---

1.2.2 The 1980s-Application-Level Integration

   By the mid-1980s, platforms that enabled output from applications to be 
   captured and placed in documents were readily available.  One application 
   such as an editor, owned the entire final document.  This is referred to 
   as an application-centric approach.  This included the data that was 
   captured from other applications.  The data was now part of the final 
   document and if we made any changes to the data in the original 
   applications from where it was captured, the changes would not be made in 
   the final document. 


 


                                             



 
   Figure 2. A 1980s Spreadsheet Program 

   Although this method was much easier than the manual cut-and-paste 
   operations, this approach had the following limitations: 



   °   The output data was static or "dead".  If changes were required, the 
       user had to return to the original application, make the changes, then 
       perform the electronic cut-and-paste operation. 

   °   Generally, only pictures or images of the document were traded across 
       computers. 

   °   Items that could be placed into a document were limited to a small set 
       of text and graphics, which only showed the data as it was at the time 
       it was captured.  There was no link back to the original data. 


   There is an additional problem with the applications-centric approach.  As 
   applications were able to support more and different kinds of data, the 
   complexity of those applications increased significantly.  Release cycles 
   became stretched, requiring more programmers, increasing test time, 
   limiting function between releases, and significantly increasing 
   development costs. 

IBM BookManager® BookServer Copyright 1989, 1999 IBM Corporation. All rights reserved.