Over
the weekend we had a mini exercise to give us a refresher on
downloading and building open source software. We were simply expected
to get any open source project, and construct it on our machine. Since I
am working with Team Obsidian, and Hunter
is still putting some finishing touches on his code before he makes it
public to us, I downloaded and built my project from last semester
again.
Last
semester I worked with a team to develop an automated testing framework
for the Google Visualization Suite, specifically focusing on Google
Charts. So, I re-downloaded and built the code again on my machine. It
went off pretty much without a hitch, since I had done the same thing a
semester ago.
With
that done, I set about on some activities for my team. I had been put
of setting up the IRC channel for our project on freenode.net. We
discussed using freenode, because it is very popular among open source
projects. So, Michael Cole
and I went onto IRC to play around and check out some channels. We
investigated #obsidian, #obsidiantest, and #obsidiantesting. The
#obsidian room had another person in it, but both other rooms were not
owned at all. We decided to register #obsidiantesting for now, and try
to obtain the room #obsidian in the future, as it does not seem to be in
active use.
After
this, I tackled the second part of our homework. This was to read the
paper “The Cathedral and the Bazaar” by Eric S. Raymond. This paper was
about the author’s experiences as he was developing a mail client using
Linus Torvalds’ approach to open source software development. These
principles basically come down to releasing early and often, and
allowing as many people to help you develop as you possibly can. He
compares this approach to a Bazaar, where anyone can set up and
contribute to the market in their own way. This is in comparison to
traditional software development, in which a few designers are
responsible for all major decisions. They decide the direction of the
project, and are ultimately responsible for the final incarnation.
He
compares and contrasts the two methods, and ultimately walks away very
impressed with the bazaar style. He numbers out notions that he
considers to be very important to this development style, from the very
easy release early and often, to other findings. One that really grabbed
my attention was the idea that if you communicate with your beta
testers like they are your greatest asset, they will become your
greatest asset. I liked this part, because he talked about many methods
of communicating with his testers, and how they met his enthusiasm with
equal enthusiasm of their own.
One
other thing I found interesting was his theory on why Linux has such a
high quality of code. He posits that because the people working on the
project are so excited about the part they are working on, they make it
much better than they would had they been forced to work on something
they were not interested in or did not love. I agree with this theory
quite a bit, and I think that it bodes well for our project. All of our
team members seem to be very enthusiastic and proactive about tackling
the parts of the project that they were assigned.
No comments:
Post a Comment