Home
Patent Search
IMT Blog
REGISTER
|
SIGN IN
United States Patent
5796967
Filepp , ; et al.
August 18, 1998
Title
Method for presenting applications in an interactive service
Abstract
A method for presenting applications in an interactive service featuring steps for generating screen displays of the service applications at the reception systems of the respective users. Steps are provided for generating the application displays as screens having a plurality of partitions, the partitions being constructed from reusable elements. In accord with the method, the screens include at least a first partition at which an application may be presented and a second, concurrently displayed partition including command functions for managing the display. The method further includes steps for providing command functions that facilitate random navigation to new applications with a variety of different procedures which the user can choose from. In (preferred) one form, the functions are presented as a command bar located at the bottom of the screen. Further, the method includes steps for opening and closing windows on the display to enable presentation of additional data relating to the presented applications. Still further, the method includes steps for providing additional partitions for concurrently displaying other applications, which may include advertising.
Inventors:
Filepp; Robert
(Springfield,
NJ
)
, Appleman; Kenneth H.
(White Plains,
NY
)
, Bidwell; Alexander W.
(New York,
NY
)
, Wolf; Allan M.
(Ridgefield,
CT
)
, Galambos; James A.
(Westport,
CT
)
, Meo; Sam
(New York,
NY
)
Assignee:
International Business Machines Corporation
(Armonk,
NY
)
Appl. No.:
158031
Filed:
November 26, 1993
Current U.S. Class:
715/764
715/854
709/215
709/217
Field of Search:
395/155-159,161,160,200.03,200.04,200.05,200.08,601,326,329-335,339-349,352-354
U.S. Patent Documents
4821211
April 1989
Torres
4949248
August 1990
Caro
4953159
August 1990
Hayden et al.
5050105
September 1991
Peters
5220657
June 1993
Bly et al.
5280583
January 1994
Nakayama et al.
Other References
McGregor, "Designing User Interface Tools for the X Window System", Compcon Spring '89 IEEE Computer Society Int'l Conference, pp. 243-246., 1989. .
Dunwoody et al, "A Dynamic Profile of Window System Usage", Computer Workstations Conference, pp. 90-99., 1988. .
Gancarz, "Uwm: A User Interface for X Windows", Usenix Association Summer Conference Proceedings, 1986. .
Scheifler et al, "The X Windows System", ACM Transactions on Graphics, vol. 5, No. 2, pp. 79-709., Apr. 1986. .
Microsoft Windows version 2.0 user's guide, pp. 88-91., 1987. .
"Microsoft Windows--Version 2.0", 1987, (Book 1; p. viii-xiii; Book 4, pp. 36,37,83). .
Mastering Windows.TM. 3.0 by Robert Cowart, 1990, pp. 6-27..~
Primary Examiner:
Field; Joseph H.
Attorney, Agent or Firm:
Scifo; Paul C.
Parent Case Text
RELATED APPLICATIONS
This is a division of application Ser. No. 388,156 filed Jul. 28, 1989, which issued Sep. 13, 1994, as U.S. Pat. No. 5,347,632, application Ser. No. 388,156 being a continuation in part of application Ser. No. 328,790, filed Mar. 23, 1989, abandoned, which itself was a continuation in part of application Ser. No. 219,931, filed Jul. 15, 1988 abandoned.
Claims
What we claim is:
1. A method for presenting interactive applications on a computer network, the network including a multiplicity of user reception systems at which respective users may request a multiplicity of available applications, the respective reception systems including a monitor at which the applications requested can be presented as one or more screens of display, the method comprising the steps of:
a. generating a screen display at a respective reception system for a requested application, the screen display being generated by the respective reception system from data objects having a prescribed data structure, at least some of which objects may be stored at the respective reception system, the screen display including a plurality of partitions, the partitions being constructed from objects, the objects being retrieved from the objects stored at the respective reception system, or if unavailable from the objects stored at the respective reception system, then from the network, such that at least some of the objects may be used in more than one application;
b. generating at least a first partition for presenting applications; and
c. generating concurrently with the first partition at least a second partition for presenting a plurality of command functions, the command functions including at least a first group which are selectable to permit movement between applications.
2. The method of claim 1 wherein the data structure of the objects includes a header and one or more data segments and wherein generating the second partition includes providing the first group of command functions with a first subgroup of command functions which are selectable to permit random movement between applications.
3. The method of claim 2 wherein the objects are stored at the respective reception systems in accordance with a predetermined plan, and wherein providing the first subgroup of commands includes providing a command for causing the user to be presented with a t least one procedure for navigating to a new application.
4. The method of claim 2 wherein the predetermined plan for storing objects at the respective reception systems includes providing the objects with a storage control parameter in their respective headers, and wherein providing the first subgroup of command functions includes providing at least one command for causing the user to be presented with a plurality of different procedure for navigating to a new application.
5. The method of claim 4 wherein the object storage controle parameter is dependent on the currency of the object data, and wherein providing the navigation procedures includes enabling the user to enter a character string at the reception system to randomly search the available applications for a desired application.
6. The method of claim 4 wherein providing the navigation procedures includes enabling the user to access an index of available applications from which a desired application may be selected.
7. The method of claim 4 wherein providing the navigation procedures includes enabling the user to access a directory of application subject matter from which a desired application may be selected.
8. The method of claim 4 wherein providing the navigation procedures includes enabling the user to access a physical analogy of the available applications from which a desired application may be selected.
9. The method of claims 5, 6, 7 or 8 wherein providing the navigation procedures to a new application includes presenting a window at the display in which the user is presented with multiple, interactive command functions to effect navigation.
10. The method of claim 2 wherein the objects are stored at the respective reception systems in accordance with a predetermined plan, and wherein providing the first subgroup of command functions includes providing at a command for enabling the user to progress through a sequence of applications previously designated.
11. The method of claim 10 wherein the predetermined plan for storing objects at the respective reception systems includes providing the objects with a storage control parameter in their respective headers, and wherein providing the user with a command for progressing through a sequence of applications includes enabling the user to adjust the application sequence.
12. The method of claim 1 further including generating at least a third screen partition concurrently with the first and second screen partitions for presenting a second application and wherein the data structure of the objects includes a header and one or more data segments.
13. The method of claim 12 wherein the objects are stored at the respective reception systems in accordance with a predetermined plan, and wherein the predetermined plan for storing objects at the respective reception systems includes providing the objects with a storage control parameter in their respective headers, and wherein presenting a third screen partition includes presenting the second application as advertising.
14. The method of claim 1 further including generating one or more window partitions that overlays at least a portion of the application partition, the one or more windows for presenting data associated with the application displayed and wherein the data structure of the objects includes a header and one or more data segments, and wherein the objects are stored at the respective reception systems in accordance with a predetermined plan, which includes providing the objects with a storage control parameter at their respective headers.
15. The method of claim 14 wherein generating window partitions includes providing the window partitions with fields for conducting interactive procedures associated with an underlying application.
16. The method of claim 15 wherein generating window partitions includes providing the window partitions with interactive fields for conducting transactional procedures.
17. The method of claim 1 wherein generating the first and second screen partitions includes generating the respective partitions at fixed, predetermined regions of the display screen, the second partition being arranged as a command bar and wherein the data structure of the objects includes a header and one or more data segments, and wherein the objects are stored at the respective reception systems in accordance with a predetermined plan, which includes providing the objects with a storage control parameter at their respective headers.
Description
BACKGROUND OF THE INVENTION
1.Field of Use
This invention relates generally to a method for presenting applications in a distributed processing, interactive computer network intended to provide very large numbers of simultaneous users; e.g. millions, access to an interactive service having large numbers; e.g., thousands, of applications which include pre-created, interactive text/graphic sessions; and more particularly, to a method for presenting applications, the method featuring step for generating a screen display at respective user reception systems, the screen display including a plurality of partitions for concurrently presenting at least a user-requested application and a group of command functions for managing the display, the group of command functions including a subgroup of functions for randomly selecting applications for display with a variety of different procedures, the method also including steps for opening and closing windows on the display for presenting data relating to the displayed applications, and further partitions for concurrently displaying, for example, an additional application which may include advertising.
2. Prior Art
Interactive computer networks are not new. Traditionally they have included conventional, hierarchical architectures wherein a central, host computer responds to the information requests of multiple users. An illustration would be a time-sharing network in which multiple users, each at a remote terminal, log onto a host that provides data and software resource for sequentially receiving user data processing requests, executing them and supplying responses back to the users.
While such networks have been successful in making the processing power of large computers available to many users, problems have existed with them. For example, in such networks, the host has been required to satisfy all the user data processing requests. As a result, processing bottlenecks arise at the host that cause network slowdowns and compel expansion in computing resources; i.e., bigger and more complex computer facilities, where response times are sought to be held low in the face of increasing user populations.
Host size and complexity, however, are liabilities for interactive networks recently introduced to offer large numbers of the public access to transactional services such as home shopping, banking, and investment maintenance, as well as informational services concerning entertainment, business and personal matters. As can be appreciated, commercial interactive networks will have to provide attractive services at low cost and with minimal response times in order to be successful. Unlike military and governmental networks where, because of the compulsory nature of the service performed, costs, content and efficiency are of secondary concern, in commercial services, since use is predominantly elective, and paid for by the consumer, costs will have to be held low, content made interesting and response times reduced in order to attract and hold both users who would subscribe to the service and merchandisers who would rely on it as a channel of distribution for their good and services.
In this regard, it has been found particularly important that the service interface; i.e., screen presentation and manipulation facility, enable the user to quickly and easily understand and control what is displayed. Since an interactive services represent an array of informational and transactional services intended to be embraced in a matrix of entertainment, it is essential that the user find the service simple and enjoyable to work with. If the user finds the service awkward in expression or difficult to manipulate, interest and participation soon wane, rendering it impossible to maintain the broad base subscriber and merchandiser support necessary to sustain viability.
However, providing clarity of presentation and ease of use along with comprehensive services and economy of expression, a combination essential to success, is difficult. In the past service offerors have relied on full-screen, continuous-stream displays that scroll past the user and require user recall for integration of multi-screen matter. In addition, the command structure for controlling such services has tended to be either ponderous, discrete, full-screen menus that interrupt subject matter presentation and stream of impression, or cryptic, key-combination commands that rely on user memory for application. Accordingly, interacting with such services can be more work than play, tending, at best to blunt enthusiasm, or at worst discourage it. Still further, the over reliance on user memory tends not only to color the experience as laborious, but also to both slowdown presentation as users attempt to fully understand the matter on the first pass, or require the user review the material in several passes so that a clear impression can be taken, both of which drive up the service usage costs rendering the service commercially unattractive.
SUMMARY OF INVENTION
Accordingly, it is an object of this invention to provide a method for presenting applications in an interactive service which enables clarity of application expression.
It is another object of this invention to provide a method for presenting applications in an interactive service which enables ease of service manipulation.
It is still another object of this invention to provide a method for presenting applications in an interactive service that places reduced reliance on user memory for application comprehension and exercise of service control commands.
It is yet another object of this invention to provide a method for presenting applications in an interactive service which enables concurrent, on-screen display of applications and service control commands.
It is a further object of this invention to provide a method for presenting applications in an interactive service which enables forward and reverse movement between multiple application display screens.
It is a still further object of this invention to provide a method for presenting applications in an interactive service which permits multiple procedures for navigating to successive applications.
And, it is still further object of this invention to provide a method for presenting applications in an interactive service that enables concurrent display of multiple applications, at least one of which may include advertising.
Briefly, the method for presenting applications in an interactive service in accordance with this invention achieves the above-noted and other objects by featuring steps for generating a screen displays at respective user reception systems, the screen display including a plurality of partitions for concurrently presenting at least a user-requested application and a group of command functions for managing the display. In accordance with the invention, the command functions include a subgroup of functions that facilitate random navigation to new applications, at the user's behest, employing a variety of different procedures. As well, the method includes steps for opening and closing windows on the display to enable presentation of additional data relating to the presented applications. Still further, the method includes steps for providing additional partitions for concurrently displaying other applications, which may include advertising.
In preferred form, the method features steps for presenting the command function in a command bar fixed-located on the display screen; e.g., at the screen bottom, concurrently with the displayed application so that the user can conduct display control functions while viewing an application. Additionally, in preferred form, the command bar features functions for progressing forward and backward in the application, thus reducing the need for user reliance on memory or repeat of the entire application to aid comprehension.
Also in preferred form, the method features steps for presenting a subgroup of commands at the command bar for enabling the user to randomly navigate to other available applications. Particularly, the navigation subgroup includes a command entitled "Path" which enables the user to sequence through a list of user-designated preferred applications, that, in effect, "program", progress through the service, thus enabling the user to easily and quickly review the service, as for example, on a daily basis, for news, financial, sports, business and other information.
Still further, the method features a navigation subgroup command entitled "Jump" which opens a window at the display concurrent with the application, which enables the user to select a new application for display based on a review of the available applications using either a string-descriptor search, an alphabetical search, a subject category search or a physical analogy; e.g., application to departments of a store, etc. Also in preferred form, the method features steps for opening windows over the currently displayed application to present further information concerning the application or facilitate the undertaking of interactive operation with respect to the application. As will be appreciated, this again reduces need for user reliance on memory, as it enables the providing of a reference that keeps the user oriented in progressing through a service session.
Additionally, the method of the present invention enables yet additional application to be concurrently presented at the display screen; as for example applications concerning advertising which may be of interest to the user. In these advertising-related application, the user can either obtain additional information or, if desired, undertake transactional event; as for example, buying goods or services.
DESCRIPTION OF THE DRAWINGS
The above and further objects, features and advantages of the invention will become clear from the following more detailed description when read with reference to the accompanying drawings in which:
FIG. 1 is a block diagram of the interactive computer network in which the application-presentation method of the present invention may be employed;
FIG. 2 is a schematic diagram of the network illustrated in FIG. 1;
FIGS. 3a and 3b are plan views of a display screen for a user reception system employed in a network in which the application-presentation method of the present invention may be practiced;
FIGS. 4a, 4b, 4c and 4d are schematic drawings that illustrate the structure of objects, and object segments that may be used in a network in which the application-presentation method of the present invention may be employed;
FIG. 5a is a schematic diagram that illustrates the configuration of the page template object which might be used for presentation of an application in a network in which the application-presentation method of the present invention may be practiced;
FIG. 5b is a schematic diagram that illustrates page composition which might be used for presentation of applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 6 is a schematic diagram that illustrates the protocol which might be used by a reception system for supporting applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 7 is a schematic diagram that illustrates major layers for a reception system which might be used for supporting applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 8 is a block diagram that illustrates native code modules for a reception system which might be used for supporting applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 9 is a schematic diagram that illustrates an example of a partitioned application to be processed by a reception system which might be used for supporting applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 10 illustrates generation of a page with a page processing table for a reception system which might be used for supporting applications in a network in which the application-presentation method of the present invention may be practiced;
FIG. 11 is a flow diagram for an aspect of the navigation method of a reception system which might be used for supporting applications in a network in which the application-presentation method of the present invention may be practiced.
DESCRIPTION OF THE PREFERRED EMBODIMENT
GENERAL SYSTEM DESCRIPTION
FIGS. 1 and 2 show a network in which the method of the current invention for presenting applications might be used. As seen the network, designated 10, includes a plurality of reception units within a reception layer 401 for displaying information and providing transactional services. In this arrangement, many users each access network 10 with a conventional personal computer; e.g., one of the IBM or IBM-compatible type, which has been provided with application software to constitute a reception system (RS) 400.
As seen in FIG. 1, interactive network 10 uses a layered structure that includes an information layer 100, a switch/file server layer 200, and cache/concentrator layer 300 as well as reception layer 401. This structure maintains active application databases and delivers requested parts of the databases on demand to the plurality of RSs 400, shown in FIG. 2. As seen in FIG. 2, cache/concentrator layer 300 includes a plurality of cache/concentrator units 302, each of which serve a plurality of RS 400 units over lines 301. Additionally, switch/file server layer 200 is seen to include a server unit 205 connected to multiple cache/concentrator units 302 over lines 201. Still further, server unit 205 is seen to be connected to information layer 100 and its various elements, which act as means for producing, supplying and maintaining the network databases and other information necessary to support network 10. Continuing, switch/filer layer 200 is also seen to include gateway systems 210 connected to server 205. Gateways 210 couple layer 200 to other sources of information and data; e.g., other computer systems. As will be appreciated by those skilled in the art, layer 200, like layers 401 and 300, could also include multiple servers, gateways and information layers in the event even larger numbers of users were sought to be served.
Continuing with reference to FIG. 2, in preferred form, each RS 400 is seen to include a personal computer 405 having a CPU 410 including a microprocessor (as for example, one of the types made by INTEL Corporation in its X'86 family of microprocessors), companion RAM and ROM memory and other associated elements, such as monitor 412 with screen 414 and a keyboard 424. Further, personal computer 405 may also include one or two floppy disk drives 416 for receiving diskettes 426
containing application software used to support the interactive service and facilitate the interactive sessions with network 10. Additionally, personal computer 405 would include operating systems software; e.g., MS-DOS, supplied on diskettes 428
suitable for the personal computer being used. Personal computer 405 still further may also include a hard-disk drive 420 for storing the application software and operating system software which may be transferred from diskettes 426 and 428
respectfully.
Once so configured, each RS 400 provides: a common interface to other elements of interactive computer network 10; a common environment for application processing; and a common protocol for user-application conversation which is independent of the personal computer brand used. RS 400 thus constitutes a universal terminal for which only one version of all applications on network 10 need be prepared, thereby rendering the applications interpretable by a variety of brands of personal computers.
RS 400 formulated in this fashion is capable of communication with the host system to receive information containing either of two types of data, namely objects and messages. Objects have a uniform, self-defining format known to RS 400, and include data types, such as interpretable programs and presentation data for display at monitor screen 414 of the user's personal computer 405. Applications presented at RS 400 are partitioned into objects which represent the minimal units available from the higher levels of interactive network 10 or RS 400. In this arrangement, each application partition typically represents one screen or a partial screen of information, including fields filled with data used in transactions with network 10. Each such screen, commonly called a page, is represented by its parts and is described in a page template object, discussed below.
Applications, having been partitioned into minimal units, are available from higher elements of network 10 or RS 400, and are retrieved on demand by RS 400 for interpretive execution. Thus, not all partitions of a partitioned application need be resident at RS 400 to process a selected partition, thereby raising the storage efficiency of the user's RS 400 and minimizing response time. Each application partition is an independent, self-contained unit and can operate correctly by itself. Each partition may refer to other partitions either statically or dynamically. Static references are built into the partitioned application, while dynamic references are created from the execution of program logic using a set of parameters, such as user demographics or locale. Partitions may be chosen as part of the RS processing in response to user created events, or by selecting a key word of the partitioned application (e.g., "JUMP" or "INDEX," discussed below), which provides random access to all services represented by partitioned applications having key words.
Objects provide a means of packaging and distributing partitioned applications. As noted, objects make up one or more partitioned applications, and are retrieved on demand by a user's RS 400 for interpretive execution and selective storage. All objects are interpreted by RS 400, thereby enabling applications to be developed independently of the personal computer brand used.
Objects may be nested within one another or referenced by an object identifier (object-id) from within their data structure. References to objects permit the size of objects to be minimized. Further, the time required to display a page is minimized when referenced objects are stored locally at RS 400 (which storage is determined by prior usage meeting certain retention criteria to be described more fully below), or have been pre-fetched, or in fact, are already used for the current page.
Objects carry application program instructions and/or information for display at monitor screen 414 of RS 400. Application program objects, called pre-processors and post-processors, set up the environment for the user's interaction with network
10 and respond to events created when the user inputs information at keyboard 424 of RS 400. Such events typically trigger a program object to be processed, causing one of the following: sending of transactional information to the coapplications in one layer of the network 10; the receiving of information for use in programs or for presentation in application-dependent fields on monitor screen 414; or the requesting of a new objects to be processed by RS 400. Such objects may be part of the same application or a completely new application.
The RS 400 supports a protocol by which the user and the partitioned applications communicate. All partitioned applications are designed knowing that this protocol will be supported in RS 400. Hence, replication of the protocol in each partitioned application is avoided, thereby minimizing the size of the partitioned application.
RS 400 includes a means to communicate with network 10 to retrieve objects in response to events occurring at RS 400 and to send and receive messages.
RS 400 includes a means to selectively store objects according to a predetermined storage criterion, thus enabling frequently used objects to be stored locally at the RS, and causing infrequently used objects to forfeit their local storage location. The currency of objects stored locally at the RS 400 is verified before use according to the object's storage control parameters and the storage criterion in use for version checking.
Selective storage tailors the contents of the RS 400 memory to contain objects representing all or significant parts of partitioned applications favored by the user. Because selective storage of objects is local, response time is reduced for those partitioned applications that the user accesses most frequently.
Since much of the application processing formerly done by a host computer in previously known time-sharing networks is now performed at the user's RS 400, the higher elements of network 10, particularly layer 200, have as their primary functions the routing of messages, serving of objects, and line concentration. The narrowed functional load of the higher network elements permits many more users to be serviced within the same bounds of computer power and I/O capability of conventional host-centered architectures.
Network 10 provides information on a wide variety of topics, including, but not limited to news, industry, financial needs, hobbies and cultural interests. Network 10 thus eliminates the need to consult multiple information sources, giving users an efficient and timesaving overview of subjects that interest them.
The transactional features of interactive network 10 saves the user time, money, and frustration by reducing time spent traveling, standing in line, and communicating with sales personnel. The user may, through RS 400, bank, send and receive messages, review advertising, place orders for merchandise, and perform other transactions.
In preferred form, network 10 provides information, advertising and transaction processing services for a large number of users simultaneously accessing the network via the public switched telephone network (PSTN), broadcast, and/or other media with their RS 400 units. Services available to the user include display of information such as movie reviews, the latest news, airlines reservations, the purchase of items such as retail merchandise and groceries, and quotes and buy/sell orders for stocks and bonds. Network 10 provides an environment in which a user, via RS 400 establishes a session with the network and accesses a large number of services. These services are specifically constructed applications which as noted are partitioned so they may be distributed without undue transmission time, and may be processed and selectively stored on a user's RS 400 unit.
SYSTEM CONFIGURATION
As shown in FIG. 1, interactive computer network 10 includes four layers: information layer 100, switch and file server layer 200, concentrator layer 300, and reception layer 401.
Information layer 100 handles: (1) the production, storage and dissemination of data and (2) the collection and off-line processing of such data from each RS session with the network 10 so as to permit the targeting of information and advertising to be presented to users and for traditional business support.
Switch and file server layer 200 and cache/concentrator layer 300 together constitute a delivery system 20 which delivers requested data to the RSs 400 of reception layer 401 and routes data entered by the user or collected at RSs 400 to the proper application in network 10. With reference to FIG. 2, the information used in a RS 400 either resides locally at the RS 400, or is available on demand from the cache/concentrator 300 or the file server 205, via the gateway 210, which may be coupled to external providers, or is available from information layer 100.
There are two types of information in the network 10 which are utilized by the RS 400: objects and messages.
Objects include the information requested and utilized by the RS 400 to permit a user to select specific parts of applications, control the flow of information relating to the applications, and to supply information to the network. Objects are self-describing structures organized in accordance with a specific data object architecture, described below. Objects are used to package presentation data and program instructions required to support the partitioned applications and advertising presented at a RS 400. Objects are distributed on demand throughout interactive network 10. Objects may contain: control information; program instructions to set up an application processing environment and to process user or network created events; information about what is to be displayed and how it is to be displayed; references to programs to be interpretively executed; and references to other objects, which may be called based upon certain conditions or the occurrence of certain events at the user's personal computer, resulting in the selection and retrieval of other partitioned applications packaged as objects.
Messages are information provided by the user or the network and are used in fields defined within the constructs of an object, and are seen on the user's RS monitor 412, or are used for data processing at RS 400. Additionally, and as more fully described hereafter, messages are the primary means for communication within and without the network. The format of messages is application dependent. If the message is input by the user, it is formatted by the partitioned application currently being processed on RS 400. Likewise, and with reference to FIG. 2, if the data are provided from a co-application database residing in delivery system 20, or accessed via gateway 210 or high function system 110 within the information layer 100, the partitioned application currently being processed on RS 400 causes the message data to be displayed in fields on the user's display monitor as defined by the particular partitioned application.
All active objects reside in file server 205. Inactive objects or objects in preparation reside in producer system 120. Objects recently introduced into delivery system 20 from the producer system 120 will be available from file server 205, but, may not be available on cache/concentrator 302 to which the user's RS 400 has dialed. If such objects are requested by the RS 400, the cache/concentrator 302 automatically requests the object from file server 205. The requested object is routed back to the requesting cache/concentrator 302, which automatically routes it to the communications line on which the request was originally made, from which it is received by the RS 400.
The RS 400 is the point of application session control because it has the ability to select and randomly access objects representing all or part of partitioned applications and their data. RS 400 processes objects according to information contained therein and events created by the user on personal computer 405.
Applications on network 10 act in concert with the distributed partitioned applications running on RS 400. Partitioned applications constructed as groups of objects and are distributed on demand to a user's RS 400. An application partition represents the minimum amount of information and program logic needed to present a page or window, i.e. portion of a page presented to the user, perform transactions with the interactive network 10, and perform traditional data processing operations, as required, including selecting another partitioned application to be processed upon a user generated completion event for the current partitioned application.
Objects representing all or part of partitioned applications may be stored in a user's RS 400 if the objects meet certain criteria, such as being non-volatile, non-critical to network integrity, or if they are critical to ensuring reasonable response time. Such objects are either provided on diskettes 426 together with RS 400 system software used during the installation procedure or they are automatically requested by RS 400 when the user makes selections requiring objects not present in RS
400. In the latter case, RS 400 requests from cache/concentrator layer 300 only the objects necessary to execute the desired partitioned application.
Reception system application software 426 in preferred form is provided for IBM and IBM-compatible brands of personal computers 405, and all partitioned applications are constructed according to a single architecture which each such RS 400
supports. With reference to FIG. 2, to access network 10, a user preferably has a personal computer 405 with at least 512K RAM and a single disk drive 416. The user typically accesses network 10 using a 1,200 or 2,400 bps modem (not shown). To initiate a session with network 10, objects representing the logon application are retrieved from the user's personal diskette, including the R.S. application software, which was previously set up during standard installation and enrollment procedures with network 10. Once communication between RS 400 and cache/concentrator layer 300 has been established, the user begins a standard logon procedure by inputting a personal entry code. Once the logon procedure is complete, the user can begin to access various desired services (i.e., partitioned applications) which provide display of requested information and/or transaction operations.
APPLICATIONS AND PAGES
Applications, i.e., information events, are composed of a sequence of one or more pages opened at screen 414 of monitor 412. This is better seen with reference to FIGS. 3a and 3b were a page 255 is illustrated as might appear at screen 414 of monitor 412. With reference to FIG. 3a, in accordance with the invention, each page 255 is formatted with a service interface having page partitions 250, 260, 280, and 290 (not to be confused with application partitions). Window page partitions 275, well known in the art, are also available and are opened and closed conditionally on page 255 upon the occurrence of an event specified in the application being run. Each page partition 250, 260, 280, and 290 and window 275 is made up of a page element which defines the content of the partition or window.
In preferred form, each page 255 includes: a header page partition 250, which has a page element associated with it and which typically conveys information on the page's topic or sponsor; one or more body page partitions 260 and window page partitions 275, each of which is associated with a page element which as noted gives the informational and transactional content of the page. For example, a page element may contain presentation data selected as a menu option in the previous page, and/or may contain prompts to which a user responds in pre-defined fields to execute transactions. As illustrated in FIG. 3b, the page element associated with body page partition 260 includes display fields 270, 271, 272. A window page partition 275
seen in FIG. 3a represents the same informational and transactional capability as a body partition, except greater flexibility is provided for its location and size.
Continuing with reference to FIG. 3a, in accordance with the invention, advertising 280 is provided over network 10, and, like page elements, also include information for display on page 255, and may be included in any partition of a page. Advertising 280 is presented to the user on an individualized basis from queues of advertising object identifications (ids) that are constructed off-line by business system 130, and sent to file server 205 where they are accessible to each RS 400.
Individualized queues of advertising object ids are constructed based upon data collected on the partitioned applications that were accessed by a user, and upon events the user generated in response to applications. The data are collected and reported by RS 400 to a data collection co-application in file server 205 for later transmission to business system 130. In addition to application access and use characteristics, a variety of other parameters, such as user demographics or postal ZIP code, may be used as targeting criteria. From such data, queues of advertising object ids are constructed that are targeted to either individual users or to sets of users who fall into certain groups according to such parameters. Stated otherwise, the advertising presented is individualized to the respective users based on characterizations of the respective users as defined by the interaction history with the service and such other information as user demographics and locale. As will be appreciated by those skilled in the art, conventional marketing analysis techniques can be employed to establish the user characterizations based on the collected application usage data above noted and other information.
Also with reference to FIG. 3b, the service interface is seen to include a command region 285 which enables the user to interact with the network RS 400 and other elements of network 10, so as to cause such operations as navigating from page to page, performing a transaction, or obtaining more information about other applications. As shown in FIG. 3b, interface region 285 includes a command bar 290 having a number of commands 291-298 which the user can execute. The functions of commands
291-298 are discussed in greater detail below.
NETWORK OBJECTS
As noted above, in conventional time-sharing computer networks, the data and program instructions necessary to support user sessions are maintained at a central host computer. However, that approach has been found to create processing bottlenecks as greater numbers of users are connected to the network; bottlenecks which require increases in processing power and complexity; e.g., multiple hosts of greater computing capability, if the network is to meet demand. Further, such bottlenecks have been found to also slow response time as more users are connected to the network and seek to have their requests for data processing answered.
The consequences of the host processing bottlenecking is to either compel capital expenditures to expand host processing capability, or accept longer response times; i.e., a slower network, and risk user dissatisfaction.
However, even in the case where additional computing power is added, and where response time is allowed to increase, eventually the host becomes user saturated as more and more users are sought to be served by the network. The network described above, however, is designed to alleviate the effects of host-centered limitations, and extend the network saturation point. This objective is achieved by reducing the demand on the host for processing resources by structuring the network so that the higher network levels act primarily to maintain and supply data and programs to the lower levels of the network, particularly RS 400, which acts to manage and sustain the user screen displays.
More particularly, the described network features procedures for parsing the network data and program instructions required to support the interactive user sessions into packets, referred to as objects, and distributing them into the network where they can be processed at lower levels, particularly, reception system 400.
The screens presented at the user's monitor are each divided into addressable partitions shown in FIG. 3a, and the display text and graphics necessary to make up the partitions, as well as the program instructions and control data necessary to deliver and sustain the screens and partitions are formulated from pre-created objects. Further, the objects are structured in accordance with an architecture that permits the displayed data to be relocatable on the screen, and to be reusable to make up other screens and other sessions, either as pre-created and stored sessions or interactive sessions, dynamically created in response to the user's requests.
As shown in FIG. 4c, the network objects are organized as a family of objects each of which perform a specific function in support of the interactive session. More particularly, in accordance with the preferred form, the network object family is seen to include 6 members: page format objects 502, page element objects 504, window objects 506, program objects 508, advertisement objects 510 and page template objects 500.
Within this family, page format objects 502 are designed to define the partitioning 250 to 290 of the monitor screen shown in FIG. 3a. The page format objects 502 provide a means for pre-defining screen partitions and for ensuring a uniform look to the page presented on the reception system monitor. They provide the origin; i.e., drawing points, and dimensions of each page partition and different values for presentation commands such as palette and background color.
Page format objects 502 are referenced whenever non-window data is to be displayed and as noted ensure a consistent presentation of the page. In addition, page format objects 502 assures proper tessellation or "tiling" of the displayed partitions.
Page element objects 504, on the other hand, are structured to contain the display data; i.e., text and graphic, to be displayed which is mapped within screen partitions 250 to 290, and to further provide the associated control data and programs. More specifically, the display data is described within the object as NAPLPS data, and includes, PDI, ASCII, Incremental Point and other display encoding schemes. Page element objects also control the functionality within the screen partition by means of field definition segments 516 and program call segments 532, as further described in connection with the description of such segments hereafter. Page element objects 504 are relocatable and may be reused by many pages. To enable the displayable data to be relocated, display data must be created by producers in the NAPLPS relative mode.
Continuing with reference to FIG. 4c, window objects 506 include the display and control data necessary to support window partitions 275 best seen in FIG. 3a. Windows contain display data which overlay the base page and control data which supersede the base page control data for the underlying screen during the duration of the window. Window objects 506 contain data which is to be displayed or otherwise presented to the viewer which is relatively independent from the rest of the page. Display data within windows overlay the base page until the window is closed. Logic associated with the window supersedes base page logic for the duration of the window. When a window is opened, the bit map of the area covered by window is saved and most logic functions for the overlaid page are deactivated. When the window is closed, the saved bit map is swapped onto the screen, the logic functions associated with the window are disabled, and prior logic functions are reactivated.
Windows are opened by user or program control. They do not form part of the base page. Windows would typically be opened as a result of the completion of events specified in program call segments 532.
Window objects 506 are very similar in structure to page element objects 504. The critical difference is that window objects 506 specify their own size and absolute screen location by means of a partition definition segment 528.
Program objects 508 contain program instructions written in a high-level language called TRINTEX Basic Object Language, i.e., TBOL, described in greater detail hereafter, which may be executed on RS 400 to support the application. More particularly, program objects 508 include interpretable program code, executable machine code and parameters to be acted upon in conjunction with the presentation of text and graphics to the reception system monitors.
Program objects 508 may be called for execution by means of program call segments 532, which specify when a program is to be executed (event), what program to execute (program pointer), and how programs should run (parameters).
Programs are treated as objects to conform to the open-ended design philosophy of the data object architecture (DOA), allowing the dissemination of newly developed programs to be easily and economically performed. As noted above, it is desirable to have as many of these program objects staged for execution at or as close to RS 400 as possible.
Still further, advertising objects 510 include the text and graphics that may be presented at ad partition 280 presented on the monitor screen as shown in FIG. 3b.
Finally, the object family includes page template objects 500. Page template objects 500 are designed to define the components of the full screen presented to the viewer. Particularly, page template objects 500 include the entry point to a screen, the name of the page format objects which specify the various partitions a screen will have and the page element object that contain the display data and partitioning parameters for the page.
Additionally, page template object 500 includes the specific program calls required to execute the screens associated with the application being presented to the user, and may serve as the means for the user to selectively move through; i.e., navigate the pages of interest which are associated with various applications. Thus, in effect, page template objects 500 constitute the "recipe" for making up the collection of text and graphic information required to make the screens to be presented to the user.
Objects 500 to 510 shown in FIG. 4c are themselves made up of further sub-blocks of information that may be selectively collected to define the objects and resulting pages that ultimately constitute the application presented to the user in an interactive text and graphic session.
More specifically and as shown schematically in FIG. 4a, objects 500 to 510 are predefined, variable length records consisting of a fixed length header 551 and one or more self-defining record segments 552 a list of which is presented in FIG. 4c as segment types 512 to 540.
In accordance with this design, and as shown in FIG. 4b, object header 551 in preferred form is 18 bytes in length and contains a prescribed sequence of information which provides data regarding the object's identification, its anticipated use, association to other objects, its length and its version and currency.
More particularly, each of the 18 bytes of object header 551 are conventional hexadecimal, 8 bit bytes and are arranged in a fixed pattern to facilitate interpretation by network 10. Particularly, and as shown in FIG. 4b, the first byte of header 551; i.e., byte 1, identifies the length of the object ID in hexadecimal. The next six bytes; i.e., bytes 2 to 7, are allocated for identifying access control to the object so as to allow creation of closed user groups to whom the object(s) is to be provided. As will be appreciated by those skilled in the art, the ability to earmark objects in anticipation of user requests enables the network to anticipate requests and pre-collect objects from large numbers of them maintained to render the network more efficient and reduce response time. The following 4 bytes of header 551; bytes 8 to 11, are used to identify the set of objects to which the subject object belongs. In this regard, it will be appreciated that, again, for speed of access and efficiency of selection, the objects are arranged in groups or sets which are likely to be presented to user sequentially in presenting the page sets; i.e., screens that go to make up a session.
Following identification of the object set, the next byte in header 551; i.e., byte 12, gives the location of the subject object in the set. As will be appreciated here also the identification is provided to facilitate ease of object location and access among the many thousands of objects that are maintained to, thereby, render their selection and presentation more efficient and speedy.
Thereafter, the following byte of header 551; i.e., byte 13, designates the object type; e.g., page format, page template, page element, etc. Following identification of the object type, two bytes; i.e., bytes 14, 15, are allocated to define the length of the object, which may be of whatever length is necessary to supply the data necessary, and thereby provides great flexibility for creation of the screens. Thereafter, in accordance with the preferred form, a single byte; i.e., byte 16, is allocated to identify the storage characteristic for the object; i.e., the criterion which establishes at what level in network 10 the object will be stored, and the basis upon which it will be updated. At least a portion of this byte; i.e, the higher order nibble (first 4 bits reading from left to right) is associated with the last byte; i.e., byte 18, in the header which identifies the version of the object, a control used in determining how often in a predetermined period of time the object will be updated by the network.
Following storage characteristic byte 16, header 551 includes a byte; i.e., 17, which identifies the number of objects in the set to which the subject object belongs. Finally, and as noted above, header 551 includes a byte; i.e., 18, which identifies the version of the object. Particularly the object version is a number to establish the control for the update of the object that are resident at RS 400.
As shown in FIG. 4a, and as noted above, in addition to header 551, the object includes one more of the various segment types shown in FIG. 4c.
Segments 512 to 540 are the basic building blocks of the objects. And, as in the case of the object, the segments are also self-defining. As will be appreciated by those skilled in the art, by making the segments self-defining, changes in the objects and their use in the network can be made without changing pre-existing objects.
As in the case of objects, the segments have also been provided with a specific structure. Particularly, and as shown in FIG. 4a, segments 552 consists of a designation of segment type 553, identification of segment length 554, followed by the information necessary to implement the segment and its associated object 555; e.g., either, control data, display data or program code.
In this structure, segment type 553 is identified with a one-byte hexadecimal code which describes the general function of the segment. Thereafter, segment length 554 is identified as a fixed two-byte long field which carries the segment length as a hexadecimal number in INTEL format; i.e., least significant byte first. Finally, data within segments may be identified either by position or keyword, depending on the specific requirements of the segment.
The specific structure for the objects and segments in shown in FIG. 4c and is described below. In that description the following notation convention is used: ##STR1##
The structure for objects is:
PAGE TEMPLATE OBJECT,
[<header> (compression descriptor) <page format call> (page element call) . . . (program call) . . . (page element selector) (system table call) . . . external reference) (keyword/navigation) . . . ];
As noted above, page format objects 502 are designed to define the partitioning 250 to 290 of monitor screen 414 shown in FIG. 3a.
PAGE FORMAT OBJECT,
[<header> (compression descriptor) (page defaults) <partition definition>];
PAGE ELEMENT OBJECT,
[<header> (compression descriptor) (presentation data) . . . (program call) . . . (custom cursor) . . . (custom text) . . . (field definition) . . . (field-level program call) . . . (custom cursor type 2) . . . (custom graphic) . . . (field definition type 2) . . . (array definition) . . . (inventory control)];
Page element objects, as explained, are structured to contain the display data; i.e., text and graphics, to be presented at screen partitions 250 to 290.
WINDOW OBJECT,
[<header> (compression description) <partition definition> (page element call) (presentation data) . . . (program call) . . . (custom cursor) . . . (custom text) . . . (custom cursor type 2) . . . (custom graphic) . . . (field definition) . . . (field level program call) . . . (field definition type 2) . . . (array definition) . . . (inventory control)];
As noted, window objects include display and control data necessary to support window partition at screen 414.
PROGRAM OBJECTS,
[<header> (compression descriptor) <program data> . . . ].
Program objects, on the other hand, contain program instructions written in higher-level language which may be executed at RS 400 to support the application.
ADVERTISEMENT OBJECT,
[<header> (compression descriptor) (presentation data) . . . (program call) . . . (custom cursor) . . . (custom text) . . . (field definition) . . . (field-level program call) . . . (custom cursor type 2) . . . (custom graphic) . . . (field definition type 2) . . . (array definition) . . . (inventory control)];
and as can be seen, advertisement objects are substantially the same as page element objects, with the difference being that, as their name implies, their subject matter is selected to concern advertising.
Continuing, the structure for the object segments follows from the above description, and is as described more fully in parent application Ser. No. 388,156 now issued as U.S. Pat. No. 5,347,632the contents of which patent are incorporated herein by reference.
NETWORK MESSAGES
In addition to the network objects, and the display data, control data, and the program instructions they contain as previously described, network 10 also exchanges information regarding the support of user sessions and the maintenance of the network as "messenger". Specifically, messages typically relate to the exchange of information associated with initial logon of a reception system 400 to network 10, dialogue between RS 400 and other elements and communications by the other network elements amongst themselves.
To facilitate message exchange internally, and through gateway 210 to entities externally to network 10, a protocol termed the "Data Interchange Architecture" (DIA) is used to support the transport and interpretation of information. More particularly, DIA enables: communications between RS 400 units, separation of functions between network layers 100, 200, 300 and 401; consistent parsing of data; an "open" architecture for network 10; downward compatibility within the network; compatibility with standard industry protocols such as the IBM System Network Architecture; Open Systems Interconnections standard; support of network utility sessions; and standardization of common network and application return codes.
Thus DIA binds the various components of network 10 into a coherent entity by providing a common data stream for communications management purposes. DIA provides the ability to route messages between applications based in IBM System Network Architecture (SNA), (well known in the art, and more fully described in Data and Computer Communications, by W. Stallings, Chapter 12, McMillian Publishing, Inc. (1985)) and non-SNA reception system applications; e.g. home computer applications. Further, DIA provides common data structure between applications run at RS 400 units and applications that may be run on external computer networks; e.g. Dow Jones Services, accessed through gateway 210. As well, DIA provides support for utility sessions between backbone applications run within network 10. A more detailed description of network messaging in provided in parent application Ser. No. 388,156 now issued as U.S. Pat. No. 5,347,632, the contents of which patent are incorporated herein by reference.
OBJECT LANGUAGE
In accordance with the design of network 10, in order to enable the manipulation of the network objects, the application programs necessary to support the interactive text/graphic sessions are written in a high-level language referred to as "TBOL", (TRINTEX Basic Object Language, "TRINTEX" being the former company name of one of the assignees of this invention). TBOL is specifically adapted for writing the application programs so that the programs may be compiled into a compact data stream that can be interpreted by the application software operating in the user personal computer, the application software being designed to establish the network Reception System 400 previously noted and described in more detail hereafter.
The Reception System application software supports an interactive text/graphics sessions by managing objects. As explained above, objects specify the format and provide the content; i.e., the text and graphics, displayed on the user's screen so as to make up the pages that constitute the application. As also explained, pages are divided into separate areas called "partitions" by certain objects, while certain other objects describe windows which can be opened on the pages. Further, still other objects contain TBOL application programs which facilitate the data processing necessary to present the pages and their associated text and graphics.
As noted, the object architecture allows logical events to be specified in the object definitions. An example of a logical event is the completion of data entry on a screen; i.e., an application page. Logical events are mapped to physical events such as the user pressing the <ENTER> key on the keyboard. Other logical events might be the initial display of a screen page or the completion of data entry in a field. Logical events specified in page and window object definitions can be associated with the call of TBOL program objects.
RS 400 is aware of the occurrence of all physical events during the interactive text/graphic sessions. When a physical event such as depression of the forward <TAB> key corresponds to a logical event such as completion of data entry in a field, the appropriate TBOL program is executed if specified in the object definition. Accordingly, the TBOL programs can be thought of as routines which are given control to perform initialization and post-processing application logic associated with the fields, partitions and screens at the text/graphic sessions.
Reception System 400 run time environment uses the TBOL programs and their high-level key-word commands called verbs to provide all the system services needed to support a text/graphic session, particularly, display management, user input, local and remote data access.
TBOL programs have a structure that includes three sections: a header section in which the program name is specified; a data section in which the data structure the program will use are defined; and a code section in which the program logic is provided composed of one or more procedures. More specifically, the code section procedures are composed of procedure statements, each of which begins with a TBOL key-word called a verb.
The name of a procedure can also be used as the verb in a procedure statement exactly as if it were a TBOL key-word verb. This feature enables a programmer to extend the language vocabulary to include customized application-oriented verb commands.
Continuing, TBOL programs have a program syntax that includes a series of "identifiers" which are the names and labels assigned to programs, procedures, and data structures.
An identifier may be up to 31 characters long; contain only uppercase or lowercase letters A through Z, digits 0 through 9, and/or the special character underscore (.sub.--); and must begin with a letter. Included among the system identifiers are: "header section identifiers" used in the header section for the program name; "data section identifiers" used in the data section for data structure names, field names and array names; and finally, "code section identifiers" used in the code section for identification of procedure names and statement labels. A more detailed description of TBOL is provided in parent application Ser. No. 388,156 now issued as U.S. Pat. No. 5,347,632, the contents of which patent are incorporated herein by reference.
RECEPTION SYSTEM OPERATION
RS 400 of computer system network 10 uses software called native code modules (described below) to enable the user to select options and functions presented on the monitor of personal computer 405, to execute partitioned applications and to process user created events, enabling the partitioned application to interact with network 10. Through this interaction, the user is able to input data into fields provided as part of the display, or may individually select choices causing a standard or personalized page to be built (as explained below) for display on the monitor of personal computer 405. Such inputs will cause RS 400 to interpret events and trigger pre-processors or post-processors, retrieve specified objects, communicate with system components, control user options, cause the display of advertisements on a page, open or close window partitions to provide additional navigation possibilities, and collect and report data about events, including certain types of objects processed. For example, the user may select a particular option, such as opening or closing window partition 275, which is present on the monitor screen 414 and follow the selection with a completion key stroke, such as ENTER. When the completion keystroke is made, the selection is translated into a logical event that triggers the execution of a post-processor (i.e., a partitioned application program object) to process the contents of the field.
In accordance with the invention, functions supporting the user-partitioned application interface can be performed using the command bar 290, or its equivalent using pull down windows or an overlapping cascade of windows. These functions can be implemented as part of the RS native functions or can be treated as another partition(s) defined for every page for which an appropriate set of supporting objects exist and remain resident at RS 400. If the functions are part of RS 400, they can be altered or extended by verbs defined in the RS virtual machine that permit the execution of program objects to be triggered when certain functions are called, providing maximum flexibility.
To explain the functions the use of a command bar is assumed. Command bar 290 is shown in FIGS. 3a and 3b and includes a NEXT command 291, a BACK command 292, a PATH command 293, a MENU command 294, an ACTION command 295, a JUMP command 296, a HELP command 297, and an EXIT command 298.
NEXT command 291 causes the next page in the current page set to be built. If the last page of a page set has already been reached, NEXT command 291 is disabled by RS 400, avoiding the presentation of an invalid option.
BACK command 292 causes the previous page of the current page set to be built. If the present page is the first in the page set, BACK command 292 is disabled, since it is not a valid option.
A filter program can be attached to both the NEXT or BACK functions to modify their implicit sequential nature based upon the value of the occurrence in the object set id.
PATH command 293 causes the next page to be built and displayed from a list of pages that the user has entered, starting from the first entry for every new session.
MENU command 294 causes the page presenting the previous set of choices to be rebuilt.
ACTION command 295 initiates an application dependent operation such as causing a new application partition to be interpreted, a window partition 275 to be opened and enables the user to input any information required which may result in a transaction or selection of another window or page.
JUMP command 296 causes window partition 275 to be opened, allowing the user to input a keyword or to specify one from an index that may be selected for display.
HELP command 297 causes a new application partition to be interpreted such as a HELP window pertaining to where the cursor is positioned to be displayed in order to assist the user regarding the present page, a particular partition, or a field in a page element.
EXIT command 298 causes a LOGOFF page template object (PTO) to be built, and a page logoff sequence to be presented at RS 400 monitor screen 414.
NAVIGATION INTERFACE
Continuing, as a further feature, network 10 includes an improved procedure for searching and retrieving applications from the store of applications distributed throughout network 10; e.g., server 205, cache/concentrator 302 and RS 400. More specifically, the procedure features use of pre-created search tables which represent subsets of the information on the network arranged with reference to the page template objects (PTO) and object-ids of the available applications so that in accordance with the procedure, the relevant tables and associated objects can be provided to and searched at the requesting RS 400 without need to search the entire store of applications on the network. As will be appreciated, this reduces the demand on the server
205 for locating and retrieving applications for display at monitor 412.
In conventional time-sharing networks that support large conventional databases, the host receives user requests for data records; locates them; and transmits them back to the users. Accordingly, the host is obliged to undertake the data processing necessary to isolate and supply the requested information. And, as noted earlier, where large numbers of users are to be served, the many user requests can bottleneck at the host, taxing resources and leading to response slowdown.
Further, users have experienced difficulty in searching databases maintained on conventional time-sharing networks. For example, difficulties have resulted from the complex and varied way previously known database suppliers have organized and presented their information. Particularly, some database providers require searching be done only in selected fields of the database, thus requiring the user to be fully familiar with the record structure. Others have organized their databases on hierarchial structures which require the user understand the way the records are grouped. Still further, yet other database suppliers rely upon keyword indices to facilitate searching of their records, thus requiring the user to be knowledgeable regarding the particular keywords used by the database provider.
Network 10, however, is designed to avoid such difficulties. In the preferred form, the network includes procedures for creating preliminary searches which represent subsets of the network applications users are believed likely to investigate. Particularly, in accordance with these procedures, for the active applications available on network 10, a library of tables is prepared, and maintained within each of which a plurality of so called "keywords" are provided that are correlated with page template objects and object-ids of the entry screen (typically the first screen) for the respective application. In the preferred embodiment, approximately 1,000 tables are used, each having approximately 10 to 20 keywords arranged in alphabetical order to abstract the applications on the network. Further, the object-id for each table is associated with a code in the form of a character string mnemonic which is arranged in a set of alphabetically sequenced mnemonics termed the sequence set so that on entry of a character string at an RS 400, the object-id for the relevant keyword table can be obtained from the sequence set. Once the table object-id is identified, the keyword table corresponding to the desired subset of the objects and associated applications can then be obtained from network 10. Subsequently the table can be presented to the user's RS 400, where the RS 400 can provide the data processing required to present the potentially relevant keywords, objects and associated applications to the user for further review and determination as to whether more searching is required. As will be appreciated, this procedure reduces demand on server 205 and thereby permits it to be less complex and costly, and further, reduces the likelihood of host overtaxing that may cause network response slowdown.
As a further feature of this procedure, the library of keywords and their associated PTOs and objects may be generated by a plurality of operations which appear at the user's screen as different search techniques. This permits the user to select a search technique he is most comfortable with, thus expediting his inquiry.
More particularly, the user is allowed to invoke the procedure by calling up a variety of operations. The various operations have different names and seemingly present different search strategies. Specifically, the user may invoke the procedure by initiating a "Jump" command at RS 400. Thereafter, in connection with the Jump operation, the user, when prompted, may enter a word of the user's choosing at monitor screen 414 relating to the matter he is interested in locating; i.e., a subject matter search of the network applications. Additionally, the users may invoke the procedure by alternatively calling up an operation termed "Index" with selection of the Index command. When selected, the Index command presents the user with an alphabetical listing of keywords from the tables noted above which the user can select from; i.e., an alphabetical search of the network applications. Further, the user may evoke the procedure by initiating an operation termed "Guide." By selecting the Guide command, the user is provided with a series of graphic displays that presents a physical description of the network applications; e.g., department floor plan for a store the user may be electronically shopping in. Still further, the user may invoke the procedures by initiating an operation termed "Directory." By selecting the Directory command, the user is presented with the applications available on the network as a series of hierarchial menus which present the content of the network information in commonly understood categories. Finally, the user may invoke the procedure by selecting the "Path" command, which accesses a list of keywords the user has previously selected; i.e., a personally tailored form of the Index command described above. As described hereafter, Path further includes a Viewpath operation which permits the user to visually access and manage the Path list of keywords. In preferred form, where the user has not selected a list of personalized keywords, a default set is provided which includes a predetermined list and associated applications deemed by network 10 as likely to be of interest to the user.
This ability to convert these apparently different search strategies in a single procedure for accessing pre-created library tables is accomplished by translating the procedural elements of the different search techniques into a single set of procedures that will produce a mnemonic; i.e., code word, which can first be searched at the sequence set, described above to identify the object-id for the appropriate library table and, thereafter, enable access of the appropriate table to permit selection of the desired keyword and associated PTO and object-ids. That is to say, the reception system native code simply relates the user-entered character string, alphabetical range, category, or list item of respectively, "Jump", "Index", "Directory", or "Path" to the table codes through the sequence set, so that the appropriate table can be provided to the reception system and application keyword selected. Thus, while the search techniques may appear different to the user, and in fact accommodate the user's preferences and sophistication level, they nonetheless invoke the same efficient procedure of relying upon pre-created searches which identify related application PTOs and object-ids so that the table and objects may be collected and presented at the user's RS 400 where they can be processed, thereby relieving server 205.
In preferred form, however, in order to enhance presentation speed the Guide operation is specially configured. Rather than relating the keyword mnemonic to a sequence set to identify the table object-id and range of keywords corresponding to the entry PTO and associated object-ids, the Guide operation presents a series of overlapping windows that physically describe the "store" in which shopping is being conducted or the "building" from which information is being provided. The successive windows increase in degree of detail, with the final window presenting a listing of relevant keywords. Further, the PTO and object-ids for the application entry screen are directly related to the graphic presentation of the keywords. This eliminates the need to provide variable fields in the windows for each of the keywords and enables the entry screen to be correlated directly with the window graphic. As will be appreciated, this reduces the number of objects that would otherwise be required to be staged at RS 400 to support pretention of the keyword listing at monitor screen 414, and thus speeds network response.
A more detailed understanding of the procedure may be had upon a reading of the following description and review of accompanying FIGS. 2, 3a and particularly FIG. 11 which presents a flow diagram for the Jump sequence of the search procedure.
To select a particular partitioned application from among thousands of such applications residing either at the RS 400 or within delivery system 20, network 10 avoids the need for a user to know or understand, prior to a search, the organization of such partitioned applications and the query techniques necessary to access them. This is accomplished using a collection of related commands, as described below.
In accordance with the invention, the Jump command 296 as seen in FIG. 3a, can be selected, by the user from command bar 290. When Jump command 296 is selected, a window partition 275 is opened. In window 275, the user is presented and may select from a variety of displayed options that include among others, the Directory command, the Index command, and the Guide command, which when selected, have the effect noted above. Additionally, the user can select a command termed Viewpath which will presents the keywords that currently make up the list of keywords associated with the user's Path command, and from which list the user can select a desired keyword. Still further, and with reference FIG. 11, which shows the sequence where a user offers a term to identify a subject of interest, the user may enter a keyword at display field 270 within window partition 275 as a "best guess" of the mnemonic character string that is assigned to a partitioned application the user desires (e.g., the user may input such english words as "news," "pet food," "games," etcetera). Where the user enters a character string it is displayed in field 270, and then searched by RS 400 native code (discussed below) against the sequence sets above noted to identify the object-id for the appropriate table of keywords (not shown) that RS 400 may request from host 205. While as noted above, a table may include 10 to 20 keywords, in the preferred embodiment, for the sake of speed and convenience, a typical keyword table includes approximately 12 keywords.
If the string entered by the user matches a keyword existing on one of the keyword tables, and is thus associated with a specific PTO, RS 400 fetches and displays associated objects of the partitioned applications and builds the entry page in accordance with the page composition dictated by the target PTO.
If the string entered by the user does not match a specific keyword, RS 400 presents the user with the option of displaying the table of keywords approximating the specific keyword. The approximate keywords are presented as initialized, cursorable selector fields of the type provided in connection with a Index command. The user may then move the cursor to the nearest approximation of the mnemonic he originally selected, and trigger navigation to the PTO associated with that keyword, navigation being as described hereafter in connection with the RS 400 native code.
If, after selecting the Jump command, the user selects the Index command, RS 400 will retrieve the keyword table residing at RS 400, and will again build a page with initialized, cursorable fields of keywords. The table fetched upon invoking the Index command will be comprised of alphabetic keywords that occur within the range of the keywords associated with the page template object (PTO) from which the user invoked the Index command. As discussed above, the user may select to navigate to any of this range of PTOs by selecting the relevant keyword from the display. Alternatively, the user can, thereafter, select another range of alphabetical keywords by entering an appropriate character string in a screen field provided or move forward or backward in the collection by selecting the corresponding option.
By selecting the Directory command, RS 400 can be caused to fetch a table of keywords, grouped by categories, to which the PTO of the current partitioned application (as specified by the object set field 630 of the current PEO) belongs. Particularly, by selecting the Directory command, RS 400, is causes to displays a series of screens each of which contains alphabetically arranged general subject categories from which the user may select. Following selection of a category, a series of keywords associated with the specified category are displayed in further screens together with descriptive statements about the application associated with the keywords. Thereafter, the user can, in the manner previously discussed with regard to the Index command, select from and navigate to the PTOs of keywords which are related to the present page set by subject.
The Guide command provides a navigation method related to a hierarchical organization of applications provided on network 10, and are described by a series of sequentially presented overlaying windows of a type known in the art, each of which presents an increasing degree of detail for a particular subject area, terminating in a final window that gives keywords associated with the relevant applications. The Guide command makes use of the keyword segment which describes the location of the PTO in a hierarchy (referred to, in the preferred embodiment, as the "BFD," or Building-Floor-Department) as well as an associated keyword character string. The BFD describes the set of menus that are to be displayed on the screen as the sequence of pop-up windows. The Guide command may be invoked by requesting it from the Jump window described above, or by selecting the Menu command on Command Bar 290. As noted above, in the case of the Guide command, the PTO and object-ids for the application entry screen are directly associated with the graphic of the keyword presented in the final pop-up window. This enables direct access of the application entry screen without need to access the sequence set and keyword table, and thus, reduces response time by reducing the number of objects that must be processed at RS 400.
Activation of the Path command accesses the user's list of pre-selected keywords without their display, and permits the user to step through the list viewing the respective applications by repeatedly invoking the Path command. As will be appreciated, the user can set a priority for selecting keywords and viewing their associated applications by virtue of where on the list the user places the keywords. More specifically, if the user has several application of particular interest; e.g., news, weather, etc., the user can place them at the top of the list, and quickly step through them with the Path command. Further, the user can view and randomly access the keywords of his list with the Viewpath operation noted above. On activation of Viewpath, the user's Path keywords are displayed and the user can cursor through them in a conventional manner to select a desired one. Further, the user can amend the list as desired by changing the keywords on the list and/or adjusting their relative position. This is readily accomplished by entering the amendments to the list presented at the screen 414 with a series of amendment options presented in a conventional fashion with the list. As noted, the list may be personally selected by the user in the manner described, or created as a default by network 10.
Collectively, the Jump command, Index command, Directory command, Guide command, and Path command as described enable the user to quickly and easily ascertain the "location" of either the partitioned application presently displayed or the "location" of a desired partitioned application. "Location," as used in reference to the preferred embodiment means the specific relationships that a particular partitioned application bears to other such applications, and the method for selecting particular partitioned applications from such relationships. The techniques for querying a database of objects, embodied in network 10 is an advance over the prior art, insofar as no foreknowledge of either database structure or query technique or syntax is necessary, the structure and search techniques being made manifest to the user in the course of use of the commands.
RS APPLICATION PROTOCOL
RS protocol defines the way the RS supports user application conversation (input and output) and the way RS 400 processes a partitioned application. Partitioned applications are constructed knowing that this protocol will be supported unless modified by the application. The protocol is illustrated FIG. 6. The boxes in FIG. 6 identify processing states that the RS 400 passes through and the arrows indicate the transitions permitted between the various states and are annotated with the reason for the transition.
The various states are: (A) Initialize RS, (B) Process Objects, (C) Interpretively Execute Pre-processors, (D) Wait for Event, (E) Process Event, and (F) Interpretively Execute Function Extension and/or Post-processors.
The transitions between states are: (1a) Logon Page Template Object Identification (PTO-id), (1b) Object Identification, (2) Trigger Program Object identification (PO-id) & return, (3) Page Partition Template (PPT) or Window Stack Processing complete, (4) Event Occurrence, and (5) Trigger PO-id and Return.
Transition (la) from Initialize RS (A) to Process Objects (B) occurs when an initialization routine passes the object-id of the logon PTO to object interpreter 435, when the service is first invoked. Transition (1b) from Process Event (E) to Process Objects (B) occurs whenever a navigation event causes a new page template object identification (PTO-id) to be passed to object interpreter 435; or when a open window event (verb or function key) occurs passing a window object-id to the object interpreter 435; or a close window event (verb or function key) occurs causing the current top-most window to be closed.
While in the process object state, object interpreter 435 will request any objects that are identified by external references in call segments. Objects are processed by parsing and interpreting the object and its segments according to the specific object architecture. As object interpreter 435 processes objects, it builds a linked list structure called a page processing table (PPT), shown in FIG. 10, to reflect the structure of the page, each page partition, Page Element Objects (PEOs) required, program objects (POs) required and each window object (WO) that could be called. Object interpreter 435 requests all objects required to build a page except objects that could be called as the result of some event, such as a HELP window object.
Transition (2) from Process Objects (B) to Interpretively Execute Pre-processors (C) occurs when the object interpreter 435 determines that a pre-processor is to be triggered. Object processor 436 then passes the object-id of the program object to the TBOL interpreter 438. TBOL interpreter 438 uses the RS virtual machine to interpretively execute the program object. The PO can represent either a selector or an initializer. When execution is complete, a transition automatically occurs back to Process Objects (B).
Selectors are used to dynamically link and load other objects such as PEOs or other PDOs based upon parameters that they are passed when they are called. Such parameters are specified in call segments or selector segments. This feature enables RS 400 to conditionally deliver information to the user base upon predetermined parameters, such as his personal demographics or locale. For example, the parameters specified may be the transaction codes required to retrieve the user's age, sex, and personal interest codes from records contained in user profiles stored at the switch/file server layer 200.
Initializers are used to set up the application processing environment for a partitioned application and determine what events RS 400 may respond to and what the action will be.
Transition (3) from Process Objects (B) to Wait for Event (D) occurs when object interpreter 435 is finished processing objects associated with the page currently being built or opening or closing a window on a page. In the Wait for Event state (D), an input manager, which in the preferred form shown includes keyboard manager 434 seen in FIG. 8, accepts user inputs. All keystrokes are mapped from their physical codes to logical keystrokes by the Keyboard Manager 434, representing keystrokes recognized by the RS virtual machine.
When the cursor is located in a field of a page element, keystrokes are mapped to the field and the partitioned external variable (PEV) specified in the page element object (PEO) field definition segment by the cooperative action of keyboard manager, 434 and display manager 461. Certain inputs, such as RETURN or mouse clicks in particular fields, are mapped to logical events by keyboard manager 434, which are called completion (or commit) events. Completion events signify the completion of some selection or specification process associated with the partitioned application and trigger a partition level and/or page level post-processor to process the "action" parameters associated with the user's selection and commit event.
Such parameters are associated with each possible choice or input, and are set up by the earlier interpretive execution of an initializer pre-processor in state (C). Parameters usually specify actions to perform a calculation such as the balance due on an order of several items with various prices using sales tax for the user's location, navigate to PTO-id, open window WO-id or close window. Actions parameters that involve the specification of a page or window object will result in transition (1b) to the Process Objects (B) state after the post-processor is invoked as explained below.
Function keys are used to specify one or more functions which are called when the user strikes these keys. Function keys can include the occurrence of logical events, as explained above. Additionally, certain functions may be "filtered", that is, extended or altered by SET.sub.-- FUNCTION or TRIGGER.sub.-- FUNCTION verbs recognized by the RS virtual machine. Function keys cause the PO specified as a parameter of the verb to be interpretively executed whenever that function is called. Applications use this technique to modify or extend the functions provided by the RS.
Transition (5) from Process Event (E) to Interpretively Execute Pre-processors (F) occurs when Process Event State determines that a post-processor or function extension PDO is to be triggered. The id of the program object is then passed to the TBOL interpreter 438. The TBOL interpreter 438 uses the RS virtual machine to interpretively execute the PO. When execution is complete a transition automatically occurs back to Process Event (E).
RECEPTION SYSTEM SOFTWARE
The reception system 400 software is the interface between the user of personal computer 405 and interactive network 10. The object of reception system software is to minimize mainframe processing, minimize transmission across the network, and support application extendibility and portability.
RS 400 software is composed of several layers, as shown in FIG. 7. It includes external software 451, which is composed of elements well known to the art such as device drivers, the native operating systems; e.g., MS-DOS, machine-specific assembler functions (in the preferred embodiment; e.g., CRC error checking), and "C" runtime library functions; native software 420; and partitioned applications 410.
Again with reference to FIG. 7, native software 420 is compiled from the "C" language into a target machine-specific executable, and is composed of two components: the service software 430 and the operating environment 450. Operating environment
450 is comprised of the Logical Operating System 432, or LOS; and a multitasker 433. Service software 430 provides functions specific to providing interaction between the user and interactive network 10, while the operating environment 450 provides pseudo multitasking and access to local physical resources in support of service software 430. Both layers of native software 420 contain kernel, or device independent functions 430 and 432, and machine-specific or device dependent functions 433. All device dependencies are in code resident at RS 400, and are limited to implementing only those functions that are not common across machine types, to enable interactive network 10 to provide a single data stream to all makes of personal computer which are of the IBM or IBM compatible type. Source code for the native software 420 is included in parent application Ser. No. 388,156 now issued as U.S. Pat. No. 5,347,632, the contents of which patent are incorporated herein by reference. Those interested in a more detailed description of the reception system software may refer to the source code provided in the referenced patent.
Service software 430 is comprised of modules, which are device-independent software components that together obtain, interpret and store partitioned applications existing as a collection of objects. The functions performed by, and the relationship between, the service software 430 module is shown in FIG. 8 and discussed further below.
Through facilities provided by LOS 432 and multitasker 433, here called collectively operating environment 450, device-independent multitasking and access to local machine resources, such as multitasking, timers, buffer management, dynamic memory management, file storage and access, keyboard and mouse input, and printer output are provided. The operating environment 450 manages communication and synchronization of service software 430, by supporting a request/response protocol and managing the interface between the native software 420 and external software 437.
Applications software layer 410 consists of programs and data written in an interpretive language, "TRINTEX Basic Object Language" or "TBOL," described above. TBOL was written specifically for use in RS 400 and interactive network 10 to facilitate videotext-specific commands and achieve machine-independent compiling. TBOL is constructed as objects, which in interaction with one another comprise partitioned applications.
RS native software 420 provides a virtual machine interface for partitioned applications, such that all objects comprising partitioned applications "see" the same machine. RS native software provides support for the following functions: (1) keyboard and mouse input; (2) text and graphics display; (3) application interpretation; (4) application database management; (5) local application storage; (6) network and link level communications; (7) user activity data collection; and (8) advertisement management.
With reference to FIG. 8, service software 430 is comprised of the following modules: start-up (not shown); keyboard manger 434; object interpreter 435; TBOL interpreter 438; object storage facility 439; display manager 461; data collection manager 441; ad manager 442; object/communications manager interface 443; link communications manager 444; and fatal error manager 469. Each of these modules has responsibility for managing a different aspect of RS 400.
Startup reads RS 400 customization options into RAM, including modem, device driver and telephone number options, from the file CONFIG.SM. Startup invokes all RS 400 component startup functions, including navigation to the first page, a logon screen display containing fields initialized to accept the user's id and password. Since Startup is invoked only at initialization, for simplicity, it has not been shown in FIG. 8.
The principal function of keyboard manger 434 is to translate personal computer dependent physical input into a consistent set of logical keys and to invoke processors associated with these keys. Depending on the LOS key, and the associated function attached to it, navigation, opening of windows, and initiation of filter or post-processor TBOL programs may occur as the result input events handled by the keyboard manger 434. In addition, keyboard manger 434 determines inter and intra field cursor movement, and coordinates the display of field text and cursor entered by the user with display manager 461, and sends information regarding such inputs to data collection manager 441.
Object interpreter 435 is responsible for building and recursively processing a table called the "Page Processing Table," or PPT. Object interpreter 435 also manages the opening and closing of windows at the current page. Object interpreter 435
is implemented as two sub-components: the object processor 436 and object scanner 437.
Object processor 436 provides an interface to keyboard manger 434 for navigation to new pages, and for opening and closing windows in the current page. Object processor 436 makes a request to object storage facility 439 for a page template object (PTO) or window object (WO), as requested by keyboard manger 434, and for objects and their segments which comprise the PTO or WO returned by object storage facility 439 to object processor 436. Based on the particular segments comprising the object(s) making up the new PTO or WO, object processor 436 builds or adds to the page processing table (PPT), which is an internal, linked-list, global data structure reflecting the structure of the page or page format object (PFO), each page partition or page element object (PEO), and program objects (POs) required and each window object (WO) that could be called. Objects are processed by parsing and interpreting each object and its segment(s) according to their particular structure as formalized in the data object architecture (DOA). While in the process object state, (state "B" of FIG. 6), object processor 436 will request any objects specified by the PTO that are identified by external references in call segments (e.g. field level program call
518, page element selector call 524, page format call 526 program call 532, page element call 522 segments) of such objects, and will, through a request to TBOL interpreter 438, fire initializers and selectors contained in program data segments of all PTO constituent program objects, at the page, element, and field levels. Object processor 436 requests all objects required to build a page, except objects that could only be called as the result of some event external to the current partitioned application, such as a HELP window object. When in the course of building or adding to the PPT and opening/closing WOs, object processor encounters a call to an "ADSLOT" object id, the next advertisement object id at ad manager 442 is fetched, and the identified advertisement object is retrieved either locally, if available, or otherwise from the network, so that the presentation data for the advertisement can be sent to display manager 461 along with the rest of the presentation data for the other objects to enable display to the user. Object processor 436 also passes to data collection manager 441 all object ids that were requested and object ids that were viewed. Upon completion of page or window processing, object processor 436 enters the wait for event state, and control is returned to keyboard manger 434.
The second component of object interpreter 435, object scanner 437, provides a file-like interface, shared with object storage facility 439, to objects currently in use at RS 400, to enable object processor 436 to maintain and update the PPT. Through facilities provided by object scanner 437, object processor recursively constructs a page or window in the requested or current partitioned application, respectively.
Object storage facility 439 provides an interface through which object interpreter 435 and TBOL interpreter 438 either synchronously request (using the TBOL verb operator "GET") objects without which processing in either module cannot continue, or asynchronously request (using the TBOL verb operator "FETCH") objects in anticipation of later use. Object storage facility 439 returns the requested objects to the requesting module once retrieved from either local store 440 or interactive network
10. Through control structures shared with the object scanner 437, object storage facility determines whether the requested object resides locally, and if not, makes an attempt to obtain it from interactive network 10 through interaction with link communications manager 444 via object/communications manager interface 443.
When objects are requested from object storage facility 439, only the latest version of the object will be provided to guarantee currency of information to the user. Object storage facility 439 assures currency by requesting version verification from network 10 for those objects which are available locally and by requesting objects which are not locally available from delivery system 20 where currency is maintained.
Version verification increases response time. Therefore, not all objects locally available are version checked each time they are requested. Typically, objects are checked only the first time they are requested during a user session. However, there are occasions, as for example in the case of objects relating to news applications, where currency is always checked to assure integrity of the information.
The frequency with which the currency of objects is checked depends on factors such as the frequency of updating of the objects. For example, objects that are designated as ultrastable in a storage control parameter in the header of the object are never version checked unless a special version control object sent to the RS as part of logon indicates that all such objects must be version checked. Object storage facility 439 marks all object entries with such a stability category in all directories indicating that they must be version checked the next time they are requested.
Object storage facility 439 manages objects locally in local store 440, comprised of a cache (segmented between available RAM and a fixed size disk file), and stage (fixed size disk file). Ram and disk cached objects are retained only during user sessions, while objects stored in the stage file are retained between sessions. The storage control field, located in the header portion of an object, described more fully hereafter as the object "storage candidacy", indicates whether the object is stageable, cacheable or trashable.
Stageable objects must not be subject to frequent change or update. They are retained between user sessions on the system, provided storage space is available and the object has not discarded by a least-recently-used (LRU) algorithm of a conventional type; e.g., see Operating System Theory, by Coffman, Jr. and Denning, Prentice Hall Publishers, New York, 1973, which operates in combination with the storage candidacy value to determine the object storage priority, thus rendering the stage self-configuring as described more fully hereafter. Over time, the self-configuring stage will have the effect of retaining within local disk storage those objects which the user has accessed most often. The objects retained locally are thus optimized to each individual user's usage of the applications in the system. Response time to such objects is optimized since they need not be retrieved from the interactive computer system.
Cacheable objects can be retained during the current user session, but cannot be retained between sessions. These objects usually have a moderate update frequency. Object storage facility 439 retains objects in the cache according to the LRU storage retention algorithm. Object storage facility 439 uses the LRU algorithm to ensure that objects that are least frequently used forfeit their storage to objects that are more frequently used.
Trashable objects can be retained only while the user is in the context of the partitioned application in which the object was requested. Trashable objects usually have a very high update frequency and must not be retained to ensure that the user has access to the most current data.
More particularly and, as noted above, in order to render a public informational and transactional network of the type considered here attractive, the network must be both economical to use and fast. That is to say, the network must supply information and transactional support to the user at minimal costs and with a minimal response time. These objectives are sought to be achieved by locating as many information and transactional support objects which the user is likely to request, as close to the user as possible; i.e., primarily at the user's RS 400 and secondarily at delivery system 20. In this way, the user will be able to access objects required to support a desired application with minimal intervention of delivery system 20, thus reducing the cost of the session and speeding the response time.
However, the number of objects that can be maintained at RS 400 is restricted by at least two factors: the RS 400 storage capacity; i.e., RAM and disk sizes, and the need to maintain the stored objects current.
In order to optimize the effectiveness of the limited storage space at RS 400, the collection of objects is restricted to those likely to be requested by the user; i.e., tailored to the user's tastes--and to those least likely to be time sensitive; i.e., objects which are stable. To accomplish this, objects are coded for storage candidacy to identify when they will be permitted at RS 400, and subject to the LRU algorithm to maintain presence at RS 400. Additionally, to assure currency of the information and transaction support provided at RS 400, objects are further coded for version identification and checking in accordance with a system of priorities that are reflected in the storage candidacy coding.
Specifically, to effect object storage management, objects are provided with a coded version id made up of the storage control byte and version control bytes identified above as elements of the object header, specifically, bytes 16 and 18 shown in FIG. 4b. In preferred form, the version id is comprised of bytes 16 and 18 to define two fields, a first 13 bit field to identify the object version and a second three bite field to identify the object storage candidacy.
In this arrangement, the storage candidacy value of the object is addressed to not only the question of storage preference but also object currency. Specifically, the storage candidacy value establishes the basis upon which the object will be maintained at RS 400 and also identifies the susceptibility of the object to becoming stale by dictating when the object will be version checked to determine currency.
The version value of the object on the other hand, provides a parameter that can be checked against predetermined values available from delivery system 20 to determine whether an object stored at RS 400 is sufficiently current to permit its continued use, or whether the object has become stale and needs to be replaced with a current object from delivery system 20.
Still further, object storage management procedure further includes use of the LRU algorithm, for combination with the storage and version coding to enable discarding of objects which are not sufficiently used to warrant retention, thus personalizing the store of objects at RS 400 to the user's tastes. Particularly, object storage facility 439, in accordance with the LRU algorithm maintains a usage list for objects. As objects are called to support the user's applications requests, the objects are moved to the top of a usage list. As other objects are called, they push previously called objects down in the list. If an object is pushed to the bottom of the list before being recalled, it will be forfeited from the list if necessary to make room for the next called object. As will be appreciated, should a previously called object be again called before it is displaced from the list, it will be promoted to the top of the list, and once more be subject to depression in the list and possible forfeiture as other objects are called.
As pointed out above, in the course of building the screens presented to the user, objects will reside at various locations in RS 400. For example, objects may reside in the RS 400 RAM where the object is supporting a particular application screen then running or in a cache maintained at either RAM or disk 424 where the object is being held for an executing application or staged on the fixed size file on disk 424 noted above where the object is being held for use in application likely to be called by the user in the future.
In operation, the LRU algorithm is applied to all these regions and serves to move an object from RAM cache to disk cache to disk file, and potentially off RS 400 depending on object usage.
With regard to the storage candidacy value, in this arrangement, the objects stored at RS 400 include a limited set of permanent objects; e.g., those supporting logon and logoff, and other non-permanent objects which are subject to the LRU algorithm to determine whether the objects should be forfeited from RS 400 as other objects are added. Thus, in time, and based on the operation of the LRU algorithm and the storage candidacy value, the collection of objects at RS 400 will be tailored to the usage characteristics of the subscriber; i.e., self-configuring.
More particularly, the 3-bit field of the version id that contains the storage candidacy parameter can have 8 different values. A first candidacy value is applied where the object is very sensitive to time; e.g., news items, volatile pricing information such as might apply to stock quotes, etc. In accordance with this first value, the object will not be permitted to be stored on RS 400, and RS 400 will have to request such objects from delivery system 20 each time it is accessed, thus, assuring currency. A second value is applied where the object is sensitive to time but less so than the first case; e.g., the price of apples in a grocery shopping application. Here, while the price might change from day to day, it is unlikely to change during a session. Accordingly the object will be permitted to persist in RAM or at the disk cache during a session, but will not be permitted to be maintained at RS 400 between sessions.
Continuing down the hierarchy of time sensitivity, where the object concerns information sufficiently stable to be maintained between sessions, a third storage candidacy value is set to permit the object to be stored at RS 400 between sessions, on condition that the object will be version check the first time it is accessed in a subsequent session. As will be appreciated, during a session, and under the effect of the LRU algorithm, lack of use at RS 400 of the object may result in it being forfeited entirely to accommodate new objects called for execution at RS 400.
Still further, a fourth value of storage candidacy is applied where the object is considered sufficiently stable as not to require version checking between sessions; e.g., objects concerning page layouts not anticipated to change. In this case, the storage candidacy value may be encoded to permit the object to be retained from session to session without version checking. Here again, however, the LRU algorithm may cause the object to forfeit its storage for lack of use.
Where