US20060253785A1 - Remote-agent-object based multilevel broweser - Google Patents
Remote-agent-object based multilevel broweser Download PDFInfo
- Publication number
- US20060253785A1 US20060253785A1 US11/482,947 US48294706A US2006253785A1 US 20060253785 A1 US20060253785 A1 US 20060253785A1 US 48294706 A US48294706 A US 48294706A US 2006253785 A1 US2006253785 A1 US 2006253785A1
- Authority
- US
- United States
- Prior art keywords
- document
- search
- documents
- client
- multilevel
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Abandoned
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/951—Indexing; Web crawling techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/10—Text processing
- G06F40/12—Use of codes for handling textual entities
- G06F40/14—Tree-structured documents
- G06F40/143—Markup, e.g. Standard Generalized Markup Language [SGML] or Document Type Definition [DTD]
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99933—Query processing, i.e. searching
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99933—Query processing, i.e. searching
- Y10S707/99934—Query formulation, input preparation, or translation
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10—TECHNICAL SUBJECTS COVERED BY FORMER USPC
- Y10S—TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y10S707/00—Data processing: database and file management or data structures
- Y10S707/99931—Database or file accessing
- Y10S707/99933—Query processing, i.e. searching
- Y10S707/99935—Query augmenting and refining, e.g. inexact access
Definitions
- a current web page loaded into a browser usually includes one or more hyperlinks. If the desired word is on a page pointed to by any of the hyperlinks of the currently loaded page, the “find in page” feature would turn up a negative result. This forces the user to select each hyperlink and then perform a separate “find in page” search on each page referenced by the current page. In some cases this can be most distressing to the user, for example when the current page contains very many hyperlinks or has a richly link-nested structure. In such cases the user is forced to spend undue time searching for the desired word.
- Remote objects can be used to specify remote agents.
- a remote agent is a program that runs on a remote server node on behalf of a client.
- a remote agent can be defined as a static remote agent or a mobile agent.
- a mobile agent is a process that migrates from a first machine to a second machine to a third machine. For example, the first machine passes an object to the second machine and the object is executed in the second machine. After some processing, the mobile agent migrates to the third machine. In some cases the mobile agent clones itself or otherwise spawns off more agent objects to execute on the third machine.
- Mobile agents can be used to specify a query or to carry information, for example.
- a remote agent object is static if it only runs on the second machine but never migrates or clones itself to run on a third machine.
- FIG. 2 is a block diagram representing an embodiment of a browser module with multilevel search capabilities.
- FIG. 6 is a flow chart illustrating a method of searching a document.
- FIG. 9 is a flowchart illustrating a recursive multilevel browser operation.
- the memory 115 may be implemented using a single memory device or may be implemented with a plurality memory devices arranged in a hierarchical cache based structure.
- the memory 115 may also be implemented as any combination of SRAM, ROM, EEPROM, and DRAM, and disk for example.
- the memory 115 includes bit-codings for a browser module 120 .
- the browser module encoded into the memory 115 is designed in accordance with the present invention. In systems where the memory 115 includes DRAM or SRAM and a disk, the browser program will typically be loaded from the disk into the DRAM or SRAM and executed out of the DRAM or SRAM. If the browser is stored in EEPROM or ROM, it may be loaded into RAM or executed directly from the nonvolatile memory.
- the web browser 200 is coupled to a network 205 .
- the network 205 may be any type of network such as a wireline, wireless, or optical network.
- the network 205 represents the Internet, a corporate intranet, a PCS data services network, or any other combination circuit switched and/or packet switched networks. In order to describe a preferred embodiment, without limitation, the network 205 will be considered to be the Internet henceforth.
- the markup language parser 220 parses the loaded web page.
- the markup language parser 220 optionally first checks the currently loaded page for the keyword. This is useful especially for searches where the current page is not a results page but a page of text mixed with hyperlinks.
- the keyword is not found in the currently parsed results page.
- the parser detects each hyperlink in the results page and activates each one.
- the parser parses the returned page and searches for the keyword. If the keyword is found, the browser either starts building a “filtered results page” or displays the returned page to the user with the keyword highlighted. In this second embodiment, the user can click “find next” to continue the search.
- the browser while the user is viewing a given page, the browser makes one or more subsequent network accesses in the background. Hence while the user pauses to review the use of the keyword in a given page, the browser is automatically making one or more other accesses to load up a next page for searching and reviewing.
- the operation of the web browser 200 is explained in further detail in connection with FIGS. 3-6 .
- the user wishes to search all web pages starting at a route web page for a word or a concept.
- a concept may be specified as a Boolean keyword search or a natural language search expression, or a compartmentalized search entry-form, for example.
- the user enters the concept and indicates to search to the deepest level found on the present server. That is, hyperlinks will be followed until they lead to a different server or a server operated by another network entity.
- the search operates similarly to the one described above, except the tree structure specification of the search differs.
- the multilevel GUI dialog window 300 includes a search parameters window 305 .
- the user may enter a word into this dialog box similarly to prior art “find in page” interfaces.
- this window may also be used to enter Boolean keyword searches an/or natural language concept searches.
- the search parameters window 305 may be designed to accept one or more search parameters to be used with any desired search methodology.
- the multilevel GUI dialog window 300 also includes a “number of levels” window.
- This window allows a user to specify the number of levels of hyperlinks to search for the search parameters.
- the number of levels may indicate a number of levels to characterize, or may be set to “max domain.”
- the “max domain” setting instructs the markup language processor 235 to keep searching without a specified depth, but not to follow any hyperlinks leading out of a given domain. This is useful for searching an entire web site.
- An optional “local limit” button 325 may be used to allow the user to set the “number of levels” parameter to max, while limiting the search to a local network domain.
- the local network domain may optionally be specified in terms of a URL. For example, the local domain may involve all network addresses whose root is http://www.abc_company.com/.
- the multilevel GUI 300 also includes a “find next” button. This is used to allow the user to seek the next instance of the search criterion being met. This is similar to prior art “find next” buttons that work in document word searches.
- the “find next” button of the present invention is able to work across hyperlink boundaries. Moreover, it can be used to search for concepts other than simple words. This function may be used in a single document as well. In a preferred embodiment, the present invention will highlight or otherwise move to a paragraph whose words satisfy a Boolean keyword expression.
- One example how the options menu may be used is to enable a user to find a PowerPointTM presentation about a certain subject.
- the user may enter in a set of keywords to a network search engine. This may result in thousands of hits.
- the user may specify to build a map of all pages reachable from up to three levels of hyperlinks from any of the pages returned in the search results.
- the user specifies to build a sitemap file containing all pages including hyperlinks to files with a .ppt extension.
- the map page is constructed, the user may click on a given link in the map page to access the link.
- the results can be showed to the user one at a time, and the user can press the “find next” button 320 to move to the next node having a .ppt file.
- the browser preferably continues the search for one or more subsequent results in the background. More details regarding methods to construct map page documents are provided in connection with FIGS. 4 and 5 .
- a multilevel search interface is coupled to a user.
- the multilevel search interface may be a constructed similarly to the GUI window 300 .
- the step 405 may involve a VXML IVR interface or a more specialized interface including special buttons. Also “hot keys” or “short cuts” as are known in the art may be employed.
- the step 405 accepts user inputs to allow the user to interact with the method 400 in order to specify a set of multilevel search parameters. These preferably involve at least some of the parameters discussed in connection with FIG. 3 .
- the node action may involve logging information into a sitemap document.
- the sitemap document may similarly be constructed as a filtered results summary document, whereby the filter uses a browser-controlled multilevel set of search criteria.
- the markup language processor 235 works with the markup language parser to continue to search through all nodes and keeps a record of a tree structure or list of hyperlinks to nodes that meet the search criterion. In this case the method usually continues along without the need for the user to issue the “find next” command.
- Other node actions may involve downloading files of certain types or printing out web pages. Further details of building sitemap documents and filtered results summary documents are described in connection with FIG. 5 .
- control next passes to a fourth step 420 .
- a tree recursion is performed to select a next local node.
- the next local node is generally the next hyperlink in a given base page.
- the tree may be searched using a depth-first or a breadth-first technique for example.
- the step 420 generally involves maintaining a return-stack as is known in the art in order to move back up the tree when a boundary is reached.
- a condition 425 is evaluated.
- the condition 425 determines whether all of the nodes specified by the multilevel search have been processed. If not, control passes back into the step 415 . If so, control passes to an exit point 430 .
- the user will either have viewed all of the results, or will have a sitemap or filtered results document.
- the sitemap and filtered results document structures include a structured set of hyperlinks to all pages meeting the search criteria. In some cases the search criterion may be left open to cause a sitemap to be built to all pages in a given domain. Also, if certain options such as print or download options were specified, those actions will also have been completed or may still be underway when the exit point 430 is reached.
- the present invention contemplates a wide variety of possible node actions that be performed during the step 415 .
- a user may wish to sequentially step through/view all .gif or all mpg image files on pages meeting certain criteria.
- a user may also wish to build a local map of those pages and be able to view the images at a later time. In such a case the process builds an image tree. Preferably this operation may occur in the background so the user can move to another window and perform other tasks or perform other tasks with the same browser during these background operations.
- the markup processor 235 searches for headings tags found in the parsed markup file stream. When a heading is found, it is loaded into the site-map file. This provides a table of contents structure for the web site. Additional mark-up may be added by the markup processor 225 make the output look like a table of contents, for example.
- the generated map is preferably presented as an HTML or a given type of XML document.
- the markup language processor 235 preferably translates each heading into a hyperlink tag structure marked with appropriate entry point to the heading in the document in which it was found. This way the user viewing the map may select a given heading in order to access a portion of a document corresponding to the heading.
- a “filtered results page” is similar to a sitemap but only includes pages meeting a given set of criteria.
- a filtered results page may be applied to a search engine's results pages or to a web site.
- the filtering operation builds a map but only includes pages of particular interest to a user.
- the sitemap or filtered results page may be arranged as a list of hyperlinks with ancillary information, and in other cases the sitemap or filtered results page may be arranged according to a tree structure.
- the sitemap output is formatted to look like a table of contents.
- the filtered results are preferably formatted to look like a search engine results page.
- the concepts of a sitemap, a filtered results page, and a filtered sitemap are all special cases of the concept of a sitemap. Filters may be applied to prune out unwanted results using browser-controlled, second level searching techniques.
- the method 500 corresponds to one exemplary embodiment of the step 415 of the method 400 . That is, the method 500 is preferably wrapped into a control loop as specified by the method 400 .
- the step 410 also includes the step of setting up an output document to hold the sitemap. This is typically performed by opening a new markup language output document to hold the sitemap information.
- the step 410 then also involves writing a set of document tags and any other document header information needed to initialize a new markup document.
- the markup language processor 235 is operative to write output to an output file structure 240 .
- the step 410 opens an HTML document.
- the step 410 involves opening an XML document with an associated document type definition.
- the method 400 is applied with the step 410 augmented as described above.
- the method 400 proceeds whereby the step 415 involves linking a node into a tree structure.
- the step 420 is preferably performed using a depth-first ordering policy.
- sitemap or filtered sitemap building operations when the step 415 is called in the method 400 , the method 500 as described below is invoked.
- a local node is accessed and opened.
- the local node corresponds to the current markup document in the search.
- the local node starts at the root node of the search and is thenceforth determined by the step 420 .
- the information of the local node is parsed.
- the local node generally corresponds to a markup document.
- Control next passes to a third step 515 that scans the parsed document for specified types of information. This may involve search-text, tag types (e.g. headings tags), file types or other specified criteria. Note that the steps 510 and 515 may be merged so that parsing and searching are performed in a pipelined sequential order.
- the block diagram of FIG. 5 represents only one possible ordering, and other orderings would be apparent to skilled artisans.
- the information extracted from the local-node document is passed through a tag translation.
- the tag translation may generally involve switching from one set of tags to another. This step may also involve encapsulating a first set of tags into another set of tags.
- a null translation may also be applied, whereby no translation is performed.
- An example of a tag translation is when each headings tag is being added to the site-map output document. These tags may be encapsulated in hyperlink tags and URL/URI information may be added to the hyperlink tags so that when a user selects a heading, the browser brings up the local-node document and displays the portion of that document containing the heading to the user.
- An example where a null translation may be used is when the parsed object corresponds to a downloadable file whose file name is already nested within in a hyperlinked set of tags.
- An intelligent browser 705 is connected via a network connection to a server 710 .
- the network connection may be defined at a given network layer, and in the illustrative embodiment 700 , the network connection is assumed to be an application layer connection that uses the HTTP or the HTTPD protocol.
- the server 710 is coupled to an agent sandbox 715 .
- the agent sandbox 715 is preferably implemented as a JavaTM virtual machine in the same computer or computer subnetwork as the sever 710 .
- the agent sandbox provides an execution environment for remote agents (e.g., mobile agents). In some embodiments, mobile agents are not used, but the client passes a static object in a remote method invocation. The object then executes in the sandbox 715 to perform client-defined operations.
- the client-defined class server 725 is optional, and serves as a repository for client-defined classes. This allows the a customized client request to be made to be very compact. For example the client can define a customized search method and submit a method invocation along with a set of search parameters. The agent sandbox 715 will then load the client-defined class from the client-defined class server 725 . This is especially useful for cases where the client browser 705 is housed in a wireless device or some other bandwidth limited (or bandwidth-cost limited) device.
- the client preferably passes a reference to the page being viewed from the server 710 to the proxy server and the proxy server allows the client's object to execute.
- the client-defined object carries out any of the methods and/or functions defined in connection with FIGS. 2-6 .
- the markup language processor 235 is optional and the functions carried out by the markup language processor 235 are carried out completely or partially by the object created in the multilevel object factory 805 . For this reason, the markup language parser 220 is also optional in the smart browser 800 .
- the blocks 230 and 805 are included in the plug-in module.
- the blocks 220 and 235 may also be included in some embodiments of the plug-in module. Again, these blocks are optional when remote-agent-objects are used.
- the multilevel object factory preferably generates the remote agent object and passes the object to a remote server (e.g., 710 , 715 , 720 , or 725 ) using a distributed object protocol such as CORBA or RMI (remote method invocation).
- a remote server e.g., 710 , 715 , 720 , or 725
- the remote agent object is passed to the server 710 .
- the server 710 is the server that supplies the search engine results page or other company related web information as discussed in the examples in connection with FIGS. 2-6 .
- classes are then loaded to support the object from any of the servers 710 , 715 , 720 and/or 725 . Once loaded, the remote agent object is allowed to execute in the agent sandbox 715 .
- the remote agent object created in the multilevel object factory 805 may be a mobile agent or a static agent. In other words, the object created in the multilevel object factory 805 may or may not be able to transit to another server node in the network. In all cases, executable code is allowed to execute remotely on at least one network server.
- the remote agent object is allowed to orchestrate a multilevel operation such as the multilevel “find in page” operation as previously described.
- the object can optionally clone itself or create sub-objects and pass them to other machines pointed to by hyperlinks in the graph that defines the multilevel search (e.g., a tree N levels deep).
- a client-server method 900 of intelligent web browsing is illustrated in flow chart form.
- the method 900 is representative of a system based method carried out between an intelligent web browser client and a remote server.
- the method 900 can also be broken down into submethods carried out respectively on the intelligent web browser client and the server.
- a web browser 800 displays a web page to the user. This typically involves a graphical representation of data but could also use other media such as an audio speech interface.
- a set of multilevel criteria is received from the interface 230 .
- the multilevel object factory uses the present web page information and the multilevel search definition to cause a remote agent object to be created.
- the remote agent object can be as simple as a method invocation and a parameter list.
- the remote agent parses web page information based on input URL data. This web page initially corresponds to the web page being viewed in the browser 705 , 800 .
- the remote agent causes a hyperlink to be accessed. In the case of a company web site, the hyperlinks followed in the step 935 may be local to the server (e.g., 710 ).
- the remote agent object provides support for a multilevel browser operation. For example, the remote agent object assists a multilevel “find in page” search.
- a step 940 an end condition is evaluated to determine whether all of the links in the multilevel operation have been followed. If the end condition is not satisfied, control passes to a step 945 where an the search parameters are modified. This generally corresponds to switching the current URL to the URL pointed to by the link followed in the step 935 .
- Bookkeeping data is also updated to maintain position in the tree graph defining the multilevel browser operation.
- step 945 When the step 945 is executed, control passes back to the step 915 . This time, instead of being executed in the client, the step 915 is executed in the server. This happens, for example, when another remote agent object is to be created and passed to another server as indicated by the hyperlink. In many iterations of this loop, no extra remote agent object is spawned. In such iterations, the step 915 is considered to be a null pass-through event.
- results are coupled back to the user. These results may be generated by more than one remote agent, in which case the book keeping data is used with the graph structure of the multilevel search to properly order the outputs for the user.
- the multiple passes through the step 915 cause a new remote agents to be spawned and in other cases cause a single remote agent to migrate to a different server node.
- agents are only selectively spawned or migrated if a decision is made to do so. This decision preferably takes into account the multilevel graph structure and network delays or router cost parameters. It deemed to be desirable to provide non-null invocations of the step 915 only when network bandwidth costs warrant it. Otherwise too much overhead is incurred in the steps 915 , 920 , and 925 .
- the method 900 with repeated occurrences of the steps 915 , 920 and 930 eliminated is very similar to the web browser of the parent application. The difference is that this version of the method 900 runs on a remote server in the agent sandbox 715 as opposed to the smart browser 200 . This potentially reduces the number of network transactions needed to support multilevel browsing operations.
- the same functionality described in connection with FIGS. 2-6 can be performed by the smart browser 800 and the method 900 .
- the method 900 makes use of the network system configuration 700 .
Abstract
Description
- This application is a continuation in part (CIP) of copending U.S. patent application, Ser. No. 09/512,046, entitled “Web browser with multilevel functions,” filed Feb. 24, 2000, and assigned to art group unit 2755.
- 1. Field of the Invention
- This invention relates generally to network client and server devices and programs. More particularly, the invention relates to a web browser and server-side systems that provide automated multilevel-search-related functions.
- 2. Description of the Related Art
- The Internet enables companies and individuals to post information using an interconnected set of web pages. A first web page is connected to a second web page using a hyperlink. The HTTP protocol is used to transfer data across the Internet from one machine to another so that when a user clicks on a hyperlink in a web page, the web page referenced by the hyperlink is accessed. The action of a user clicking on a sequence of links to move from one web page to another is known as “surfing the web” or “web browsing.” Typically a client software system known as a “web browser” is used to this end. Search engines are provided to help a user find web pages related to a specific topic. Typically a search engine maintains a database of web pages and performs searches based on keywords found in the web pages. In most cases the search engine is implemented within a web server as opposed to a client device.
- While web browsers are very useful, prior art web browsers are in many ways limited. For example, most browsers contain a search mechanism known as “find in page.” In the Microsoft Internet Explorer, the “find in page” feature works by having a use activate the “edit” menu and then the “Find (on this page)” submenu. Alternatively the user may select CRTL+F in order to activate the “find in page” menu. When the user selects “find in page” a dialog box appears. A user is able to enter a word into this dialog box. If the word is in the markup file currently loaded into the web browser, the word will be highlighted to the user.
- While the “find in page” feature found in prior art web browsers is useful it lacks desirable functionality. For example, a current web page loaded into a browser usually includes one or more hyperlinks. If the desired word is on a page pointed to by any of the hyperlinks of the currently loaded page, the “find in page” feature would turn up a negative result. This forces the user to select each hyperlink and then perform a separate “find in page” search on each page referenced by the current page. In some cases this can be most distressing to the user, for example when the current page contains very many hyperlinks or has a richly link-nested structure. In such cases the user is forced to spend undue time searching for the desired word. It would be desirable to have a browser or a browser plug-in that would facilitate word and concept searches through multilevel document structures. Such a browser would increase the productivity of network users. It would also be desirable to allow a user to print pages linked to a given base level document or perform other page-specific functions to a nested structure of linked pages without needing to manually access and specify operations for each page.
- Another area where a more powerful web browser is needed is caused by poor server systems. For example, when visiting a certain company's web site user may be presented with a search engine specific to that web site. In some cases these search engines produce a lot of results that are irrelevant. In many cases the keywords requested in the search do not even appear in the returned web pages. For example a user may run a search and receive 450 results, but only three of these are of interest. To process such an output, the user must spend undue time hunting for the three documents of interest among the 450 results. For example these 450 results are displayed 20 per results page. The user then views each results page, clicks on each link that may be of interest, and views the contents thereof. When each page is accessed, the user uses a “find-in-page” search to determine whether the keyword is actually found in the document and if so, the context of its use. It is also problematic that the user has to select a “next” link to get another page of 20 page titles to view. Sometimes it can take tens of seconds or more to access the next set of titles. It would be desirable to have a web browser with built-in intelligence that could automate the process of wading through irrelevant search results. Other related client-side search acceleration techniques are also needed to allow a user control over searching instead of being limited by the abilities of a distant server's search facilities.
- The concept of metadata aids in the computer manipulation of web-based information such as web sites. XML (extensible markup language) uses metadata to describe the content of resources such as web pages. A dialect of XML is called RDF (resource description framework). RDF supports the notion of a pointer so that graphs may be represented in a computer readable format. For example, a web site may be described using an RDF description. The RDF description can be written to contain a set of searchable properties to describe each web page in the web site hierarchy. The properties of each RDF-described web page can include a pointer to a page pointed to by that page. In this way a computer manipulatable site map representation may be maintained by the a server. When this information is made available to another computer, the other computer can then read the site map and perform operations thereon. This can improve the ability of a search engine to search a site, for example.
- Another technology of interest is remote object technology. Remote object technology is defined form use with object-oriented programming environments. One example of an object oriented programming environment is the Java™ from Sun Microsystems Inc. Java™ defines objects that execute over a “virtual machine.” The virtual machine is a software model of a generic machine. A Java™ object is compiled into a virtual machine language known as “bytecodes.” The bytecodes are then translated by the virtual machine (or a just-in-time compiler) to run on the native processor of a target platform. An example of a distributed object technology is RMI™ (remote method invocation). RMI™ is a Java™ based distributed object technology. Another example of a distributed object technology is CORBA (common object request broker architecture). The aforementioned distributed object technologies are known in the art.
- Distributed object technology allows object-oriented classes to be defined that include a server-side remote object and a client-side object stub, also called a “proxy.” The server-side remote object implements one or more services and a communication protocol to communicate with the client-side stub. The client-side stub provides the client with an API (“interface” in object-oriented programming terminology) to call functions (i.e., “invoke methods” in object oriented programming terminology). When a method is invoked on the client-side stub, a remote procedure call and a set of parameters are marshaled into a message and sent across a communication channel to the server-side remote object. The server-side remote object then receives the message, unmarshals the parameters, invokes the corresponding method on behalf of the client, marshals a set of results into a message, and sends the message back to the client.
- Remote objects can be used to specify remote agents. A remote agent is a program that runs on a remote server node on behalf of a client. A remote agent can be defined as a static remote agent or a mobile agent. A mobile agent is a process that migrates from a first machine to a second machine to a third machine. For example, the first machine passes an object to the second machine and the object is executed in the second machine. After some processing, the mobile agent migrates to the third machine. In some cases the mobile agent clones itself or otherwise spawns off more agent objects to execute on the third machine. Mobile agents can be used to specify a query or to carry information, for example. A remote agent object is static if it only runs on the second machine but never migrates or clones itself to run on a third machine.
- It would be desirable to have a smart browser technology that could allow a client to control multilevel operations on a remote server site. For example it would be desirable for a client to specify an operation such as generating a list of all pdf files downloadable from a given company's web site. It would be desirable for a smart browser to specify a search for all documents on a web site that contain a keyword or meet a set of criteria. It would be desirable to have fast methods to go through a set of results returned by a search engine. It would be desirable to provide a technique whereby a smart browser could interact with a server to customize, manipulate, and/or prune a set of returned results to more rapidly highlight and locate desired information. It would be desirable to have a distributed object framework whereby a smart browser could identify a set of multilevel operations and have these multilevel operations carried out on behalf of the client on a remote server. It would be desirable to have a system whereby a client could specify a mobile agent to carry out client-defined filtering and/or multilevel operations. It would be desirable for the client-created mobile agent to execute remotely on one or more network servers to carry out the tasks as specified by the client.
- The present invention overcomes difficulties with prior art web browsers by providing novel searching techniques. Instead of considering a network search engine to be an external server, the web browser of the present invention takes on search engine functionality. While prior art web browsers offer “find in page” word-search capabilities, the web browser of the present invention allows users to search for words and/or concepts using a multilevel search. In a multilevel search, a user specifies a set of search parameters plus a number of levels to search. In some case the number of levels to search may be set by a default value or may be derived intelligently on-the-fly.
- The smart web browser of the present invention uses RDF and/or distributed object technology to perform multilevel operations. In a preferred embodiment, the smart web browser instantiates a client side proxy object that communicates with a server side remote object, using for example the RMI or the CORBA distributed object protocols. The server side remote object presents an interface to the client side remote object. The interface presented by the server allows the client to submit executable objects. That is, the server-side remote object provides an execution environment for objects submitted by the client for execution. The client then uses the client-side stub (proxy) to send an object that provides a software definition for a multilevel operation. The server-side remote object then allows the object submitted by the client to run. The submitted object causes multilevel web browser operations to be executed in the server on behalf of the client. The client then marshals the results and returns them to the client. In some case the server returns a URI (uniform resource identifier) or a URL (uniform resource locator) and the client subsequently access the results using standard HTTP or other web based application layer protocols. The remote runtime environment where the remote agent runs is called an agent sandbox. The agent sandbox is preferably implemented over a virtual machine.
- In one embodiment, for example, the smart web checks a currently loaded page for a search term. If the term is not found, the smart browser passes a search request object to the remote server using a remote method invocation. The search request is packaged as Java bytecodes defining an executable multilevel search operation. The server then checks each page hyperlinked to the current page. If any of the linked pages contain the desired word or concept, that page is flagged for the user to view. The search may be performed in the server with the assistance of an RDF description of each page. The words or related information may be similarly stored in alternate vectored formats for fast database searching. The user can then press a “find next” button to continue the search. Alternatively all of the results can be logged and loaded up for viewing similar to a search engine results page.
- In accordance with an aspect of the present invention, while the user is viewing a first result, the server can optionally seek out further results in the background. This methodology is useful because documents in the search may need to be gathered from diverse network addresses and may incur delays. Multilevel searches involving web pages or other web resources located on the same server can be searched quickly. When a hyperlink leads from a first server to a second server, the first server creates a distributed object and forward the search request to the second server. In some cases the search request is modified by the first server, for example by indicating a number levels the search has already undergone. In accordance with another aspect of the invention the user may select an option that causes some or all of the pages within a linked tree of documents to be printed without the need to view each page and then manually select “print page.”
- Other aspects of the present invention allow a user to build a sitemap to characterize and otherwise extract information from a linked structure of web pages. Another aspect of the present invention enables enhanced searching of both single-level and multilevel documents. The “find in page” search is expanded to allow users to specify combinatorial searches that are applied to document substructures.
- The present invention allows a client and a server to interact in a novel way so that the client and server can cooperate to perform client-defined multilevel browsing operations. This takes control away from the server and puts it in the hands of an intelligent client such as a smart web browser. Presently a user is at the mercy of a poor search engine. Using prior art methods, when a search engine is incorporated into a web site, the user has no option but to select the provided search engine. Even when the user can select search engines, the search engine often provides unrelated results. The present invention puts controls in the hands of the user that automate the processes involved in sifting through results and looking for specific types of documents that may be found on a given web site. The present invention also allows a user to automatically characterize the contents of a web site by building a sitemap or a filtered site map. The processes and structures of the present invention vastly improve user productivity for performing web-related tasks.
- The present invention also makes use of distributed object technology and/or mobile agent technology to enhance the speed at which client-initiated multilevel browsing operations can be processed. As will be discussed, the present invention minimizes the amount of network bandwidth needed to support multilevel browsing and reduces overall network bandwidth requirements in many cases. The present invention reduces the amount of network traffic generated to the client. This is useful, for example in minimizing the amount of data packets transmitted to and from the client in a wireless system.
- The various novel features of the present invention are illustrated in the figures listed below and described in the detailed description which follows.
-
FIG. 1 is a block diagram illustrating a computer system in accordance with the present invention that us configured to with a multilevel browser program. -
FIG. 2 is a block diagram representing an embodiment of a browser module with multilevel search capabilities. -
FIG. 3 is a block diagram illustrating an exemplary GUI window used to control multilevel search-based browsing in accordance with the present invention. -
FIG. 4 is a flow chart illustrating methods of browsing involving an automated recursive tree-walking process. -
FIG. 5 is a flow chart illustrating a sub-method of building a site-map document. -
FIG. 6 is a flow chart illustrating a method of searching a document. -
FIG. 7 is a block diagram illustrating a client-server configuration within a network using distributed objects to support multilevel browsing operations. -
FIG. 2 is a block diagram representing an embodiment of a browser module with remote-agent-objected based multilevel search capabilities. -
FIG. 9 is a flowchart illustrating a recursive multilevel browser operation. -
FIG. 1 is a block diagram representing an embodiment of acomputerized device 100 designed in accordance with the present invention. The computerized device may be embodied as a desktop computer, a handheld computer, a networked web appliance, a cellular or PCS phone, or may be integrated into another type of system such as a home entertainment system. Skilled artisans will readily realize thedevice 100 may be implemented in various types of housings and may be embedded into various types of systems without departing from the scope of the present invention. - The
computerized device 100 includes aprocessor 105. Theprocessor 105 may be implemented using a standard microprocessor such as a Pentium™ processor from Intel Corp., a RISC type processor, or any other processor or microcontroller. Theprocessor 105 is coupled to abus 110. Thebus 110 may be implemented in some systems using a hierarchical plurality of busses connected via bus bridges. Such hierarchical bussing systems are common in workstations and PC's. For example thebus 110 may include a local bus bridged to a PCI bus, and the PCI bus may be bridged to an ISA bus. All such configurations, to include a simple external processor bus are examples of thegeneric bus 110. Coupled to thebus 110 is amemory 115. Again thememory 115 may be implemented using a single memory device or may be implemented with a plurality memory devices arranged in a hierarchical cache based structure. Thememory 115 may also be implemented as any combination of SRAM, ROM, EEPROM, and DRAM, and disk for example. Thememory 115 includes bit-codings for abrowser module 120. The browser module encoded into thememory 115 is designed in accordance with the present invention. In systems where thememory 115 includes DRAM or SRAM and a disk, the browser program will typically be loaded from the disk into the DRAM or SRAM and executed out of the DRAM or SRAM. If the browser is stored in EEPROM or ROM, it may be loaded into RAM or executed directly from the nonvolatile memory. - The
memory 115 also preferably includes anoperating system 122. The operating system provides software to allow application programs such as theweb browser 120 to communicate with hardware resources. In specialized devices the application functionality of theweb browser 120 may be merged with theoperating system 122, or a dedicated micro kernel may be developed. Even in specialized hand held devices it is preferable to separate hardware layer and application layer functions using a standard version of theoperating system 122. Theoperating system 122 is preferably implemented as a Windows NT™, Windows 2000™, or Windows CE™ operating systems from Microsoft, Inc., Solaris™ from Sun, or any other commercially available operating system. Various other operating systems are being developed for the wireless handheld market. The selection of theoperating system 122 is a design detail that does not weigh into the scope of the present invention. - Also coupled to the
bus 110 is anetwork interface 125. Thenetwork interface 125 may be a network interface card (NIC), an analog modem, or a faster modem such as an xDSL or cable modem. If thedevice 100 is a handheld device, the network interface may include a cellular or PCS connection or any other type of wireless network interface. In such systems the network interface may include cellular telephone functionality as well as wireless data services. A mobile IP protocol is preferably supported in the network interface of such systems. - The
device 100 also includes an I/O system 130. The I/O system 130 preferably involves a display monitor and a set of display drivers. The I/O system also preferably involves an input device such as a mouse and/or keyboard/keypad. In some embodiments, the I/O system 130 may involve an audio interface. In such systems web content is provided to the user using text-to-speech and inputs are accepted via speech recognition. In general, the I/O system provides a means for the user to interact with theweb browser module 120. Any subset of the components of thecomputerized device 100 may optionally be integrated onto a single IC die. - The operation of the
computerized device 100 follows standard computer techniques. The novel aspect of thecomputerized device 100 is encoded into thebrowser module 120. Thebrowser module 120 and its operation are discussed in connection with the subsequent set of figures. Hence the details of program execution-operation of thecomputerized device 100 is understood to a skilled artisan in connection withFIGS. 2-6 . - Referring now to
FIG. 2 , block diagram representative of an embodiment of a multilevel-search web browser 200 is illustrated. Theweb browser 200 is coupled to anetwork 205. Thenetwork 205 may be any type of network such as a wireline, wireless, or optical network. Thenetwork 205 represents the Internet, a corporate intranet, a PCS data services network, or any other combination circuit switched and/or packet switched networks. In order to describe a preferred embodiment, without limitation, thenetwork 205 will be considered to be the Internet henceforth. - The
web browser 200 is coupled to thenetwork 205 using aprotocol stack 210. Theprotocol stack 210 may be part of the browser software in specialized devices, but is preferably supplied by thecomputer operating system 122. In most embodiments theprotocol stack 210 is a part of theoperating system 122 or thenetwork interface 125 as opposed to theweb browser module - The
web browser 200 interfaces to theprotocol stack 210 via anapplication layer interface 215. The application layer interface typically implements at least aspects of the hypertext transfer protocol (HTTP). As web browsing evolves, it is contemplated that other protocols may replace HTTP. Hence theapplication layer interface 215 may in general implement any application layer interface. Theapplication layer interface 215 may likewise implement session layer and presentation layer functionality in accordance with the OSI model. - The application layer interface is operatively coupled to a markup language parser. The markup language parser parses the application layer information. In current systems the application layer data typically involves HTML web pages. A newer technology known as XML allows customized markup languages to be designed. Various dialects of XML are being developed. Hence the language parser may parse HTML, any type of XML, or more generally any type of markup or document-structure language. For example the
language parser 220 may parse VoiceXML, also known as VXML. In preferred embodiments, the markup language parser parses a markup language document that is received via the application layer interface. - The
markup language parser 220 is coupled to a browser GUI (graphical user interface). The GUI may involve a standard web browser GUI as is currently provided by the Netscape Navigator™ or the Microsoft Explorer™. Alternatively, the browser GUI may be specialized for use with hand held devices. The Spyglass Prism™ is an example of a web browser designed for use with small-screened hand-held devices. Also, the browser GUI may involve a voice interface as is common on VXML based systems. Hence the browser GUI may use audio signals instead of graphic signals. In general, the browser GUI is a user interface that allows a user to interact with the web browser to enter network requests and to receive (view or hear) information obtained from the network. The browser GUI may involve a mouse, pointer device, a speech recognition interface, a set of buttons, or standard windows based display. - Coupled to the
browser GUI 225 is a multilevelsearch control interface 230. Themultilevel search interface 230 allows a user to enter information specific to a browser-initiated multilevel search. A preferred embodiment of the multilevel search interface is discussed in further detail in connection withFIG. 3 . Themultilevel search interface 230 is typically implemented in the same technology as thebrowser GUI 225. Themultilevel search interface 230 may be an integral part of thebrowser GUI 225 or may be implemented as a plug-in, for example using JAVA™ technology from Sun Microsystems, Inc. Themultilevel search interface 230 may also be implemented using XML or VXML voice I/O or may otherwise be specialized for use with small screen devices using dedicated physical buttons or other specialized input means. - The
multilevel search interface 230 is operatively coupled to amarkup language processor 235. Themarkup language processor 235 is also operatively coupled to themarkup language parser 220. The markup language processor is optionally operatively coupled to anoutput document 240. Theoutput document 240 may be stored in thememory 115, to include disk. The output document is also coupled to thebrowser GUI 225. It should be noted that a given embodiment may omit certain connections and may use some connections shown as bidirectional in only one direction. The embodiment shown may be pruned down to meet such embodiments or other couplings may be added without departing from the scope and spirit of the present invention. - In operation, the
browser 200 receives a web page via theprotocol stack 210 into theapplication layer interface 215. For example, the web page may be an HTML web page containing search results from a web server's search facility. To understand the invention, consider an example where the returned results include 400 hits, listed at 20 hits per page. Each listed “hit” corresponds to web page title or content-summary information and a hyperlink to a web site. In this example the search engine returned a lot of “junk,” i.e., pages that do not even contain the keywords specified in the initial search. The user then specifies via the multilevel search interface 235 a “one-level search” with a “next-to-end” and a keyword. The “one level search” means to check each hyperlink referenced in the results page for the keyword, but not to check any further. That is, no hyperlinks on any pages referenced from the results page will be accessed. The “next-to-end” parameter instructs the browser to follow the “next” link all the way to the end. Recall that the search engine returns 20 results per page and the user needs to click the “next” hyperlink at the bottom of the page to get to the next 20 results. - With the search specified, the
markup language parser 220 parses the loaded web page. In one embodiment themarkup language parser 220 optionally first checks the currently loaded page for the keyword. This is useful especially for searches where the current page is not a results page but a page of text mixed with hyperlinks. In this example the keyword is not found in the currently parsed results page. Hence the parser detects each hyperlink in the results page and activates each one. Next it parses the returned page and searches for the keyword. If the keyword is found, the browser either starts building a “filtered results page” or displays the returned page to the user with the keyword highlighted. In this second embodiment, the user can click “find next” to continue the search. - In a preferred embodiment, while the user is viewing a given page, the browser makes one or more subsequent network accesses in the background. Hence while the user pauses to review the use of the keyword in a given page, the browser is automatically making one or more other accesses to load up a next page for searching and reviewing. The operation of the
web browser 200 is explained in further detail in connection withFIGS. 3-6 . - A second example of the use of the browser involves evaluating a web site. Suppose a user visits a web site run by a company. The user is performing research about this company and its products. In some cases the company has a search engine and in other cases it does not. All too often the search engine provided by the company is inefficient. An example use of the present invention for processing results of an inefficient search engine is given above.
- In the second example, the user wishes to search all web pages starting at a route web page for a word or a concept. A concept may be specified as a Boolean keyword search or a natural language search expression, or a compartmentalized search entry-form, for example. The user enters the concept and indicates to search to the deepest level found on the present server. That is, hyperlinks will be followed until they lead to a different server or a server operated by another network entity. The search operates similarly to the one described above, except the tree structure specification of the search differs.
- The user may also select other options such as “download pdf files” or “print all pages” or “print all pages meeting search criteria.” Similarly the search may stop when a pdf file is found or a page with the keyword is found, and the user may then make the print or download decision. In another embodiment the system reports the total number of pages to print or pdf files to download before going ahead with those operations. This prevents undue waste in cases where there is more information than anticipated. The present invention also optionally builds a tree-structured sitemap of the web site and optionally lists specified types of resources found. For example each node may include a title, number and titles of pdf files, number of links, number of occurrences of keywords, whether or not the page matches a Boolean criterion, or other user-specified information to be collected. A method for building a sitemap is discussed in more detail in connection with
FIGS. 4 and 5 . The sitemap information is sent to theoutput document 240. Theoutput document 240 is read by thebrowser GUI 225 for interactive user viewing and manipulation. - Referring now to
FIG. 3 , an exemplary multilevelGUI dialog window 300 is illustrated. In a preferred embodiment themultilevel GUI window 300 is implemented as a computer window superimposed on a computer display monitor. In alternative embodiments the multilevel GUI is implemented using dedicated keys, a VXML IVR (interactive voice response) interface, or a combination thereof. In general any user interface technology may be used. - The multilevel
GUI dialog window 300 includes asearch parameters window 305. The user may enter a word into this dialog box similarly to prior art “find in page” interfaces. In some embodiments, this window may also be used to enter Boolean keyword searches an/or natural language concept searches. Thesearch parameters window 305 may be designed to accept one or more search parameters to be used with any desired search methodology. - The multilevel
GUI dialog window 300 also includes a “number of levels” window. This window allows a user to specify the number of levels of hyperlinks to search for the search parameters. In applications where a web site is being characterized, the number of levels may indicate a number of levels to characterize, or may be set to “max domain.” The “max domain” setting instructs themarkup language processor 235 to keep searching without a specified depth, but not to follow any hyperlinks leading out of a given domain. This is useful for searching an entire web site. An optional “local limit”button 325 may be used to allow the user to set the “number of levels” parameter to max, while limiting the search to a local network domain. The local network domain may optionally be specified in terms of a URL. For example, the local domain may involve all network addresses whose root is http://www.abc_company.com/. - The
GUI 300 optionally includes a 1→→N window 315. This window allows a user to specify how many “next page” links to follow. For example in a results page, the search may be specified as one level deep, except the search will follow the “next page” link until the end of the results page is reached. Alternatively the user may only desire to evaluate the first ten pages, especially if the pages are arranged from most probable to least probable as is common with most search engines. This parameter is generally set to a number of “next page” links to follow. The default is preferably set to “max.” - The
multilevel GUI 300 also includes a “find next” button. This is used to allow the user to seek the next instance of the search criterion being met. This is similar to prior art “find next” buttons that work in document word searches. The “find next” button of the present invention is able to work across hyperlink boundaries. Moreover, it can be used to search for concepts other than simple words. This function may be used in a single document as well. In a preferred embodiment, the present invention will highlight or otherwise move to a paragraph whose words satisfy a Boolean keyword expression. - The print on/off
control 300 may involve an options pull-down menu. The print on/off may be used to specify to print all pages scanned in a search. For example a user may wish to print all pages from a company's web site. Pull-down options allow a user to be more selective in only printing pages meeting the search criterion. - The
options menu 335 is preferably a pull down menu providing further options. For example a user may with to build a tree structured map of a given web site. Certain information may be desired about each node in the map. For example the number and titles of pdf files linked to at each node may be of interest. Also, the user may with to download all pdf files meeting a certain criterion. The options menu allows for future expansion. Various other web-based functions may be automated using the options menu. Other options possibilities will become apparent in connection with the discussions ofFIGS. 4 and 5 . TheGUI 300 may be developed using known techniques to provide control of the functionality discussed in connection withFIGS. 4 and 5 . Likewise, other options may be added to save users time in web-page manipulation oriented tasks by following the general teachings of the present invention. - One example how the options menu may be used is to enable a user to find a PowerPoint™ presentation about a certain subject. The user may enter in a set of keywords to a network search engine. This may result in thousands of hits. Now the user may specify to build a map of all pages reachable from up to three levels of hyperlinks from any of the pages returned in the search results. In this case the user specifies to build a sitemap file containing all pages including hyperlinks to files with a .ppt extension. When the map page is constructed, the user may click on a given link in the map page to access the link. Alternatively the results can be showed to the user one at a time, and the user can press the “find next”
button 320 to move to the next node having a .ppt file. While the user is reviewing a first result the browser preferably continues the search for one or more subsequent results in the background. More details regarding methods to construct map page documents are provided in connection withFIGS. 4 and 5 . - Referring now to
FIG. 4 , amethod 400 is illustrated in block diagram form. This method is performed by theweb browser 200 and is coded into thebrowser module 120. Note that themethod 400 may also be implemented by a processor plug-in module. In XML based systems, themethod 400 may be optionally encoded directly into an XML accessed by an XML based browser. That is, in XML systems, browser functionality may be added by XML documents containing program code without the need for a plug-in module. All such embodiments are contemplated by the present invention. - In a
first step 405, a multilevel search interface is coupled to a user. The multilevel search interface may be a constructed similarly to theGUI window 300. In a given system thestep 405 may involve a VXML IVR interface or a more specialized interface including special buttons. Also “hot keys” or “short cuts” as are known in the art may be employed. Thestep 405 accepts user inputs to allow the user to interact with themethod 400 in order to specify a set of multilevel search parameters. These preferably involve at least some of the parameters discussed in connection withFIG. 3 . - In a
second step 410 the multilevel search parameters are accepted. These multilevel search parameters are used to configure a search program that is executed by themarkup language processor 235. In athird step 415 the multilevel search parameters are applied to a local node. The local node corresponds to the web page that has most recently been accessed and has been processed by themarkup language parser 220. For example, theweb browser 200 displays the root node's web page on thedisplay 130 whilesteps step 415 may be optionally skipped or executed on the root node. - The
step 415 involves evaluating the markup document corresponding to the local node and performing a node action. The node action may involve searching the local node for a word, concept, link to file type, or other designated set of search criteria. This generally involves parsing the local node document, extracting tagged substructures, and interpreting them in the context of the markup grammar. That is, thestep 415 generally can search text and can search for text within specific types of tags. This allows the search engine to be able to specifically search for hyperlinks to certain file types, extract headings, and provide other functions as are subsequently discussed. In one example the node action is to display the part of the page meeting the search criteria to the user and wait until the “find next” button is pressed before continuing. While waiting for the “find next” button to be pressed, this exemplary node action continues the search in the background to prevent the user from having to wait for subsequent hyperlinks to be accessed. - In another embodiment, the node action may involve logging information into a sitemap document. The sitemap document may similarly be constructed as a filtered results summary document, whereby the filter uses a browser-controlled multilevel set of search criteria. In this type of embodiment the
markup language processor 235 works with the markup language parser to continue to search through all nodes and keeps a record of a tree structure or list of hyperlinks to nodes that meet the search criterion. In this case the method usually continues along without the need for the user to issue the “find next” command. Other node actions may involve downloading files of certain types or printing out web pages. Further details of building sitemap documents and filtered results summary documents are described in connection withFIG. 5 . - Depending on the multilevel search configuration, either after the “find next” command is issued or automatically, control next passes to a
fourth step 420. In the step 420 a tree recursion is performed to select a next local node. For example, the next local node is generally the next hyperlink in a given base page. The tree may be searched using a depth-first or a breadth-first technique for example. Thestep 420 generally involves maintaining a return-stack as is known in the art in order to move back up the tree when a boundary is reached. For example when a three-level search is underway, the current local node is a level-two node, and all of the hyperlinks in the level-two node are exhausted, the next local node is the next hyperlink in the parent level-one node. Recursive tree and linked list traversing algorithms are well known in the art of computer science and any such algorithm may be used to implement thestep 420. Before any hyperlink is accessed, it is preferably compared to a history list and skipped if it would cause a loop. This ensures the integrity of a loop-free tree structure. - After the
step 420, acondition 425 is evaluated. Thecondition 425 determines whether all of the nodes specified by the multilevel search have been processed. If not, control passes back into thestep 415. If so, control passes to anexit point 430. Depending on the embodiment and/or multilevel search settings, at the end of themethod 400 the user will either have viewed all of the results, or will have a sitemap or filtered results document. The sitemap and filtered results document structures include a structured set of hyperlinks to all pages meeting the search criteria. In some cases the search criterion may be left open to cause a sitemap to be built to all pages in a given domain. Also, if certain options such as print or download options were specified, those actions will also have been completed or may still be underway when theexit point 430 is reached. - The present invention contemplates a wide variety of possible node actions that be performed during the
step 415. For example a user may wish to sequentially step through/view all .gif or all mpg image files on pages meeting certain criteria. A user may also wish to build a local map of those pages and be able to view the images at a later time. In such a case the process builds an image tree. Preferably this operation may occur in the background so the user can move to another window and perform other tasks or perform other tasks with the same browser during these background operations. Other examples of node actions that can be performed during thestep 415 include but are not limited to: 1) display text to user, wait for a “find next” command, and keep doing this until the end of each document is reached; 2) link URL of current node to a site-map structure; 3) link URLs of files found on current node and having a specified type(s) to the sitemap structure, for example to .pdf or .ppt files; 4) generate a headings list for the current node in the site-map structure; 5) print out current web page; 5) download one or more linked files of a specified file type to a local directory; 6) display an image file; 7) add the URL and title information to a filtered search results output list-page. - Number 4) above is an example of mark-up specific searching. In this example the search parameter involves not a keyword but a tag type. The
markup processor 235 searches for headings tags found in the parsed markup file stream. When a heading is found, it is loaded into the site-map file. This provides a table of contents structure for the web site. Additional mark-up may be added by themarkup processor 225 make the output look like a table of contents, for example. The generated map is preferably presented as an HTML or a given type of XML document. Themarkup language processor 235 preferably translates each heading into a hyperlink tag structure marked with appropriate entry point to the heading in the document in which it was found. This way the user viewing the map may select a given heading in order to access a portion of a document corresponding to the heading. - Referring now to
FIG. 5 , an embodiment of a method of building asitemap 500 is illustrated in block diagram form. Themethod 500 can also be used to generate a filtered results page. A “filtered results page” is similar to a sitemap but only includes pages meeting a given set of criteria. In general a filtered results page may be applied to a search engine's results pages or to a web site. The filtering operation builds a map but only includes pages of particular interest to a user. In some cases the sitemap or filtered results page may be arranged as a list of hyperlinks with ancillary information, and in other cases the sitemap or filtered results page may be arranged according to a tree structure. In a preferred embodiment, the sitemap output is formatted to look like a table of contents. The filtered results are preferably formatted to look like a search engine results page. In general the concepts of a sitemap, a filtered results page, and a filtered sitemap are all special cases of the concept of a sitemap. Filters may be applied to prune out unwanted results using browser-controlled, second level searching techniques. - The
method 500 corresponds to one exemplary embodiment of thestep 415 of themethod 400. That is, themethod 500 is preferably wrapped into a control loop as specified by themethod 400. When configured in this way, thestep 410 also includes the step of setting up an output document to hold the sitemap. This is typically performed by opening a new markup language output document to hold the sitemap information. Thestep 410 then also involves writing a set of document tags and any other document header information needed to initialize a new markup document. In this case themarkup language processor 235 is operative to write output to anoutput file structure 240. In a preferred embodiment, thestep 410 opens an HTML document. In another preferred embodiment thestep 410 involves opening an XML document with an associated document type definition. - To build a sitemap, then, the
method 400 is applied with thestep 410 augmented as described above. Themethod 400 proceeds whereby thestep 415 involves linking a node into a tree structure. During site-mapping operations, thestep 420 is preferably performed using a depth-first ordering policy. During sitemap or filtered sitemap building operations, when thestep 415 is called in themethod 400, themethod 500 as described below is invoked. - In a step 505 a local node is accessed and opened. The local node corresponds to the current markup document in the search. The local node starts at the root node of the search and is thenceforth determined by the
step 420. In asecond step 510 the information of the local node is parsed. The local node generally corresponds to a markup document. Control next passes to athird step 515 that scans the parsed document for specified types of information. This may involve search-text, tag types (e.g. headings tags), file types or other specified criteria. Note that thesteps FIG. 5 represents only one possible ordering, and other orderings would be apparent to skilled artisans. - Control next passes to a
fourth step 520. In thefourth step 520 the information extracted from the local-node document is passed through a tag translation. The tag translation may generally involve switching from one set of tags to another. This step may also involve encapsulating a first set of tags into another set of tags. A null translation may also be applied, whereby no translation is performed. An example of a tag translation is when each headings tag is being added to the site-map output document. These tags may be encapsulated in hyperlink tags and URL/URI information may be added to the hyperlink tags so that when a user selects a heading, the browser brings up the local-node document and displays the portion of that document containing the heading to the user. An example where a null translation may be used is when the parsed object corresponds to a downloadable file whose file name is already nested within in a hyperlinked set of tags. - Control next passes to a
fifth step 525 where the selected parsed and translated objects are written to the output markup file structure. After the writing is complete, control is determined by adecision 530. If the end tags of the parsed document are encountered control passes to thestep 420. Otherwise control passes back to thestep 515 where scanning through the parsed document continues. In an alternative embodiment control passes back to thestep 510 and continued parsing is interleaved with the selecting, translation and writing operations. - Referring now to
FIG. 6 , a method of searching adocument 600 is illustrated in flow chart form. This method improves on prior art “find in page” search functions and is also used in multilevel searches. In a first step a plurality of search parameters are accepted from a user. In asecond step 610, a document is parsed and subsections are extracted therefrom. If the document is already open, parsing typically begins from the point where the user presently has the text-input marker or cursor. A subsection may be a paragraph in a word processing document, a cell in a spreadsheet, or any specified tagged object type in a markup document. The subsection may be tag-type independent, i.e., the subsection may be a paragraph or a heading. - In a
third step 615 the subsection is treated as a sub-document and a search criterion is applied to the subsection. For example the search be specified as “dog AND tail” or “dog WITHIN 10 WORDS of tail.” Thestep 615 compares each parsed subsection to the search criterion. In afourth step 620 match results are indicated to the user. For example the subsection of the document matching the search criterion is displayed on theoutput screen 130. If no match is found, thestep 620 performs a null operation. If no match is found, or the user hits the “find next” button, control next passes to afifth step 630. In thesixth step 630, the parser advances to the next subsection and control next passes to thesecond step 610. If this search technique is used within a multilevel document structure and a hyperlink is encountered, thesixth step 630 performs the node-advance similarly to thestep 420. - It should be noted that while prior art document search mechanisms provide “find in page,” they lack the ability for a user to search subsections of the document using a more sophisticated search as described above. This inability leads to user inefficiencies and lack of control in manipulating single and hyperlinked document structures. This aspect of the present invention hence solves an important problem with a latent need.
- The foregoing description of the invention in connection with
FIGS. 1-6 was provided in the parent application.FIGS. 7-9 below and the description thereof are added in the CIP application. Some of the multilevel browser embodiments of the parent are directed to methods whereby web links are accessed to perform individual legs of multilevel searches. The disclosure below identifies remote-object-object based client-side and server-side structures and methods that allow the multilevel search operations to be performed more quickly and with less overhead. All of the methods discussed in connection withFIGS. 1-6 may be carried out using the remote-agent-object based approaches defined below. The approaches described in connection withFIGS. 7-9 may be considered to be alternative embodiments of aspects of the parent invention. - Referring now to
FIG. 7 , a block diagram of anetwork configuration 700 is illustrated. Anintelligent browser 705 is connected via a network connection to aserver 710. The network connection may be defined at a given network layer, and in theillustrative embodiment 700, the network connection is assumed to be an application layer connection that uses the HTTP or the HTTPD protocol. Theserver 710 is coupled to anagent sandbox 715. Theagent sandbox 715 is preferably implemented as a Java™ virtual machine in the same computer or computer subnetwork as thesever 710. The agent sandbox provides an execution environment for remote agents (e.g., mobile agents). In some embodiments, mobile agents are not used, but the client passes a static object in a remote method invocation. The object then executes in thesandbox 715 to perform client-defined operations. - In some embodiments the
agent sandbox 715 may reside in a separate server from theserver 710. For example, thebrowser 705 may correspond to a wireless client. In such cases, if theserver 710 does not host an agent sandbox, it may still be desirable to execute the client-specified object from a server remote from theserver 710. This is because the link between the server hosting theagent sandbox 715 may have a much higher bandwidth connection to theserver 710 than thewireless client 705. In another example, theclient 705 may be in the U.S. while theserver 710 is in China. In this example theserver 710 does not host theagent sandbox 710, so another server in China that does host the an agent sandbox and also has a high speed connection to theserver 710 is selected to run the executable object generated by theclient 705. - Coupled to the
server 710 via a network connection (e.g., an HTTP connection) is aclass server 720. Theclass server 720, for example comprises a repository of Java™ classes whose methods can be invoked by the client's object. A class loading methodology is preferably used so that the client can invoke methods with predefined interfaces. This allows the client-generated object to be small and compact. That is, method invocations are high level. In a preferred embodiment, theserver 710 includes a class server such as theclass server 720. The optional class server within theserver 710 preferably serves as a cache for commonly used classes. Pre-built objects corresponding to commonly used parametric searches and such are maintained in theserver 710 and/or the sandbox's (715) runtime environment. - Also coupled to the
server 710 is a client-defined class server. The client-definedclass server 725 is optional, and serves as a repository for client-defined classes. This allows the a customized client request to be made to be very compact. For example the client can define a customized search method and submit a method invocation along with a set of search parameters. Theagent sandbox 715 will then load the client-defined class from the client-definedclass server 725. This is especially useful for cases where theclient browser 705 is housed in a wireless device or some other bandwidth limited (or bandwidth-cost limited) device. - In operation, the
client browser 705 submits a request to theserver 710. Theserver 710 acts as the server as discussed in connection withFIGS. 1-6 . Like the browser described in connection withFIGS. 1-6 , thebrowser 705 is an intelligent browser that specifies multilevel operations. Instead of thebrowser 705 activating a plurality of hyperlinks, the browser submits a remote-agent object to theserver 710. Theserver 710 then builds the object and passes it to thesandbox 715 for execution. In systems where theserver 710 does not host thesandbox 715, a separate network connection to a proxy server that does host thesandbox 715 may be used (not shown). In this case the client preferably passes a reference to the page being viewed from theserver 710 to the proxy server and the proxy server allows the client's object to execute. The client-defined object carries out any of the methods and/or functions defined in connection withFIGS. 2-6 . By removing the need for the intelligent browser to activate hyperlinks while performing the multilevel operations, a significant amount of network traffic can be reduced. - Note that the
server 710 still may be as deficient as the servers as discussed inFIGS. 2-6 . By providing theagent sandbox 715, the server allows thesmart client 705 to probe for and specify exactly what it needs. Theserver 710 does not need to be modified further. That is, the same set of web pages and built-in search engine provided by theserver 710 may be used. Theagent sandbox 715 is a generic piece of software that is added. This additional piece of software allows thesmart browser 705 to reformat, filter and otherwise perform multilevel browser operations on information located at theserver 710. In a preferred embodiment theagent sandbox 715 resides on thesame server 710 as the related web site pages. This allows multilevel searches to be carried out without generating any further requests across the network, thereby speeding multilevel browsing operations considerably. As previously discussed, when theserver 710 does host theagent sandbox 715, a nearby server that does host theagent sandbox 715 is selected. Here the term “nearby” preferably represents a network metric such as a router cost function. - Referring now to
FIG. 8 , a smart browser according to the present invention is illustrated in block diagram form. Thebrowser 800 may be implemented by augmenting a standard browser with a plug in module. Note the commonality between thesmart browser 800 and thebrowser 200 ofFIG. 2 . Only one new block is added, amultilevel object factory 805. All of the previous connections between blocks remain in tact, although many of these connections are optional and can be omitted in a given embodiment. Themultilevel object factory 805 has optional coupling to themarkup language parser 220, themarkup language processor 235, theoutput document 240 and themultilevel search interface 230. In some embodiments one or more of these couplings may be present. The multilevel object factory is also coupled to theapplication layer interface 215. In thesmart browser 800, themarkup language processor 235 is optional and the functions carried out by themarkup language processor 235 are carried out completely or partially by the object created in themultilevel object factory 805. For this reason, themarkup language parser 220 is also optional in thesmart browser 800. - When the present invention is implemented as a plug-in module, the
blocks blocks - In operation, the
smart browser 800 acts to perform essentially the same functionality as thebrowser 200 and the operations discussed in connection withFIGS. 2-6 . However, thesmart browser 800 performs these operations in a different way. The browser returns a page of information from thenetwork 205 and displays it via thebrowser GUI 225. The user uses themultilevel search interface 230 to specify a multilevel operation such as a multilevel search or a multilevel print as previously described. Instead of having themarkup language processor 235 orchestrate the set of multilevel operations by activating a plurality of hyperlinks, the request is forwarded from themultilevel search interface 230 to themultilevel object generator 805. - The
multilevel object factory 805 is operative to specify the request for a multilevel operation. The multilevel object factory generates a remote agent object. The remote agent object is preferably an object-oriented program that invokes one or more methods to perform the multilevel operation (e.g., a multilevel search or multilevel “find in page”). The methods invoked by the object are defined by an interface. The methods are instantiated by objects defined by classes. As discussed in connection withFIG. 7 , the class descriptions defining the instantiations of the methods may reside on an external server such as any of the servers (or server processes) 710, 715, 720, or 725. - The multilevel object factory preferably generates the remote agent object and passes the object to a remote server (e.g., 710, 715, 720, or 725) using a distributed object protocol such as CORBA or RMI (remote method invocation). In a preferred embodiment, the remote agent object is passed to the
server 710. In this example theserver 710 is the server that supplies the search engine results page or other company related web information as discussed in the examples in connection withFIGS. 2-6 . In the preferred embodiment, classes are then loaded to support the object from any of theservers agent sandbox 715. - Note that the remote agent object created in the
multilevel object factory 805 may be a mobile agent or a static agent. In other words, the object created in themultilevel object factory 805 may or may not be able to transit to another server node in the network. In all cases, executable code is allowed to execute remotely on at least one network server. Once in theagent sandbox 715, the remote agent object is allowed to orchestrate a multilevel operation such as the multilevel “find in page” operation as previously described. The object can optionally clone itself or create sub-objects and pass them to other machines pointed to by hyperlinks in the graph that defines the multilevel search (e.g., a tree N levels deep). - An advantage to performing the multilevel browsing operations using remote agent objects is to offload the requirement for the markup language processor to activate a full set of hyperlinks as defined by the search graph of the
multilevel search interface 230. The search can be carried out much faster if it is conducted at the target server or servers. That is, if the multilevel graph spans multiple servers, it may be advantageous to spawn off sub-search objects to run on each target machine. Mobile agent technology is known, as are programming practices in CORBA and RMI used to create processes that run on remote servers. The present invention can be embodied using known remote object and mobile (or static) agent programming practices to implement the methodologies and systems of the present invention. Moreover, server-side metadata such as RDF descriptions can be used for example, to supply sitemaps and assist in locating web-based information in some cases. These advantages are especially useful for cases when thesmart browser 800 is implemented in a wireless device. - As previously discussed, some embodiments include the
markup language parser 220, themarkup language processor 235, and the multilevel object factory 810. In such embodiments, themarkup language parser 220 is preferably operative to parse a markup language document received via theapplication layer interface 215. The parsed document is passed to themarkup language processor 235 to create a first-level hyperlink map. The hyperlink map is used to represent the set of hyperlinks pointed to by the page in the browser. The hyperlink map is then used as a state parameter for the remote agent object. The remote agent object uses this hyperlink map as a launch-off point for performing a multi-level browser operation. When the remote agent object executes in theagent sandbox 715, it will follow down the links in the hyperlink map and will then repeat the process of themarkup language parser 220 and themarkup language processor 235 to recurse to lower levels. In some embodiments the remote agent object will spawn off new remote objects when the hyperlink graph points to different network nodes. - In other embodiments, the remote agent object waits until it is at the remote server and then performs the operations of the
markup language parser 220 and themarkup language processor 235. Note that these elements (220 and 235) will be needed in the remote agent object to carry out successive levels of the multilevel operation. Moreover, this approach requires less state to be sent with the remote agent object, thus saving bandwidth. In such embodiments, all that is needed is a URL to the current web page (or point therein) and a search definition. With server-side class loading, the as little as a single method invocation is required in the remote agent object created in thefactory 805. This allows compact agent objects to be transmitted in support of multilevel browser operations. - From
FIG. 8 and the previous discussion ofFIGS. 2-6 , it can be seen that thebrowser 800 is representative of an embodiment of a smart browser, also called an intelligent browser. The browser includes theapplication layer interface 215 that is operative to receive at least one data packet. This data packet includes at least a portion of a target web page. The target web page may be a standard HTML web page, or may correspond to more general types of web data such as XML data or a VXML transaction, for example. The web page is presented to the user using the user interface. The user then uses the multilevelsearch control interface 230 to define a multilevel browser operation such as a filtering operation or a multilevel “find in page” operation, for example. Similarly, the multilevel browser operation may be specified to list or print all files of a certain type reachable from links in a finite graph of hyperlinks (e.g., a tree) reachable from the web page. Themultilevel object factory 805 is coupled to receive a first input relating to the target web page and a second input from the multilevel search control interface. The multilevel object factory is operative to specify a remote object agent that orchestrates a multilevel browser operation based upon the first and second inputs. The remote object agent is exported from theweb browser 800 to execute on a network server such as theserver - It should be noted that the
intelligent browser 800 may be implemented in a wireless device. In such systems, the I/O system ofFIG. 1 includes wireless interfaces. These interfaces may include a cellular connection, a packet-over-cellular service, a packet radio connection (e.g., wireless LAN or WAN) and an electronic positioning system receiver. The electronic positioning system receiver may be embodied as a global positioning system (GPS) receiver, or an indoor positioning system receiver (e.g., using electronic tag location methods), or a combination of both. In accordance with an aspect of the present invention, the electronic positioning system is used to derive a parameter used in the multilevel search. This parameter reports the locality of the intelligent client. This information may be used to cause a multilevel search to be limited to return results from web sites representing entities within a specified geographical area. - Referring now to
FIG. 9 , a client-server method 900 of intelligent web browsing is illustrated in flow chart form. Themethod 900 is representative of a system based method carried out between an intelligent web browser client and a remote server. Themethod 900 can also be broken down into submethods carried out respectively on the intelligent web browser client and the server. - In a
first step 905, theweb browser 800 displays a web page to the user. This typically involves a graphical representation of data but could also use other media such as an audio speech interface. In astep 910, a set of multilevel criteria is received from theinterface 230. In astep 915, the multilevel object factory uses the present web page information and the multilevel search definition to cause a remote agent object to be created. As discussed earlier, the remote agent object can be as simple as a method invocation and a parameter list. - In a
step 920 the remote agent object is transmitted from theclient step 925 the remote agent is built at the remote server (e.g., 710). This may involve class loading and object building, or may involve referencing an already existing object, for example when the remote agent corresponds to a well-known method invocation. In a preferred embodiment an object is instantiated to support the remote agent functionality. In some embodiments a set of bytecodes are passed to a server-side remote object that provides an execution environment for the bytecodes. - In a
step 930, the remote agent parses web page information based on input URL data. This web page initially corresponds to the web page being viewed in thebrowser step 935, the remote agent causes a hyperlink to be accessed. In the case of a company web site, the hyperlinks followed in thestep 935 may be local to the server (e.g., 710). - As discussed in connection with
FIGS. 3-6 , the remote agent object provides support for a multilevel browser operation. For example, the remote agent object assists a multilevel “find in page” search. In astep 940 an end condition is evaluated to determine whether all of the links in the multilevel operation have been followed. If the end condition is not satisfied, control passes to astep 945 where an the search parameters are modified. This generally corresponds to switching the current URL to the URL pointed to by the link followed in thestep 935. Bookkeeping data is also updated to maintain position in the tree graph defining the multilevel browser operation. - When the
step 945 is executed, control passes back to thestep 915. This time, instead of being executed in the client, thestep 915 is executed in the server. This happens, for example, when another remote agent object is to be created and passed to another server as indicated by the hyperlink. In many iterations of this loop, no extra remote agent object is spawned. In such iterations, thestep 915 is considered to be a null pass-through event. - In cases where the
step 915 is executed, then thestep 920 is executed to pass the newly created remote agent to its destination for execution. For iterations where thestep 915 is a null pass-through step, thestep 920 is also a null pass-through step. Likewise, for iterations where thestep 915 is executed, thestep 925 is also executed in the destination server. Thesteps steps composition step 950. - Note from the above, that the
method 900 can be implemented by a single remote agent object to as many as one remote agent object per hyperlink in the multilevel search graph. Depending on the graph structure and client and/or server settings, a given embodiment of themethod 900 might generate one or some intermediate number of remote agent objects. - In the
step 950, results are coupled back to the user. These results may be generated by more than one remote agent, in which case the book keeping data is used with the graph structure of the multilevel search to properly order the outputs for the user. Note in some embodiments the multiple passes through thestep 915 cause a new remote agents to be spawned and in other cases cause a single remote agent to migrate to a different server node. In a preferred embodiment, agents are only selectively spawned or migrated if a decision is made to do so. This decision preferably takes into account the multilevel graph structure and network delays or router cost parameters. It deemed to be desirable to provide non-null invocations of thestep 915 only when network bandwidth costs warrant it. Otherwise too much overhead is incurred in thesteps - Note that the
method 900 with repeated occurrences of thesteps method 900 runs on a remote server in theagent sandbox 715 as opposed to thesmart browser 200. This potentially reduces the number of network transactions needed to support multilevel browsing operations. The same functionality described in connection withFIGS. 2-6 can be performed by thesmart browser 800 and themethod 900. Themethod 900 makes use of thenetwork system configuration 700. - Although the present invention has been described with reference to specific embodiments, other embodiments may occur to those skilled in the art without deviating from the intended scope. Therefore, it is to be understood that the invention herein encompasses all such embodiments that do not depart from the spirit and scope of the invention as defined in the appended claims.
Claims (21)
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US11/482,947 US20060253785A1 (en) | 2000-02-24 | 2006-07-10 | Remote-agent-object based multilevel broweser |
Applications Claiming Priority (3)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
US09/512,046 US6819339B1 (en) | 2000-02-24 | 2000-02-24 | Web browser with multilevel functions |
US09/702,455 US7107536B1 (en) | 2000-02-24 | 2000-10-31 | Remote-agent-object based multilevel browser |
US11/482,947 US20060253785A1 (en) | 2000-02-24 | 2006-07-10 | Remote-agent-object based multilevel broweser |
Related Parent Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/702,455 Continuation US7107536B1 (en) | 2000-02-24 | 2000-10-31 | Remote-agent-object based multilevel browser |
Publications (1)
Publication Number | Publication Date |
---|---|
US20060253785A1 true US20060253785A1 (en) | 2006-11-09 |
Family
ID=33418516
Family Applications (5)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/512,046 Expired - Fee Related US6819339B1 (en) | 2000-02-24 | 2000-02-24 | Web browser with multilevel functions |
US09/702,455 Expired - Fee Related US7107536B1 (en) | 2000-02-24 | 2000-10-31 | Remote-agent-object based multilevel browser |
US10/900,180 Expired - Fee Related US7721214B2 (en) | 2000-02-24 | 2004-07-28 | Web browser with multilevel functions |
US11/482,947 Abandoned US20060253785A1 (en) | 2000-02-24 | 2006-07-10 | Remote-agent-object based multilevel broweser |
US12/759,452 Abandoned US20100199206A1 (en) | 2000-02-24 | 2010-04-13 | Web browser with multilevel functions |
Family Applications Before (3)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US09/512,046 Expired - Fee Related US6819339B1 (en) | 2000-02-24 | 2000-02-24 | Web browser with multilevel functions |
US09/702,455 Expired - Fee Related US7107536B1 (en) | 2000-02-24 | 2000-10-31 | Remote-agent-object based multilevel browser |
US10/900,180 Expired - Fee Related US7721214B2 (en) | 2000-02-24 | 2004-07-28 | Web browser with multilevel functions |
Family Applications After (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
US12/759,452 Abandoned US20100199206A1 (en) | 2000-02-24 | 2010-04-13 | Web browser with multilevel functions |
Country Status (1)
Country | Link |
---|---|
US (5) | US6819339B1 (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030188004A1 (en) * | 2002-02-01 | 2003-10-02 | John Fairweather | System and method for navigating data |
US20100223562A1 (en) * | 2009-02-27 | 2010-09-02 | Amadeus S.A.S. | Graphical user interface for search request management |
US20130013665A1 (en) * | 2011-07-06 | 2013-01-10 | Swee Huat Sng | Sandboxed Daemon Process Invocation through HTTP |
Families Citing this family (77)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US7047033B2 (en) * | 2000-02-01 | 2006-05-16 | Infogin Ltd | Methods and apparatus for analyzing, processing and formatting network information such as web-pages |
US6819339B1 (en) * | 2000-02-24 | 2004-11-16 | Eric Morgan Dowling | Web browser with multilevel functions |
EP1306745B1 (en) * | 2000-07-13 | 2011-08-24 | Fuji Xerox Co., Ltd. | Web data output managing device and web data output method |
US20020078019A1 (en) * | 2000-10-02 | 2002-06-20 | Lawton Scott S. | Method and system for organizing search results into a single page showing two levels of detail |
AU2002243448A1 (en) | 2000-10-24 | 2002-06-24 | Singingfish.Com, Inc. | Method of sizing an embedded media player page |
US8122236B2 (en) | 2001-10-24 | 2012-02-21 | Aol Inc. | Method of disseminating advertisements using an embedded media player page |
US6917937B1 (en) * | 2000-11-01 | 2005-07-12 | Sas Institute Inc. | Server-side object filtering |
US7925967B2 (en) | 2000-11-21 | 2011-04-12 | Aol Inc. | Metadata quality improvement |
US20040030683A1 (en) * | 2000-11-21 | 2004-02-12 | Evans Philip Clark | System and process for mediated crawling |
US20020069260A1 (en) * | 2000-12-01 | 2002-06-06 | Tagg James S. | Systems and methods for printing website data |
US7100114B1 (en) * | 2000-12-11 | 2006-08-29 | International Business Machines Corporation | System and process for creating bookmark web pages using web browser intermediaries |
EP1233333A1 (en) * | 2001-02-19 | 2002-08-21 | Hewlett-Packard Company | Process for executing a downloadable service receiving restrictive access rights to al least one profile file |
US7861252B2 (en) * | 2001-03-21 | 2010-12-28 | Andrzej Uszok | Intelligent software agent system architecture |
US7080328B1 (en) * | 2001-03-28 | 2006-07-18 | Ebay, Inc. | Graphical user interface for filtering a population of items |
US7334196B2 (en) * | 2001-06-25 | 2008-02-19 | Siemens Medical Solutions Health Services Corporation | User interface display navigation and item selection system |
US7089233B2 (en) * | 2001-09-06 | 2006-08-08 | International Business Machines Corporation | Method and system for searching for web content |
US20030055827A1 (en) * | 2001-09-20 | 2003-03-20 | International Business Machines Corporation | Apparatus and method for performing background web searches |
JP2003242135A (en) * | 2002-02-20 | 2003-08-29 | Minolta Co Ltd | Printing data preparing program |
US7162686B2 (en) * | 2002-04-19 | 2007-01-09 | Computer Associates Think, Inc. | System and method for navigating search results |
US20030202504A1 (en) * | 2002-04-30 | 2003-10-30 | Avaya Technology Corp. | Method of implementing a VXML application into an IP device and an IP device having VXML capability |
US20030231216A1 (en) * | 2002-06-13 | 2003-12-18 | International Business Machines Corp. | Internet navigation tree with bookmarking and emailing capability |
US7191186B1 (en) * | 2002-11-27 | 2007-03-13 | Microsoft Corporation | Method and computer-readable medium for importing and exporting hierarchically structured data |
US7676462B2 (en) * | 2002-12-19 | 2010-03-09 | International Business Machines Corporation | Method, apparatus, and program for refining search criteria through focusing word definition |
US7577568B2 (en) * | 2003-06-10 | 2009-08-18 | At&T Intellctual Property Ii, L.P. | Methods and system for creating voice files using a VoiceXML application |
US7162473B2 (en) * | 2003-06-26 | 2007-01-09 | Microsoft Corporation | Method and system for usage analyzer that determines user accessed sources, indexes data subsets, and associated metadata, processing implicit queries based on potential interest to users |
US7225187B2 (en) * | 2003-06-26 | 2007-05-29 | Microsoft Corporation | Systems and methods for performing background queries from content and activity |
US7331038B1 (en) * | 2003-07-02 | 2008-02-12 | Amazon.Com, Inc. | Predictive prefetching to improve parallelization of document generation subtasks |
US8437345B2 (en) * | 2003-07-09 | 2013-05-07 | Hitachi, Ltd. | Terminal and communication system |
US7831581B1 (en) | 2004-03-01 | 2010-11-09 | Radix Holdings, Llc | Enhanced search |
US7539982B2 (en) * | 2004-05-07 | 2009-05-26 | International Business Machines Corporation | XML based scripting language |
US7827182B1 (en) * | 2004-06-02 | 2010-11-02 | Cisco Technology, Inc | Searching for a path to identify where to move entries among hash tables with storage for multiple entries per bucket during insert operations |
US20060020465A1 (en) * | 2004-07-26 | 2006-01-26 | Cousineau Leo E | Ontology based system for data capture and knowledge representation |
US9031898B2 (en) * | 2004-09-27 | 2015-05-12 | Google Inc. | Presentation of search results based on document structure |
US8595223B2 (en) | 2004-10-15 | 2013-11-26 | Microsoft Corporation | Method and apparatus for intranet searching |
US7277890B2 (en) * | 2004-12-01 | 2007-10-02 | Research In Motion Limited | Method of finding a search string in a document for viewing on a mobile communication device |
US7921091B2 (en) * | 2004-12-16 | 2011-04-05 | At&T Intellectual Property Ii, L.P. | System and method for providing a natural language interface to a database |
US7565357B2 (en) * | 2004-12-30 | 2009-07-21 | Alcatel Lucent | Multi-sensor communication system |
US8719265B1 (en) * | 2005-11-07 | 2014-05-06 | Google Inc. | Pre-fetching information in anticipation of a user request |
US8307275B2 (en) * | 2005-12-08 | 2012-11-06 | International Business Machines Corporation | Document-based information and uniform resource locator (URL) management |
US7924986B2 (en) * | 2006-01-27 | 2011-04-12 | Accenture Global Services Limited | IVR system manager |
US20080016462A1 (en) * | 2006-03-01 | 2008-01-17 | Wyler Eran S | Methods and apparatus for enabling use of web content on various types of devices |
US8396849B2 (en) * | 2006-03-30 | 2013-03-12 | Microsoft Corporation | Automatic browser search provider detection and usage |
US9633356B2 (en) | 2006-07-20 | 2017-04-25 | Aol Inc. | Targeted advertising for playlists based upon search queries |
US9323867B2 (en) | 2006-08-03 | 2016-04-26 | Microsoft Technology Licensing, Llc | Search tool using multiple different search engine types across different data sets |
US20080033993A1 (en) * | 2006-08-04 | 2008-02-07 | International Business Machines Corporation | Database Access Through Ontologies With Semi-Automatic Semantic Mapping |
US20080133460A1 (en) * | 2006-12-05 | 2008-06-05 | Timothy Pressler Clark | Searching descendant pages of a root page for keywords |
US20080140670A1 (en) * | 2006-12-08 | 2008-06-12 | Joseph Pally | Netvariables in a recursive browser system |
US7836039B2 (en) * | 2006-12-12 | 2010-11-16 | International Business Machines Corporation | Searching descendant pages for persistent keywords |
JP4367495B2 (en) * | 2007-02-15 | 2009-11-18 | ブラザー工業株式会社 | Image forming apparatus |
US20080267145A1 (en) * | 2007-04-27 | 2008-10-30 | Symbol Technologies, Inc. | Methods and Apparatus for Managing Wireless Networks Using Geographical-Level and Site-Level Visualization |
US8898128B2 (en) * | 2007-05-07 | 2014-11-25 | Nokia Corporation | Content storing device query |
FI124000B (en) * | 2007-12-11 | 2014-01-31 | Documill Oy | Method and arrangement for processing data retrieval results |
US8126869B2 (en) * | 2008-02-08 | 2012-02-28 | Microsoft Corporation | Automated client sitemap generation |
US8051370B2 (en) * | 2008-04-23 | 2011-11-01 | Microsoft Corporation | Intelligent autocompletion |
US8395643B2 (en) * | 2008-05-20 | 2013-03-12 | Microsoft Corporation | Motion-based data review and zoom |
US9081956B2 (en) * | 2008-05-26 | 2015-07-14 | Trusteer Ltd. | Remote DOM access |
EP2169570A1 (en) * | 2008-09-25 | 2010-03-31 | Infogin LTD | Mobile sites detection and handling |
KR101545582B1 (en) * | 2008-10-29 | 2015-08-19 | 엘지전자 주식회사 | Terminal and method for controlling the same |
US20100192224A1 (en) * | 2009-01-26 | 2010-07-29 | International Business Machines Corporation | Sandbox web navigation |
US9391825B1 (en) | 2009-03-24 | 2016-07-12 | Amazon Technologies, Inc. | System and method for tracking service results |
JP5526947B2 (en) * | 2010-03-31 | 2014-06-18 | 富士通株式会社 | Search program, search device, and search method |
US9356991B2 (en) * | 2010-05-10 | 2016-05-31 | Litera Technology Llc | Systems and methods for a bidirectional multi-function communication module |
US9955008B2 (en) * | 2010-11-15 | 2018-04-24 | Genesys Telecommunications Laboratories, Inc. | Method for interacting with a multimedia presentation served by an interactive response unit |
US9251268B2 (en) | 2010-12-01 | 2016-02-02 | Microsoft Technology Licensing, Llc | Automated target specific format conversion of context information from a user query |
CN102591887B (en) * | 2011-01-18 | 2016-07-06 | 腾讯科技(深圳)有限公司 | Network data pre-head method and system |
US20150199357A1 (en) * | 2011-04-14 | 2015-07-16 | Google Inc. | Selecting primary resources |
US8688668B1 (en) | 2011-07-15 | 2014-04-01 | Adobe Systems Incorporated | Method and apparatus for improved navigation among search results |
US8959082B2 (en) | 2011-10-31 | 2015-02-17 | Elwha Llc | Context-sensitive query enrichment |
US9813429B2 (en) * | 2012-01-03 | 2017-11-07 | International Business Machines Corporation | Method for secure web browsing |
US9044143B2 (en) | 2012-02-10 | 2015-06-02 | Intermountain Invention Management, Llc | Fluidic esophageal heat sensors |
US9262593B2 (en) | 2012-06-25 | 2016-02-16 | Microsoft Technology Licensing, Llc | Client services for web-based applications |
JP2018507579A (en) * | 2014-12-31 | 2018-03-15 | レベル スリー コミュニケーションズ,エルエルシー | Network address resolution |
KR20170095632A (en) * | 2016-02-15 | 2017-08-23 | 한국전자통신연구원 | Face recognition method |
US10289432B2 (en) * | 2017-03-02 | 2019-05-14 | Salesforce.Com, Inc. | Adaptively linking data between independent systems based on a uniform resource locator |
US10554701B1 (en) | 2018-04-09 | 2020-02-04 | Amazon Technologies, Inc. | Real-time call tracing in a service-oriented system |
IT201900005778A1 (en) * | 2019-04-17 | 2020-10-17 | Roberto Montelatici | INDEPENDENT INTERNET SEARCH METHOD |
US11829734B2 (en) * | 2022-01-11 | 2023-11-28 | Lemon Inc. | Graphical user interface generator using markup language |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5793964A (en) * | 1995-06-07 | 1998-08-11 | International Business Machines Corporation | Web browser system |
US6144973A (en) * | 1996-09-06 | 2000-11-07 | Kabushiki Kaisha Toshiba | Document requesting system and method of receiving related document in advance |
US6151624A (en) * | 1998-02-03 | 2000-11-21 | Realnames Corporation | Navigating network resources based on metadata |
US6304864B1 (en) * | 1999-04-20 | 2001-10-16 | Textwise Llc | System for retrieving multimedia information from the internet using multiple evolving intelligent agents |
US6405192B1 (en) * | 1999-07-30 | 2002-06-11 | International Business Machines Corporation | Navigation assistant-method and apparatus for providing user configured complementary information for data browsing in a viewer context |
US6549941B1 (en) * | 1999-11-22 | 2003-04-15 | Alexa Internet | Software system and methods for resubmitting form data to related web sites |
US6721747B2 (en) * | 2000-01-14 | 2004-04-13 | Saba Software, Inc. | Method and apparatus for an information server |
US7020679B2 (en) * | 2000-05-12 | 2006-03-28 | Taoofsearch, Inc. | Two-level internet search service system |
US7107536B1 (en) * | 2000-02-24 | 2006-09-12 | Eric Morgan Dowling | Remote-agent-object based multilevel browser |
Family Cites Families (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5696963A (en) * | 1993-11-19 | 1997-12-09 | Waverley Holdings, Inc. | System, method and computer program product for searching through an individual document and a group of documents |
US5701451A (en) | 1995-06-07 | 1997-12-23 | International Business Machines Corporation | Method for fulfilling requests of a web browser |
US5903892A (en) * | 1996-05-24 | 1999-05-11 | Magnifi, Inc. | Indexing of media content on a network |
US6026409A (en) * | 1996-09-26 | 2000-02-15 | Blumenthal; Joshua O. | System and method for search and retrieval of digital information by making and scaled viewing |
US6006252A (en) * | 1996-10-08 | 1999-12-21 | Wolfe; Mark A. | System and method for communicating information relating to a network resource |
US6381637B1 (en) * | 1996-10-23 | 2002-04-30 | Access Co., Ltd. | Information apparatus having automatic web reading function |
US6282548B1 (en) * | 1997-06-21 | 2001-08-28 | Alexa Internet | Automatically generate and displaying metadata as supplemental information concurrently with the web page, there being no link between web page and metadata |
US6295526B1 (en) * | 1997-10-14 | 2001-09-25 | Bellsouth Intellectual Property Corporation | Method and system for processing a memory map to provide listing information representing data within a database |
US6202061B1 (en) * | 1997-10-24 | 2001-03-13 | Pictra, Inc. | Methods and apparatuses for creating a collection of media |
US6415278B1 (en) * | 1997-11-14 | 2002-07-02 | Adobe Systems Incorporated | Retrieving documents transitively linked to an initial document |
US6324552B1 (en) * | 1997-11-25 | 2001-11-27 | International Business Machines Corporation | Browsers for focused access of data |
US5930472A (en) | 1998-04-29 | 1999-07-27 | Motorola, Inc. | Method and apparatus in a wireless communication system for splitting a browser functionality between a wireless client and an infrastructure portion |
US6334145B1 (en) * | 1998-06-30 | 2001-12-25 | International Business Machines Corporation | Method of storing and classifying selectable web page links and sublinks thereof to a predetermined depth in response to a single user input |
US6278993B1 (en) * | 1998-12-08 | 2001-08-21 | Yodlee.Com, Inc. | Method and apparatus for extending an on-line internet search beyond pre-referenced sources and returning data over a data-packet-network (DPN) using private search engines as proxy-engines |
US6338059B1 (en) * | 1998-12-17 | 2002-01-08 | International Business Machines Corporation | Hyperlinked search interface for distributed database |
US6434552B1 (en) * | 1999-01-29 | 2002-08-13 | Hewlett-Packard Company | Method for data retrieval |
US6631496B1 (en) * | 1999-03-22 | 2003-10-07 | Nec Corporation | System for personalizing, organizing and managing web information |
US6175830B1 (en) * | 1999-05-20 | 2001-01-16 | Evresearch, Ltd. | Information management, retrieval and display system and associated method |
US6476827B1 (en) * | 1999-08-31 | 2002-11-05 | Xoucin, Inc. | Method and apparatus for creating and maintaining a scrapbook of information pages |
US6125395A (en) * | 1999-10-04 | 2000-09-26 | Piiq.Com, Inc. | Method for identifying collections of internet web sites with domain names |
US7216115B1 (en) * | 1999-11-10 | 2007-05-08 | Fastcase.Com, Inc. | Apparatus and method for displaying records responsive to a database query |
US6313855B1 (en) * | 2000-02-04 | 2001-11-06 | Browse3D Corporation | System and method for web browsing |
US6516312B1 (en) * | 2000-04-04 | 2003-02-04 | International Business Machine Corporation | System and method for dynamically associating keywords with domain-specific search engine queries |
JP2007146003A (en) * | 2005-11-28 | 2007-06-14 | Toyo Ink Mfg Co Ltd | Ink composition |
-
2000
- 2000-02-24 US US09/512,046 patent/US6819339B1/en not_active Expired - Fee Related
- 2000-10-31 US US09/702,455 patent/US7107536B1/en not_active Expired - Fee Related
-
2004
- 2004-07-28 US US10/900,180 patent/US7721214B2/en not_active Expired - Fee Related
-
2006
- 2006-07-10 US US11/482,947 patent/US20060253785A1/en not_active Abandoned
-
2010
- 2010-04-13 US US12/759,452 patent/US20100199206A1/en not_active Abandoned
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5793964A (en) * | 1995-06-07 | 1998-08-11 | International Business Machines Corporation | Web browser system |
US6144973A (en) * | 1996-09-06 | 2000-11-07 | Kabushiki Kaisha Toshiba | Document requesting system and method of receiving related document in advance |
US6151624A (en) * | 1998-02-03 | 2000-11-21 | Realnames Corporation | Navigating network resources based on metadata |
US6304864B1 (en) * | 1999-04-20 | 2001-10-16 | Textwise Llc | System for retrieving multimedia information from the internet using multiple evolving intelligent agents |
US6405192B1 (en) * | 1999-07-30 | 2002-06-11 | International Business Machines Corporation | Navigation assistant-method and apparatus for providing user configured complementary information for data browsing in a viewer context |
US6549941B1 (en) * | 1999-11-22 | 2003-04-15 | Alexa Internet | Software system and methods for resubmitting form data to related web sites |
US6721747B2 (en) * | 2000-01-14 | 2004-04-13 | Saba Software, Inc. | Method and apparatus for an information server |
US7107536B1 (en) * | 2000-02-24 | 2006-09-12 | Eric Morgan Dowling | Remote-agent-object based multilevel browser |
US7020679B2 (en) * | 2000-05-12 | 2006-03-28 | Taoofsearch, Inc. | Two-level internet search service system |
Cited By (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030188004A1 (en) * | 2002-02-01 | 2003-10-02 | John Fairweather | System and method for navigating data |
US7555755B2 (en) * | 2002-02-01 | 2009-06-30 | John Fairweather | System and method for navigating data |
US20100223562A1 (en) * | 2009-02-27 | 2010-09-02 | Amadeus S.A.S. | Graphical user interface for search request management |
US9459765B2 (en) * | 2009-02-27 | 2016-10-04 | Amadeus S.A.S. | Graphical user interface for search request management |
US10353544B2 (en) | 2009-02-27 | 2019-07-16 | Amadeus S.A.S. | Graphical user interface for search request management |
US20130013665A1 (en) * | 2011-07-06 | 2013-01-10 | Swee Huat Sng | Sandboxed Daemon Process Invocation through HTTP |
US8825748B2 (en) * | 2011-07-06 | 2014-09-02 | Sharp Laboratories Of America, Inc. | Sandboxed daemon process invocation through HTTP |
Also Published As
Publication number | Publication date |
---|---|
US7721214B2 (en) | 2010-05-18 |
US20100199206A1 (en) | 2010-08-05 |
US7107536B1 (en) | 2006-09-12 |
US6819339B1 (en) | 2004-11-16 |
US20040267739A1 (en) | 2004-12-30 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US7107536B1 (en) | Remote-agent-object based multilevel browser | |
KR100900743B1 (en) | Data sharing system, method and software tool | |
US7770105B2 (en) | Communication apparatus, system and web page processing method | |
JP3548098B2 (en) | Method and system for providing a native language query service | |
US7027975B1 (en) | Guided natural language interface system and method | |
KR100297632B1 (en) | Method and apparatus for transparently accessing multiple data stream types from an HTML browser | |
US6292802B1 (en) | Methods and system for using web browser to search large collections of documents | |
US5793966A (en) | Computer system and computer-implemented process for creation and maintenance of online services | |
TW571204B (en) | Content publication system for supporting real-time integration and processing of multimedia content including dynamic data, and method thereof | |
JP4437918B2 (en) | Apparatus and method for selectively retrieving information and subsequently displaying the information | |
US20060161685A1 (en) | Client aware extensible markup language content retrieval and integration in a wireless portal system | |
US20020054090A1 (en) | Method and apparatus for creating and providing personalized access to web content and services from terminals having diverse capabilities | |
US20030088639A1 (en) | Method and an apparatus for transforming content from one markup to another markup language non-intrusively using a server load balancer and a reverse proxy transcoding engine | |
US20020078105A1 (en) | Method and apparatus for editing web document from plurality of web site information | |
WO2004064357A2 (en) | Data conversion server for voice browsing system | |
KR20020004931A (en) | Conversational browser and conversational systems | |
JP2005031995A (en) | User interface control device and method | |
WO2005052811A1 (en) | Searching in a computer network | |
KR20030064828A (en) | Method and system for fulfilling requests for information from a network client | |
JPH11502346A (en) | Computer system and computer execution process for creating and maintaining online services | |
US7970936B2 (en) | In a world wide web communications network simplifying the uniform resource locators (URLS) displayed in association with received web documents | |
US20030158894A1 (en) | Multiterminal publishing system and corresponding method for using same | |
JP2004513419A (en) | Method and system for automating internet interactions | |
WO2002087135A2 (en) | System and method for adapting information content for an electronic device | |
JPH11167584A (en) | Page shift method and its execution device and medium recording page shift processing program and data |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
AS | Assignment |
Owner name: NEXTWAVE SOULUTIONS, LP, COSTA RICA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:DOWLING, ERIC M.;REEL/FRAME:020270/0407 Effective date: 20071213 Owner name: NEXTWAVE SOULUTIONS, LP,COSTA RICA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:DOWLING, ERIC M.;REEL/FRAME:020270/0407 Effective date: 20071213 |
|
AS | Assignment |
Owner name: RPX-NW ACQUISITION LLC, DELAWARE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NEXTWAVE SOLUTIONS, LP;REEL/FRAME:021679/0145 Effective date: 20080911 Owner name: NEXTWAVE SOLUTIONS, LP, COSTA RICA Free format text: SECURITY AGREEMENT;ASSIGNOR:RPX-NW ACQUISITION LLC;REEL/FRAME:021679/0155 Effective date: 20081008 Owner name: RPX-NW ACQUISITION LLC,DELAWARE Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:NEXTWAVE SOLUTIONS, LP;REEL/FRAME:021679/0145 Effective date: 20080911 Owner name: NEXTWAVE SOLUTIONS, LP,COSTA RICA Free format text: SECURITY AGREEMENT;ASSIGNOR:RPX-NW ACQUISITION LLC;REEL/FRAME:021679/0155 Effective date: 20081008 |
|
STCB | Information on status: application discontinuation |
Free format text: ABANDONED -- FAILURE TO RESPOND TO AN OFFICE ACTION |
|
AS | Assignment |
Owner name: RPX CORPORATION, CALIFORNIA Free format text: ASSIGNMENT OF ASSIGNORS INTEREST;ASSIGNOR:RPX-NW ACQUISITION LLC;REEL/FRAME:027082/0368 Effective date: 20110923 |