, existing authentication methods for refuting an identity lack any true security basis, and newer technologies, such as Jott and Twitter, are similarly vulnerable to impersonation.

It is said that imitation is the sincerest form of flattery. Unfortunately, online social networks provide no method for distinguishing an impersonation from the real thing. While your online words and actions may circulate for years, so do those of an impersonator.

This article originally appeared in Security Focus.

Copyright © 2007, SecurityFocus