Browse Prior Art Database

Semantic annotated human task meta-data and UI generator

IP.com Disclosure Number: IPCOM000179700D
Original Publication Date: 2009-Feb-23
Included in the Prior Art Database: 2009-Feb-23
Document File: 7 page(s) / 61K

Publishing Venue

IBM

Abstract

Our invention, Semantic annotated human task meta-data and UI generator, contains two key ideas:

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 54% of the total text.

Page 1 of 7

Semantic annotated human task meta -data and UI generator

Standards like BPEL, schema, WSDL exist today for specifying business processes and their associated human tasks interace. And there are many user interface generators that can generate the corresponding user interfaces for handling these human tasks.

For example:

A document routing business process contains an approval step (human task) in the workflow. This approval task has two input data:
document title - string
document author - string
And one output data
approval status - boolean
reason - string

There are many existing user interface generators that can read the BPEL / schema / wsdl and generate an user interface to something like this:

Document Title: Semantic annotated data

Document author: Tinny Ng

Approval status

Reason

Good proposal

Complete

Document Title: Semantic annotated data

Document author: Tinny Ng

Approval status

Reason

True False

Good proposal

Complete

True

Good proposal

Complete

The generated user interface can only recognize that the approval status is a boolean and assign a checkbox, or a radio button, or a field for getting user input. However, ideally, this approval status is a decision making input and should be implemented as an explicit action button in the generated user interface:

Document Title: Semantic annotated data

Document author: Tinny Ng

Reason

Good proposal

Approve Reject

Using existing technologies (BPEL / schema / WSDL), there is no way for the generator to know the intented purpose of each variable behind scene. It does not know when to generate a regular checkbox / radio button / field for a boolean versus an action button. Some booleans indeed need a checkbox; while some booleans are decision making action. Business company may need to hire developers to implement a customized user interface that best fits the original design of the human task manually.

Another example. For the same approval step, the intended usage for the "Reason" field may be for rejecting the document only and business end users are not required to enter a reason for approving. In

1

Page 2 of 7

this case, the

generated UI wants to show the field onl

y

i

f

"Reject" is clicked.

Document Title: Semantic annotated data

Document author: Tinny Ng

Approve Reject

Reason:

Need rework

OK Cancel

Then it's difficult to generate these kinds of relationship automatically simply based on information from WSDL / BPEL / schema. Developers must step in and customize the user interface programming code manually.

Our invention:
Our invention, Semantic annotated human task meta-data and UI generator, contains two key ideas:

Semantic annotated human task meta-data which extends the existing business process specification


1.

BPEL / schema / WSDL.
There are two aspects of the annotation.

One is to indicate the usage of the data itself such as whether a boolean data is a decision


1.

point. We picked decision point be the first needed annotation beca...