US20120331498A1 - Cooperative system for measuring electronic media - Google Patents

Abstract

A cooperative electronic media measurement system is disclosed that measures the electronic media usage of one or more individuals by a media research entity. The media research entity defines at least one specialized task useful to the media research entity. The at least one specialized task is performed by a media handler provided by another entity and concerns a handling of electronic media by the media handler. The individual is tracked with at least one software agent of the media research entity. At least one software agent obtains a product of the specialized task performed by the media handler via at least one defined interface that provides interoperability between the media handler and the at least one software agent. Cooperative media handlers automatically obtain information from, or otherwise obtain information about, presented media objects including identification tags, if present, and other information, for collection by the research data collection agents.

Description

CROSS-REFERENCE TO RELATED APPLICATIONS
• This application is a continuation of pending U.S. patent application Ser. No. 10/026,872, filed Dec. 21, 2001, which is a continuation of U.S. patent application Ser. No. 08/880,371, filed Jun. 23, 1997, each incorporated by reference herein.
FIELD OF THE INVENTION
• The present invention relates generally to a system for measuring a population's exposure to and interactions with electronic media (hereinafter, “electronic media measurement systems”), and more particularly, to a cooperative electronic media measurement system using media handlers to extract information from, or otherwise obtain information about, presented media objects, including identification tags, if present, for collection by software agents on behalf of a centralized media research facility.
BACKGROUND OF THE INVENTION
• The success of any advertising campaign depends on the accurate placement of advertisements within media, and the verification that specific advertising messages were presented in accord with a predefined media plan. Generally, an advertising campaign is targeted for one or more segments of a population, with media planners determining the best media vehicles to reach the target audience. In this manner, the advertiser seeks to find the most efficient media to minimize the cost to deliver a desired audience.
• Thus, prior to executing a given advertising campaign, media planners use syndicated research, such as Nielsen ratings, to determine the best media vehicles to reach a target audience. In addition, media planners utilize other information sources to research and compare the costs associated with reaching an audience through each available media vehicle. During a given advertising campaign, it is helpful to measure the target audience's exposure to the advertising messages, since media planners might make corrections in order to optimize the execution of the media plan. Likewise, after a given advertising campaign, media planners often analyze the execution of the campaign to confirm that the advertising messages reached the targeted audience to determine the accuracy of the campaign's messages in reaching the targeted audience.
• As audiences have fragmented, due to the increasing number of available channels and online options, it has become increasingly challenging for media planners to determine which media vehicles provide the best avenue to a given audience. The Internet, in particular, provides advertisers with many media options and is becoming ubiquitously available in an expanding variety of personal electronic devices, far beyond its initial limited availability to users via computer terminals and desktop computers. As with other media, advertising has become an important part of Internet revenue models. Much of the Internet's value to the advertising community is due to its enormous and evolving diversity of advertising formats, including the banner ad and Java applets, and its capability to deliver customized and relevant advertising to end users. For a more detailed discussion of advertising media, see D. Jugenheimer et al., Advertising Media Strategy and Tactics (W.C.B. Brown & Benchmark, 1992), incorporated by reference herein.
• Thus, the Internet provides an efficient mechanism for matching the advertising message to the appropriate segment of the audience. Such diverse advertising formats, however, present challenges for measuring a population's exposure to and interactions with such advertisements. While the success of the Internet can be attributed, in large part, to its open media standards that permit the creation and delivery of content having diverse formats across many platforms, there is currently no user-centric system capable of adequately measuring the diverse media formats across the growing variety of Internet-enabled consumer platforms, consistent with the needs of the advertising community.
• Generally, a given population's exposure to and interactions with media is measured by knowing the television channels and other information sources that the members of the population select. This can be performed either as a census, where the choices of the entire population are collected, or as a sample, where a statistically valid sub-population or panel is chosen to represent the entire population. Nielsen Media Research, for example, uses a panel of households, known as “Nielsen Families,” for measuring television viewing. Such panels enable research companies to correlate demographics, such as age, gender, income and education, with choice of content.
• Conventional content frequently contains, or is associated with, metadata that provides information about the content. For example, many broadcasters transmit information with conventional programming to help identify the content, for example, by program and episode. Nielsen, for example, extracts such accompanying information for measurement purposes to track the programs viewed by certain members of a panel. In addition, smart electronic program guides use such accompanying information to help individuals or their agents find content of interest.
• Similarly, the World Wide Web Consortium (W3C), has endorsed the Platform for Internet Content Selection (PICS), which is an open standard for tagging information and coding content on the Internet. The PICS standard is designed to allow software to automatically filter content that individuals choose not to receive, such as violent content, according to a ratings system. The PICS standard provides parents and other individuals with the ability to select categories of content that can be automatically blocked, in a similar manner to V-chip technology, for conventional programming. While the PICS standard allows an entire web site or static pages to be rated, the PICS standard does not permit tagging content on an object level.
• In addition, traditional electronic advertising, such as television and radio advertisements, have unique identification codes, or Industry Standard Commercial Identification (ISCI) codes, which are used for handling, broadcasting, storing and retrieving commercials. Under the ISCI standard, an ISCI alpha prefix and an ISCI numeric code identify each commercial. An ISCI prefix is assigned by ISCI to national and regional advertisers and advertising agencies. The ISCI code may be used in any manner, at the discretion of the prefix owner, provided the code consists of four letters followed by four numbers. Although ISCI codes are not presently encoded as computer readable data with each advertisement, they might evolve to do so for Internet advertising, to better manage advertising on the Internet.
• There exists both “pull” and “push” models for delivering Internet content. On traditional web sites, individuals “pull” content by browsing. These web sites can use tools to analyze the “hits” to their sites in real-time. Additionally, there exist “push” models of content delivery, such as provided by PointCast™. PointCast™ is a webcasting service that “pushes” or streams a variety of information, including editorial and advertising content, to a receiving software component, such as their proprietary screensaver, or Microsoft's Internet Explorer browser, version 4.0. Presently, each PointCast™ subscriber self-reports demographic information. Therefore, PointCast™ can provide advertisers with user-centric information about advertising exposure. Self-monitoring, however, is contrary to advertising industry guidelines, which express a preference for measurement by a disinterested third party. In addition, the PointCast™ system is limited to measuring only electronic media distributed by PointCast™.
• Even assuming that an independent auditor verified such site-centric measurements, the measurements often do not accurately reflect the activity of individuals. For example, many of the “hits” on a web site are associated with electronic agents that perform functions on the Internet on behalf of individuals. Examples of such electronic agents include web robots, issued by search engines such as those provided by Infoseek Corp. to index the contents of the Internet, and personal agents that automatically retrieve information from the Internet that matches the specified preferences of an individual. Thus, such electronic agents increase the traffic count of the respective web site, as they are not necessarily representative of an individual viewing Internet content. For example, an agent might download the entire contents of a site, while the user only views a single article. Proxy servers, on the other hand, which cache or copy Internet content to a local server or hard disk drive for subsequent access, can decrease the traffic count of a given web site. Proxy servers are used to reduce access time by storing a copy of information that was recently downloaded from a site. Thus, upon a subsequent request, the information can be delivered from the local server rather than the Internet without the knowledge of the web site traffic counter.
• While conventional electronic media measurement systems, such as Nielsen Media Research's PeopleMeter™, have successfully measured traditional media, such as television and radio, such systems are not easily extendable to the Internet environment. In addition, the site-centric measurement approaches discussed above have proven unsatisfactory. In order to accurately measure a population's exposure to and interactions with such electronic media, a user-centric measurement approach is needed which is based on a panel chosen to be statistically representative of the total population of interest. Current user-centric Internet measurement systems, however, such as the NPD Group's PC Meter™, are based on interception and interpretation of electronic media presented to members of a panel. Such interception techniques, however, rely on observing calls by software applications to the operating system and require privileged access into operating system internals. Furthermore, PC Meter™ is currently limited to household users of the Windows™ operating system, which may not be statistically representative of the total population of interest. For a more detailed discussion of the PC Meter™ system, see PCT Published Application Number WO 96/41495.
• It is believed that observing operating system internals will become increasingly challenging, if not impossible, with the trend towards more secure operating systems and communication security. Windows NT™ from Microsoft™, for example, implements a concentric ring structure of ascending privilege with an outermost ring of lowest privilege and an innermost ring of highest privilege, from which applications are excluded, based on the processor ring architecture specified by Intel Corporation. As security services become more available to Internet applications, both for computer-to-computer communications and application-to-application communications, much of this internal traffic will be encrypted. In addition, such operating system monitoring techniques will be challenging to implement within the many proprietary implementations of Internet-enabled devices, such as WebTV™. Even assuming that such user-centric measurement systems are successful in obtaining access to these communications, it is very challenging to understand what the intercepted messages mean.
• A recent industry article indicates that a new company, Relevant Knowledge Inc., of Atlanta, Ga., has developed a real-time approach to compete with the PC Meter™ system. Although Relevant Knowledge Inc. did not comment publicly for the article, it does not appear that Relevant Knowledge Inc. is using a cooperative approach. Rather, it appears that Relevant Knowledge Inc. is monitoring information, using interception and interpretation, and leveraging the communication capabilities of the Internet to distribute their monitoring software to their panel members and to collect data in real-time.
• As apparent from the above-described deficiencies with conventional electronic media measurement systems, a need exists for a universal system for measuring electronic media having diverse formats, including television, radio, Internet, and online services, across a plurality of platforms. A further need exists for a cooperating system that extends the open media standards of the Internet to measure a population's exposure to and interactions with such electronic media. Yet another need exists for a system to measure traditional television, radio, cable television, digital satellite programming and advertising delivered to households that use Internet-enabled computers and appliances for viewing, listening and interacting with such content.
SUMMARY OF THE INVENTION
• Generally, a method and apparatus are disclosed for measuring the electronic media usage of one or more individuals by a media research entity. The media research entity defines at least one specialized task useful to the media research entity. The at least one specialized task is performed by a media handler provided by another entity and concerns a handling of electronic media by the media handler. The individual is tracked with at least one software agent of the media research entity. At least one software agent obtains a product of the specialized task performed by the media handler via at least one defined interface that provides interoperability between the media handler and the at least one software agent.
• The media research entity can assign one or more software agents, hereinafter referred to as research data collection (RDC) agents, to measure an individual's exposure to and interactions with electronic media. Thus, each research data collection agent serves as an intermediary between the local environment of an individual and the central media research entity. In one embodiment, a research data collection agent is associated with each computing device utilized by a given individual. Thus, an individual might be assigned multiple research data collection agents, if required, to measure the individual's use of electronic media across multiple devices. Furthermore, a single research data collection agent may serve a plurality of individuals utilizing the same computing device.
• According to a further aspect of the invention, cooperative media handlers are utilized by the computing device of an individual to present media to an individual and to extract information from, or otherwise obtain information about, the media objects, including identification tags, if present, for collection by the research data collection agents. As used herein, the term cooperating system means a system that relies on the media handler software which presents the media to cooperate by gathering and developing information about media activities of end users, and transmitting such activity information to the media research controller, either directly or indirectly via the research data collection agents, in effect acting as agents of the media research controller, as opposed to conventional approaches of intercepting and interpreting the media activities of end users.
• As used herein, the term media handler includes persistent software components which extend the capabilities of a software application or operating system to present media objects of a particular media type to an individual and autonomous software components, such as Java applets, which may only temporarily extend the capabilities of the host to present media. In addition, the term media handler includes software applications that generate media experiences in real-time, such as video games, and resident software components, such as the PointCast™ agent, which present media to end users. In addition, the term media handler includes those portions of an Internet browser, also called “viewers” and “plug-ins”, that are capable and responsible for decoding specific media types, such as JPEG images, and using the resources of their host to present the media to the end user. The media handlers may be mobile, moving from one host computer to another, or stable, anchored to one host.
• Thus, in addition to their primary function of presenting media, the cooperative media handlers also serve as software agents for the research data collection agents, by gathering and deriving relevant information about the media presentation on behalf of the research data collection agents and then sending this information to the research data collection agents. The media handlers are in the best position to provide information about the media objects presented to the individuals, since the media handler can decode the particular media type, and determine what is presented to the individual, when it is presented and how the individual interacts with the object. In this manner, the media handler can report on an individual's exposure to and interactions with a media object, such as zooming in on a particular feature of a media object or rotating the object, and the source of the control signals, such as a specific individual or process. In the case of stable, installed media handlers, such as video games or screensaver clients; or in the case of mobile, transient media handlers, such as a Java applet advertisement, they become cooperative through the implementation of an application programming interface (API), and communicate to the research data collection agents via their host. In all of these instances, the media handler's host might contribute contextual information, such as the web site of origin, to the media handler along with the media object.
• The research data collection agents and the cooperative media handlers (the hosted processes) can communicate using at least one defined interface, such as an application programming interface (API) to define the function calls which the hosted processes and their hosts use to communicate and share resources and services, or a communication protocol. In this manner, different companies can develop interoperable research data collection agents, cooperative media handlers and hosts.
• When electronic media is received by a cooperative media handler, the media handler automatically extracts information from, or otherwise obtains information about, the media, including an identification tag, if present, for transmission to a research data collection agent. In addition to the extracted identification tag, if present, the cooperative media handler preferably transmits any content metadata associated with the media object, including program, episode and version for television and radio programming; identifiers of the media handler and the host of the media handler; timestamps related to the presentation; any contextual information which has been made available to the media handler, such as an indication of the media source for television and radio programming, or a source computer's domain name or IP address for a web site, or the context in which it was presented; presentation information; user interaction information; and content rating information, such as PICS, if available.
• A more complete understanding of the present invention, as well as further features and advantages of the present invention, will be obtained by reference to the following detailed description and drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
• FIG. 1 is a schematic block diagram illustrating a suitable communications network for interconnecting a panel member with one or more content providers and a media research controller of an electronic media measurement system;
• FIG. 2 illustrates the communications and cooperative relationship between the various entities shown inFIGS. 1,3 and4;
• FIG. 3 is a schematic block diagram of the media research controller ofFIG. 1;
• FIG. 4 is a schematic block diagram of a panel member-computing device ofFIG. 1;
• FIG. 5 illustrates a sample table from the panel member database table ofFIG. 2;
• FIG. 6 illustrates a sample table from the registered media database table ofFIG. 2;
• FIG. 7 illustrates a sample table from the encryption keys database table ofFIG. 2;
• FIG. 8 illustrates a sample table from the log entry database table ofFIG. 2;
• FIG. 9 illustrates a sample table from the local panel member database table ofFIG. 3;
• FIG. 10 illustrates a sample table from the local encryption keys database table ofFIG. 3;
• FIG. 11 illustrates a sample log entry object from the media queues ofFIG. 3;
• FIG. 12ais a flow chart describing an exemplary tag registration and encoding process as implemented by the media research controller ofFIG. 3;
• FIGS. 12band12cillustrate an exemplary identification tag format definition for a real identification tag and a dummy identification tag, respectively;
• FIG. 13 is a flow chart describing a data importing process as implemented by the media research controller ofFIG. 3;
• FIG. 14 is a flow chart illustrating a reporting process as implemented by the media research controller ofFIG. 3;
• FIG. 15aprovides an overview of the processes shown inFIGS. 15bthrough15fon the media queues ofFIG. 4;
• FIG. 15bis a flow chart describing a logging process as implemented by the research data collection agent ofFIG. 4;
• FIG. 15cis a flow chart describing a filter process as implemented by the research data collection agent ofFIG. 4;
• FIG. 15dis a flow chart describing a tag decoding subroutine as implemented by the research data collection agent ofFIG. 4;
• FIG. 15eis a flow chart describing a create-dispatch object process as implemented by the research data collection agent ofFIG. 4;
• FIG. 15fis a flow chart describing a dispatch process as implemented by the research data collection agent ofFIG. 4;
• FIG. 16 is a flow chart describing a suitable host tracking process as implemented by the panel member-computing device ofFIG. 4; and
• FIG. 17 is a flow chart describing a cooperative media handler process as implemented by the panel member-computing device ofFIG. 4.
DETAILED DESCRIPTION
• FIG. 1 shows an illustrative network environment for transferring media information, such as video, audio and data, from one ormore content providers110,120, to apanel member205 operating a panel member-computing device400, discussed further below in conjunction withFIG. 4, over one or more external networks, such asnetwork130. In an alternative or supplemental embodiment, discussed further below, the panel member-computing device400 may be configured to receive media information from one or more local sources, such as prerecorded media obtained from a CD-ROM or DVD player, or to generate media objects in real-time, for example, by means of a video game generating an advertising image with the name of the panel member for placement on a wall inside a virtual world, or a combination thereof. According to a feature of the present invention, the media information presented to the panel member by means of the panel member-computing device400 is measured by amedia research controller300, discussed further below in conjunction withFIG. 3, for subsequent reporting to one ormore research customers150. It is noted that the reports may be generated by themedia research controller300 in real time, or historically, or both.
• Theexternal networks130 shown inFIG. 1 include the Internet, the Public Switched Telephone Network (“PSTN”) and networks for the delivery of radio and television programming, such as the Digital Satellite Service (“DSS™”), cable television (“CATV”) and other over-the-air transmission networks for broadcast television, radio and satellite communications. The PSTN, as used herein, includes the combination of local and long distance wire or wireless facilities and switches, as well as cellular network systems. The Internet, as used herein, includes the World Wide Web (the “Web”) and other systems for storing and retrieving information using the Internet.
• According to a feature of the present invention, themedia research controller300 registers advertisements and other media for subsequent measurement and provides a unique identification tag that may be added to, or associated with, the existing media object for identification purposes. In addition, the present invention extracts information from, or otherwise obtains information about, presented media objects, including metadata or other information associated with a given media object, for later collection by themedia research controller300 even when the media object has not been previously registered and tagged by themedia research controller300. In this manner, the present invention measures a panel member's exposure to and interactions with all electronic media, regardless of whether the media has been previously registered or tagged for identification purposes. The panel members are preferably chosen for their demographics and have agreed to participate in a research panel to have their electronic media measured. The panel is preferably constructed so that it is representative of and projectable to the entire population. It is noted that if a census is desired, as opposed to a panel, the panel simply consists of the entire population of interest.
Entity Relationships and Interfaces
• As shown conceptually inFIG. 2, the present invention preferably consists of a number of cooperating entities to measure electronic media presented to apanel member205 by means of the panel member-computing device400, namely, a centralizedmedia research controller300, a researchdata collection agent1500 and acooperative media handler1700. As previously indicated, themedia research controller300 measures a panel's exposure to and interactions with electronic media for subsequent reporting to one ormore research customers150. As used herein, the term cooperating system means a system that relies on the media handler software which presents the media to cooperate, by gathering and developing information about media activities of end users, and transmitting such activity information to the media research controller, either directly or indirectly via the research data collection agents, in effect acting as agents of the media research controller, as opposed to conventional approaches of intercepting and interpreting the media activities of end users.
• To this end, themedia research controller300 preferably assigns one or more software agents, referred to as researchdata collection agents1500, discussed further below in conjunction withFIG. 15, to measure each panel member's exposure to and interactions with electronic media. Thus, each researchdata collection agent1500 serves as an intermediary between the local environment of apanel member205 and the centralmedia research controller300. In one embodiment, a researchdata collection agent1500 is associated with eachcomputing device400 utilized by a givenpanel member205. Thus, a single panel member might be assigned multiple research data collection agents, if required to measure the panel member's use of electronic media across multiple devices. Furthermore, a single researchdata collection agent1500 may serve a plurality of panel members utilizing thesame computing device400. It is not necessary for the researchdata collection agent1500 to be proximal to the associated panel member or the panel member's physical environment. Since the researchdata collection agent1500 is user-centric, theagent1500 might be a component of the person's net-centric environment and be activated as a side effect of the person obtaining access to network resources, or theagent1500 may reside on a smartcard carried by the individual.
• In addition, thecooperative media handlers1700, discussed below in conjunction withFIG. 17, are utilized by the panel member-computing device400 to present media to thepanel member205 and to extract information from, or otherwise obtain information about, presented media objects, including identification tags, if present, for collection by the researchdata collection agents1500. Thus, in addition to their primary function of presenting media, thecooperative media handlers1700 also serve as software agents for the researchdata collection agents1500, by gathering and deriving relevant information about the media presentation on behalf of the research data collection agents and then sending this information to the research data collection agents. It is noted that while some of the media handlers might be implemented ascooperative media handlers1700, as described herein, the present invention is not dependent upon all of the media handlers being implemented in a cooperative manner. However, the extent of research data collection agent's1500 ability to measure a panel member's exposure to and interactions with electronic media is limited to media presented by means of acooperative media handler1700.
• As used herein, a software agent is a software component that has the ability and authority to act on behalf of a controlling entity, generally instantiated as a process on a host computer. Each agent's controlling entity provides the agent with resources, such as encryption keys and a digital certificate for identification, and delegates specific tasks to the agent. The software agents manage these tasks, and operate in an autonomous or semi-autonomous mode with respect to its controlling entity. The software agents may be mobile, moving from one host computer to another, or stable, anchored at one host.
• The software agent's host preferably provides the software agent with necessary computational resources, such as processor cycles, memory and communication. In the illustrative embodiment, discussed further below, an Internet browser software product, such as Netscape Navigator™ or Microsoft Internet Explorer™, resident on the panel member-computing device400, serves as thehost1600 for the researchdata collection agent1500 and thehost1600′ for thecooperative media handler1700 and provides them with the required computational resources. In alternate embodiments, the host for one or both of the research data collection agent and the cooperative media handlers may be embodied as an operating system or a virtual machine, such as the Java virtual machine. The researchdata collection agent1500 preferably communicates with themedia research controller300 by means of the host's secure communication capabilities, such as SSL or S/MIME. In addition, the researchdata collection agent1500 relies on itshost1600 to pass through data sent by thecooperative media handler1700 for collection by a researchdata collection agent1500.
• The communications between the various entities are preferably sufficiently encrypted to protect the privacy of thepanel members205, the proprietary information and competitive interests of themedia research controller300 and the validity of the collected data, as would be apparent to a person of ordinary skill. Furthermore, to protect the privacy of non-panel members, the cooperative features of thehosts1600,1600′, as described herein, are preferably initially disabled when distributed to end-users, and are only subsequently enabled when specifically authorized by a panel member who has agreed to participate in a research panel.
• In some embodiments, the researchdata collection agent1500 is not proximal to the panel member(s) it tracks, or the panel member's physical environment. For example, the research data collection agent might be hosted by the central computer of an online service. In other embodiments, the research data collection agent is hosted separately from the cooperative media handlers. For example, the researchdata collection agent1500 might be located in a wristwatch worn by the panel member. Thus, the host services and resources may reside locally or be accessed by means of a network connection, or a combination thereof, as would be apparent to a person of ordinary skill. It is noted that the researchdata collection agents1500 and thecooperative media handlers1700 might typically share the same host or might be hosted separately, for example, in an embodiment where the media researchdata collection agent1500 associated with apanel member205 is resident on a smartcard or wristwatch carried by an individual.
• As illustrated inFIG. 2, the present invention relies on a number of cooperative interfaces between the various entities. A media object is presented to apanel member205 by thecooperative media handler1700 using a conventional bi-directional user interface210. Thecooperative media handler1700 transmits identification tags and other information extracted from, or otherwise obtained about, media objects for collection by a researchdata collection agent1500, by means of acooperative interface220. Upon receipt of an extracted identification tag or other obtained information from amedia handler1700, the researchdata collection agent1500 sends a message to themedia research controller300, preferably over anexternal network130, containing the extracted identification tag and other obtained information.
• In addition, as previously indicated, the researchdata collection agents1500 and thecooperative media handlers1700 also require interfaces to theirrespective hosts1600,1600′. In a preferred embodiment, the researchdata collection agents1500 and the cooperative media handlers1700 (the hosted processes) use an application programming interface (API) to define the function calls which the hosted processes and their hosts, such as thehost1600, use to communicate and share resources and services. In this manner, different companies can develop interoperable research data collection agents, cooperative media handlers and hosts.
Entity Functions
• As previously indicated, when electronic media is received by acooperative media handler1700 for presentation, themedia handler1700 automatically extracts information from, or otherwise obtains information about, presented media object, including an identification tag, if present, and other relevant information, as available, for transmission to a researchdata collection agent1500. In addition to the extracted identification tag, if present, thecooperative media handler1700 preferably transmits any content metadata included in or associated with the media object, including program and episode information; an identification of media handler and its host; any contextual information which has been made available to themedia handler1700, such as an indication of the media source for television and radio programming, or a source computer's domain name or IP address for a web site; any presentation information or user interaction information, and content rating information, such as PICS, if available. Each of these information types is discussed further below in conjunction withFIG. 8.
• A researchdata collection agent1500 preferably commences tracking of a panel member, and begins collecting such transmissions from thecooperative media handlers1700, when a panel member enters the scope of the researchdata collection agent1500 and has been identified and authenticated, for example, following a log on procedure. Similarly, the researchdata collection agent1500 suspends tracking of apanel member205 when the person leaves the scope of the remotemedia research agent1500, such as following a log off procedure. The research data collection agent (i) creates log entry objects from transmissions received from the cooperative media handlers and places the log entry objects into an unfiltered media queue, (ii) confirms the integrity of the messages and filters out unnecessary log entry objects from the unfiltered media queue to create a filtered media queue, (iii) creates dispatch objects using objects from the filtered media queue and places created dispatch objects into a dispatch queue, and (iv) transmits dispatch objects from the dispatch queue to the media research controller, when resources are available.
• Thecooperative media handlers1700 preferably are not aware of whether or not a researchdata collection agent1500 is present and measuring the content presented to a panel member, for the same reason that the identities of Nielsen families are carefully concealed from the media sources that Nielsen measures, to prevent manipulation of the content presented to the panel members. Thus, thecooperative media handlers1700 preferably transmit information extracted from, or otherwise obtained about, presented media objects, for collection by a researchdata collection agent1500, whether or not a researchdata collection agent1500 is present. For efficiency purposes, however, when thesame host1600 is hosting both the researchdata collection agent1500 andmedia handler1700, the host may inhibit thecooperative media handler1700 from transmitting information extracted from or obtained about media objects when a researchdata collection agent1500 is not present, is not in a tracking mode or the panel members are not in scope.
• According to a feature of the present invention, themedia handlers1700 thus cooperate with the researchdata collection agents1500 by gathering and deriving relevant information about the media presentation on behalf of the research data collection agents and then sending this information to the research data collection agents. Themedia handlers1700 are in the best position to provide information about the media objects presented to thepanel members205, since the media handler is responsible for decoding or generating specific media objects, they determine what is presented, how it is presented, when it is presented and how thepanel member205 interacts with the presented media. Additionally, themedia handlers1700 can extract metadata from the media object and glean contextual information about the media object from the media handler's host. Through their cooperation and assistance, the research data collection agent can gain access to information that might not be available in any other way.
• Since themedia research controller300 can measure a panel's exposure to and interactions with thecooperative media handlers1700 themselves, in addition to exposure to and interactions with the underlying media objects of interest, providers of media handlers are motivated to develop media handlers which cooperate with the software processes of the present invention, and thereby ensure that the media handlers provide the required functionality. In this manner, media handler developers can receive custom research reports to analyze the panel's exposure to and interactions with such media handlers.
Media Terminology
• A media type is a means to represent media information, such as an image or a sound. For example, the PNG (Portable Network Graphics) file format is a media type for representing computer images. As used herein, a media object is an item or instance of a media type. Currently, there are hundreds of media types in use. Typically, media types are implemented as files, and increasingly as objects, and allow for some means of attaching metadata, or information about the data. In some cases, the metadata is implemented as a text field that resides in a header preceding the content data and can be extracted prior to decoding the content data. Generally, extraction of the header data is not computationally intensive, and can take place prior to or during the decoding and display or playback of the content.
• Every media type must have a corresponding media handler, located at the end-user's client side to present a given media object to an individual. As used herein, the term media handler includes persistent software components which extend the capabilities of a software application or an operating system to present media objects of a particular media type to an individual and autonomous software components, such as Java applets, which may only temporarily extend the capabilities of the host to present media. In addition, the term media handler includes software applications that generate media experiences in real-time, such as video games, and resident software components, such as the PointCast™ agent, which present media to end users. In addition, the term media handler includes those portions of an Internet browser, also called “viewers” and “plug-ins”, that are capable and responsible for decoding specific media types, such as JPEG images, and using the resources of their host to present the media to the end user. The media handlers may be mobile, moving from one host computer to another, or stable, anchored to one host.
• Typically, an Internet browser software product has a collection of internal media handlers, such as JPEG and GIF decoders. When a browser encounters a media object of a given media type, the browser transfers the media object to the appropriate media handler for decoding. In addition, most browsers incorporate “plug-in” architecture, which allows third parties to develop new media types and distribute corresponding media handlers as “plug-ins.” For example, Netscape Navigator™ currently has over two hundred compatible plug-ins available.
• It is noted that for continuous media, such as audio and video, identification tags are preferably placed at periodic intervals throughout the duration of the media or on an associated data channel. Since the identification tags might be utilized by hostile software to automatically remove advertising, the identification tags are preferably placed in both the advertising, as well as the associated advertising-supported media objects to discourage such automatic removal. In one preferred embodiment, discussed below in conjunction withFIG. 12, real identification tags are placed in media that is to be measured (and a fraction that is not), and dummy identification tags are placed in other content. As discussed further below, in order to permit the researchdata collection agents1500 to be tasked by themedia research controller300 to report only registered media, the researchdata collection agent1500 can preferably distinguish real identification tags from dummy identification tags, and, if desired, return only real identification tags to themedia research controller300.
Media Research Controller
• FIG. 3 is a block diagram showing the architecture of an illustrativemedia research controller300. Themedia research controller300 preferably includes certain standard hardware components, such as a central processing unit (CPU)310, a random access memory (RAM)320, a read only memory (ROM)330, aclock340, acommunications port350, and adata storage device360. TheCPU310 is preferably linked to each of the other listed elements, either by means of a shared data bus, or dedicated connections, as shown inFIG. 3.
• TheCPU310 may be embodied as a single commercially available processor, such as Intel's Pentium 100 MHz P54C microprocessor, Motorola's 120 MHz PowerPC 604 microprocessor or Sun Microsystem's 166 MHz UltraSPARC-I microprocessor. Alternatively, theCPU310 may be embodied as a number of such processors operating in parallel, on one or more distributed processing nodes. Thedata storage device360 and/orROM330 are operable to store one or more instructions, as discussed below in conjunction withFIGS. 12 through 14, which theCPU310 is operable to retrieve, interpret and execute. TheCPU310 preferably includes a control unit, an arithmetic logic unit (ALU), and a CPU local memory storage device, such as, for example, an instruction cache or a plurality of registers, in a known manner. The control unit is operable to retrieve instructions from thedata storage device360 orROM330. The ALU is operable to perform a plurality of operations needed to carry out instructions. The CPU local memory storage device is operable to provide high-speed storage used for storing temporary results and control information.
• As discussed further below in conjunction withFIGS. 5 through 8, thedata storage device360 includes acentral database370 for storing a panel member database table500, a registered media database table600, an encryption keys database table700 and a log entry database table800. The panel member database table500 preferably stores personal and demographic information for each member of the panel. The registered media database table600 preferably stores information about each media object that is registered with themedia research controller300, including the assigned tag identification number. The encryption keys database table700 preferably stores the public key/private key pairs that are utilized in the illustrative embodiment to implement secure communications. The log entry database table800 preferably stores information that has been extracted from or obtained about media objects presented to panel members and provided to themedia research controller300 by the remotemedia research agents1500.
• In addition, as discussed further below in conjunction withFIGS. 12 through 14, thedata storage device360 preferably includes a registration andtag encoding process1200, adata importing process1300 and areporting process1400. Generally, the registration andtag encoding process1200 registers a particular media object, and then places an assigned tag identification number in the registered media object, or otherwise associates the tag identification number with the media object. Thedata importing process1300 receives log dispatches from the remotemedia research agents1500, verifies the authenticity and integrity of the received messages and places confirmed entries in the log entry table800. Thereporting process1400 preferably generates reports in real-time or historically, to suit the needs of aparticular research customer150.
• Thecommunications port350 connects themedia research controller300 to theexternal networks130, thereby linking the media research controller to each remote media research agent, as shown inFIGS. 1 and 2. Thecommunications port350 preferably includes multiple communication channels for simultaneously connecting themedia research controller300 to multiple research data collection agents.
Panel Member-Computing Device
• FIG. 4 is a block diagram showing the architecture of an illustrative panel member-computing device400. The panel member-computing device400 may be embodied as any device which presents media to individuals, including, for example, an Internet-enabled device, such as a network computer, a set-top box, a television, a telephone, pager, or personal digital assistant. The panel member-computing device400 typically includes certain standard hardware components, such as a central processing unit (CPU)410, a random access memory (RAM)420, a read only memory (ROM)430, aclock440, acommunications port450, and adata storage device460. Each of thesecomponents410,420,430,440,450 and460 may be identical to the corresponding components described above in conjunction withFIG. 2. In addition, the panel member-computing device400 may include one ormore output devices470, such as a video display card and video monitor, a sound card and speaker, and one ormore multimedia players480, such as a CD-ROM or DVD device.
• As discussed further below in conjunction withFIGS. 9 through 11, thedata storage device460 includes a local panel member database table900, a local encryption keys database table1000 andmedia queues490. The local panel member database table900 preferably stores an indication of whether each panel member associated with a given research data collection agent is currently within the scope of the agent. The local encryption keys database table1000 preferably stores each of the public key values which are required by the researchdata collection agent1500 to determine if a given extracted tag is a real tag or a dummy tag and for secure communications with themedia research controller300. Themedia queues490 preferably store information that has been extracted from or obtained about media objects presented to panel members and provided to the research data collection agent by the media handlers. As discussed further below in conjunction withFIGS. 15athrough15f, themedia queues490 preferably consist of anunfiltered media queue1150, a filtered media queue1170, each containing log entry objects1100, and adispatch queue1190, containing dispatch objects1130.
• In addition, as discussed further below in conjunction withFIGS. 15 through 17, thedata storage device460 preferably includes a research data collection (RDC)agent1500, a RDC/cooperativemedia handler host1600 and one or morecooperative media handlers1700. The research data collection (RDC)agent1500 preferably includes (i) alogging process1510 to create log entry objects from transmissions received from the cooperative media handlers and places the log entry objects into an unfiltered media queue, (ii) afiltering process1530 to confirm the integrity of the messages and filter out unnecessary log entry objects from the unfiltered media queue, using adecoding subroutine1550 to create a filtered media queue, (iii) a createdispatch process1580 to create dispatch objects using objects from the filtered media queue and that places created dispatch objects into a dispatch queue, and (iv) adispatch process1595 to transmit dispatch objects from the dispatch queue to the media research controller, when resources are available. The RDC/cooperativemedia handler host1600 preferably provides resources and services1675 to hosted processes and executes atracking process1605 to notify the researchdata collection agents1500 when a panel member enters or exits the agent's scope. Thecooperative media handlers1700 preferably receive and decode media objects and extract information from, or otherwise obtain information about, presented media objects, including identification tags, if present, for collection by researchdata collection agents1500.
• It is noted that the research datacollection agent process1500 and related database tables900,1000,1100 have been shown as part of the panel member-computing device400 for illustrative purposes only, and could be resident on a device physically remote from thepanel member205 in alternate embodiments, as previously indicated, such as part of the panel member's net-centric environment which is activated as a side effect of the person obtaining access to network resources, or resident or hosted within a wristwatch that the panel member wears.
• Thecommunications port450 connects the panel member-computing device400 to theexternal networks130, thereby linking thecomputing device400 to themedia research controller300 andcontent providers110,120, as shown inFIGS. 1 and 2. Thecommunications port450 preferably includes multiple communication channels for simultaneous connections.
Database Tables
• FIG. 5 illustrates an exemplary panel member database table500 that preferably stores personal and demographic information for each member of the panel. The panel member database table500 maintains a plurality of records, such as records505-520, each associated with a different panel member. For each panel member identified by a panel member identifier infield540, the panel member database table500 includes the panel member's name, sex, age, city and state infields545 through565, respectively. In addition, the panel member database table500 includes an indication of the member's education level and income infields570 and575, the associated research data collection agent infield580, and the member's email address infield585. The panel member identifier stored infield540 may be utilized, for example, to index the log entry table800, discussed below in conjunction withFIG. 8.
• FIG. 6 illustrates an exemplary registered media database table600 which preferably stores information about each media object which is registered with themedia research controller300, including the assigned tag identification number. The registered media database table600 maintains a plurality of records, such as records605-620, each associated with a different registered media object. For each registered media object identified by a registered media identification number (M₁) infield640, the registered media database table600 includes an indication of the entity which registered the object, the associated agency and the media type infields645 through655, respectively. In addition, an encrypted and digitally signed version of registered media identification number preferably serves as the identification tag, C₂, and is recorded infield660. Finally, the anticipated starting and ending dates for which the media object will be distributed are recorded infields665 and670.
• FIG. 7 illustrates an exemplary encryption keys database table700 that preferably stores the public key/private key pairs which are utilized by themedia research controller300 in the illustrative embodiment to implement encrypted communications with the various entities and other security features. In a preferred embodiment, themedia research controller300 generates public key/private key pairs and securely distributes the various public keys to the research data collection agents of the panel members. In some embodiments, the agents may be provided with initial public keys prior to distributing the agent. Thus, the encryption keys database table700 maintains a plurality of records, such as records705-720, each associated with a different public key/private key pair. For each pair, identified by a key pair identifier infield740, the encryption keys database table700 includes an indication of the key pair owner infield745, namely, the entity which holds the private key, such as the research data collection agent (RDCA)1500 or the media research controller (MRC)300. In addition, the encryption keys database table700 includes the corresponding public key and private key values infields750 and755, respectively.
• FIG. 8 illustrates an exemplary log entry database table800 which preferably stores information which has been extracted from or obtained about media objects presented to panel members and provided to themedia research controller300 by the research data collection agents, in a manner described further below. The log entry database table800 maintains a plurality of records, such as records805-820, each associated with a different log entry. For each log entry identified by an entry identifier infield830, the log entry database table800 includes an identifier of the researchdata collection agent1500,cooperative media handler1700, andmedia handler host1600′ associated with the log entry object infields835,840 and842, respectively. In this manner, theresearch media controller300 can generate custom reports to analyze a panel's exposure to and interactions with themedia handlers1700 and media handler hosts1600′ themselves, in addition to the underlying media objects of interest. In addition, each logged entry includes an identifier of the particular media object and panel member associated with the entry infields845 and850, as well as date and time stamps infields855 and860. Each entry preferably also includes any metadata, contextual information, presentation information, and user interactions that were received in the log entry object, infields865,870,875 and880, respectively.
• As shown inFIG. 8, the metadata infield865 may include program identification information and the corresponding media type. The contextual information infield870 may include the URL and other information indicating the source of the corresponding media object, or the context in which it was presented to the panel member. The presentation information infield875 may include the language and format in which the media object was presented to the panel member, or other information indicating how the objected was presented to the panel member, when alternative presentations are possible. Finally, the user interactions infield880 may specify how the panel member interacted with the object, including whether the panel member zoomed in on portions of the media object, or rotated the object, as well as the size of the media object and any user inputs.
• FIG. 9 illustrates an exemplary local panel member database table900 which preferably stores a semaphore indicating whether each panel member associated with a given research data collection agent is currently within the scope of the agent. The local panel member database table900 maintains a plurality of records, such as records905-915, each associated with a different panel member. For each panel member identified by a panel member identifier infield940, the local panel member database table900 contains a semaphore infield945 indicating whether or not the panel member is in scope.
• FIG. 10 illustrates an exemplary local encryption keys database table1000 which preferably stores each of the encryption key values which are required by the researchdata collection agent1500 to determine if a given extracted tag is a real tag or a dummy tag and for secure communications with themedia research controller300. The local encryption keys database table1000 maintains a plurality of records, such as records1005-1020, each associated with a different encryption key. For each public key identified by a local key identifier infield1040, the local encryption keys database table1000 includes a key pair identifier infield1045 and the key value infield1050.
• FIG. 11 illustrates an exemplarylog entry object1100 from themedia queues490, which preferably stores information which has been extracted from or obtained about a given media object presented to one or more panel members and provided to the researchdata collection agent1500 by thecooperative media handlers1700. Thelog entry object1100 consists of a plurality of elements, including a date/time stamp1102,panel member interactions1104,media handler identifier1106,presentation information1108,identification tag1110,panel member identifier1112,contextual information1114,metadata1116 and the mediahandler host identifier1118. Each of these information types has been described above in conjunction withFIG. 8. In addition, thelog entry object1100 can include a copy of the media object itself, if desired.
Processes
• As discussed above, themedia research controller300 preferably executes a tag registration andencoding process1200, shown inFIG. 12a, to register a particular media object, and to place an assigned tag identification number in the registered media object, or on a simultaneous channel. As illustrated inFIG. 12a, themedia research controller300 begins the processes embodying the principles of the present invention duringstep1210 upon receipt of a request to register a particular media object.
• Thereafter, themedia research controller300 assigns a media identification key, M₁, duringstep1215 and creates a record of the media object in the registered media database table600. In one embodiment, the identification tags are based on an extended version of the ISCI standard. In a preferred embodiment, themedia research controller300 utilizes a doubly encrypted identification tag to protect the security of the media identification key, which is preferably never made available outside themedia research controller300. Thus, the assigned media identification key, M₁, is preferably strongly encrypted with a private key, K₁, duringstep1220 to produce a payload, C₁. Thereafter, the payload, C₁, is digitally signed with a private key, K₂, to create a digital signature, DS₁, duringstep1225 and then the digital signature, DS₁, is prepended to the payload, C₁, duringstep1230 to produce a new message, M₂, shown inFIG. 12b. It is noted that the new message, M₂, may optionally include bit padding to increase C₁to a predefined length. Finally, the new message, M₂, is encrypted with a private key, K₃, duringstep1235 to produce a real tag, C₂. For a more detailed discussion of suitable encryption and security techniques, see B. Schneier, Applied Cryptography (2d ed. 1997), incorporated by reference herein.
• Thereafter, the real tag, C₂, is placed in the media object duringstep1240, or on a simultaneous channel, and also placed infield660 of the registeredmedia database600. The identification tag can be bound to the content in a variety of ways. Preferably, the advertisement or other media is represented as an object, and has been designed to implement the cooperative approach associated with the present invention by reserving a field or sub-object for storing the identification tag. It is important that the identification tag does not interfere in any way with the normal use of the media by any media handler or media handler host that has not implemented the cooperative API described herein. Binding can be generally achieved since the most prevalent media file formats and datatypes support extension through user-defined chunks or objects. As a last resort, the identification tag might be stored within an embedded comment field, into its URL, or even through steganography (hidden codes).
• Finally, the media object is released for distribution duringstep1245, before program control ends duringstep1250.
• As discussed above, dummy identification tags may be utilized in some content to prevent hostile software from automatically removing advertising media objects. To promote the use of dummy identification tags, they are preferably constructed simply as a random sequence of n bits, as shown inFIG. 12c. In an alternate embodiment, the functionality for generating both real and dummy identification tags and encoding the identification tags into the media objects can be performed by third parties, such as advertising agencies or commercial producers, including edit houses or production facilities.
• As discussed above, themedia research controller300 preferably executes adata importing process1300, shown inFIG. 13, to receive log dispatches from the remotemedia research agents1500, verify the authenticity and integrity of the received messages and place confirmed entries in the log entry table800. As illustrated inFIG. 13, themedia research controller300 executes the data importing process upon receipt of a log dispatch from a research data collection agent duringstep1310.
• Thereafter, the data importing process verifies the authenticity and integrity of the dispatch duringstep1320 and then decrypts the dispatch duringstep1330. Finally, the data importing process places all confirmed entries from the dispatch in the log entry database table800 duringstep1340 before program control terminates duringstep1350.
• As previously indicated, themedia research controller300 preferably executes a real-time reporting process1400, or a non-real-time reporting process1400′, to generate reports suited to the needs of aparticular research customer150. As shown inFIG. 14, thedata importing process1300, discussed above, will create records of data received from the researchdata collection agents1500 in acentral database370, from which the reports will be generated. The reports will be formatted to present available data, in a conventional manner, according to the needs one ormore research customers150. Thecustomers150 can access the reports over theexternal networks130, or the reports can be printed off-line and provided to thecustomer150, as would be apparent to a person of ordinary skill.
Research Data Collection Agent
• As shown inFIG. 15a, the researchdata collection agent1500 preferably includes alogging process1510, afilter process1530, a create-dispatch object process1580 and adispatch process1595, discussed further below in conjunction withFIGS. 15b,15c,15eand15f, respectively. Thelogging process1510 is executed by the researchdata collection agent1500 to receive messages fromcooperative media handlers1700 and create entry objects in anunfiltered media queue1150. Thefilter process1530 is executed to filter out dummy tags and other unnecessary information from theunfiltered media queue1150 to create a filtered media queue1170. Finally, the researchdata collection agent1500 executes the create-dispatch object process1580 to place the entries from the filtered media queue1170 into adispatch queue1190 for transmission to themedia research controller300, by the dispatch process1195 when resources are available. It is noted that the researchdata collection agent1500 may also receive information regarding specific transactions made by an associated panel member, from a cooperative financial or end-user transaction process, as would be apparent to a person of ordinary skill, based on the disclosure herein.
• As previously indicated, the researchdata collection agent1500 executes alogging process1510, shown inFIG. 15b, to receive messages from thecooperative media handler1700 and to create entries in theunfiltered media queue1150. Thus, thelogging process1510 is entered duringstep1515 upon receipt by the researchdata collection agent1500 of a message from acooperative media handler1700 by means of thehost1600 of themedia handler1700, containing an extracted identification tag, C₂′, if available, and other information. Thereafter, thelogging process1510 creates alog entry object1100 with the identification tag and other received information in theunfiltered media queue1150 duringstep1520, which is added to the tail of theunfiltered media queue1150 duringstep1525 before program control terminates duringstep1528.
• As discussed above, the researchdata collection agent1500 periodically executes afilter process1530, shown inFIG. 15c, to filter out dummy tags and other unnecessary information from theunfiltered media queue1150 to create a filtered media queue1170. As shown inFIG. 15c, thefilter process1530 performs a test duringstep1532 to determine if theunfiltered media queue1150 is empty. If it is determined duringstep1532 that theunfiltered media queue1150 is empty, then program control ends duringstep1548. If, however, it is determined duringstep1532 that theunfiltered media queue1150 is not empty, then alog entry object1100 is retrieved from the head of theunfiltered media queue1150 duringstep1534.
• Thereafter, a test is performed duringstep1535 to determine whether themedia research controller300 has instructed the researchdata collection agent1500 to return information only about registered media. If it is determined duringstep1535 that themedia research controller300 has not instructed the researchdata collection agent1500 to return information only about registered media, then program control proceeds to step1540. If, however, it is determined duringstep1535 that themedia research controller300 has instructed the researchdata collection agent1500 to return information only about registered media, then adecoding subroutine1550, discussed below in conjunction withFIG. 15d, is executed duringstep1536 to determine if the received tag, C₂′, stored in thecurrent entry object1100 of theunfiltered media queue1150, contains a real identification tag, which is generally associated with registered media.
• After thesubroutine1550 executes, a test is performed duringstep1538 to determine if the received tag, C₂′, stored in the currentlog entry object1100, contains a real identification tag. If it is determined duringstep1538 that the received tag, C₂′, stored in the currentlog entry object1100, does not contain a real identification tag, then program control returns directly to step1532 to process additional log entry objects1100, if any. If, however, it is determined duringstep1538 that the received tag, C₂′, stored in thelog entry object1100, does contain a real identification tag, then the currentlog entry object1100 is placed in the filtered media queue1170 duringstep1540. Thereafter, program control returns to step1532 to process additional log entry objects1100, if any, in the manner described above.
• In this manner, for efficiency purposes, if the researchdata collection agent1500 is tasked by the media research controller to report only registered media, then the researchdata collection agent1500 preferably only returns log entry objects1100 containing real identification tags to themedia research controller300. The researchdata collection agent1500 preferably does not know, however, whether or not the media object containing a real identification tag is an advertisement. As previously indicated, real identification tags are preferably placed in a small percentage of the advertising-subsidized media objects as a countermeasure against hostile advertising filtering software. Thus, if the research data collection agent's local encryption keys K₂and K₃are compromised and obtained by advertising filtering software, the advertising filtering software erroneously filters out some non-advertising media objects. It is noted that advertising filtering software is further discouraged by the preferred selection of challenging encryption techniques, since the researchdata collection agent1500 can evaluate the authenticity of a received tag hours after the media is presented, while the advertising filtering software typically would need to decrypt the identification tags in real-time.
• As previously indicated, thefilter process1530 executes adecoding subroutine1550, shown inFIG. 15d, duringstep1536 to determine if the received tag, C₂′, stored in the current record of theunfiltered media log1100, contains a real identification tag. Thedecoding subroutine1550 is entered duringstep1554, where the received tag, C₂′, is decrypted to create a message, M₂′, using the value of the public key, K₃, stored in the local encryption keys database table1000. Thedecoding subroutine1550 then creates a digital signature, DS₁′, from the first in hits of the message, M₂′, and creates C₁′ from the remainder duringstep1558.
• Thereafter, a test is performed duringstep1562 to determine if the digital signature, DS₁′, is a valid digital signature of C₁′. If it is determined duringstep1562 that the digital signature, DS₁′, is a valid digital signature of C₁′ then the decoding subroutine returns an indication duringstep1564 to thefilter process1530 that the received tag is a real identification tag. If, however, it is determined duringstep1562 that the digital signature, DS₁′, is not a valid digital signature of C₁′ then the decoding subroutine returns an indication duringstep1568 to thefilter process1530 that the received tag is an invalid or dummy identification tag. In order to implement the security algorithms discussed inFIG. 15d, the researchdata collection agent1500 preferably has access to the necessary cryptographic services through its host, and has received the required public keys, K₂and K₃, by means of a secure distribution method from themedia research controller300.
• As discussed above, the researchdata collection agent1500 executes a create-dispatch object process1580, shown inFIG. 15e, to place entries from the filtered media queue1170 into adispatch queue1190 for transmission by thedispatch process1190, shown inFIG. 15f, to themedia research controller300, when resources are available. Initially, the create-dispatch object process1580 performs a test duringstep1581 to determine if the filtered media queue1170 is empty. If it is determined duringstep1581 that the filtered media queue1170 is empty, then program control ends duringstep1582.
• If, however, it is determined duringstep1581 that the filtered media queue1170 is not empty, then a further test is performed duringstep1583 to determine if adispatch object1130 already exists. If it is determined duringstep1583 that adispatch object1130 already exists, then program control proceeds directly to step1585. If, however, it is determined duringstep1583 that adispatch object1130 does not already exist, then anew dispatch object1130 is created duringstep1584.
• Thereafter, alog entry object1100 is retrieved from the head of the filteredmedia queue1150 duringstep1585, and placed in thedispatch object1130. Thereafter, a test is performed duringstep1587 to determine if the filtered media queue1170 is empty or if thedispatch object1130 is full. If it is determined duringstep1585 that the filtered media queue1170 is not empty and that thedispatch object1130 is not full, then program control returns to step1581 to continue processing additional log entry objects1100 from the filtered media queue1170, if any.
• If, however, it is determined duringstep1585 that the filtered media queue1170 is empty or that thedispatch object1130 is full, then thedispatch object1130 will be compressed and digitally signed duringstep1588. Thereafter, thecompressed dispatch object1130 will preferably be encrypted duringstep1589 using the public key, K₄, of themedia research controller300. Finally, the compressed andencrypted dispatch object1130 will be added to the tail of thedispatch queue1190 duringstep1590 and program control returns to step1581 and continues in the manner discussed above.
• As previously indicated, adispatch process1595, shown inFIG. 15f, transmitsdispatch objects1130 from thedispatch queue1190 to themedia research controller300, when resources are available. Thus, a test is initially performed duringstep1596 to determine if thedispatch queue1190 is empty. If it is determined duringstep1596 that thedispatch queue1190 is empty, then program control terminates duringstep1597. If, however, it is determined duringstep1596 that thedispatch queue1190 is not empty, then adispatch object1130 is removed from the head of thedispatch queue1190 duringstep1598 and sent to themedia research controller300 duringstep1599. Thereafter, program control returns to step1596 to processadditional dispatch objects1130, if any, in the manner described above.
Host Process(es)
• As previously indicated, the RDC/cooperativemedia handler host1600 preferably executes atracking process1605, shown inFIG. 16, to notify the hosted researchdata collection agents1500 when a panel member enters or exits the agent's scope. In addition, as discussed above, the researchdata collection agents1500 and thecooperative media handlers1700 preferably rely on services and resources provided by the host processes for processing, security, storage and communication. In the illustrative embodiment discussed herein, an Internet browser software product, such as Netscape Navigator™ or Microsoft Internet Explorer™, resident on the panel member-computing device400, serves as the host for both the researchdata collection agent1500 and thecooperative media handler1700 and provides them with computational resources.
• Thus, as shown inFIG. 16, anillustrative tracking process1605 continuously reports when a panel member enters or exits its scope duringstep1610. Once thetracking process1605 determines that a panel member has entered the scope of a research data collection agent, a message is sent to the hosted researchdata collection agent1500 duringstep1620 monitoring the panel member indicating that a panel member has entered the agent's scope.
• Thereafter, thetracking process1605 continuously monitors the panel member duringstep1630 until the panel member leaves the scope of the hosted researchdata collection agent1500. If it is determined duringstep1630 that the panelist has left the scope of the hosted researchdata collection agent1500, then a message is sent to the hosted researchdata collection agent1500 duringstep1640 indicating that a panel member has left the agent's scope. A test is then performed duringstep1645 to determine if tracking should be suspended. If it is determined duringstep1645 that tracking should not be suspended, then program control returns to step1610 to continue tracking, in the manner described above. If, however, it is determined duringstep1645 that tracking should be suspended, then program control terminates duringstep1650.
Cooperative Media Handlers
• As discussed above, the cooperativemedia handler process1700, shown inFIG. 17, preferably presents media objects and extracts information from, or otherwise obtains information about, presented media objects, including identification tags, if present, for collection by researchdata collection agents1500, whether or not a researchdata collection agent1500 is actually present. In an alternate implementation, thecooperative media handler1700 can transmit all information obtained about a media object directly to themedia research controller300 and perform all other tasks associated with the researchdata collection agents1500. In other words, thecooperative media handler1700 can directly serve as the software agent of themedia research controller300, without the need of an intermediary researchdata collection agent1500, as would be apparent to a person of ordinary skill.
• In one embodiment, themedia handler1700 is provided with one or more remotely configurable settings which may be dynamically specified by the researchdata collection agent1500 or themedia research controller300 to help filter out information which is not of interest to the researchdata collection agent1500. Alternatively, themedia handler1700 can query its host to identify the information of interest to the researchdata collection agent1500. If, however, the host does not want to inform themedia handler1700 that no researchdata collection agent1500 is present, the host might have themedia handler1700 pass everything. Thus, themedia handler1700 preferably reports information according to instructions received from the researchdata collection agent1500 via its host.
• As shown inFIG. 17, acooperative media handler1700 is initiated by itshost1600′, such as the illustrative Internet browser, upon receipt of a media object having a media type that is compatible with the media handler. Thereafter, themedia handler process1700 decodes and parses the media object duringstep1705, in a known manner. A test is then performed duringstep1710 to determine if there is media object information associated with the media object, such as an identification tag, metadata or contextual information. As previously indicated, metadata may include program identification information and the corresponding media type and contextual information may include the URL and other information indicating the source of the corresponding media object, or the context in which it was presented to the end user. If it is determined duringstep1710 that there is media object information associated with the media object, then a test is performed duringstep1785 to determine if a message has already been created for the researchdata collection agent1500. If it is determined duringstep1785 that a message has already been created for the researchdata collection agent1500, then program control proceeds directly to step1795. If, however, it is determined duringstep1785 that a message has not yet been created for the researchdata collection agent1500, then a message is created duringstep1790.
• Thereafter, the media object information identified duringstep1710 is added to the agent message duringstep1795. Thereafter, a further test is performed duringstep1720 to determine if there is media to present to a user. If it is determined duringstep1720 that there is no media to present to a user, then program control proceeds directly to step1740. If, however, it is determined duringstep1720 that there is media to present to a user, then a conventional presentation process is performed duringstep1725 to present the media. A test is then performed duringstep1730 to determine if there is presentation information to report to the researchdata collection agent1500. As previously indicated, presentation information may include the language and format in which the media object was presented to the panel member, or other information indicating how the object was presented to the panel member, when alternative presentations are possible. If it is determined duringstep1730 that there is no presentation information to report to the researchdata collection agent1500, then program control proceeds directly to step1760. Tf, however, it is determined duringstep1730 that there is presentation information to report to the researchdata collection agent1500, then the presentation information identified duringstep1730 is added to the message duringstep1735.
• A test is performed duringstep1740 to determine if user interactions are received during the presentation of the media object. If it is determined duringstep1740 that user interactions are received during the presentation of the media object, then program control proceeds directly to step1760. If, however, it is determined duringstep1740 that user interactions are received during the presentation of the media object, then the user interactions are received and processed duringstep1745. A test is then performed duringstep1750 to determine if there is user interaction information to report to the researchdata collection agent1500. As previously indicated, user interaction information may specify how the panel member interacted with the object, including whether the panel member zoomed in on portions of the media object, or rotated the object, as well as the size of the media object and any user inputs.
• If it is determined duringstep1750 that there is no user interaction information to report to the researchdata collection agent1500, then program control proceeds directly to step1760. If, however, it is determined duringstep1750 that there is user interaction information to report to the researchdata collection agent1500, then the user interaction information identified duringstep1750 is added to the message duringstep1755.
• A test is performed duringstep1760 to determine if the presentation of the media object is finished. If it is determined duringstep1760 that the presentation of the media object is not finished, then program control will return to step1705 to continue processing in the manner described above. If, however, it is determined duringstep1760 that the presentation of the media object is finished, then a test is performed duringstep1765 to determine if there is a message to be sent to the researchdata collection agent1500. If it is determined duringstep1765 that there is a message to be sent to the researchdata collection agent1500, then the message is completed duringstep1770, for example, by adding media handler identifying information, and a final time and date stamp. Thereafter, the completed message is sent to the researchdata collection agent1500. Program control will then terminate duringstep1780.
• It is to be understood that the embodiments and variations shown and described herein are merely illustrative of the principles of this invention and that various modifications may be implemented by those skilled in the art without departing from the scope and spirit of the invention.

Claims (1)

1. A method of measuring the electronic media usage of an individual by a media research entity, including the steps of:

defining at least one specialized task useful to the media research entity for a media handler provided by a second entity to perform, the at least one specialized task concerning a handling of electronic media by the media handler;

tracking the individual with at least one software agent of the media research entity, wherein at least one of the software agents obtains a product of the specialized task performed by the media handler via at least one defined interface that provides interoperability between the media handler and the at least one software agent; and

receiving from at least one of the software agents the product of the specialized task.

Images