Paper #: 5.3.5
Title:    Streaming Queries over Streaming Data

1. Problems
-Technology exists that processes streaming queries over static data, or static queries over streaming data, but streaming queries over streaming data
has been a significant problem. Rather than simply registering constantly reporting queries, queries should be able to access recent data, or
selections in time of data.

2. New Idea and Strengths
-This paper proposes a system called PSoup, which parses both queries and data to maintain a simultaneous relationship to stream both. When a query is
registered, all new data is applied across the query. The query also traverses previously stored data, marking where the data fits the query. The
queries are registered, so that whoever created the query can access it in the future.
-The idea of the repeatable query makes sense, especially with data streams. Most users will be accessing the same data, or at least within the
parameters of their choosing. And since the queries are constantly updated with new data, repeat accesses are extremely quick.

3. Weaknesses and Extensions
-This seems like a massive amount of data to process, especially as new queries are being added. I was suprised that PSoup did as well in their tests
as it did, to be honest. They do make the assumption, however, that new queries will be minimal. The system may drag in situations of high density
query addition.
-They don't discuss what happens as query requests become more frequent, how the system would be able to handle constant updates. For example, in
situations like continuously updating weather, this system might not work as well. Although perhaps ideal for their given example of a PDA checking
for updates on given queries, as a general application of constant streaming the system is not perfect.