Скачать 0.92 Mb.
T H E H O N E Y N E T P R O J E C T® | KYE paper - DRAFT
Know Your Enemy:
The Social Dynamics of Hacking
The Honeynet Project
Thomas J. Holt– The Spartan Devils Chapter of the Honeynet Project
Max Kilger – The Spartan Devils Chapter of the Honeynet Project
The Know Your Enemy paper series is a collection of professional papers designed to provide in-depth insight into current security threats, as well as technical discussions of the application and analysis of new research tools and technologies. The focus on technological challenges and solutions is critical to improve security, though it is also vital to recognize the human element behind any technical attack. Attackers have various motives, skills, and relationships, though these may often be ignored when focusing on security solutions to defeat their efforts. The human element must, however, be given greater consideration in order to improve our knowledge of all facets of attackers. In fact, individuals can engage in attacks from around the world and target any resource from critical infrastructure to financial institutions using varied techniques. Though a substantive body of research examines both the continuing evolution of attack techniques in the attacker community and technical solutions to mitigate these tactics, there is a relative lack of research on the social dynamics and human aspects behind these actions. Such information is vital, however, to improve our understanding of the nature of attackers around the world. In this paper, we will explore the various facets of the global hacker community, including the distribution of skill, social relationships between actors, and common motives of attackers. We will also explore the prospective future of attacks and attacker behavior in order to improve our general understanding of the hacker community.
The Composition of Skill in the Hacker Community
The development of the Internet and modem technology transformed the nature of hacking from local computer enthusiasts working together to share information into a worldwide network of skilled and unskilled actors with diverse interests and capabilities. While there are myriad definitions for hacker (Bachmann, 2010; Jordan & Taylor, 1998; Schell & Dodge, 2002; Taylor, 1999), the most comprehensive and accurate terms identify hackers as individuals with an interest in technology who use their knowledge to access computers and devices with or without authorization from the owner (Schell & Dodge, 2002). This definition recognizes that hacking involves an application of knowledge and may done legitimately with permission from a system manager, or maliciously without approval. It is unknown how many individuals are actually involved in hacking due to the secretive nature of this subculture. For example, Jordan and Taylor (1998) estimated that there are at least 100,000 hackers worldwide, though this figure has undoubtedly increased substantially over the last 15 years. Despite difficulties in quantifying the hacker population, there is substantive evidence that the hacker community is a strong meritocracy where individuals are judged on their skill and ability to manipulate technologies in ways never before seen or intended. Those who can devise unique tools and identify new vulnerabilities garner respect from their peers and develop a reputation for skill and ability within the subculture.
The distribution of skill within the hacker community is pyramidal in nature (see Figure 1). At the top resides a very small number of skilled actors who have substantive abilities to identify new vulnerabilities, create exploits, and implement new programs that can be used for various attacks. These individuals are what Holt and Kilger (2008) refer to as “makecraft” hackers because of their capacity to make whole-cloth tools that otherwise did not exist. Other researchers may call these individuals elite hackers, or black/white/grey hat hackers (Furnell, 2002; Jordan and Taylor, 1998) depending on their ethical outlook. Attackers that possess these high level skills pose the greatest threat because they have both the creativity and knowledge base to identify unknown exploits and unusual attack vectors that may have been otherwise ignored.
Below these high skill actors resides a larger population of semi-skilled actors who can recognize and use various tools and exploits, though they often do not have the technical proficiency or interest to generate these tools on their own. As a result, they may utilize tools and tactics from the makecrafters in order to engage in various attacks. These individuals may also be called “techcrafters” as they can implement and adapt existing tools to suit their needs, but generally do not create these tools on their own (Holt & Kilger, 2008). This population of hackers have become extremely important in the hacker community over the last decade with the emergence of malware and stolen data markets (Chu, Holt, & Ahn, 2010; Franklin, Paxon, Perrig, & Savage, 2007). The middle tier of skilled hackers can often buy tools from extremely proficient actors, and apply these resources in the course of attacks. In turn, the information or data they acquire can be resold to others for a profit.
Innovators and Creative Developers Game Changer
Applied Skills Though Not Thought Leaders
Feed Off Top Tiers
Figure 1: The Distribution of Skill in the Hacker Community
Finally, the bottom of the pyramid is populated by low or unskilled hackers who may otherwise be referred to as script kiddies or noobs (Furnell, 2002; Jordan & Taylor, 1998). These actors have little understanding of the mechanics of an attack or compromise, and depend entirely upon the ingenuity of other hackers in order to engage in attacks. They recognize and value engaging in different compromises, but do not have a functional appreciation for the ways that an exploit actually impacts system processes. As a consequence, their attacks are often of little consequence and pose at best a nuisance to computer security personnel and administrators.
At the same time, script kiddies can serve as an attack point for more skilled hackers. For instance, many script kiddies will download copies of existing malware in the hopes of using it for their own agendas. These kits may, however, be joined with rootkit malware to infect that script kiddie’s computer by a more skilled actor (see Chu et al., 2010). In turn, these infected computers can be used as an effective launch site, or monitored in order to mask the activities of skilled hackers. In addition, these infected computers may act as an early warning system for more skilled hackers. When authorities investigate the script kiddie, this alerts the more skilled hacker that they may need to move on from this source and alter their malware coding in the process to reduce the likelihood of detection.
It is also important to note that the emergence of the on-line black market have enabled low skill attackers to pay for hacking services managed by moderate and highly-skilled actors. For instance, a mid-tier hacker operating a large botnet can lease out their infrastructure to others for DDoS attacks, spam distribution, or proxy servers to low or unskilled actors. This means the script kiddie can now more effectively engage in attacks while at the same time create a lucrative business model for the more proficient actors in the community (see Chu et al., 2010; Holt & Lampke, 2010). This evolution in information sharing and resource generation has had a profound impact on the hacker community, and on the nature of computer security and investigation as a whole.
The Global Distribution of Skill
The distribution of skill appears to be somewhat consistent in industrialized nations including China, Russia, the United States, and other early adopters of the Internet. It is not clear, however, if these populations are truly equal across place. For instance, there are certain attacks that can be regularly associated with actors from a given nation, such as the use of web defacements by hackers in Turkey, Pakistan, and India (Holt, 2009b). Similarly, there is substantive evidence that a number of actors in Brazil, China, Russia, and Romania are involved in the creation and distribution of malware and sophisticated zero day exploits targeting financial institutions and government agencies. As a consequence, there is a need for substantive research to assess the size and skill of hacker populations across place. A small number of studies have tried to accomplish such a task with extremely limited success such as Chiesa, Ducci, and Ciapi (2008) recently attempted to survey a wide population of hackers across the globe. They received 216 completed surveys from over 20 countries, though the participants were not evenly distributed across geographies and had to voluntarily participate. As a result, this large sample is not likely representative of the larger hacker community, especially active criminal hackers. Thus, there is a need for expansive study in this area to identify the distribution of ability within and across different geographies.
It is also not clear how hacker populations in nations that have just come on-line in the last decade, such as northern and central Africa and parts of Latin America, may engage in different forms of attacks. Since the Internet enables geographically dispersed groups to share information and attack tools, it is plausible that it will take less time for emerging hacker communities to expand their overall efficacy. Kilger (2007) has suggested that there is a potential emerging serious threat as a consequence of some of these developing countries coming online. Countries where there are a large proportion of individuals living in poor economic conditions with little hope of mobility and significant amounts of time on their hands and a digital window into a global economy with electronic access to vast amounts of capital would seem to be a fertile breeding ground for cybercriminal activity. This is evident in the emergence of 419 scammers operating out of Nigeria and other parts of Africa where economic opportunities are scarce and success can be made through persistent operations against victim populations (Adineran, 2007; Holt & Graves, 2007; King & Thomas, 2010; Warner, 2011).
In fact, Holt (2008) tracked the creation of various pieces of free malware moved across multiple nations with attribution to various creators in a relatively short period of time. One stand-alone DDoS attack tool called Try2DDOS which was created by a French hacker appeared as a download in diverse hacker sites across the globe, including Argentina, China, Ecuador, Guatemala, and Russia (see Figure 2). This demonstrates that hacker groups look at websites across disparate geographic places in order to find new tools, and will share resources with others when it appears to be effective. In turn, this exchange of exploits may enhance the skill-sets of hacker communities over time as they study the exploit examples they have acquired and may facilitate and accelerate the evolution of more highly skilled hacking groups at a faster pace than seen in previous decades.
Figure 2: Examining the Distribution of Malware Around the World
Social Relationships in the Hacker Subculture
Hackers are not born, but rather emerge slowly from the confluence of native technical aptitude, access to technology, and prolonged virtual and face-to-face socialization with others who share similar norms and values. Often researchers focus upon virtual socialization experiences that help instill the norms and values of this subculture, but some of the most important relationships develop in face-to-face settings (Holt, 2007). The different environments in which these social processes occur play an important role in the early formative years of persons who eventually identify themselves as hackers. In this section, some of the social processes and forces involved in shaping both hacking and the hacker community are discussed.
The importance of technology for hackers often emerges early in youth. Many who become involved in the hacker community report developing an interest in technology at an early age. Hackers report gaining access to computers in their early teens or even younger for hackers in the late 1990s to the present (Bachmann, 2010; Holt, 2007). Simply utilizing computers in public cafes and schools can also help pique a hackers' interest in technology (Holt, 2009b). In fact, in nations where home Internet access is expensive, like Turkey and Iran, hackers often report using computers in cafes and other public locales in order to connect with others. Identifying peers who share their affinity for technology on or off-line is also extremely valuable because it helps to maintain their interests.
Hackers maintain loose peer associations with individuals in on-line environments that may be useful in the development of their skill and ability (Holt, 2009, 2010; Holt & Kilger, 2008; Meyer, 1989; Schell & Dodge, 2002; Taylor, 1999). There are myriad communities operating via computer-mediated communications across the globe for hackers at every skill level to identify others who share their interests, including Internet Relay Chat (IRC), forums, blogs, social networking sites, and other on-line environments (Holt, 2007, 2009a, b, 2010). Hackers have operated in Bulletin Board Systems (BBS) since the late 70s and early 1980s to provide information, tools, and techniques on hacking (Meyer, 1989; Scott, 2005). The content was posted in plain text, and occasionally featured images and art made from ASCII text, in keeping with the limitations of technology at the time. These sites allowed asynchronous communications between users, in that they could post a message and respond to others. In addition, individuals hosted downloadable content including text files and tutorials, though some also hosted pirated software and material called warez (Meyers, 1989). The BBS became an important resource for new hackers since experienced technology users and budding hackers could share detailed information about systems they explored and discuss their exploits (Landreth, 1984).
The BBS allowed hackers to form groups with private networks and password protected boards to keep out the uninitiated and maintain privacy (Landreth, 1984; Meyer, 1989). Closed BBS were initially local in nature based on telephone area codes, but changed with time as more individuals obtained computers and sought out others on-line. Local hacker groups grew to prominence as a result of BBS based on their exploits and intrusions into sensitive computer systems, such as the Masters of Disaster and the Legion of Doom (Slatalla & Quittner, 1995). As a result, it is common for modern hackers to belong to multiple forums and websites in order to gain access to pivotal resources on-line (see Figure 3; Holt et al., 2009).
As on-line communications services evolve, so too do the communication practices of hacker groups. In fact, the global distribution of the hacker population has resulted in regional variations in CMCs based on local preferences. For instance, Russian hacker groups appear to utilize IRC, forums, and some social networking sites like Vkontakte or LiveJournal in order to connect with others (Holt et al., 2009), though they utilize ICQ for instant messaging and private exchanges (Chu et al., 2010). Turkish hackers, however, prefer MSN messenger, forums, and email and find IRC to be of limited value (Holt, 2009b). Finally, some Chinese hackers appear to use Baidu and QQ more frequently in order to communicate (Holt, Soles, & Leslie, 2008). Thus, security researchers and law enforcement must consider the local communications preferences of a given nation in order to properly investigate malicious actor groups.
In addition, communication services can aid in the identification of links between hackers within and across various groups. In fact, some hackers maintain membership in multiple hacking groups and identifying friendship networks can provide important clues to the social dynamics of both intra-group as well intergroup interactions. Figure 3 below illustrates social ties among a number of active Russian hacking groups. The thickness of the connecting line represents the strength of social ties and communications among the different groups (Holt et al, 2009).
Figure 3: Visualized Relationships between
|Draft Paper Commissioned for Conference Please Do Not Circulate||Draft of a working paper for education at brown university|
|This is a draft paper and still in progress. Please contact if you like to cite. Comments are welcome||White paper for Foresight Institute draft 12/01 Stephen L. Gillett|
|2001. Type of document: Draft rivo id number: 109 Aarts, P. G. Guidelines for Programmes Psychosocial and Mental Health Care Assistance in (Post) Disaster and Conflict Areas. Draft. 2001||Midterm (March 20) (30%), Term Paper (April 9) (30%) and Final (May 8) (40%). Paper (12 pages maximum) – on major current policy issue facing one of the countries studied. Please submit to ta by email and note class title and number. Readings|
|This Information Paper is intended to provide information to the Productivity Commission. The Paper does not represent the policy position of the Ministerial Council on Consumer Affairs or the individual members of the Council||Not on long sheets of paper—but if you are experienced in giving oral reports, and long sheets of paper work for you, do it in the way that works best for you, but do not read|
|Handbook of Paper and Board Edited by Herbert Holik Handbook of Paper and Board. H. Holik (Ed.) © 2006 wiley-vch verlag GmbH & Co. KgaA, Weinheim||Draft – not for quotation|