Title: System and method of integrating software components
Abstract: A system for integrating services in a system includes a plurality of services, wherein each service includes a software component associated with the system and provides at least one functionality. A user interface, such as a web browser, displays the plurality of services, wherein each service has a link associated with it for accessing the service directly. A scripting shell, responsive to selection of a first service and a second service from the plurality of services, pipes the first and second services together to form a scenario such that output from the first service is provided as input to the second service. A link may be associated with scenario for accessing the scenario directly.
Patent Number: 6,973,655 Issued on 12/06/2005 to Jacquin, et al.
| Inventors:
|
Jacquin; Thierry (Gieres, FR);
Gastaldo; Michel (Montbonnot, FR)
|
| Assignee:
|
Xerox Corporation (Stamford, CT)
|
| Appl. No.:
|
017600 |
| Filed:
|
December 18, 2001 |
| Current U.S. Class: |
719/310 |
| Intern'l Class: |
G06F 009/46 |
| Field of Search: |
719/328,331,332,320,310
717/113,109,105,100
345/762,760
709/219,223
715/513
|
References Cited [Referenced By]
U.S. Patent Documents
| 5386568 | Jan., 1995 | Wold et al.
| |
| 5448739 | Sep., 1995 | Jacobson.
| |
| 5485617 | Jan., 1996 | Stutz et al.
| |
| 5579469 | Nov., 1996 | Pike.
| |
| 5761684 | Jun., 1998 | Gibson.
| |
| 5794006 | Aug., 1998 | Sanderman.
| |
| 5848291 | Dec., 1998 | Milne et al.
| |
| 5867651 | Feb., 1999 | Dan et al.
| |
| 6012098 | Jan., 2000 | Bayeh et al.
| |
| 6133915 | Oct., 2000 | Arcuri et al.
| |
| 6211874 | Apr., 2001 | Himmel et al.
| |
| 6237030 | May., 2001 | Adams et al.
| |
| 6421716 | Jul., 2002 | Eldridge et al.
| |
| 6437805 | Aug., 2002 | Sojoodi et al.
| |
| 6438615 | Aug., 2002 | Faustini.
| |
| 6516349 | Feb., 2003 | Lieberman.
| |
| 6643708 | Nov., 2003 | Francis et al.
| |
| 6681386 | Jan., 2004 | Amin et al.
| |
| 6738964 | May., 2004 | Zink et al.
| |
Other References
L. Carr, et al, "Link Services or Link Agents ?", ACM, 1998, pp. 113-122.
|
Primary Examiner: Lao; Sue
Attorney, Agent or Firm: Walder; Jeannette M
Claims
1. A computer implemented interactive method of integrating and customizing document-based
services in a knowledge management system, comprising:
providing a plurality of document-based services, wherein each document-based
service comprises a software component associated with the system and provides
at least one functionality for processing at least a portion of a document;
wherein each software component includes a control panel for defining and inputting
commands, input parameters and output parameters associated with the software component;
wherein input parameters comprise a document-based string parameter;
displaying the plurality of document-based services in a user interface, wherein
each document-based service has a link associated with it for accessing the document-based
service directly;
responsive to user selection of a first document-based service from the plurality
of document-based services, displaying the first document-based service's control
panel in the user interface;
responsive to user input of at least one command and at least one input parameter
in the control panel of the first document-based service, caching any results objects
for further usage within a session memory, displaying any output generated by the
document-based first service in the user interface, and providing handles to specify
paths to the results objects within the session memory;
responsive to user selection of a second document-based service from the plurality
of document-based services, displaying the second document-based service's control
panel in the user interface; and
responsive to user selection of at least a portion of the output generated by
the first document-based service, at least one command and at least one input parameter
in the control panel of the second document-based service, piping the at least
a portion of the output generated by the first document-based service to the selected
input parameter of the second document-based service and displaying any output
generated by the second document-based service in the user interface.
2. The method of claim 1, further comprising:
responsive to user selection of a third document-based service from the plurality
of document-based services, displaying the third service's control panel in the
user interface; and
responsive to user selection of at least a portion of the output generated by
the second document-based service, at least one command and at least one input
parameter in the control panel of the third document-based service, piping the
at least a portion of the output generated by the second document-based service
to the selected input parameter of the third document-based service and displaying
any output generated by the third document-based service in the user interface.
3. The method of claim 1, further comprising:
responsive to user selection of a fourth document-based service from the plurality
of document-based services, displaying the fourth service's control panel in the
user interface; and
responsive to user selection of at least a portion of the output generated by
the first document-based service, at least one command and at least one input parameter
in the control panel of the fourth document-based service, piping the at least
a portion of the output generated by the first document-based service to the selected
input parameter of the fourth document-based service and displaying any output
generated by the fourth document-based service in the user interface.
4. The method of claim 1, wherein the user interface comprises a browser and
wherein each link comprises an http path from the browser to its corresponding service.
5. The method of claim 1, further comprising:
providing a filter control comprising a plurality of selectable criteria, wherein
the criteria refer to conditions defined for each of the document-based services;
responsive to user selection of at least one criteria in the filter control;
executing the filter control and displaying any results, wherein the results
comprise a subset of the plurality of document-based services.
6. The method of claim 1, wherein the plurality of services comprise a hyperbolic
tree view builder, a mail sender, a PDF converter, a search engine, a document
summarizer, a database, a term extractor, a terms extractor, a translator, an image
extractor and a dictionary.
7. The method of claim 1, wherein each document-based service includes a filter
control for selecting criteria pertaining to the document-based service.
8. The method of claim 1, wherein the output parameters comprise at least one
of a string, a document URL and a document pointer.
9. the method of claim 1, wherein the control panel defines the document-based
service's functionality and service properties.
10. A computer implemented interactive system for integrating and customizing
document-based services in a knowledge management system, comprising:
a plurality of document-based services, wherein each document-based service comprises
a software component associated with the system and provides at least one functionality
for processing at least a portion of a document;
wherein each software component includes a control panel for defining and inputting
commands, input parameters and output parameters associated with the software component;
wherein input parameters comprise at least one of a string, a document URL and
a document pointer;
a user interface for displaying the plurality of document-based services, wherein
each document-based service has a link associated with it for accessing the document-based
service directly;
responsive to user selection of a first document-based service from the plurality
of document-based services, displaying the first document-based service's control
panel in the user interface;
a first framework, responsive to user input of at least one command and at least
one input parameter in the control panel of the first document-based service, for
displaying any output generated by the first document-based service in the user interface;
a second framework, responsive to user selection of a second document-based service
from the plurality of document-based services, for displaying the second document-based
service's control panel in the user interface; and
a scripting shell, responsive to user selection of at least a portion of the
output generated by the first document-based service, at least one command and
at least one input parameter in the control panel of the second document-based
service, for piping the at least a portion of the output generated by the first
document-based service to the selected input parameter of the second document-based
service and displaying any output generated by the second document-based service.
Description
BACKGROUND OF THE INVENTION
This invention relates generally to systems and methods for providing software
services, and more particularly, to a system and method of integrating and customizing
software components using a user interface such as a browser.
Numerous software products providing a variety of functionalities are available
for users to purchase and use. Despite the abundance of software products available,
many users must customize their software installations in order to meet their particular
requirements. Other users are unable to find a complete solution to their particular
requirements in any single software product and try to integrate several existing
software packages to achieve the desired functionality. Even if a user is able
to integrate several software packages together, the cost may be greater than the
customer's budget permits. Also, all of the software functionalities in each of
the software products may not be used, further exacerbating the cost issue.
Even users who purchase only a single software component frequently use only
a fraction of the software package's functionalities. This is less problematic
to the user if the software package is modestly priced. However, if the software
package is expensive (based on the user's budget), the user will be less inclined
to purchase the package. Some vendors, in recognition of this problem, offer software
on a shared basis. For example, a complete software package may be loaded on a
server located on a network, such as the Internet, and users given access to the
package. The user would then be billed on the basis of the number of software functionalities
used and the time logged into the server.
Knowledge management customer requirements, in particular, often do not
correspond to existing software packages. Knowledge management, by its very nature,
has fluid requirements. Seldom does a single software package meet the knowledge
management customer's requirements, and if so, not for long. The only economically
viable approach is for the knowledge management customer to integrate several components
to form a dedicated solution corresponding to the desired solution.
One of the big difficulties for web portal designers is keeping new and up-to-date
functionalities on their portals. Portals are static from the final user's point
of view, in the sense that they offer a limited number of functions that are often
not renewed for months. Updating data is relatively straightforward. However, keeping
up with the latest revisions to software components (such as plug-ins) or new ones
being created (such as for multimedia, security, encryption, generation of forms
for billing, etc.) is often problematic.
Several systems have been implemented in an attempt to solve these problems
for users. For example, Hewlett-Packard's e-speak is one particular instance of
a distributed components architecture using CORBA (Common Object Request Broker
Architecture developed by an industry consortium known as the Object Management
Group). CORBA is an architecture that enables pieces of programs, called objects,
to communicate with one another regardless of what programming language they were
written in or what operating system they are running on. e-speak offers web-access
from a browser for registration and discovery of services, and for interaction
through a dedicated client and an edited XML document. Registered services cannot
be accessed directly through the web-access, but only through registered servlets.
e-speak provides users with a means to discover services, with a user-level consideration
of the description/discovery language. e-speak also offers an easy service registration
for external services that can only be accessed using a servlet, i.e., with an
http interface.
Another system is an internal product development project by Xerox the Enterprise
Output Management (EOM) Solutions Platform for constructing solutions that manage
the execution of output jobs directed to a collection of output facilities in an
enterprise. The collection of output facilities can include any combination of
local printers, viewers, Web publishers, archives, FAX machines, CD-ROM writers,
EDI drivers, or remote print submission services (among others). A major element
of the EOM Solutions Platform is an Integration Framework called DSDIF ("Document
Systems Domain Integration Framework"). A framework, in an object-oriented system,
is a set of classes that embodies an abstract design for solutions to a number
of related problems. DSDIF enables components and services belonging to the EOM
Solutions Platform to be integrated with each other into an overall solution. This
allows many customized solutions to be obtained, with minimal implementation effort,
from a common set of services and components.
There is a need for a system and method of integrating and customizing software
components easily in a minimum amount of time. There is also a need for a system
and method which employs a convenient user interface such as a browser. There is
a further need for a system and method which provides users with the ability to
implement their own scenarios involving a set of software components.
SUMMARY OF THE INVENTION
A system for integrating services in a system, according to the invention, includes
a plurality of services, wherein each service comprises a software component associated
with the system and provides at least one functionality. A user interface, such
as a web browser, displays the plurality of services, wherein each service has
a link associated with it for accessing the service directly. A scripting shell,
responsive to selection of a first service and a second service from the plurality
of services, pipes the first and second services together to form a scenario such
that output from the first service is provided as input to the second service.
A link may be associated with scenario for accessing the scenario directly. A method
of integrating services in a system, according to the invention, includes providing
a plurality of services in a user interface, selecting a first service from the
plurality of services, selecting a second service from the plurality of services,
and piping the first and second services together to form a scenario such that
output from the first service is provided as input to the second service.
A scenario may be any number of services piped or chained together. In one embodiment,
the services are chained together such that the output of one service is applied
to the input of the next. For example, if a scenario includes services A and B,
the output of A is provided to the input of B (alternatively, the input of B may
also include additional input). If the user wishes to add a third service to the
scenario, the user may choose, for example, as input of C, to provide: the output
of A, the output of B, or the output of A and B. When the user is satisfied with
the configuration of the final scenario, the system can establish a link to the
scenario directly. Thus, the scenario now appears to be a "single" service, a black
box, with input and outputs.
In another aspect of the invention, a method of accessing a service in a system,
includes providing a plurality of services, wherein each service comprises a software
component associated with the system and provides at least one functionality; displaying
the plurality of services in a browser, wherein each service has an http link associated
with it for accessing the service directly; selecting a service from the plurality
of services; and accessing the service from the user interface.
To assist a user in selecting from a large number of services in the system, a
filter control may be used to enable the user to select among a plurality of selectable
criteria, wherein the criteria refer to conditions defined for each of the services.
Responsive to selection of at least one criteria in the filter control, the filter
control generates any results and the user interface displays the results, wherein
the results comprise a subset of the plurality of services. The user can then use
the scripting shell to pipe selections from the results together.
The system may be used to locate services, experiment with a selected service,
design a scenario (a set of services piped together), validate and save the scenario.
The system and method of integrating services can be used as a display shelf for
software technologies, allowing users to discover, test and demonstrate combinations
of the individual services. The system and method may be used as a tool to interactively
design software; it may be used as a dynamic portal to build services on demand.
The user interface may be a web browser, but other types of interfaces may also
be used. A browser is generally a program which allows a person to read hypertext
and provides some means of viewing the contents of nodes (or pages) and of navigating
from one node to another. When the user interface is a web browser, the system
offers an intuitive way of selecting services and indicating which data have to
be considered as input and output. The system takes services, which are software
components, and documents and registers them in the system. The web browser displays
the services in a web page. The services may be remotely located and may be invoked
via an http interface. The system enables the creation, testing and validation
of scenarios, which are an association of a set of services, with data transferred
between services. A scenario can be seen as a customized software application involving
several services.
Services may be written in any description language, for example, Java or
a Java interface. Using a language such as Java makes documentation and registration
of services in the system relatively straightforward; the only effort involved
in adding a service generally consists of documenting the service and writing a
main function for the service invocation. Additionally, the output of a service
may be structured or it may be treated as a string. For example, a "wizard" has
been developed for an example system known as "Kasba." The Kasba Wizard allows
users to plug in a new service to the Kasba system in less than an hour by following
a series of prompts and inputting in requested information.
The system and method of integrating services may be used in many different applications.
For example, a user can build his/her own service, putting together (composing)
existing services, for example on a web portal. The system and method of integrating
services may be used by a software developer to quickly design and implement a
scenario involving several software components. A solution designer/integrator
could use the system and method of integrating components in order to: evaluate
a component in a targeted context by replacing one component for another; test
the new component in the targeted context; understand the transfer of data between
the components; predict problems in terms of volume of exchanged data; and identify
where the data needed is not available, etc., to find the right sequence of components
and quickly design a dedicated scenario for a given situation, e.g., customized
for a given presentation.
The system and method of integrating services may be used to solve knowledge
management requirements. In the area of knowledge management, data is often poorly
structured and typed (very often the data is represented only as a string) and
often the data is very heterogeneous (requiring the manipulation of entities of
very different nature), and the interpretation of the data is not unique and depends
on the context. The knowledge management user plays a role in the analysis of knowledge
data by the selection and combination of data. The system and method of integrating
services gives the user an opportunity to specify and easily implement the selection
and treatment of the knowledge data.
The system and method of integrating services makes the creation of scenarios
quick and easy. Scenario designers can quickly identifying missing and/or important
components. This aspect is useful to the designer, because identifying the value
of some components is often only possible when used in combination with other components;
many software components cannot be evaluated in isolation. By using the system
and method of integrating services, the user may make the decision to buy a software
component after having seen it working in some scenarios (possibly by purchasing
a limited evaluation license, which is enough to plug the component into the system).
In this way, the ultimate user has an opportunity to evaluate the components before
making the decision, and limits the user's a priori investment.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of a system for integrating software components;
FIG. 2 is a display of a group of services available in a system for integrating
software components;
FIG. 3 is a filter control panel for selecting among the services in FIG. 2;
FIG. 4 is a control panel for an emailer service;
FIG. 5 is a filter control panel for an emailer service;
FIG. 6 is a list of service properties for an emailer service;
FIG. 7 is a list of service parameters for an emailer service;
FIG. 8 is a control panel for a dictionary service;
FIG. 9 is a filter control panel for a dictionary service;
FIG. 10 is a list of service properties for a dictionary service;
FIG. 11 is a list of service parameters for a dictionary service;
FIG. 12 is a control panel for a summarizer service;
FIG. 13 is a filter control panel for a summarizer service;
FIG. 14 is a list of service properties for a summarizer service;
FIG. 15 is a list of service parameters for a summarizer service;
FIG. 16 is a block diagram of an exemplary scenario built using the system of
FIG. 1; and
FIG. 17 is a block diagram of how the system of FIG. 1 would be implemented
on a network.
DETAILED DESCRIPTION OF THE EMBODIMENTS
The invention may be implemented in numerous environments. For convenience, the
invention will be described with more particularity with regard to an implementation
in the area of knowledge management. Referring to FIG. 1, a block diagram of a
system for integrating services is shown and referred to by reference numeral
100.
System
100 includes a plurality of services
10,
12,
14,
a framework
20, an http-Gateway
40 and a scripting shell
50.
System
100 communicates over a network using a bus
30. System
100
includes a plurality of services
10,
12,
14. Service
10
is a search engine product called askOnce; service
12 is a Docushare database
account; service
14 is a recommender service. Each service is connected
or plugged into a bus
30 (which may be for example a Jini bus)
30
using frameworks
20.
System
100 relies, for its infrastructure in order to connect to a
network (which may be an internet, the Internet or an intranet), on an existing
platform (or bus). For this embodiment Jini has been chosen, but many other buses
or platforms could be used to plug-in the services. On top of the bus
30,
the gateway
40 supports discovery and activation of the services
10,
12,
14 and the interactions with the services conforming to the framework
20. Gateway
40 enables users to activate an individual service directly.
Scripting shell
50 provides the scenario scripting, i.e., the creation of
an integration of services according to some user input (script, store, replay,
share, publish as template) into a scenario. Services
10,
12,
14
can be integrated on the same bus
30 using another framework
25 and
applications
60 can involve the same services outside system
100,
either relying on framework
20 or any other existing framework
25.
In this embodiment, the bus
30 is a Jini bus (software from Sun Microsystems
that simplifies the connection and sharing of devices, such as printers and disk
drives, on a network) and the software components are written in Java or have a
Java interface. Also in this embodiment the http gateway
40 provides a direct
path from the user interface (here the web browser) to the software components
10,
12,
14. Shell
50 and framework
20 are written
in Java. The Java introspection mechanisms provide a convenient method of understanding
the input and output parameters between software components and provide support
to transform the output of one component into the input of another.
The system
100 may be implemented on any network, such an internet, the
Internet or an intranet. If, for example, implemented on the Internet
80,
a user would access the system
100 through a work station or personal computer
90 with a user interface, such as a web browser (See FIG. 17). Upon accessing
the system
100, the set of services available would be displayed. Referring
to FIG. 2, a display of a group of services available for integration in system
100 is shown. In a web browser, the user can load a page and can see all
the registered services. The available services displayed are: Mail Sender, Search
Engine on computer/Zdnet, Search engine on general/OpenDirectory, Web Assistant,
Terms extraction with strategies, Knowledge Pump Communities, Dictionary, SearchNRank,
Various Kasba Facilities, Search engine on /france/AlloCine, Html Document editor,
Docushare on docushare.grenoble.xrce.xerox.com, Document Language Identifier, Image
Composer, Image Extractor, Document Categorizer, PdfConverter, Search engine on
/general/Google, Document Souls Visualizer, DocuShare on lautaret.grenoble.xrce.xerox.com:1707,
Terms Extractor, Knowledge Profiling, Knowledge Profile Repository and Document
Summarizer. Each service contains a link for accessing each service directly. For
example, when a user "clicks" on the link to Mail Sender, the functionality for
Mail Sender: control panel (FIG.
4,) Filter Control (FIG. 5), Service Properties
(FIG. 6) and Service Parameters Description (FIG. 7) would be displayed.
If the user does not wish to access any individual service directly, the user
can proceed to use the lookup, discovery and filtering functionality. Also displayed
on the web page is Filter Control
200 (see FIG. 3). Filter Control
200,
while not necessary to operation of system
100, is a feature which helps
a user select among a large number of services offering multiple functionality.
Filter Control
200 contains several filters which the user may use in the
process of selecting services to integrate. Filter Control
200 includes
locator
210, constraints
220, selector
230,
240,
250,
from which the user may select using keywords. For example, from locator
210,
the user may select "the first locator discovered on the LAN", "all the locators
on the LAN" or one of a specific identified location on the LAN. From constraints
220, the user may select particular name, domain, version, manufacturer,
etc. Or the user may leave constraint
220 empty and search on all. From
selector
230,
240,
250, the user may select domain, location,
version, vendor, manufacturer, model, name, serial number from the pull down menu.
The user can input keywords of his/her own choosing in box
232,
242,
252.
After the user completes the Filter Control
200, the user selected filter
services
260 and the system filters and constrains, producing a result.
The result will typically be one or more services, but may be the null set, depending
on the filtering criteria selected. Once a non-empty result is achieved, the user
may experiment with the resulting service(s), sending some input as documented
and validating that the service corresponds to the user's need. Alternatively,
the user may use the results to start creating a customized scenario.
Assume, for example, the user has a series of documents that need to be reviewed,
summarized, a report written and sent to a client. No single service provides all
these functions, so the user must use the system
100 to create a scenario
which performs all of these functions. In FIG. 2, the user would first select and
then access Document Summarizer by clicking on the link. The control panel
300
for Document Summarizer is then displayed to the user (see FIG. 12). The user selects
"0" in command box
302 for send a request. In box
304, the user selects
the language of the article to be summarized; the default is English, but French,
Spanish, German, Italian and Portuguese may also be selected. Boxes
306,
308 and
310 enable the user to input the document or text to be summarized:
box
306 for string text, box
308 for the URL of the document, box
310 for the file location if the document is a file (such as stored in a
directory on the user's personal computer or work station). If the user changes
his/her mind, box
320 "clear results" may be selected and all values input
return to empty. Note that control panel
300 also includes pull-down menus
312 (language),
314 (text),
316 (URL), and
318 (files).
The pull-down menus are used to display the output from a previous service. For
example, if the user had selected "Search engine on/general/Google" as the first
service, then the system will display the search results (output of Google) in
the the appropriate pull-down menu.
The summarizer service contains its own filter control
350 (see FIG. 13)
from which the user can filter and constrain the Document Summarizer functions.
Filter control
350 contains similar selection boxes from which the user
may filter and constrain the summarizer service: locator
352, constraints
354, selector
356,
358 and
360. Upon completion of
the desired filtering, the user selects filter service
362. FIG. 14 lists
the particular service properties for Document Summarizer and FIG. 15 lists the
service parameters descriptions for Document Summarizer.
If the user is satisfied with his/her selections on the Document Summarizer service,
he/she may provide input in the form of the documents to be summarized. Once the
user inputs the documents to Document Summarizer, the output is a list of document
summaries, one for each document. As part of the scenario creation feature of the
system, system
100 will automatically suggest as input of the lastly selected
service the output of the previous one. (Note also, that if there were multiple
prior services in a scenario, any of the output from any of the prior services
could be selected for the input to this service.) Users are free to use this input
or put in their own input, or both. Going back and forward in the browser, the
user accesses the step-by-step mode where each step corresponds to the invocation
of a service.
Suppose the user submits several documents to Document Summarizer and Document
Summarizer returns paragraph summaries of each. Suppose the user wishes to check
the definition of some of the words in the summaries returned. The user can then
submit the words to the dictionary service. While system
100 will (by the
default setting) takes the output of the document summarizer service and pipes
it into the input of the dictionary service, in this case the user may wish to
lookup only a few words in the dictionary. When dictionary service is selected,
control panel
400 (see FIG. 8) and filter control
450 (see FIG. 9)
are displayed to the user.
Dictionary control panel
400 includes a command line
402
which includes two options: "0" for lookup each word of a text in a dictionary,
and "1" for hyperlinking each work of a text towards its translation. Output format
404 enables the user to select text or html for the output. Text may be
input via text using box
406 or a via a URL using box
408. Box
410
enables the user to select the language of the output. Box
412 enables the
user to specify the input language. Box
420 enables the user to clear the
results and start over. Note that control panel
400 may also include pull-down
menus for providing the output of the summarizer service directly as input.
Filter control
450 similarly contains locator
452, constraints
454, selector
456,
458 and
460. The filter control
450 may be used to select the dictionary. Filter services
462 causes
the dictionary service to return the definitions. FIG. 10 displays the service
parameters for dictionary service and FIG. 11 displays the service parameters for
dictionary service.
Once the user has checked definitions and completed the summary, the user may
write his/her report. This may be done offline or by using an editor service (if
one is provided). In this system, the only editor is HTML Document Editor. If that
is acceptable for the user's requirements, the user may invoke that service, create
a report, attach summaries and definitions. System
100 will pipe the output
of the document summarizer and the dictionary into the input of the HTML document
editor, as well as the input provided directly by the user. The user then decides
to send the report to a distant colleague. The user invokes the Mail Sender.
System
100 takes the output of the HTML document editor and pipes it
into the attachment section of the Mail Sender. When the Mail Sender is selected,
control panel
500 (see FIG. 4) and filter control
550 (see FIG. 5)
are displayed. Control panel
500 includes command line
502 for sending
an email, box
504 for inputting the text of a message, box
506 for
listing any blind copy recipients, box
508 for listing the subject of the
email, box
510 for listing the prebody of the message, box
510 for
inputting the "to" recipient of the email, box
514 for inserting attachments,
box
516 for listing copy recipients, box
516 for listing the body
of the message, box
518 for listing the sender of the email. Box
522
clears all input data. Filter control
550 includes locator
552, constraints
554, selectors
553,
558 and
560. Box
562 selects
filter services. FIG. 6 lists the service properties for Mail Sender and FIG. 7
lists the service parameters description for the Mail Sender.
The user has created a scenario which is a combination of document summarizer,
dictionary, HTML document editor and mail sender all piped together in accordance
with the user's criteria. Referring to FIG. 16, the resulting scenario
280
includes document summarizer service
282, dictionary service
284,
HTML document editor service
286 and mail sender service
388. In
this scenario, the user has optionally selected three inputs: input
289
for inputting documents to be summarized; input
290 for inputting words
to the dictionary; input
291 for inputting text to HTML document editor.
In addition to user input
290, dictionary
284 is configured to receive
input directly from the output of document summarizer
282 and HTML document
editor
286. In addition to user input
291, HTML document editor
286
is configured to receive input from the output of document summarizer
282
and dictionary
284. In scenario
280 the only input provided to mail
sender service is the output of HTML document editor
286. However, the user
could come back at a later date and modify scenario
280 to provide additional
inputs to the various services or remove those not needed. Scenario
280
has four outputs: the output
292 from HTML document editor
286, output
294 from document summarizer
282, output
295 from dictionary
284, and output
296 from mail sender
288. When the user is
happy with scenario
288, he/she may bookmark it. Reloading the bookmark
instantiates the execution of the whole row of services.
Designing, implementing and testing a scenario involving several components
today are a highly time consuming exercise. Consequently, software engineers tend
to reduce the time they spend for these stages when the project is not big enough,
with some risks to freeze a solution that is not a good one because of rapid decisions,
and often the risk not to satisfy the customer requirements at the end. Another
consequence is the need to have big enough projects to invest on, because the cost
of all these preliminary stages cannot be paid otherwise. By using the system and
method of the invention, the software engineer (and even the final user) can discover,
experiment and link services in few minutes. It becomes very natural to change
the scenario until a satisfactory solution is found. Changing one component to
another can be done as many times as necessary. Software components may be easily
added to the system. Indeed, users may suggest that additional components be added.
Or, some of the user created scenarios may be added to the system.
There are several ways to use the system to discover available services. For
example, the user may browse in a kind of forest, where services are classified
according to several perspectives, knowing that the same service can be reached
via different paths. This user can also provide a (set of) keyword(s) to search
for specific services. Then all the services compatible with these keywords are
listed. As soon as these services are found to be compatible, an interface to test
these services is provided to the user. This interface allows the user to invoke
several services at the same time, if the services are close to one another.
In this case the interface is the intersection of the respective interfaces,
and
it is presented as soon as it covers all the mandatory parameters for the respective
services. For example, to perform several searches on different sources in parallel,
one may enter "search" as a keyword. This automatically selects several wrappers,
with a form allowing the user to invoke all of them. Basically the only mandatory
parameter for a search is the string the user wants to search, so it will be a
parameter required in the form. The more precise the service description, the more
limiting the set of compliant services, and the more the user can specify a larger
set of parameters which can be used for all the selected services. Assume a user
wants to interact with a single service. The user can specify all parameters available
for the service. Now assume the user wants to interact with two services using
the same parameters. The user can specify only those parameters that are common
to both services. Using the same example, if the user restricts the search to Google,
the user has access to more parameters than searching on Google, AltaVista and
Excite at the same time.
An aspect of the invention is the ability to share and replay registered scenarios.
A scenario is an integrated set of services. If the system administrator registers
the scenario on the system, other users can replay the registered scenario. Alternatively,
the system administrator may permit a user to register scenarios only for his/her
own use later. When a user is happy with a scenario, the user can store it. Then
the registered scenario can be replayed it in two modes; the first mode is executing
it in one shot, while the second mode is a step-by-step mode. Using the second
mode, the user can decompose a scenario, clicking on the back and forward buttons
in the user interface. This technique can be used to debug scenarios as well as
to learn by examples. This is particularly true for the users who are not the author
of the scenario.
Using a debug mode, it is easy to develop and extend an existing scenario.
The system can be used by several users, each authoring a part corresponding to
her/his competencies, jointly creating a collaborative work.
Another aspect of the system is that composite scenarios may be built. Just
as one service may be embedded in another, one scenario may be embedded in another,
resulting in very complex scenarios. A library of scenarios could be developed
and could be used on demand.
To facilitate the building of composite scenarios, a structured repository where
scenarios are stored and managed may be incorporated in the system. This structured
repository can be used as a place to capture a memory of practices and scenarios,
which can be explored each time someone needs to design and deploy a new scenario.
A recommender service could be combined with the system (if a recommender service
is not one of the offered services) to submit, comment and recommend scenarios
to the community of users. This could be beneficial for solution designer or software
integrator groups, where it is often difficult to capture and build on top of previous realizations.
An aspect of the invention is that the system and method of integrating services
is that it supports the discovery, interaction and chaining of services into scenarios,
and the execution of scenarios to demonstrate a global behavior. This functionality
could be provided as a two-level structure or a single level structure. For example,
a scenario builder would address the first phase: services discovery and scenarios
scripting; and a "player" addresses the scenario execution.
By implementing the system as a two-level structure, the different requirements
of the different types of users of the system may be addressed. For example, using
the two-level concept, a software integrator could use the builder to design the
scenario corresponding to the final user requirements, before presenting the scenario
(in the form of a "player") to the customer who will validate the result. The "player"
user interface could be simplified and polished, while keeping the possibility
of capturing the user parameters, and possibly to provide interaction options depending
on the considered scenarios. Taking the example of an information search, the user
should indicate the terms to be searched. Then, using a form, the user could check
some boxes corresponding to the search results he/she wants to be reprocessed and
for example summarized for an insertion in a report document.
The framework
20 need include only what is required from a service, a
service description language, introspection, interaction in order to be a part
of the system. Existing software frameworks may be extended with the system of
the invention. For example, e-speak, UDDI (Universal Description, Discovery and
Integration: a Web-based distributed directory that enables business to list themselves
on the Internet and discover each other, similar to a traditional phone book's
yellow and white pages), SOAP (Simple Object Access Protocol provides a way for
applications to communicate with each other over the Internet, independent of platform.
SOAP relies on XML to define the format of the information and then adds the necessary
HTTP headers to send it) may be extended by the invention to provide discovery,
interaction and scripting capabilities. Because the invention requirements in terms
of framework are minimal, it will generally be easy to augment an existing integration
framework, such as e-speak, with these functionalities.
In addition to services such as software components, devices such as printers,
facsimile machines, etc. may be plugged into the system in the same way as software
services. Also, the integration of third party services and devices (e.g., phone),
in complement to other internal services such as document.com or the Community
Wall may be added to the system.
The system and method of the invention integrates the three phases necessary
(and requiring significant effort) when using e-speak web access (and UDDI): discover
existing services by browsing; build a request corresponding to a call: most of
the time, consist in writing an XML document valid in the discovered schema, that
represents one interaction with a service; and write a client application that
chains tested interactions.
DSDIF and the invention are complementary, and may be integrated together.
The invention typically supports discovering which services are available, adding
a new service, testing how it can be integrated with other services, as well as
designing and validating some scenarios. DSDIF can be used to build a "production
software", with robustness, strict semantics and proved performances. An analogy
can be made, considering DSDIF as another framework on other bus(es), this set
constituting an operating system for programming. The user is a programmer who
reads the specifications and produces code for using a set of services in a scheduled
sequence. When a given service makes some sense for an end-user, it could provided
(using the invention) to view, interact with and chain features on top, similar
to providing a set of commands on top of a subsystem. To do that, the DSDIF service
must conform to the framework of the invention. Then the prototyping at functional
level can be done by interactively scripting, before coding the application within SIF.
In general, it is anticipated that only a limited effort will be required to
add
a service to the system, corresponding to write a "main" function and document
the service. However, the effort to plug-in a service depends on the nature of
the service itself. If the service is a basic input/output stream, let say equivalent
to a UNIX command, the effort corresponds to less than an hour of work. If the
service has a complex API such as Docushare, it can take one week or more to get
the right plug-in.
In the example of the invention which implements a system for a knowledge management
customer, it may be important to provide a knowledge profile service. This service
is useful because it provides a comparable representation for non-homogeneous data.
Without constraining data models, it provides a way to compare most data, as long
as there is a dedicated feature to profile a data published by a service. This
can be seen as the basic knowledge metadata, or eventually a cast operation.
The knowledge profile service provides knowledge profiles for documents, users,
communities of interest and web sources (e.g., Open Directory). The knowledge profile
is based on the textual data associated with the profiled object. Therefore it
is orthogonal to e.g., demographic profiles about a user or document meta-data.
The current knowledge profile associated with an object is a set of term-weight
vectors. Knowledge profiles can be compared in order to measure their similarity
in terms of information content/interest. The knowledge profile service provides
two families of functionality. First, an operations interface permits searching
for knowledge profiles of specific objects as well as comparing/ranking them against
others. Second, a registration interface permits knowledge profile providers to
add new or update existing profiling information (aka term-weight vectors) of profiled objects.
A knowledge profile service also provides a contextual filtering algorithm. One
of the problems with knowledge services is the number of potential results for
an operation. Retrieving documents from a search or from a repository is not limited
by the number of results. Chaining such operations leads quickly to an exponential
tree in which some filtering is needed. Providing the right generic filtering is
impossible most of the time for a programmer. The right filtering depends on the
user, or on a community. So applying the rank feature of a knowledge profile service
against the context to consider, before cutting less relevant branches is often
the most appropriate, generic and low-coast filtering technique.
It will be appreciated that the present invention may be readily implemented
in
software using software development environments that provide portable source code
that can be used on a variety of hardware platforms. Alternatively, the disclosed
system may be implemented partially or fully in hardware using standard logic circuits.
Whether software or hardware is used to implement the system varies depending on
the speed and efficiency requirements of the system and also the particular function
and the particular software or hardware systems and the particular microprocessor
or microcomputer systems being utilized.
The invention has been described with reference to a particular embodiment. Modifications
and alterations will occur to others upon reading and understanding this specification
taken together with the drawings. The embodiments are but examples, and various
alternatives, modifications, variations or improvements may be made by those skilled
in the art from this teaching which are intended to be encompassed by the following claims.
*
