I chose to do my reading for the evening on the past, present, and future of database technology. The article As we may think… is an entertaining read both because of how right and wrong it is. The article, written in 1945 by Vannevar Bush, highlights the improvements and innovations humanity could expect to see in the years following World War 2. A large part of the article is focused on the hypothetical “memex” machine, which operates very similarly to a computer connected to the worldwide web. A key part of the memex, is the relational structure of the information (such as books and files) located within it, which could be called, similar to how a database can be queried. What this article shows is how even in the 1940’s, well before Codd’s work on the relational model of data, people were concerned about data storage and anticipated improved systems for accessing and storing information. Something that stood out to me was how different Bush thought the future might be in certain areas. For example, he places a lot of importance on the shrinking of photos, still believes books will be the primary file used in these “memexes”, and sees voice transcription and in-body electrical signals as being more important than text (which is still a dominant form of interaction in both people and computers). To me, this highlighted how technology can drastically change the direction of the future and how innovations of tomorrow can easily replace innovations of today.
I also read the biography of Edgar Codd in an effort to better understand the current database management system. While Codd’s life was varied and interesting to read about, a few pieces of information stuck out. First was the fact that products attempting databases already existed but were not particularly useful and required a lot of skill to operate. This kind of makes me wonder how long database technology had been attempted before Codd’s model. Another piece was that despite Codd creating a useful foundation for creating databases through the relational model of data, IBM failed to capitalize on its own employee’s discovery until other companies had released products and his idea, despite being widely used now, faced a good deal of criticism at the time. This seems like a catastrophe on the part of IBM especially considering the massive success of its two rivals, Oracle and Ingres, now. I was also surprised that Cobb did not appear to financially benefit from his idea in any major way initially, but still promoted it.
Finally, I looked to the future of databases in the article Why Hadoop is the future of the database. This article highlighted some issues with the current use of traditional relational database systems as well as one of the newer systems being used to fix that issue. One major issue of the traditional database is that large quantities of data cannot be effectively analyzed. Facebook began using a system called Hadoop which is now being used in several websites that collect massive amounts of data. Hadoop at a basic level involves splitting data into several computers which run analysis processes on each simultaneously to process large datasets. Something interesting in the article was the announcement of Pivotal HD which is a program designed to decrease the runtime of working with Hadoop data. The new program would also work with SQL commands, meaning even with innovation in the field, SQL is still a very relevant language!