Browse Prior Art Database

Weighting file edit history to determine major updates Disclosure Number: IPCOM000016323D
Original Publication Date: 2002-Sep-16
Included in the Prior Art Database: 2003-Jun-21

Publishing Venue



Following creation, documents may undergo many changes and bear little resemblance to the original document. Under these circumstances, when trying to contact someone regarding the contents, going to the creator is not going to be useful. File tracking and version control systems (such as CVS and CMVC) will usually list a change history for a file, starting with the name of the person who created the file, and the name of every subsequent person who made changes. Sometimes (as with CMVC), a comment can be attached to an update, but it is left to the user to sift through potentially hundreds of comments to find out who made the significant changes (those people are probably most knowledgeable about the document) and when. A solution to this problem is to 'weight' each of the changes to the file in such a way that bigger or more important changes are given bigger weights. When trying to determine who made major updates to a file, or when they were made, a user could examine the changes with the largest weights. The weighting algorithm would depend on the type of data being stored, and what is deemed to be an important change. An example is to base a weighting system on an HTML document. By performing a comparison between the file before and after it was edited, a list of HTML elements which have changed could be created. Now, any changes to elements under the 'HEAD' element, which contains information directed to the browser (such as keywords) may not be of importance, but the addition of an 'H2' element, would indicate that a major new section has been added. The fact that the ordering of elements in an unordered list ('UL' tag) has changed may not be important, but it may be very important if the ordering has changed in an ordered list ('OL' tag). Using this kind of system, it would be possible to assign weights to each such change, and calculate a 'score' for that edit, which would be saved and later compared to scores from other edits. 1