Enterprise Mashups       


 Chandrasekhar (Chandra) Narayanaswami photo

Enterprise Mashups - overview

Enterprise Mashups (2007-2013)  

Chandra Narayanaswami, Danny Soroker, Dan Coffman, Aaron Zinman, Jingtoa Wang, Jonathan Munson

Soon after seeing a series of mashups on Google maps in 2007, we conceived of and led innovations around a series of “enterprise substrates” (OrgMaps, EventMaps, ProductMaps, BookMaps) which could be mashed up with interesting data sets on top.

One example was Organizational Maps that used a tree map visual rendering to show the organizational structure.  On it were overlaid various enterprise pieces of information or HR information that were aggregated up the tree hierarchy to quickly show. Examples include patents issued, financial information, performance ratings, delinquencies in HR processes, etc. This work was done in the early days of Web 2.0 and a significant amount of work had to be done to have fast renderings (See reference 1). This work was also used to render product catalog hierarchies from retailers.  The dynamic information that was overlaid included sales units, sales dollars, item promotions, stock-outs, excess inventory, ratings and reviews, etc. https://www.forbes.com/2008/04/21/ibm-social-networks-tech-enter-cx_ag_0422ibm.html

This mashup was also shown at the 2008 ACM Programming Contest Finals to the participants – where the participant teams were represented as geographical hierarchy.

Another substrate was multi-track conference events where the calendars for the event were represented in a zoomable canvas (see attached figure).  This work was deployed at ACM CHI conferences in 2010 and 2011 (and IBM’s Lotusphere in the same years). As one zoomed in more details were made visible and search functions (by author, organization, topic, etc.) would highlight the results on the substrate (much like Google Maps searches).  Other features included popularity heat maps based on attendance/#users adding the event to their personal calendars,  ability to comment on specific sessions by twitter handles.

Yet another substrate (BookMaps) was based on covers of magazines and journals much like those still found in university libraries. The canvas is constructed based on the user’s preferences and the overlaid information included reviews and comments by other readers.  Zooming in presents more details, for example going from the cover of the journal to the article and then optionally into reviews with the idea that any person who read the article from say ACM or IEEE and cared to review would leave their review behind for subsequent readers.