The Deep Web (also called the Deepnet, the Invisible Web, the Undernet or the hidden Web) is World Wide Web content that is not part of the Surface Web, which is indexed by standard search engines.
It should not be confused with the dark Internet, the computers that can no longer be reached via Internet, or with the distributed filesharing network Darknet, which could be classified as a smaller part of the Deep Web.
Mike Bergman, founder of BrightPlanet, credited with coining the phrase, said that searching on the Internet today can be compared to dragging a net across the surface of the ocean: a great deal may be caught in the net, but there is a wealth of information that is deep and therefore missed. Most of the Web’s information is buried far down on dynamically generated sites, and standard search engines do not find it. Traditional search engines cannot “see” or retrieve content in the deep Web—those pages do not exist until they are created dynamically as the result of a specific search. The deep Web is several orders of magnitude larger than the surface Web.
- Deep Resources
- Crawling the deep Web
- Classifying resources
Estimates based on extrapolations from a study done at University of California, Berkeley in 2001, speculate that the deep Web consists of about 7,500 terabytes. More accurate estimates are available for the number of resources in the deep Web: He detected around 300,000 deep web sites in the entire Web in 2004, and, according to Shestakov, around 14,000 deep web sites existed in the Russian part of the Web in 2006.
Bergman, in a seminal paper on the deep Web published in the Journal of Electronic Publishing, mentioned that Jill Ellsworth used the term invisible Web in 1994 to refer to websites that were not registered with any search engine. Bergman cited a January 1996 article by Frank Garcia :
“It would be a site that’s possibly reasonably designed, but they didn’t bother to register it with any of the search engines. So, no one can find them! You’re hidden. I call that the invisible Web.”
Another early use of the term Invisible Web was by Bruce Mount and Matthew B. Koll of Personal Library Software, in a description of the @1 deep Web tool found in a December 1996 press release.
The first use of the specific term deep Web, now generally accepted, occurred in the aforementioned 2001 Bergman study.
Deep Web resources may be classified into one or more of the following categories:
- Dynamic content: dynamic pages which are returned in response to a submitted query or accessed only through a form, especially if open-domain input elements (such as text fields) are used; such fields are hard to navigate without domain knowledge.
- Unlinked content: pages which are not linked to by other pages, which may prevent Web crawling programs from accessing the content. This content is referred to as pages without backlinks (or inlinks).
- Private Web: sites that require registration and login (password-protected resources).
- Contextual Web: pages with content varying for different access contexts (e.g., ranges of client IP addresses or previous navigation sequence).
- Limited access content: sites that limit access to their pages in a technical way (e.g., using the Robots Exclusion Standard, CAPTCHAs, or no-cache Pragma HTTP headers which prohibit search engines from browsing them and creating cached copies).
- Non-HTML/text content: textual content encoded in multimedia (image or video) files or specific file formats not handled by search engines.
- Text content using the Gopher protocol and files hosted on FTP that are not indexed by most search engines. Engines such as Google do not index pages outside of HTTP or HTTPS.
To discover content on the Web, search engines use web crawlers that follow hyperlinks through known protocol virtual port numbers. This technique is ideal for discovering resources on the surface Web but is often ineffective at finding deep Web resources. For example, these crawlers do not attempt to find dynamic pages that are the result of database queries due to the infinite number of queries that are possible. It has been noted that this can be (partially) overcome by providing links to query results, but this could unintentionally inflate the popularity for a member of the deep Web.
In 2005, Yahoo! made a small part of the deep Web searchable by releasing Yahoo! Subscriptions. This search engine searches through a few subscription-only Web sites. Some subscription websites display their full content to search engine robots so they will show up in user searches, but then show users a login or subscription page when they click a link from the search engine results page.
DeepPeep, Intute, Deep Web Technologies, and Scirus are a few search engines that have accessed the deep web. Intute ran out of funding and is now a temporary static archive as of July, 2011.
Crawling the deep Web
Researchers have been exploring how the deep Web can be crawled in an automatic fashion. In 2001, Sriram Raghavan and Hector Garcia-Molina presented an architectural model for a hidden-Web crawler that used key terms provided by users or collected from the query interfaces to query a Web form and crawl the deep Web resources. Alexandros Ntoulas, Petros Zerfos, and Junghoo Cho of UCLA created a hidden-Web crawler that automatically generated meaningful queries to issue against search forms. Several form query languages (e.g., DEQUEL) have been proposed that, besides issuing a query, also allow to extract structured data from result pages. Another effort is DeepPeep, a project of the University of Utah sponsored by the National Science Foundation, which gathered hidden-Web sources (Web forms) in different domains based on novel focused crawler techniques.
Commercial search engines have begun exploring alternative methods to crawl the deep Web. The Sitemap Protocol (first developed by Google) and mod oai are mechanisms that allow search engines and other interested parties to discover deep Web resources on particular Web servers. Both mechanisms allow Web servers to advertise the URLs that are accessible on them, thereby allowing automatic discovery of resources that are not directly linked to the surface Web. Google’s deep Web surfacing system pre-computes submissions for each HTML form and adds the resulting HTML pages into the Google search engine index. The surfaced results account for a thousand queries per second to deep Web content. In this system, the pre-computation of submissions is done using three algorithms: (1) selecting input values for text search inputs that accept keywords, (2) identifying inputs which accept only values of a specific type (e.g., date), and (3) selecting a small number of input combinations that generate URLs suitable for inclusion into the Web search index.
Automatically determining if a Web resource is a member of the surface Web or the deep Web is difficult. If a resource is indexed by a search engine, it is not necessarily a member of the surface Web, because the resource could have been found using another method (e.g., the Sitemap Protocol, mod oai, OAIster) instead of traditional crawling. If a search engine provides a backlink for a resource, one may assume that the resource is in the surface Web. Unfortunately, search engines do not always provide all backlinks to resources. Even if a backlink does exist, there is no way to determine if the resource providing the link is itself in the surface Web without crawling all of the Web. Furthermore, a resource may reside in the surface Web, but it has not yet been found by a search engine. Therefore, if we have an arbitrary resource, we cannot know for sure if the resource resides in the surface Web or deep Web without a complete crawl of the Web.
Most of the work of classifying search results has been in categorizing the surface Web by topic. For classification of deep Web resources, Ipeirotis et al. presented an algorithm that classifies a deep Web site into the category that generates the largest number of hits for some carefully selected, topically-focused queries. Deep Web directories under development include OAIster at the University of Michigan, Intute at the University of Manchester, Infomine at the University of California at Riverside, and DirectSearch (by Gary Price). This classification poses a challenge while searching the deep Web whereby two levels of categorization are required. The first level is to categorize sites into vertical topics (e.g., health, travel, automobiles) and sub-topics according to the nature of the content underlying their databases.
The more difficult challenge is to categorize and map the information extracted from multiple deep Web sources according to end-user needs. Deep Web search reports cannot display URLs like traditional search reports. End users expect their search tools to not only find what they are looking for quickly, but to be intuitive and user-friendly. In order to be meaningful, the search reports have to offer some depth to the nature of content that underlie the sources or else the end-user will be lost in the sea of URLs that do not indicate what content lies beneath them. The format in which search results are to be presented varies widely by the particular topic of the search and the type of content being exposed. The challenge is to find and map similar data elements from multiple disparate sources so that search results may be exposed in a unified format on the search report irrespective of their source.
The lines between search engine content and the deep Web have begun to blur, as search services start to provide access to part or all of once-restricted content. An increasing amount of deep Web content is opening up to free search as publishers and libraries make agreements with large search engines. In the future, deep Web content may be defined less by opportunity for search than by access fees or other types of authentication.
The Interesting Part
Silk Road is an online marketplace whose operators run it as a Tor hidden service. The majority of products that sellers list on Silk Road qualify as contraband in most jurisdictions. NPR has referred to the site as the “Amazon.com of illegal drugs”; The Economist has likewise described it, in an article on Bitcoin, as “a sort of eBay for drugs hidden in a dark corner of the web known as Tor. Annual sales are estimated to be $22 million.
Buyers and sellers conduct all transactions with Bitcoin, a crypto-currency that provides strong anonymity if used correctly. Although the Bitcoin’s exchange rate may fluctuate greatly in short periods of time, most of the prices on Silk Road are bound to United States dollar to prevent too drastic inflation or deflation. Buyers can register on Silk Road for free, but sellers must purchase new accounts through auctions to mitigate the possibility of malicious individuals distributing tainted goods. The website launched in February 2011 with development having begun three months prior.
Most sellers are based in the UK and US, and offer products such as heroin, LSD, cannabis, and other drugs. However, the site’s operators prohibit goods or services intended to harm others, such as stolen credit card numbers, counterfeit currency, personal info, assassinations, weapons of mass destruction, and materials used to make such weapons. An administrator claims “over 99% of all transactions conducted within the escrow system are completed to the satisfaction of both buyer and seller, or a mutually agreed upon resolution is found.”
In reaction to a Gawker article on the marketplace, US Senators Charles Schumer and Joe Manchin sent a letter to US Attorney General Eric Holder and DEA Administrator Michele Leonhart insisting that the agency shut down the marketplace. In a press conference Schumer described Silk Road as follows:
It’s a certifiable one-stop shop for illegal drugs that represents the most brazen attempt to peddle drugs online that we have ever seen. It’s more brazen than anything else by lightyears.
Subsequently, Silk Road’s administrators posted on the Silk Road forums the following statement:
The die have been cast [sic] and now we will see how they land. We will be diverting even more effort into countering their attacks and making the site as resilient as possible, which means we may not be as responsive to messages for a while. I’m sure this news will scare some off, but should we win the fight, a new era will be born. Even if we lose, the genie is out of the bottle and they are fighting a losing War already.
After this attention, traffic to the website increased dramatically, resulting in record sales and the bitcoin saw a corresponding rise in value. The site was also used during the markup hearing for the 2011 Stop Online Piracy Act as an exemplar of the evolution of some websites to distributed networking and computer systems which by design are not blockable by domain name filtering such as proposed in SOPA.