Citizen Cyberscience Homework 1

Superhero ability

Firstly, I should tell you that my self-selected superhero power is using computers without a GUI.

My ability is a blessing and a curse. It helps me develop ideas quickly, but also make my software utterly unusable by people other than me. I love working with designers.

BOINC

There are 44 BOINC projects, according to this list. According to allprojectstats.com, there are nearly 2.5 million users, over half of them helping the SETI@home project.

The array of projects listed at BOINC is great! There are so many math-oriented ones. As a former math major, I appreciate the selection of cryptography projects. I also appreciate that, like real cryptography research, both sides of the cyptography battlefield are well-represented: the eternal struggle between inventing strong cryptography and breaking strong cryptography.

I chose to join the NFS@Home project which aims to factor large integers using a technique called the Number Field Sieve. Our computers are farmed to help with the “lattice sieving step”, which I know nothing about but the information is there for any citizen cyberscientist.

NFS@Home uses our computers to perform a semi-nefarious but legitimate research task that, as a consequence, helps push our RSA key sizes larger and larger. It makes me wonder what kind of cryptography military uses.

(Note that PrimeGrid also exists, whose goal is to identify very large prime numbers that could possibly be used in generating RSA keys. These two projects seem to be opposite in application.)

I wanted to run NFS@Home on a server I rent space on because I actually wanted to keep this task running over a somewhat long period of time, and I know that it is being under-utilized because of the paltry number of web hits I receive. Running NFS@Home on my laptop doesn’t make sense because I always turn it off and on and generally do work that requires me to push the resources to the limits.

Setting up BOINC on Ubuntu is easy, and the project is structured so that a GUI can be used to connect to and manage the remote BOINC application. It wasn’t the easiest application to set up, but editing the remote_hosts.cfg and gui_rpc_auth.cfg files in /var/lib/boinc-client were the key to getting everything to work.

Unfortunately, upon attempting to start my first task, I was notified that my system requirements were not great enough.

18-Sep-2012 13:21:55 [NFS@Home] Message from server: 16e Lattice Sieve V5 needs 1239.78 MB RAM but only 447.74 MB is available for use.
18-Sep-2012 13:21:55 [NFS@Home] Message from server: 16e Lattice Sieve needs 1239.78 MB RAM but only 447.74 MB is available for use.
18-Sep-2012 13:21:55 [NFS@Home] Message from server: 15e Lattice Sieve needs 476.84 MB RAM but only 447.74 MB is available for use.
18-Sep-2012 13:21:55 [NFS@Home] Message from server: 14e Lattice Sieve needs 476.84 MB RAM but only 447.74 MB is available for use.

This particular project is very non-visual, so I am rather comfortable in leaving it running without too great of feedback.

Zooniverse

There are 9 projects on Zooniverse, according to their project listings. Presently there are over half a million participants online.

I participated in the Whale Sounds project to classify similar sounding whale sounds. The tutorial on how to use their UI was jarring at first, but also very informative to show me how to quickly use their system. The use of spectographs helped me decide which samples to try out first. As a volunteer thinking project, it did a great job to utilize sight and sound senses and natural human pattern matching abilities that computers generally find difficult.

There weren’t many tasks to do on this project, it appeared, so I kept on getting the same samples to classify. I wasn’t too sure if that was a pre-trail stage to see if I was a well-behaved user, or if no more progress was required.