Tom White's Blog

Hadoop!

In a previous blog I wrote about Nutch's MapReduce implementation, for distributed processing of massive data sets. This, and the closely related Nutch Distributed File System (renamed Hadoop Distributed File System), have now been moved into a standalone project called Hadoop.

According to Doug Cutting, who created Hadoop (as well as Lucene and Nutch), the name comes from:

The name my kid gave a stuffed yellow elephant. Short, relatively easy to spell and pronounce, meaningless, and not used elsewhere: those are my naming criteria. Kids are good at generating such. Googol is a kid's term.

An interesting way to name projects - how do you choose names?

Comments

• It is a surpsingly challenging problem Tom.

  I usually would Google for the proposed name, and the word Java for instance. It was surprising how many times that could shoot down a project name. Google currently lists 157 pages for Hadoop alone!

  Secondly, being admittedly 'left-brained,' coming up with names is very difficult already.

  Do you think my project would be named cajo otherwise? :-)

  John

  PS Here is how I came up with the name for my project.

  Posted by: cajo on February 08, 2006 at 10:12 AM

• PS Since you asked:

  A short but interesting exposition, about attempting to definitively name things on the internet, can be found here:

  
  http://www.skyhunter.com/marcs/petnames/IntroPetNames.html

  Posted by: cajo on February 08, 2006 at 06:31 PM

• PS A short but interesting exposition on definitively naming things can be found here:
  http://www.skyhunter.com/marcs/petnames/IntroPetNames.html

  Posted by: cajo on February 08, 2006 at 06:40 PM

• Sorry about that, it appears that pressing the preview button sometimes posts the message anyway.

  John

  Posted by: cajo on February 08, 2006 at 06:42 PM

This work is licensed under a Creative Commons License.