Home
Patent Search
IMT Blog
REGISTER
|
SIGN IN
United States Patent Application
20020161745
Kind Code
A1
Call, Charles Gainor
October 31, 2002
Methods and apparatus for using the internet domain name system to disseminate product information
Abstract
Methods and apparatus for disseminating over the Internet product information produced and maintained by product manufacturers using existing universal product codes (bar codes) as access keys. A cross-referencing resource preferably implemented by the existing Internet Domain Name System (DNS) receives Internet request messages containing all or part of a universal product code value and returns the Internet address at which data or services relating to the identified product, or to the manufacturer of that product, may be obtained. By using preferred Web data storage formats and protocol which conform to XML, XLS, XLink, Xpointer, RDF and Web service standards specifications, product and company information may be seamlessly identified, retrieved and integrated with information from other sources. A "web register" module can be employed to provide an Internet interface between a shared sales Internet server and an otherwise conventional inventory control system, and operates in conjunction with the cross-referencing server to provide detailed product information to Internet shoppers who may purchase goods from existing stores via the Internet.
Inventors:
Call; Charles Gainor
(West Yarmouth, MA)
Correspondence Name and Address:
68 HORSE POND ROAD
CHARLES G. CALL
WEST YARMOUTH
MA
02673-2516
US
Series Code:
139421
Filed:
May 6, 2002
U.S. Current Class:
707/1
U.S. Class at Publication:
707/1
Intern'l Class:
G06F 007/00
Claims
What is claimed is:
1. A new use for the Internet Domain Name System which comprises, in combination, the steps of: establishing a plurality of separate hosts connected to the Internet, each of said hosts providing data relating to one or more products, each of said products being designated by a unique universal product code value, and each of said hosts being designated by a unique Internet address, storing cross-references in the Internet Domain Name System, each of said cross-references establishing a relationship between a domain name that incorporates at least a predetermined portion of one of said product code values and a corresponding Internet address of the host computer that provides data relating to the product designated by said given one of said product code values, and retrieving data via the Internet that relates to a particular product designated by a particular universal product code value by performing the steps of: transforming said particular universal product code value into a destination domain name that incorporates at least said predetermined portion of said particular one of said universal product code values, using the Internet Domain Name System to convert said destination domain name into the destination Internet address that corresponds to said destination domain name, and retrieving data relating to said particular product from the host designated by said destination domain name.
2. The new use for the Internet Domain Name System set forth in claim 1
wherein said data relating to said particular product is expressed in Extensible Markup Language.
3. The new use for the Internet Domain Name System set forth in claim 2
wherein said specific resource is a Web service provided by said host designated by said domain name.
4. The new use for the Internet Domain Name System set forth in claim 1
wherein said step of retrieving data relating to said particular product from the host designated by said destination domain name comprises, in combination, forming a Universal Resource Identifier which includes said destination domain name and further specifies a specific resource provided by said host designated by said domain name, transmitting a data request message containing said Universal Resource Identifier to said destination Internet address, and receiving response data transmitted by said host designated by said domain name in response to said data request message.
5. The new use for the Internet Domain Name System set forth in claim 4
wherein said step of forming a Universal Resource Identifier which includes said destination domain name and further specifies a specific resource provided by said host designated by said domain name includes the substep of retrieving resource identification data from said host designated by said domain name.
6. The new use for the Internet Domain Name System set forth in claim 5
wherein said resource identification data is expressed in the Web Services Description Language.
7. The new use for the Internet Domain Name System set forth in claim 5
wherein said resource identification data is expressed in the Web Services Inspection Language.
8. The new use for the Internet DomainName System set forth in claim 1
wherein said particular universal product code value is a Global Trade Item Number.
9. A system for disseminating information relating to a plurality of different manufactured products, each of said products being designated by a unique universal product code value, said system comprising, in combination: a plurality of different host computers each comprising storage means for accepting and storing information about at least a given one of said products and each having an entry point connection to the Internet that is uniquely identified by an Internet protocol address for receiving and responding to Internet messages requesting information relating to said given one of said products, registration means for storing cross-references in the Internet Domain Name System, each of said cross-references specifying an association between a product code domain name and a corresponding Internet protocol address wherein said product code domain name is an alphanumeric character string whose content may be completely derived from a specific universal product code value using a predetermined transformation process, and wherein said corresponding Internet protocol address identifies the particular entry point connection that receives and responds to Internet messages requesting information relating to a product designated by said specific universal product code value, and means for retrieving information about a selected one of said products that is designated by a selected one of said universal product code values comprising: means for deriving a target product code domain name from said selected one of said universal product code values using said predetermined transformation process, means for transmitting said target product code domain name to the Internet Domain Name System to obtain the corresponding target Internet protocol address specified by one of said cross-references, and means for transmitting an Internet message requesting information relating to said selected one of said products to said corresponding target Internet protocol address.
10. The system set forth in claim 9 wherein said means for retrieving information about a selected one of said products that is designated by a selected one of said universal product code values further comprises means for forming a Universal Resource Identifier which includes said target product code domain name and for including said Universal Resource Identifier in said Internet message requesting information.
11. The system set forth in claim 10 wherein said means for forming a Universal Resource Identifier which includes said destination domain name further comprises means for retrieving resource identification data from said host designated by said product code domain name.
12. The system as set forth in claim 10 wherein said Universal Resource Identifier specifies a Web service.
13. The system set forth in claim 12 wherein said means for forming a Universal Resource Identifier which includes said destination domain name further comprises means for retrieving resource identification data from said host designated by said product code domain name.
14. The system set forth in claim 13 wherein said resource identification data is expressed in the Web Services Description Language.
15. The system set forth in claim 13 wherein said resource identification data is expressed in the Web Services Inspection Language.
16. The system set forth in claim 9 wherein said particular universal product code value is a Global Trade Item Number.
17. A system for distributing information via the Internet relating to a manufactured product designated by a specific univeral product code value, said system comprising, in combination: means for storing said information at a computing host accessible via the Internet at a specific Internet protocol address, a first source of a specific universal product code value, means coupled to said first source for deriving a domain name character string from said specific universal product code value in accordance with a predetermined translation process, means for storing a cross-reference in the Internet Domain Name System, said cross-reference specifying an association between said domain name character string and said specific Internet protocol address, and utilization means coupled to the Internet comprising: a second source of said specific universal product code value, means coupled to said second source for deriving a target domain name character string from said specific universal product code value in accordance with said predetermined translation process, means for transmitting said target domain name character string to the Internet Domain Name System to obtain said specific Internet protocol address, and means for transmitting a request message via the Internet to said specific Internet protocol address to obtain said information relating to said manufactured product designated by said specific universal product code value.
18. The system set forth in claim 17 wherein said means for retrieving information about a selected one of said products that is designated by a selected one of said universal product code values further comprises means for forming a Universal Resource Identifier which includes said target domain name character string and for including said Universal Resource Identifier in said Internet message requesting information.
19. The system set forth in claim 18 wherein said means for forming a Universal Resource Identifier which includes said target domain name character string further comprises means for retrieving resource identification data from said host designated by said target domain name character string.
20. The system as set forth in claim 18 wherein said Universal Resource Identifier specifies a Web service.
21. The system set forth in claim 20 wherein said means for forming a Universal Resource Identifier which includes said target domain name character string further comprises means for retrieving resource description data from said host designated by said target domain name character string.
22. The system set forth in claim 21 wherein said resource description data describes said Web service and is expressed in the Web Services Description Language.
23. The system set forth in claim 21 wherein said resource description data identifies said Web service and is expressed in the Web Services Inspection Language.
24. The system set forth in claim 20 wherein said particular universal product code value is a numeric Global Trade Item Number.
Description
[0001] This is a continuation in part of U.S. application Ser. No. 09/626,186 filed on Jul. 27, 2000 which was a division of U.S. application Ser. No. 09/316,597 filed on May 21, 1999 (now U.S. Pat. No. 6,154,738 issued on Nov. 28, 2000), which was a continuation in part of U.S. application Ser. No. 09/049,426 filed on Mar. 27, 1998 entitled "Methods and Apparatus for Disseminating Product Information via the Internet" ( now U.S. Pat. No. 5,913,210 issued on Jun. 15, 1999). This application claims the benefit of the filing dates of each of the of foregoing applications, and incorporates the disclosure of each of the foregoing patents and applications herein by reference..
FIELD OF THE INVENTION
[0002] This invention relates to methods and apparatus for transferring requests for specific information to preferred sources of that information on the Internet.
BACKGROUND AND SUMMARY OF THE INVENTION
[0003] Manufacturers must provide information about their products to resellers, consumers, and others. Resellers need product information to select, promote and support the products they distribute. Consumers need information about available products to make informed buying choices. Advertisers, product analysts, manufacturer's representatives, shippers, and others also need information about the goods with which they deal.
[0004] Under current practices, product information typically originates with manufacturers and is primarily distributed in conventional print media advertising and product packaging. This information is often incomplete, difficult to update, and available only to a limited distribution. While the advent of the World Wide Web has permitted manufacturers to make detailed, up-to-date product information available via the Internet, the information describing a specific product is often difficult to locate, particularly when the URL (uniform resource locator) of the manufacturer's website is not known.
[0005] It is a general object of the present invention to transfer a request for information specified by an identifier, such as a product code, to a preferred source of that information, such as an Internet information resource devoted to the product specified by the product code which is created and maintained by the product's manufacturer.
[0006] The preferred embodiment of the present invention employs an Internet resource, called a "product code translator" that is preferrably implemented using the existing Internet Domain Name System for storing cross-references between universal product codes identifying specific products and Internet addresses specifying the locations at which information about these products may be obtained. The cross-references specify the universal product codes assigned to the participating manufacturers, such as Global Trade Item Numbers in the EAN-UCC system widely used in retail stores for barcode scanning at checkout counters, and the Internet addresses where information can be obtained about the products designated by those codes.
[0007] A preferred embodiment of the present invention takes the form of a system for disseminating information relating to different manufactured products each of which is designated by a unique universal product code value. The system employs different host computers for storing information about the products. Each computer supports an entry point connection to the Internet that is uniquely identified by an Internet protocol address for receiving and responding to Internet messages requesting information relating to one or more of the products. The server administrators register cross-references in the Internet Domain Name System. Each cross-reference specifies an association between an alphanumeric product code domain name and a corresponding Internet protocol address. The alphanumeric product code domain name is a character string whose content may be completely derived from a specific universal product code value using a predetermined transformation process. The cross-referenced corresponding Internet protocol address identifies the particular entry point connection that receives and responds to Internet messages requesting information relating to a product designated by the specific universal product code value.
[0008] Thereafter, any user connected to the Internet may retrieve information or invoke services relating to a product designated by a selected universal product code value. The user derives a target product code domain name character string from the selected universal product code value using the predetermined transformation process, and transmits the target product code domain name to the Internet Domain Name System to obtain the corresponding target Internet protocol address specified by one of the registered cross-references. The user then transmits an Internet message requesting information relating to the selected products to the target Internet protocol address.
[0009] The request message contains a Universal Resource Identifier (URI) which includes the target product code domain name character string and which identifies the specific resource reqeusted. The host designated by the Internet protocol address interprets the URI to provide the data or service specified by the URI. The identity and invocation mechanism (binding specification) for resources available at a given host may be specifed by resource description files which may be accessed at predetermined location at the host specified by the target domain name. These resource description files may advantageously include Web Service Inspection Language and Web Service Description Language documents which identify and describe product and company related Web services available at each host.
[0010] These and other objects, features and advantages of the present invention will be made more apparent through a consideration of the following detailed description of a preferred embodiment of the invention. In the course of this description, frequent reference will be made to the attached drawings.
BRIEF DESCRIPTION OF THE DRAWINGS
[0011] FIG. 1 is a schematic block diagram illustrating various instrumentalities which make use of the invention interconnected via the Internet;
[0012] FIG. 2 is a diagram illustrating the interrelationship of the principle data structures used to implement a product code translator of the type contemplated by the invention;
[0013] FIGS. 3 and 4 are a flow diagram depicting the operation of a CGI program which implements the product code translator, responding to request HTTP messages containing universal product codes and redirecting those request messages to the URLs where information about designated products may be found;
[0014] FIG. 5 is a data flow diagram illustrating the manner in which a web browser interacts with a server which acts as a product code translator by redirecting links on a merchant's web page to product information made available by the manufacturer;
[0015] FIG. 6 is a data flow diagram illustrating a further embodiment of the invention in which a web browser interacts with a domain name server which serves as a product code translator, redirecting links on a merchant's web page to XML data provided by a shared sales server used by several merchants,, the XML data from the manufacturer being displayed in accordance with an XSL stylesheet specification unique to each merchant, and the online merchant functions being implemented by the shared sales server which is connected to the conventional inventory control system operated by each of the merchants;
[0016] FIGS. 7 and 8 are a block diagrams illustrating the principle components of a typical retailer's inventory control system and the interface between that system and a shared sales server which maintains a parallel but more limited inventory database used during on-line sales transactions; and
[0017] FIGS. 9-12 are flow charts illustrating example procedures which employ the principles of the invention to disseminate information and services relating to manufactured products.
DETAILED DESCRIPTION
[0018] The preferred embodiment of the present invention takes advantage of three existing and highly successful technologies: the Internet, the Internet Domain Name System (DNS), and the EAN-UCC universal product code system. In accordance with the present invention, the Internet and the DNS are employed to provide low cost, worldwide, bidirectional communication which enables product information to be requested by and sent to any person or firm using one or more universal product codes as information access keys. The universal product code system is in widespread use to uniquely identify each of the thousands of different suppliers and millions of different items that are warehoused, sold, delivered and billed throughout commercial channels of distribution. In accordance with the invention, the Internet Domain Name System is used in a new way to act as a product code translator for resolving domain names containing all or part of a universal product code into a cross- referenced Internet addresses from which information and services relating to the products specified by those product codes may be obtained.
[0019] The present invention enables the retrieval of information about products from the source of those products, typically the manufacturer, by those who need that information, such as resellers and consumers. In accordance with the invention, any person or firm having access to the Internet and knowing the universal product code for a product may obtain information about that product from the participating manufacturer which supplies that product. The system employs the DNS as a product code translator, which is implemented by a plurality of distributed name servers but which is illustrated by the single resource seen at 101 in FIG. 1. The DNS product code translator is accessed via the Internet to perform a translation of domain names containing specified universal product codes into the corresponding Internet addresses from which information about the designated products can be obtained.
[0020] The product code translator 101 stores cross-references between product codes and Internet addresses. The product codes and the Internet addresses are provided by or on behalf of participating manufacturers and suppliers, such as the manufacturer illustrated at 103 in FIG. 1. These cross-references may then be retrieved from the cross-reference resource 101 by resellers, prospective buyers, as illustrated by the distributor 105, the retailer 107 and the customer 109 seen in FIG. 1. Internet Service Providers, as illustrated by the ISP 111 in FIG. 1, may also utilize the data provided by the product code translator 101 to provide a variety of services and functions.
[0021] Before further describing how these entities function within the system, it will be useful to clarify some of the terms which will be used in this specification:
[0022] The term "universal product codes" (lower case) is used to indicate standardized industry or inter-industry codes used to designate items, packages and services made, used, leased or sold in commerce. The term thus includes the Universal Product Codes ("U.P.C.s") used by suppliers in the United States and Canada and managed by the Uniform Code Council, Inc., 8163 Old Yankee Road, Dayton, Ohio 45458; the EAN codes used by suppliers outside the U.S. and Canada under the general direction of EAN International, rue Royale 145, 1000 Bruxelles--Belgium; and any other multi-industry or single industry standard product designation system. The fourteen-digit Global Trade Item Number or "GTIN" numbering system is a preferred form of universal product numbering system and subsumes the U.P.C. codes as well as the EAN-13 and SCC-14 codes. In database applications, each of the code values is stored in a standardized form, right justified and zero filled, in a 14 digit field. The final, lowest-order digit of the code is a check digit which is calculated from the higher order digits using a standard algorithm established by the EAN--UCC System (available at http://www.ean-int.org/cdcalcul.html).
[0023] The term "manufacturer" will be used to refer to manufacturers, suppliers, vendors, licensors and others to whom sets of universal product codes have been assigned, or their agents. Typically, this assignment takes the form of the designation of a particular value for a portion of the universal product code which is reserved for exclusive use by a particular manufacturer. For example, the entity to which a specific six-digit "company-identifier" portion of a 12-digit numeric U.P.C. code has been assigned is a "manufacturer" as that term is used in this specification. The portion of the universal product code that specifies the product's manufacturer is variously called the "company prefix," the "company code," and the "company identifier."
[0024] The term "product" is used to refer to a kind of item which is uniquely identified by a single universal product code, as opposed to a specific individual item of that kind. For example, a specific U.P.C. code is assigned by the manufacturer, Hershey Foods Corp., to different U.P.C. code is assigned to the same brand of peanut butter when packaged in containers of a different size (a different "product").
[0025] The term "Internet address" will be used to refer to the all, or a significant part of, a reference to a resource on the Internet. Such a reference may take the form of a numerical IP address or an alphanumeric Uniform Resource Locator ("URL") which may identify a file on a specified machine, a database query, a specific command output, or some other accessible Internet resource. Thus, the term "Internet address" includes such things as a specific 32-bit address of a specific computer connected to the Internet, written in decimal as "123.040.212.002". Alternatively, the term "Internet address" may refer to a domain name such as "patentsoft.com" which can be resolved into a numerical IP-address using a domain name server. In addition, an "Internet address" may take the form of the URL of a file accessible via the Internet, such as "ftp://www.sample.com/directory/filename.xxx"; a URL identifying a query processing script with passed parameters, such as "http://xxx.yyyy.com/cg- i/search%01234567890123"; or an email address such as "847563@manufacturer.com".
The Product Code Translator
[0026] The product code translator seen at 101 in FIG. 1 performs two primary functions illustrated in FIG. 2: (1) its registration handler 203
accepts cross-references submitted by manufacturers which relate their assigned universal product codes to associated Internet addresses where information relating to their products may be obtained, and (2) its query handler 204 accepts queries via the Internet 205, each query including all or part of one or more universal product codes, and returns the Internet addresses which can be used to obtain information about the products identified by those codes. The product code translator 101 may also advantageously perform other functions, examples of which are described below.
[0027] In the preferred embodiment of the invention, the existing Internet Domain Name System, including its standard methods of accepting domain name registrations from registrants and for maintaining a distributed, hierarchical mechanism for organizing the name space of the Internet, is used to advantage to perform product code to Internet address translation. However, such functions can also be performed by other means, as illustrated by the example described in conjunction with FIG. 2
of the drawings.
[0028] As seen in FIG. 1 and 2, the product code translator 101 may be advantageously implemented by a server computer which stores information in a relational database consisting of the tables depicted in FIG. 2. If desired, the product code translator 101 may be implemented with a plurality of "mirrored" servers at different locations, or clustered servers at the same location containing the same cross-referencing data to share the processing burden and provide redundant fault-tolerant reliability. In addition, different servers or sets of mirrored or clustered servers may be used to process different assigned subsets of the gamut of universal product codes. Whether one or many servers is used, each may be preferably implemented using conventional server hardware and conventional server operating system software, such as Microsoft NT Server, Netscape Application Server, SCO Unixware, Sun Enterprise Server, and the like.
[0029] The registration handler process 203, which may be implemented on a server which stores cross-references, or on a separate server operated by a central registration authority, receives each registration submission via the Internet 205 to create an incoming registration data illustrated by the data template record 207. The registration template record 207
includes several fields: an Company-ID field which holds the company-identifier portion of a universal product code in EAN format; a URL field which holds the Uniform Record Locator constituting the "base address" at which information can be retrieved about products designated by those universal product codes, company information fields which include the company's name, mailing address and email address fields so that the manufacturer submitting the registration can be identified and contacted; and a date field which specifies the date upon which the registration was first made.
[0030] The registration handler 203 may obtain the submitted data needed to create the registration template record in a variety of ways, such as accepting a HTML web page form completed and submitted via the Internet by a registrant, processing an incoming email message containing the necessary information, or receiving the needed information by telephone or regular mail.
[0031] The registration handler process preferably incorporates a mechanism or procedure for insuring that the registrant has the authority to create and alter the information being supplied. A variety of methods for enhancing the security of the registration process may be employed, including the issuance of a password at the time a range of universal product codes is first registered, with the requirement that the same password be thereafter provided by anyone who seeks to alter the information originally provided with respect to any product code within that previously registered range. The registration procedure may also require each registrant to provide an email contact address to which an email message of predetermined content is sent after the initial registration form is completed, to which the registrant must respond within a predetermined time to verify the registration. Any attempt thereafter to change the contact email address results in a message being sent to the originally registered email contact address advising that an attempt is being made to alter the registration. Finally, email confirmation may be requested from the email address registered with InterNIC for the URL to which universal product codes are to be linked. This step confirms that the person attempting a registration in fact has authority to link to that host computer and provides an additional safeguard against unauthorized submissions.
[0032] Instead of maintaining a user name and password database, the registration handler can use a database of certificates, such as Certificate Server available from Netscape, to create, sign, and manage certificates for all participating manufacturers, configuring other servers to accept only authorized user certificates. A scalable database may be used to store the status of each certificate, and the issuance and revocation of certificates can be centrally administered from the product code translator or a separate registration authority. Similar password, certificate or digital signature protection schemes may be used to provide access to certain data or to data in certain forms only to authorized requesters.
[0033] The information contained in the incoming registration template 207
is used to create records (rows) in three separate tables in the relational database: a company table 211, a URL table 213 and a cross-reference table 215. As seen in FIG. 2, the company table 211
includes a numerical company number field CoNo which is also present in the cross-reference table 215 so that each cross-reference table row can be related to a particular company description record in the company table which has the same CoNo value. The key field CoNo establishes a one-to-many relationship between the company table 211 and the cross-reference table 215 since a participating company identified by a unique CoNo value may register more than one set of universal product codes, potentially associated with different IP-addresses, requiring more than one row in the cross-reference table 215.
[0034] The Company-ID field in the registration template record is used to complete two fields, FromCode and ThruCode, in a row in cross-reference table 215. These fields specify a range of one or more consecutive universal product codes. Both of these two fields preferably stores a 64-bit integer which specifies a 14 decimal digit universal product code called a Global Trade Item Number ("GTIN") drawn from the global pool of 14-digit numbers which includes the U.P.C., EAN-13 and SCC-14 codes. In this way, all three coding systems can be accommodated by the cross-reference table 215; for example, a 12-digit U.P.C. number 7 12345
12345 9 is the same as the 13 digit EAN number 07 12345 12345 9 and the same as the 14 digit SCC-14 number 0 07 12345 12345 9. If the U.P.C. six digit company identifier 7 12345 is specified in the registration template Company-ID field, the FromCode field of the cross-reference table record would be loaded with the number 71234500000 to specify the lowest valued universal product code cross-referenced to the corresponding IP address in the IPAdr field of the table 215, and the ThruCode field would be loaded with 71234599999 to specify the highest valued universal product code cross-referenced to that IP address. The use of the low-value/high-value range specification fields in each row of the cross-reference table 215 permits different ranges of universal product codes having the same Company-ID value to be associated with different Internet addresses in the IPAdr field of the cross-reference table 215, thus enabling a single manufacturer having a single assigned Company-ID value to store information about different products designated by different sets of its universal product codes on different Internet servers, or to cross-reference non-continuous sets of universal product codes to the same or different servers. Note further that a manufacturer need not cross-reference all of its available assigned universal product codes, but may omit unused codes or codes designating products for which no information is to be made available.
[0035] The IP-address field in each row of the cross-reference table 215
holds a 32-bit IP address used to route Internet data packets to a destination computer using the TCP/IP protocol. The 32-bit IP address value in the cross-reference table 215 is obtained from the IP address field of the URL table 213, and that 32-bit address value is prefetched by querying a conventional domain name server (DNS) seen at 217 assigned to the cross-reference resource 101. The DNS 217 translates the alphanumeric URL in the URL field of the URL table 213 into the current 32-bit IP address used by Internet routers to guide data packets to the proper destination computer. The alphanumeric URL in the URL field of the URL table 213 is supplied via the registration template 207 when Internet location of the manufacturer's product description data is supplied during the registration process..
[0036] The separate URL table 213 has a one-to-many relationship to the cross-reference table 215 and uses the 32-bit IP address value as the relational key. This arrangement allows a single URL base address to be shared by a plurality of different manufacturers. Thus, for example, a single Internet service provider (ISP) may act as a shared Internet resource for storing data about a products originating from many different manufacturers. It is accordingly unnecessary for each manufacturer to operate its own server or have its own assigned URL. Instead, a manufacturer may place its product descriptions on any server having an assigned Internet address. Note that it is further unnecessary for the manufacturer to have, or supply, an assigned URL rather than a numerical IP address; however, since corresponding URL's are ordinarily available and easer to remember, and because it may be desirable to later change numerical IP addresses while retaining the same URL, the use of URL's for registration is preferred.
[0037] Note also that, because URL/IP address assignments may be added, altered or deleted on a daily basis by the URL assignment authority, updates to the DNS tables should also be reflected by automatic updates to the cross-reference table IP-Address fields. In this way, a change in URL/IP address assignments propagated in the DNS system require no additional action on behalf of the manufacturers to insure the continuing ability of the product code translator to produce the appropriate new IP addresses in response to universal product code queries. If, as discussed later, the Internet domain name system itself is used as the product code to IP address translation mechanism, updating two tables would be unnecessary.
[0038] With the foregoing as background, the registration and query/response finctions performed by the product code translator may by summarized as follows: each participating manufacturer, or someone acting on its behalf, submits a registration which generates an incoming registration template 207 containing information about the registering manufacturer, including an identification of the universal product codes which designate products for which information is to be made available, together with the URL which specifies the Internet resource which will make that product information available.
[0039] The supplied URL is stored in the URL Table 213 and converted into a numerical IP address in the IPAdr field of the URL Table 213 using an available domain name server 217. This 32 bit IP address is stored in the IPAdr field of the cross-reference record (row) in table 215, along with a specification of the universal product codes of the products described by information which is available at this IP address, the range of codes being specified by the values stored in the FromCode and ThruCode fields in the new record in cross-reference table 215.
[0040] When an incoming query is received by the query handler 204, a table lookup function is performed by searching the cross-reference table 215 for a row record or records which specify a set of universal product codes which include the code or codes specified by the query. If matching row(s) are found, the IP-address(es) found in the matching row(s) are returned to the query submitter; otherwise, a special code (such as a zero valued IP Address) is returned to indicate that information for the product code(s) of interest has not been registered.
[0041] The Internet resource which acts as the product code translator can additionally perform some or all of the following additional functions:
[0042] It can respond to a request for information about a particular participating manufacturer and return to the requester the information in the company table 211 as well as the specification of all of the registered universal product codes assigned to that participating manufacturer and the IP address (or URL) of the location where further information on the products designated by the registered universal product codes may be obtained.
[0043] The product code translator can respond to a query containing a designation of one or more universal product codes by identifying the email address of the manufacturer. The product code translator, or any other computer which obtains cross-references between universal product codes and email addresses from the product code translator, may act as an SMTP forwarding agent; for example, forwarding email which contains a designation of universal product code from a sender to an email address designated by the manufacturer of the product designated by that code. Alternatively, resellers and others may obtain email addresses from the product code translator which can be included in "mailto:" hypertext links in product listings, allowing a webpage viewer to display and complete a blank email request for information which is routed directly to the manufacturer's designated email address. The email address returned in response to a request may be a standard email address such as "upcinfo@domainname" where "domainname" is the domain name portion of the URL supplied by the manufacturer, in which case the specific universal product code would, by convention, be supplied as all or part of the "subject" of the email message sent to that address, enabling the manufacturer to identify the specific product which is the subject of the inquiry.
[0044] The product code translator can further provide all or part of the information from company table 211 to provide information about the manufacturer(s) to whom registered universal product code or codes are assigned. Note that, in general, the information which is required or recommended for inclusion with other company information may be limited to that data necessary or desirable to enable the code translator to perform its functions. Other information about the company may simply be placed in an allocated namespace on the manufacturer's server.
[0045] The cross-referencing utility can provide the entire contents of its URL table to a requesting computer, such as a search engine which can then perform conventional "web crawler" indexing of the websites specified by the listed URLs and/or IP addresses, thereby generating complete or partial indexes to all or less than all of the products whose product description locations have been registered with the product code translator.
[0046] The cross-reference table 215 can be scanned by the product code translator in response to a request for certain universal product codes only; for example, books are assigned EAN numbers which always begin with the prefix number 978 before the company-id value (publisher designation) portion of the International Standard Book Number (ISBN) which makes up the remainder of the EAN number for each book, allowing all IP addresses for information about books to be provided by the cross-referencing server to create a database or index to book information. In the same way, the cross-reference table could be scanned for product codes assigned to a particular manufacturers (e.g. book publishers) to provide a more focused index.
[0047] The product code translator, as noted earlier, may facilitate the registration process by providing a website from which HTML registration form pages may be fetched, displayed and completed using a conventional web-browser program. In addition, the product code translator may advantageously make query forms available to permit information request queries to be made directly, as well presenting informational web pages which provide instructions and guidelines for registration procedures, recommendations for the storage of information on registered information resources, and instructions and downloadable software which may be used to simplify and facilitate searches and functions performed at other resources on the Internet which utilize the services provided by the product code translator.
[0048] When, as discussed later, the Internet Domain Name System is employed to perform product code or company code translation to an Internet address, the authorized DNS registration authorities can provide informational and registration services to registrants.
Information Publication by Manufacturers
[0049] The present invention provides significant advantages and opportunities to manufacturers. Information which manufactures now distribute in other ways can be made immediately available to those who need or desire that information. Examples include text and graphics which describe and promote the sale of each product to potential buyers; product labeling information, some of which may be required to be made available to potential buyers such as product weights and volumes, ingredients, nutritional facts, dosage and use instructions, some or all of which is now included on product packaging and which can be reproduced as mixed text and graphics HTML page for viewing by distributors, retailers, advertisers, catalog publishers, potential customers and purchasers; logos, photographs of products, and other graphics files in a variety of resolutions for use by both electronic and print rendering to promote product sales, usage and support. Instructional and service information including self-help diagnostics and recommended solutions, product part lists and ordering information, product return procedures, current pricing information, identification of dealers and distributors, warranty and guarantee explanations, and support telephone numbers may be provided.
[0050] The scope and content of the information each manufacturer makes available is completely under the control of that manufacturer. In order to make this information accessible in a standard way, it is desirable that the manufacturer conform to standard resource naming conventions so that interested parties which obtain the manufacturer's registered IP address from the product code translator can find the desired information at this address. This naming convention may take numerous forms, and the following are merely exemplary:
[0051] A root directory named "upcinfo" may be created on each registered computer, and a subdirectory having a name which is the universal product code (expressed as a zero-filled, right-justified fourteen digit number) is created to hold the information concerning the product designated by that universal product code. At the minimum, each such directory includes a product home page named "info.html" which typically provides whatever general product information the manufacturer wishes to place before all interested parties. This product home page may link to additional information related to the product on other pages when appropriate.
[0052] By way of example, a product HTML home page for a book would be created by the book's publisher and could include a complete bibliographic citation identifying the title, author, book type (hardcover, paperback, etc.), recommended retail price, ISBN number, number of pages, publication year, etc. In addition, each book's home page might include an imbedded thumbnail image (JPEG or GIF file) of the book jacket, and links could be added enabling the viewer to see additional information concerning that book when available, such as an interview with the book's author, quotes from favorable reviews, book group discussion guides, a table of contents or introductory chapter, etc.
[0053] Thus, information uniquely formatted to best advantage by the manufacturer could be made available by accessing a single URL, having the same form for all products, formed by combining the IP-address obtained from the standard by concatenating a prefix and suffix. The suffix has the form:
[0054] "/upcinfo/12345678901234/info.html"
[0055] where the numerical part of the suffix is the universal product code directory name, and where the suffix is appended to the at the end of the prefix of the form:
[0056] "http://23.123.40.198"
[0057] consisting of the protocol identifier "http://" and by the 32-bit IP address from the product code translator written in its standard four decimal number format (four three digit numbers separated by periods, each of which is a value in the range 0-255 representing the binary value of one of the four 8-bit bytes making up the 32-bit IP address).
[0058] If a manufacturer stores product information in a database, the product directories and the HTML and other data files which are to be made available can be rewritten automatically under program control as the information in the manufacturer's database changes. Alternatively, a request for a particular "file," such as the web page designated "/upcinfo/product-code/info.html," may be intercepted at the manufacturer's server and handled as a database query to which the server responds by dynamically writing an HTML response page using information in the manufacturer's product database. Available database program development tools, such as Microsoft's Access 97 and Borland's Delphi 3.0, include database manipulation tools which allow programs to be readily written which automate the process of generating product description pages from an existing database.
[0059] The present invention may be employed to allow the same information found on a product's packaging to be made available to prospective online buyers. For food products, for example, in addition to the product name, logo and promotional materials, such existing packaging information typically includes an ingredient list, nutrition facts, serving suggestions and directions, recommended recipes, and product guarantee information. Over the counter pharmaceuticals, cosmetics and health care products often include further information, such as specific directions on dosage and use, warnings and instructions in the event of misuse, storage and product lifetime information, and active ingredient specifications. Frequently, this valuable information is printed on a product container or container insert which is discarded shortly after purchase. By making this information readily available to purchasers and end-users over the Internet, the manufacturer can help insure that such valuable product information, some of which may be legally required on the product's packaging, is available to the consumer at the time of an online sale and after the product has been purchased.
[0060] While information of the kind traditionally placed on product packaging already exists and can be converted by the manufacturer into a format suitable for publication on the World Wide Web, and thus made widely available at little cost, the invention allows information in other forms to be provided at low incremental cost. For example, multimedia presentations may be presented to promote, describe and support a product and its uses. User manuals and service documentation can be provided in Adobe Acrobat portable document format or the like for viewing and printing by resellers, service personnel and consumers.
[0061] It is frequently desirable to transfer to another computer data created by the manufacturer which provides limited product description information for each product offered to enable more efficient indexing, cataloging, inventory control, and other applications. By way of example, in the bookselling industry, publishers, distributors, retailers, and libraries often require a database of bibliographic information which consists, for each book, of the book title, author name(s), publisher's name, publication date, type of book (hardcover, paperback, etc.), page count, recommended retail price(s), and ISBN number (which takes the form of a subpart of the EAN universal product code). To the extent the content and format of data records which describe particular classes of products in particular industries and trade groups have been previously adopted and placed in widespread use, those structured data records may advantageously be made available utilizing the present invention. This is preferably achieved in two ways: a data record (file) containing such field- structured information about each product which is designated by a universal product code is placed by the manufacturer in the directory it creates for that product. This structured data record is given a filename indicative of the format used to store the structured data. For example, each directory bearing a name corresponding to the EAN number for a book would preferably contain a file named "biblio.dat" which contains a single structured record containing bibliographic data describing that book.
[0062] In addition, the manufacturer would place a combined file, also called "biblio.dat" in its root ".backslash.upcinfo" directory which contains all of the records for all of the products individually described in the subdirectories which have that structure in a single file. For most manufacturers, these structured data files, both individual record files in the subdirectories and the combined file in the root directory, may be automatically created and updated on a periodic or dynamic basis from the content of the manufacturer's existing database. The use of a single combined file at each server permits multi-manufacturer databases to be created by first retrieving the IP-addresses of all or part of the cross-reference table 215, and then retrieving and merging the combined data files from the "/upcinfo" directories from each identified server. Alternatively, when information about all of a given manufacturer's products of a given type is not desired, the needed individual structured data files can be retrieved from the individual product directories.
[0063] As described later in more detail, the information which the manufacturer makes available can advantageously be stored using the Extensible Markup Language (XML), which is also well suited for providing metadata which defines and describes the meaning of the various kinds of information that can be provided about individual products, groups of products, and the manufacturers and distributors from which those products are obtained.
[0064] This ability to obtain accurate and up-to-date product information from the manufacturer can substantially reduce the cost to resellers, catalog producers, and database vendors which is traditionally incurred in capturing this data by conventional means. For example, a retailer creating a computerized inventory control system for the first time with previously purchased merchandise may use a conventional hand-held barcode scanner to capture the universal product codes from all goods in inventory, and then retrieve complete and accurate product description records for each product via the Internet using the present invention.
[0065] The ability to obtain, update, and verify product description information by accessing manufacturer data can be readily included as callable functions built into inventory control and EDI software used by manufacturers, distributors, and retailers. Institutional "consumers," such as hospitals, government agencies, and libraries, may use the information to build internal databases for internal use.
[0066] The structured records noted above are typically, but not necessarily, copied into a separate database which is thereafter manipulated by the requester. Because each copied database record includes a field containing the universal product code, the ability to obtain and verify data in the remainder of the record from the manufacturer's server is retained. Note that it is possible for the user of the local database to verify, update and add to the product information specified by the universal product code at the time that data is referred to or relied upon. In addition, or in the alternative, the database can be periodically and automatically verified against current data made available by the manufacturer and updated to insure the continued completeness and accuracy of the entire local database.
[0067] The present invention enables a computer connected to the Internet to dynamically retrieve arbitrarily large quantities of data about an individual product when needed. This capability makes it unnecessary, and normally undesirable, to copy "content" into a local database which is not needed for structured indexing and retrieval purposes. Thus, again using books as example products, the local database might consist simply of title, author and publisher information to form a searchable local database. This database could be built by first obtaining all of the IP-addresses for universal product codes beginning with "978" (the EAN prefix for books) from the product code translator, retrieving the combined "biblio.dat" file from the "/upcinfo" directory at each IP-address, and extracting the universal product code, title, author and publisher data from these records to form the desired searchable local database. This database may then be rapidly searched to produce an output listing of all books meeting a specified search criteria, and complete information about each of the identified books can then be obtained using the universal product codes.
[0068] General product information indexes can be also readily be created by means of conventional "web crawler" indexing engines of the type now widely used to index World Wide Web sites. These indexing engines may scan either the product descriptions created by the manufacturer in the form of HTML or multimedia files, or the structured data files containing fielded information, or both. By limiting the scope of the information indexed to the product information data identified by the product code translator, search results produced by these product indexing systems are less likely to be obscured by references to other, less relevant information which happens to employ the term or terms used in a search request.
[0069] The principles of the invention may be applied to particular advantage by online resellers. By making detailed, accurate and up-to-date information about products which are offered readily available to interested prospective buyers, both the reseller and the manufacturer can more effectively promote the offered product to an interested buyer, and the buyer can make a more informed buying decision by obtaining more detailed information which facilitates product comparisons and matching the product's features with the buyer's needs.
[0070] In this regard, it may be noted that small retailers can employ shared software and services, and share access to product information and promotional materials made available by the manufacturer in accordance with the invention, at low costs, enabling even the smallest retailer to offer its entire inventory of products (and more) to its customers at low cost, with each product being fully described and promoted by the materials made available by the manufacturer. Similarly, small manufacturers can effectively describe and promote their products throughout a widespread distribution system by simply placing their available promotional and descriptive materials on an available shared server and registering the assigned universal product codes together with the shared server's address, for distribution by the product code translator, all at minimal cost.
[0071] In addition, the present invention may be used to advantage in combination with Electronic Data Interchange, a standard mechanism for exchanging business documents in standard format between computers. EDI systems typically use value added networks (VANs), such as the networks provided by GE, IBM Atlantis and Sterling, or EDI transfers can be made via the Internet using services such a those provided by EDI Network of Turnersville, N.J. Using EDI, manufacturers make available electronic catalog descriptions of their products being offered for distribution and resale. When a buyer selects products of interest to order from the vendor's catalog, the retailer's computer accesses the vendor's computer to transfer the U.P.C. codes to the retailer's computer without rekeying. The retailer may then issue an EDI 850 purchase order transaction which is sent to the vendor's mailbox. In addition, the EDI system may transfer limited additional information to the retailer, such as suggested retail price. When the products are shipped, an EDI 856 shipping notice is sent to the retailer containing bill of lading information (bill of lading number, carrier and weight), purchase order information, and carton contents using U.P.C. product codes and counts. The vendor also sends an EDI 810 invoice to the retailer in EDI format which enables the retailer to process the invoice and schedule payment either by check or electronic funds transfer, using an EDI 830 remittance advice transaction to give payment details for invoices being paid.
[0072] These EDI transactions enable retailers to not only automate product procurement functions but also to easily maintain an accurate inventory control system in which each product is designated by a universal product code. The present invention may be used to augment an EDI system by providing resellers and consumers with detailed product information and services for any product designated by a universal product code which is made by a participating manufacturer. As described later, a running application program such as an inventory control program may access supporting services from a manufacturer which relate to or support a given product by using the invention to convert the product code designating that product into the URI to which a Web service request is transmitted.
[0073] Internet Service Providers, such as the ISP indicated at 111 in FIG. 1, may provide shared computer services which interoperates with a reseller's inventory control system to provide customers with the information they desire before and after making purchases.
[0074] As seen in FIG. 1, and as previously discussed, a reseller (including both the example distributor 105 and the example retailer 107) may be assumed to have conventional inventory control systems, typically using EDI document processing, which includes in each case inventory data consisting of at least the universal product code for each product and, typically, count numbers indicating quantity on hand, quantity on order, quantity back-ordered, etc. This limited part of the reseller's database can be transferred from the reseller's inventory database (at 105 or 107) to an ISP 111 which serves many resellers but maintains a table of universal product codes for all goods offered by each reseller served, together with the on-hand counts for each code.
[0075] The ISP 111 hosts a website for each reseller served in conventional fashion, typically using a domain name assigned to the reseller. The ISP further makes available online merchant software which enables customers to search the reseller's website for products of interest, and view lists of products resulting from each search. Examples of such merchant software include Microsoft Site Server, available from Microsoft Corporation, and Merchantec Softcart marketed by Mercantec, Inc. of Lisle, Ill. Using the present invention, product listings presented to customers by these online merchant software systems may be enhanced with links to detailed information about any product of interest made available by participating manufacturers.. The searchable product database used by the ISP 111 may be built, as described above, using the universal product codes supplied by the retailer to access the structured data files made available by the participating manufacturers (e.g, manufacturer 103 in FIG. 1) at the IP-addresses supplied by the product code translator.
[0076] The implementation of the invention may be facilitated by supporting software which performs a number of utility functions. As noted above, programs may be readily written to automate the conversion of information stored in a manufacturer's existing product database into the form of static or dynamically generated HTML pages which can be transmitted to fulfill information requests routed to the manufacturer by the cross-referencing facility. Industry and inter-industry groups can promulgate standards and guidelines which will promote consistent formats for product descriptions which are accessed in accordance with the invention. Inventory control and online merchant software can be readily enhanced to take advantage of the availability of database records and more robust product descriptions and supporting Web services which are made available via the Internet. Product information can be made available at terminals and kiosks placed in retail stores, showrooms and public places using conventional Web browsers which execute hypertext links to product information as well as other applications programs which may invoke product and company related Web services at product-code based URIs.
Using HTTP Relocation to Redirect Product Information Request Messages
[0077] The Perl program show.pl, listed in detail in the microfiche appendix, is a CGI (Common Gateway Interface) program which executes on a Web server and which operates as a product code translator as seen at 101
in FIG. 1. This illustrative program uses a file-based database rather than the relational database depicted in FIG. 2. The database consists of a set of files, each of which is designated by a file name consisting of a company code followed by the suffix ".xrl" and each containing cross-referencing information for all product codes beginning with that company code. The Perl program show is specially adapted to locate information on books which are generated by a universal product code known as the International Standard Book Number (ISBN), a nine digit decimal number followed by a check character, used by the publishing houses, book distributors, retail bookstores and libraries to uniquely identify books. A variable number of leading digits of each ISBN designate particular publishers, with the remaining digits being assigned by that publisher to designate a particular edition of a particular book.
[0078] The Perl program show.pl processes an incoming HTTP message containing a parameter which specified the value of a universal product code (in this case, an ISBN number), performs a table lookup operation to retrieve the URL at which information about the product specified by that URL may be found, and then returns an "error" message to the requesting browser which contains that URL in the response message's "Location" response field. As specified in Section 10.11 of the Hypertext Transfer Protocol -- HTTP/1.0 specification, RFC 1945 (May 1996), the Location response-header field defines the exact location of the resource that was identified by the Request-URI. for type 3xx responses, and the location field must indicate the server's preferred URL for automatic redirection to the resource. Only one absolute URL is allowed. The response header may also include the status code 302 which, under the HTTP protocol, indicates that the target data has "moved temporarily" to the URL specified in the location field. In practice, however, it has been found that inclusion of a status code value is not necessary to enable existing web browser programs to automatically redirect the original request to the new location specified in the location response header.
[0079] The Perl CGI program processes an incoming HTTP message request directed to a URL of the form "http://www.upclink.com/cgi-bin/show?isbn=1- 234567890" which is parsed as follows: "www.upclink.com/cgi-bin" is the name of the directory holding the show.pl Perl program and "show?isbn=1234567890" calls the show.pl CGI program and passes to that program the parameter "1234567890" represented by the parameter name "isbn". The ability to execute Perl CGI programs is a common feature of most web servers, and is described, for example, in "Developing CGI Applications with Perl" by John Deep and Peter Holfelder, ISBN 0-471-14158-5 (John Wiley & Sons--1996). The Perl programming language is described in many texts including "Perl 5 Complete" by Edward S. Peschko and Michele DeWolfe, ISBN 0-07-913698-2 (McGraw Hill 1998). Hypertext Markup Language is also widely used and described, for example, in "HTML Publishing Bible" by Alan Simpson, ISBN 0-7645-3009-7 (IDG Books Worldwide 1996).
[0080] As seen in FIG. 3, the show program is entered at 301 and calls a sub routine named get_isbn which processes the incoming message. As indicated at 303, the subroutine get_isbn first loads the input string variable named $isbn with the parameter named "isbn" supplied by the calling message. The subroutine get_isbn then calls the subroutine isbn_message which returns the string "ok" if the contents of $isbn satisfy the requirements for a correct ISBN (International Standard Book Number); that is, the string must contain 10 digits as indicated by the test at 305 and, as indicated as 307, the first nine digits must translate using a predetermined algorithm (performed by the subroutine check_char) into a check digit character which matches the last (10 th) character in the incoming ISBN number stored as the string $isbn.
[0081] The algorithm for generating an ISBN check characters works as follows. First, note that an EAN numbers for books may be converted to the book's ISBN number by removing the first three digits (978) and the last digit from the EAN (the last digit is the EAN check digit, leaving a nine-digit number. For example, EAN 9780940016330 becomes ISBN 094001633
(the first nine digits without the ISBN check character. To generate the ISBN check character, each ISBN digit is multiplied by a predetermined associated weighting factor and the resulting products are added together. The weighting factors for the first nine digits begin with 10
and form the descending series 10, 9, 8 . . . 2. Thus for the nine digits 0 9 4 0 0 1 6 3 3, the products summed are0+81+32+0+0+5+24+9+6=157. This sum is divided by the number 11. (157/11=14 with 3 remainder). The remainder, if any, is subtracted from 11 to get the check digit. (11-3=8). If the check digit is 10, it is represented by the Roman numeral X.. The final ISBN in our example is accordingly 0-940016-33-8. By generating the check digit and comparing it with the received check digit, the validity of the ISBN may be verified.
[0082] If the incoming ISBN string passes all of these tests, the routine isbn_message returns "ok," otherwise, it returns an appropriate error message and the subroutine send_error_page is called at 309 to write and transmit an HTML error page to the requester, advising that the ISBN number supplied was incorrect.
[0083] When the request containing an invalid ISBN number is being supplied from a source other than the user, such as an online retailer's web site which employs the ISBN number supplied by its inventory control system, an advisory error message can also be sent directly to the retailer to indicate that an error was detected. Although such an advisory should be unnecessary, since a similar algorithm for identifying invalid ISBN numbers should be used by the retailer to validate the data before transmission, it is nonetheless desirable to report such errors to the source as well. The error report may sent as an accumulated error log file or as an immediately transmitted message sent to a predetermined error message handling routine provided at the retailer's server.
[0084] If the $isbn variable meets the length and check-digit tests, it is processed by the subroutine make_cocode which determines, for that ISBN number, how many of the leading digits constitute the "company code" which is assigned to a particular publisher. The routine make_cocode performs this operation by calling the subroutine second_hyphen_position which performs tests to determine the number of digits in the company code which can be established from the value of $isbn.
[0085] The relationship between any given ISBN and the URL which identifies the source of information about the book designated by the ISBN is selected by the party (typically the publisher or its designated agent) which controls the server which provides that information. Because different publishers and their web site hosts may use different methods for establishing URLs for their book information, the Perl script show.pl operates in different ways depending on the company code $cc which forms the leading digits of the incoming ISBN.
[0086] The show.pl script handles these differences by fetching a file of control information from a file having the name "../link/cocode.xrl" which is constructed from the company code. The portion "../link/" is a predetermined data directory available to the CGI script show.pl, "cocode" is the numerical string corresponding to the company code extracted at 311 from the ISBN by the make_cocode routine, and ".xrl" is a standard file suffix. For example, if the incoming ISBN is "8870812345" the routine make_cocode will determine that the first five digits are the company code, causing file name"../link/88708.xrl" to be stored as the variable $cstr. As seen at 313 in FIG. 1, the subroutine load_xrl loads the named control file from local disk space at the cross-referencing server which executes the Perl script. If the file named $cstr is not found, an error report is issued indicating that no cross-referencing data has been supplied by the company identified by the company code in $cstr as seen at 315 and 317.
[0087] The contents of the fetched file are then analyzed by load_xrl which fetches the first line and performs a test at 321 (in FIG. 4) to determine if the fetched file is a multiline text file with the first line holding the string "list". If not, the file contains a single line.
[0088] If the file is a single line file, as seen at 323, the contents of the single line are used in combination with the ISBN number to form the URL which specifies where the desired book information may be obtained. For example, if the first line of the control file named ". ./link/88708.xrl" contains the string "hstp://www.upclink.com/ss/ .html", the subroutine load_xrl inserts the hyphenated form of the ISBN at the position indicated by the space to form the URL "http://www.upclink.com/s- s/88708-1234-5.html". The first and second characters of the control file are used to identify variations in the manner in which the ISBN and the control file string are combined to form the URL. For example, if second character of the control file (an "s" in the example) is an "s", the ISBN is hyphenated before being substituted for the space in the string, and the "s" is replaced with a "t".
[0089] If the first line of a multiline control file contains the string "list", all of the lines of the control file are read into a hash table. Each line contains both an ISBN and the URL where information about the book identified by that ISBN can be found. A hash table lookup is then performed to find the particular line holding the target ISBN, and the desired URL is obtained from the line found as indicated at 325 in FIG. 4.
[0090] After the target URL is formed, the subroutine send_response returns an HTTP response message to the requesting browser which reads:
[0091] "Content-type: text/html.backslash.n location: $target .backslash.n.backslash.n" where the URL previously determined is substituted where $target appears. This message is interpreted by the web browser which receives the message as an indication that the requested information (requested from the URL "http://www.upclink.com/cgi-bin/show$- isbn=1234567890") has been relocated to the URL specified by $target. The requesting web browser then automatically resends the request to the location where the needed information is actually located, and does so in a way that is transparent to the user who will normally be unaware that the transmitted request has been redirected to a different location.
[0092] FIG. 5 of the drawings provides an overview of the typical operation of the CGI Perl script show.pl described above. An online shopper manipulates a web browser application program indicated at 330 to look for and purchase a particular book. In doing so, the shopper provides the browser 330 with the URL of an online book retailer which operates the online sales server seen at 332 in FIG. 5. During the browsing session, the sales server transmits a web page 334 which lists the citations to one or more books, and each of those citations anchors a hyperlink to the cross-reference server seen at 336.
[0093] The operator of the sales server which creates the book list web page needs nothing other than the ISBN number of the book to create a link. However, as part of its inventory control system database, the retail web site typically has limited additional information, such as the title, author name(s), publisher name, as well as the ISBN. Using this available information, the book listing webpage can include an informative citation to the book which forms the anchor of a hyperlink to additional information. For example, if the following citation information from the retailer's inventory control system is used to display the following book listing:
[0094] 8 Ball Chicks: A Year in the Violent World of Girl Gangs, By Gini Sikes, Anchor Paperback, ISBN: 0-385-47432-6. LearnMore
[0095] The hyperlink anchor "Learn More" at the end of the citation is formed by the following HTML:
[0096] <a
[0097] href="http://www.upclink.com/cgi-bin/show?isbn=0553571656"
[0098] >
[0099] Learn More
[0100] </a>
[0101] When the anchor text "Learn More" on the web page 334 is clicked on by the shopper, an HTTP request message 338 is sent to the href address of the cross-referencing server 336, triggering the execution of the CGI Perl program show.pl which returns the redirection error message 340, thereby informing the web-browser 330 that the desired information is at a different location. The web browser 330 immediately (and transparently to the shopper) reissues the request to the address specified in the relocation error message. This redirected request, seen at 342, is transmitted to the product information server 344 operated by the publisher, which returns a web page 346 to the browser containing the desired additional information on the book. The web page 346 typically includes an image of the book jacket, a synopsis of the book, and brief reviews, and may well contain links to additional information provided by the publisher, such as an author interview, a table of contents, or whatever else the publisher may wish to include. In this way, the buyer is provided with up-to-date and detailed book information which is equivalent to or superior to the information which may be obtained by picking up the book from the shelf of a "bricks-and-mortar" bookstore.
Image-Cued Links
[0102] Because additional information may not be available via the Internet from the publisher for all of the books in the retailer 's inventory, it is desirable to provide a mechanism which avoids suggesting that a site visitor should click on links that will ultimately prove unworkable.
[0103] The mechanism involves the prior retrieval of data describing which universal product codes have been registered before web pages containing links using these codes are generated. For example, a sales website could transmit a list of the company codes for those universal product codes of products to be offered for sale, and obtain in return a listing of those company codes which have been registered. Alternatively, a list of supported company codes could be periodically broadcast (e.g. by FTP transmission) to subscribers. The sales website could then use this list to distinguish those products for which additional information was being made available by the manufacturer from those for which there is no additional information, and include links on product lists only when they will work.
[0104] Alternatively, the web page producing site can perform a prior fetch of the needed Internet addresses from which product information may be obtained by sending a request message containing one or more universal product codes (or company codes) to the cross-referencing server and receive in return a list of the corresponding Internet addresses.. In this way, the links to additional information contained on product list pages can be refer to the manufacturer's servers directly, and can suppress the creation of links when no information has been made available.
[0105] A mechanism here called "Image-cued links" can also be used to suppress the appearance on a web page of links to unavailable product or company information. For example, an image-cued link can display a graphical icon for a book listing on web page 334 which might be either a visible button with the legend "Learn More" or an invisible (transparent or single-pixel) graphic image, depending on whether or not the ISBN for a listed book has a corresponding URL stored at the cross-referencing server 336. The HTML for an image-cued link to information about the book identified by ISBN 0821219804 might be written like this:
[0106] <a
[0107] href="http://www.upclink.com/cgi-bin/show?isbn=0821219804">
[0108] <img
[0109] src="http ://www.upclink.com/cgi-bin/button?isbn=0821219804" border="0" width="51" height="21"
[0110] >
[0111] where the "anchor" for the link to the CGI script named "show" is an imbedded image (fetched from a different CGI script named "button" which also executes on the cross-reference server). The Perl script button.pl is also reproduced in the Appendix and performs the same initial processing of the incoming ISBN number as the script show.pl. Both show.pl and button.pl determine whether data is available from which a cross-reference from an incoming ISBN number to a URL can be made. If it the cross-reference can be made, show.pl returns a relocation message containing the needed URL whereas button.pl returns the URL of an image that indicates that more information is available. If the cross-reference cannot be made, show.pl returns an error message in the form of an HTML page while button.pl returns the URL of a null image (either a transparent image or a single pixel image). Note that the content of the image file may be controlled by the web page producer since the CGI routine at the cross-referencing server (e.g. button.pl) may be unique to the caller and may hence return image URL's specified by the caller. Alternatively, a single CGI button routine can be used with the desired image URLs being passed as parameters to the cross-referencing server.
[0112] The net effect on the web page is the appearance of a button or other image inviting the site visitor to click on a link to learn more when more information is available, but to suppress the display of the button, or to display a "no information available" button or image when the cross-reference cannot be made. In this way, the site visitor is affirmatively informed when more information is available, and discouraged from looking further when no information is available, while the HTML placed on the book listing has a standard form which can be included without prior knowledge of whether needed data is available or not. Product Code Cross-referencing using the Lightweight Directory Access Protocol The Internet LDAP protocol may be used to advantage to implement the product code translation process. This protocol, developed at the University of Michigan and later further developed by Netscape Communications Corp. provides both access and update capabilities, allowing directory information to be created and managed as well as queried. LDAP is an open Internet standard, produced by the Internet Engineering Task Force (IETF), the same body responsible for creating TCP/IP, the Internet Domain Name System (DNS), and the hypertext transport protocol (HTTP). The LDAP protocol is defined in RFCs 1777 and 1778 and informational documentation is further provided in RFC 1823. The use of LDAP to provide directory lookup services via the Internet is further detailed in the literature. See, for example, Implementing LDAP by Mark Wilcox (Wrox Press--1999) and LDAP--Programming Directory Enabled Applications with Lightweight Directory Access Protocol by Tim Howes and Mark Smith ( Macmillan Technology Series--1997). Operational LDAP server software may be purchased from a variety of sources, and includes the "Netscape Directory Server" marketed by the Netscape Communications Corporation.
[0113] An LDAP server may be advantageously employed to store "entries," each of which is uniquely identified by a distinguished name (DN) which may take the form of the company code portion of the universal product code, creating a "flat namespace" in a single level tree structure, with the remainder of the entry including a string specifying the URL of the server resource from which information about products assigned that company code may be found. In one arrangement, an online merchant's server may send a request to a remote directory server using the LDAP protocol to obtain the URL at which information about a specific product is available. Next, the merchant's server could again use the LDAP protocol to fetch information about a specific product designated by the remainder of the universal product code from a second LDAP directory server at the URL specified by the first server, the second LDAP server being operated by the product manufacturer to store the URL at which data describing particular products is stored. The actual product data may advantageously be stored as XML "documents" as discussed later.
Product Code Cross-Referencing with Domain Name Servers
[0114] Cross-referencing a universal product code to the Internet address of the source of information about the product designated by that code can be advantageously performed by Internet domain name servers (DNS). Conventional Internet domain names are symbolic names (a character string) that identify different computers and resources on the Internet. While computers connected to the Internet actually use binary IP (Internet Protocol) addresses to find each other, people find words and abbreviations to be much more convenient to remember and use. The Internet domain name system assigns computers and resources a domain name that corresponds to the numerical IP address used to access that computer or resource. Each domain name must be unique, and server operators register their desired domain names with a domain name registration server which then provides the cross-references to the appropriate DNS name servers. The Internet domain name system that is currently in widespread use is described in RFCs 1034 and 1035. RFC 1034 provides an introduction to the Domain Name System (DNS), and omits many details which can be found in its companion RFC 1035 which is entitled "Domain Names--Implementation and Specification. "
[0115] Domain names are composed of a hierarchy of names that appear in descending levels from right to left. Therefore, the levels that appear at the end of URLs and E-mail addresses are the first-level and second-level domains. For example, in the domain name "patentsoft.com," the suffix "com" is the first-level domain, and "patentsoft" is the second-level domain. By creating a new first-level domain (e.g. "upc"), the universal product code, or the company code portion of a set of universal product codes, could form the second-level domain. Alternatively, a preassigned name space may be reserved for product code to Internet address conversion by using universal product code or company prefix, perhaps combined with a distinguishing prefix, and an existing top level domain name. To achieve standardization, the full 14-digit GTIN number is preferably used as the second level domain name to create a registered domain name of the form:
[0116] 5 12345678901234.info
[0117] Alternatively, a prefix or suffix might be combined with a product code or company code in the second level portion of the registered domain name as illustrated by the following examples:
[0118] upc123456789012.com
[0119] 1234567890123ean.net
[0120] 12345678901234.biz
[0121] cocode123456.com
[0122] Note that, if the company code portion universal product code is used as the second-level domain, as illustrated in the last example above, registration need only be done once for all product codes sharing that company code. Note also that, as noted earlier, it is desirable that each manufacturer respond to a request for information about that particular participating manufacturer. For example, retail merchants and distributors may advantageously use the company code portion of a universal product code to access a variety of useful information about the company generally, including contact information and distribution, shipping and discount policies. In this way, any retailer can use the web to obtain general information about a company while those retailers with established accounts with a particular vendor (as confirmed, for example, using digital signatures) may obtain private information which is hidden from the general public.
[0123] In this way, universal product codes and/or company codes can be used as domain names which are cross-referenced to IP addresses using existing DNS facilities. Thus, when a web browser issues a request directed to a URL including a company code domain name such as "123456.info," the DNS server (typically assigned by the customer's Internet service provider) responds with a corresponding IP address of an information server maintained by registered the IP address corresponding to that company code based domain name with a DNS registration authority). If the assigned DNS server doesn't already have the cross-reference between a company code domain name (or a product code domain name) and the manufacturer's server's IP address, it asks the primary DNS server that is responsible for the domain if it has the server's IP address. If the primary DNS is busy or unavailable, it will ask the secondary DNS server assigned to that domain. When the customer's DNS server gets the manufacturer's server's IP address, it can then supply the destination IP address needed to transmit a request message to the manufacturer's server to access a designated resource relating to the identified company or product.
[0124] Note that product information can be obtained using a registered domain name that contains the company code only, with the remainder of the universal product code being passed in a further portion of the URI which designates the product information resource. For example, if "123456" is the company code portion of the product code "1234567890123," the full URI designating product information might take the form:
[0125] http://1232456.info/1234567890123/thumbnail.jpg
[0126] which specifies a JPEG image file depicting a product identified by the EAN product code 1234567890123 which is available at a host identified by the domain name 123456.info. The manufacturer's server processes an HTTP request GET message containing this URI in the usual way to return the image file to the requester. The same result could be obtained by registering a domain name containing the full product code (e.g. "1234567890123.info") which cross references the same host server IP address. In this case, the HTTP request GET message would contain the URI:
[0127] http://12324567890123.info/1234567890123/thumbnail.jpg
[0128] and would produce the same result. Each product code in active use by a single manufacturer might be associated with a company code domain name, and all such domain names would be cross referenced to the same IP address. The technique of cross-referencing multiple, different domain names to the same IP address is known as "name based virtual hosting" and permits a single server to support multiple domain names which are cross-referenced in the DNS name server to a single IP address. In this way, a manufacturer to whom multiple company codes have been assigned could support multiple company code domain names, and multiple product code domain names for each company code, using a single server and a single IP address.
[0129] Alternatively, a single server may use IP-based virtual hosting to support multiple domain names which the DNS name server cross-references to different IP addresses. IP-based virtual hosts use the IP address of the connection to determine the correct virtual host to serve and accordingly need to have a separate IP address for each host. With name-based virtual hosting, the server obtains the hostname which is supplied as part of the HTTP headers. Using this technique, many different hosts can share the same IP address. Name-based virtual hosting is usually simpler, and the administrator of the manufacturer's server need only configure the HTTP Server to recognize the different hostnames. The configuration file maps each designated domain name to a root directory in the server's mass storage system. Name-based virtual hosting also eases the demand for scarce IP addresses and would typically be used unless there is a specific reason to choose IP-based virtual hosting.
[0130] In the Internet Domain Name System, different domain names (such as a group of different product and company code based domain names) can be registered in the conventional way to refer to the same IP address. Section 3.6.2, Aliases and canonical names, of RFC 1034, DOMAIN NAMES--CONCEPTS AND FACILITIES, P. Mockapetris (Nov. 1987), DNS also provides specific mechanisms which permit different domain names to be translated into the same IP address. One mechanism uses the canonical name (CNAME) resource record (RR). A CNAME RR identifies its owner name as an alias, and specifies the corresponding canonical name in the RDATA section of the RR. CNAME RRs cause special action in DNS software. When a name server fails to find a desired RR in the resource set associated with the domain name, it checks to see if the resource set consists of a CNAME record with a matching class. If so, the name server includes the CNAME record in the response and restarts the query at the domain name specified in the data field of the CNAME record. The one exception to this rule is that queries which match the CNAME type are not restarted.
[0131] For example, suppose a name server was processing a query with for 1234567890123.info and found the following resource records:
1
1234567890123.info IN CNAME PATENTSOFT.COM PATENTSOFT.COM IN A 209.179.179.18
[0132] By using registering the product code based domain name 1234567890123.info as an alias of the canonical name PATENTSOFT.COM, a request for product information specifying the URI "http://1234567890123.info/ " would be redirected by DNS to the IP address assigned to PATENTSOFT.COM.
[0133] As similar result occurs for domain names placed in pointer (PTR) resoource records that always point at the primary name and not the alias. This avoids extra indirections in accessing information. For example, the PTR RR for the above host should be:
[0134] 18.179.179.209.IN-ADDR.ARPA IN PTR 1234567890123.info
[0135] Note also that different universal product code systems, such as the UPC codes used in the United States and Canada and the EAN codes used elsewhere in world have traditionally used a different number of digits in their product code systems. All of these coding systems have been subsumed into the a 14 decimal digit universal product code called the "Global Trade Item Number" or "GTIN" whose format permits any of the earlier codes to be represented in a standard 14 digit format. Thus, any of these existing universal product codes can be represented in a standard 14 digit format recommended for use in database applications and that can also advantageously form a standard format second-level portion of a domain name, with the first level domain (also called the top level domain or TLD) being any of existing first level domains, or a new top level domain not previously allocated. At the time the domain names containing a given company code and/or a given product code is registered with the Domain Name System, appropriate procedures may be used to confirm that the applicant for DNS registration has been assigned the right to use that particular code in the existing uniform product code system, or has authority to act on behalf of the true assignee of that code. In this way, the existing universal product code registration authority retains primary responsibility for assigning codes, whereas the DNS registrar at most need only confirm the identity of the DNS registrant.
[0136] The format and usage the preferred Global Trade Item Number (GTIN) is described in the Global Trade Item Number Application Guideline promulgated by the EAN--UCC System. A GTIN (Global Trade Item Number) is used for the unique identification of trade items worldwide within the EAN--UCC System. A GTIN has a fourteen-digit data structure though its data carrier (bar code) may contain only twelve digits (the U.P.C.), thirteen digits (EAN-13) or eight digits (EAN-8). The GTIN is defined as a 14-digit number to accommodate all the different structures. The term "trade item" refers to any product or service upon which there is a need to retrieve pre-defined information; this product or service may be priced, ordered, or invoiced at any point in the supply chain. This includes individual items as well as all of their different packaging configurations. There are four data structures for the GTIN; each provides unique numbers when right-justified in a 14-digit field:
[0137] UCC.RTM.-12 (Twelve Digits)
[0138] Six digits representing the Company Prefix assigned by the UCC.RTM.
[0139] Five digits representing the Item Reference Number
[0140] One digit representing the Check Digit
[0141] Or
[0142] Eight digits representing the Company Prefix assigned by the UCC.RTM.
[0143] Three digits representing the item Reference Number
[0144] One digit representing the Check digit
[0145] EAN/UCC.RTM.-13 (Thirteen Digits)
[0146] Twelve digits containing a variable length Company Prefix with the remaining portion of the twelve digits being the Item Reference Number
[0147] One digit representing the Check Digit
[0148] EAN/UCC.RTM.-14 (Fourteen Digits)
[0149] First Digit available to indicate packaging level
[0150] Twelve digits containing a variable length Company Prefix with the remaining portion of the twelve digits being the Item Reference Number
[0151] One digit representing the Check Digit
[0152] EAN/UCC.RTM.-8 (Eight Digits)
[0153] Seven digits containing a variable length Company Prefix with the remainder of the seven digits being the Item Reference Number
[0154] One digit representing the Check Digit
[0155] If company codes rather than the full product code values are registered as part of the domain name, it is necessary for users who desire information or services relating to a product designated by a given universal product code to provide a mechanism for extracting the company code portion of from the product code in order to form the correct domain name. Although the company code for U.S. products is normally the first six digits of the UPC code, that is not true for many products (such as books originally designated by ISBN numbers and/or EAN numbers based on ISBN numbers, and products made abroad that bear EAN codes). Reliable extraction of a company code which takes these special cases into account requires a program-controlled process (see, by way of example, the discussion of the process 311 in FIG. 3 for extracting a company code from the product code for a book). Alternatively, a programmed process may first attempt to match the leading digits of a product code with a registered domain name by first testing the most likely six-digit combination to determine if it can be found in a DNS name server and, if that attempt fails, incrementally altering the number of leading digits extracted from the product code until a match is found or until it is determined that none of the permissible digits that might constitute a company code forms a match. By registering the complete product code as a part of registered domain name, this company code extraction process becomes unnecessary. In addition, the terminal digit in a complete product code is a check digit which should be retained in the domain name. The validity of a product code value which is included in a domain name that a manufacturer seeks to register may be verified by using the standard check digit algorithm at the time of registration.
[0156] The use of the domain name server system as the mechanism for cross-referencing universal product codes and Internet addresses is used in the illustrative embodiment of the invention which is depicted in FIG. 6 and described next.
Using DNS, XML, XSL/CSS and XPointers
[0157] FIG. 6 of the drawings illustrates a system which permits computers operated by large numbers of manufacturers and large numbers of online retailers to work together to provide shopping services to customers via the World Wide Web.
[0158] FIG. 6 shows a browser 410 being used by an online shopper to view products offered by a retailer which operates an inventory control system computer seen at 422. The inventory control system 422 is conventional and includes one or more conventional point of sale registers as illustrated at 421 through which sales are made to customers who visit the physical "bricks and mortar" store. In addition, however, the inventory control system is provided with a "web register" which appears to the inventory control system to function in the same way as a conventional point of sale register but which, in fact, operates through a sales server 430 which provides Internet services on a shared basis to multiple retail stores and their inventory control systems. A communications pathway connects the web register 420 and the inventory control system 422 to the shared sales server 430, with the inventory control system 420 supplying the product codes and corresponding on-hand quantities as indicated at 432 and receiving from the shared sales server order information as indicated at 434.
[0159] A product information server 423 supplies product information to the browser 410 in the form of XML data as indicated at 424 in response to requests 425. The browser 410 preferably utilizes the Internet domain name system as proposed above to convert incoming universal product codes into Internet addresses, with the domain name system consisting of an assigned domain name server 470 which receives domain names containing universal product codes in address requests 472 and returns the registered Internet addresses 474 to the browser 410. When needed, the assigned domain name server 470 obtains the registered cross-references between domain names containing universal product codes and IP addresses from the primary DNS 476 or from the secondary DNS 478.
[0160] The shared sales server 430 sends web pages 438 containing information about products available from a connected retailer to the browser 410, along with XSL (Extensible Stylesheet Language) or CSS (Cascaded Style Sheets) style specifications as seen at 440. The use of XML and XSL or CSS provides several advantages. First, the selection and rendering of the product information is controlled by the links specified in the web page 438 as produced by the sales server. For example, if the web page 438 contains a product listing web page created in response to a search request from the browser 410, each included product description may include a link to only an that portion of an XML product description which contains a brief product description and a thumbnail image of the listed product, whereas, in response to a customer's request for more detailed information, the sales server may return a web page containing an XML "Xpointer" link to detailed product information and/or to an enlarged image of the product. In both cases, the style in which the XML data is rendered by the browser (e.g. typeface, font size and color, background color, etc.) is controlled by the style specification supplied by the sale server. In this way, the same XML data may have different visual styles when included on the pages created by different retail vendors.
[0161] The XSL (Extensible Stylesheet Language) consists of consists of two parts: (1) a language for transforming XML documents, and an XML vocabulary for specifying formatting semantics. An XSL stylesheet specifies the presentation of a class of XML documents by describing how an instance of the class is transformed into an XML document that uses the formatting vocabulary. XSL is described in the World Wide Web Consortium's "Extensible Stylesheet Language Specification," a working draft of which may be found at http://www.w3.org/TR/WD-xsl (Apr. 21, 1999). An XSL stylesheet processor accepts a document or data in XML and an XSL stylesheet and produces the presentation of that XML source content as intended by the stylesheet. It is contemplated that most major web browser applications will include XSL stylesheet processors which will enable them to convert the combination of XML and XSL data into a form, such as a viewable HTML web page, as specified by the XSL.
[0162] In addition to using XSL to specify the rendering style of XML data, cascaded style sheets (CSS) can also be used as set forth in the Proposed Recommendation dated April 28, 1999 from the World Wide Web Consortium (see http://www.w3.org/TR/1999/xml-stylesheet-19990428). This specification allows a stylesheet to be associated with an XML document by including one or more processing instructions with a target of "xmi-stylesheet" in the document's prolog. The World Wide Web Consortium's recommendation regarding cascaded style sheets may be found at http://www.w3.org/TR/REC-CSS1(Jan. 11, 1999) which specifies level 1
of the Cascading Style Sheet mechanism (CSS1). CSS1 is a simple style sheet mechanism that allows authors and readers to attach style (e.g. fonts, colors and spacing) to HTML documents. The CSS 1 language is human readable and writeable, and expresses style in common desktop publishing terminology. One of the fundamental features of CSS is that style sheets cascade; authors can attach a preferred style sheet, while the reader may have a personal style sheet to adjust for human or technological handicaps. Thus product descriptions as viewed on the browser may include content from the product manufacturer, reflect a preferred rendering style specification from the online reseller, as well as the personal style preferences of the viewer.
[0163] XSL could alternatively be used at the shared sales server 430 to transform XML data fetched by the server 430 from the manufacturer's server 423 and then converted into HTML documents with CSS style sheets at the sales server 430. This has the benefit of being backwards compatible with browsers which do not include the ability to handle XSL/CSS. Alternatively, XSL conversion can be performed on the server 430
to transform XML data into XML documents with CSS style sheets. XML, unlike HTML, comes with no formatting conventions and will always need a style sheet to be displayed. This method requires that the browser have the ability to use CSS to render XML. A third alternative is to use XSL to generate HTML/CSS on the client side, a method which requires that the browser have the ability to directly use CSS and XML, which older browsers cannot do. Finally, the browser may transform XML and XSL into "CSS formatting objects". Compared to the previous method, this method is more direct as the content isn't converted to/from HTML.
[0164] The ability to select only a portion of an XML product description document for reproduction on a web page is provided by the Xpointer protocol. As explained in the World Wide Web Consortium Working Draft of Mar. 3, 1998, at http://www.w3.org/TR/WD-xptr, the XML Pointer Language (Xpointer) document specifies a language that supports addressing into the internal structures of XML documents. In particular, it provides for specific reference to elements, character strings, and other parts of XML documents, whether or not they bear an explicit ID attribute. Using Xpointer, only selected portions of an XML product description made available from the manufacturer's server need be presented on a given web page, enabling the creator of the web page which links in XML data to control the nature and extent of the information shown.
[0165] The manner in which explicit relationships between two or more data objects, such as a retailer's product list page and the product information about a product listed on that page, may be expressed as a link asserted in elements contained in XML documents. These "XLinks" is the simplest case are like the HTML links described above in that they are expressed at one end of the link only, are initiated by users to initiate travel to the other end of the link, go only to one destination (which may be determined by a DNS server or by an independent cross-referencing server), and produce an effect which is mainly determined by the browser. The functionality of links is being vastly extended, however, by the XML Linking Language (XLink) specification being developed by the World Wide Web Consortium and available at http://www.w3.org/TR/WD-xlink. As extended, the XLink specification will provide more sophisticated multi-ended and typed links which can be used to advantage to automatically incorporate linked-in product information from one or more manufacturers into displays and multimedia presentations presented by retailers and others.
[0166] As previously discussed, in addition to the use of a product code translation utility which cross-references all or part of a universal product code into an Internet address, it is desirable to establish a protocol or convention which enables a requester to specific kinds of information about identified products or companies. In a conventional HTML system, this can be done by establi