Benry’s CS 371P Blog

Predicted Number of Hours to Complete: 8 hours.
Actual Number of Hours to Complete: 8 hours.

TestDarwin.out: http://blogs.utexas.edu/bjy95/testdarwinout/

I chose to not work with a partner this week for the MatLab project and as much as I enjoyed being able to take control of my own work without consulting someone else, I noticed how many more errors I made without the careful review of a partner over my shoulder.

Pair programming is useful, and I’m glad that it’s an obligation for this course, as the quality of work is much higher when using two minds to solve a single problem.

Next week, I’ll be glad to team up with someone again.

Predicted Number of Hours to Complete: 5 hours.
Actual Number of Hours to Complete: 10 hours.

TestMatLab.out: http://blogs.utexas.edu/bjy95/testmatlabout/.

I did decently well on the test that was given last week, and I’m happy with the grade I got, but I’m curious as to what I missed exactly. I think it may have been a casting that I missed, or something small, but I felt that the test was easy to wade through, and completely fair with respect to material presented in class and on projects.

This past project, the Australian Voting project, was a huge pain in the butt to do. This had to do with the fact that we were not allowed to use std::vectors and C++ doesn’t have VLA’s (officially, but g++ has extensions that implement it). I ended up defining arrays to accommodate maximum values, which is extremely space inefficient, but it did the trick and I’m hoping that we pass Rekha’s test harness. We got by UVA’s acceptance tests, so I’m optimistic for our grade on this project.

Curse you, C++ and your lack of VLA.

Predicted Number of Hours to Complete: 5 hours.
Actual Number of Hours to Complete: 10 hours.

Execution Time: 0.476s
Rank: 134.

Voting.out: http://blogs.utexas.edu/bjy95/votingout/.
TestVoting.out: http://blogs.utexas.edu/bjy95/testvotingout/.

I read this paper (PDF) last night that creates the analogy of Kindergarten and pair programming. Dr. Lin had given us this exact same paper to read back in 315h, but it was good to have it reintroduced, especially in light of the new partners I’ve had/will have for this course.

One interesting thought from the paper was the idea that it could take hours, days or even weeks for partners’ to learn each others’ idiosyncrasies before they developed The Perfect Pair Programming Partnership. With our current rotation of partners for 371p, we usually have just two weeks to work with our partners, leaving me with the question, “How do we develop that pair programming rapport in an effective manner limited by a short time span for projects?”

I have a few ideas that I want to try out for this coming project, we’ll see how they work out.

I just turned in my Primes.txt for the second project that I worked on with Ting. You can see the results here:

Ting takes software design seriously. Period.

Everyone always talks about profiling, but Ting actually does it. Here’s a great blog post on using gprof, I’ll definitely start profiling my apps beginning with the next project. Per Dustin, I’m an Edison programmer, which is absolutely terrible. Time to start using real data to write programs.

This week, I learned more than I taught. I like it that way.

PROFILING. HERE I COME.

Predicted Number of Hours to Complete: 5 hours.
Actual Number of Hours to Complete: 10 hours.

Execution Time: 0.016s
Rank: 31.

Primes.out: http://blogs.utexas.edu/bjy95/primesout/.
TestPrimes.out http://blogs.utexas.edu/bjy95/testprimesout/.