;; This buffer is for notes you don't want to save, and for Lisp evaluation.
;; If you want to create a file, visit that file with C-x C-f,
;; then enter the text in that file's own buffer.
XPDL Work Flow
[A Generic Distributive Development Methodology and Framework]
A work flow implementation in a company involves the detailed study of the business and then a sopheticated implementation of the requirments.The develoopment and implementation of such a suite happens as a whole project and generally the system developed doesnt use much of the applications available and sucesfull and is build rigid.
The project aims at developing a methodology , which allows developers to resue the existing applications as part of the system , hence making it viable to resue the sucessful appliactions and focus only the logical flow of the process.To express this logical flow , the project uses the XML Process Descriptive Languagewhich is already becoming a standard.
The idea is to reuse the existing applications , by invoking them and analysing thier output.This concept is similar to the way shel scripts work for command line applications.where a small number of utilities are invoked using the shell script and the combination of which is a different product.Apparently there is no piping or control statements to coordinate such control.The system being build aims at achieving this atleast with the componets or applications that it controls.
This leads to two tasks , first getting the output of a program across a distributed environment and the second is to process accoringly the different MIME types the application maybe sending as output.
The Flow in and Flow out of a program is done by interfacing a bot to every application that we desire to use.The bot handles the out and then redirects one to the users browser and the other to the System's messaging server.The system's messaging server must hence be able to authenticate the messages and fowrard it to the System for managing the remaining operations.The project uses the xmmp protocol for messaging.
The handling of different MIME types is done by using two components.An Ad-Hoc databse and a configuration server.The configuration server could be ldap service that stores what kind of format is expected from which client.And in what format should the input be passed as.The configuration server just maps the rules , most generally the set of nested tags that should be searched for in the Ad-Hoc database.The Adhoc databse is just a dump of all such tags neatly organised.he tags function as the field , when the innermost tag is mentioned , in other cases , the tags that follow after the search path are treated as data.This allows for interesting data to be inserted into contents along with tags without any processing.
The implementation of the Ad-Hoc database includes a solution for quick searching.It takes considerably a long time to search for regex patterns , so using indexes is necessity.indexing can become cpu-intensive if they have to reindexed after every addition or deletion of the data.So a datawarehouse is build upon the Ad-Hoc database using conventional database , so as to provide a quick index of the entries in the Ad-Hoc databse,
The System itself consists of the XPDL