Data in blob fields are Java object serializations. The only practical way of viewing them is using Java code to deserialize. The only objects that are serialized in this way are data source and data point definitions, which are very application specific anyway.
For background... Originally, tables were defined for each type of data source and point such that the information was stored in a more traditional way, i.e. each distinct piece of data in a distinct field. However, as more data source types were added to Mango it quickly became a significant maintenance issue. The current storage technique of using blob has a number of strong advantages, with really the only downside being the fact that - as you've noticed - the content is inaccessible via SQL.