Our first four episodes on multimedia programming, which cover basic image processing, are now online—comments are welcome.

As of this morning, 102 people have signed up for our Winter 2011 offering—it will be the largest run of the course ever. Registration has now closed, but we will send out notices when we have dates for the next one.

The second part of the discussion of persistence is now online. We’re almost up to how Python’s ‘pickle’ module works; should get there over the weekend or on Monday.

The chart below shows cumulative hours on Software Carpentry since we started working on Version 4 in May. We’re not as far along as I’d like, but progress has been pretty steady…

Elango Cheran has recorded an episodes for the sets & dictionaries lecture on reconstructing phylogenetic trees. It’s a neat application of dictionaries to a problem that’s usually explained as a matrix. Please check it out and let us know what you think.

Tommy Guy has recorded four introductory episodes on matrix programming with NumPy covering:

1. Introduction
2. Basic operations
3. Indexing
4. Linear algebra

Please have a look and let us know what you think.

We’ve been having problems hosting the course’s Subversion repository at its current URL (http://software-carpentry.org/swc), and so we’ve moved it to:

http://svn.software-carpentry.org/swc

This URL points to the same repository as before. If you already have a working copy checked out you can switch it to use the new URL by running the following shell command in the top-level folder of your working copy:

svn switch --relocate http://software-carpentry.org/swc http://svn.software-carpentry.org/swc

A few weeks ago we gave a quiz to the SWC students on the first three topics of the course: Version Control, Spreadsheets, and Databases. We did this in addition to our student check-ins in order to get a sense of how the students were doing with the material. Half of the class (24 students) submitted their answers.

Results:
Everyone did really, really well! The majority of the mistakes were small SQL syntax errors, or other similarly minor things.

Problem areas:

• Spreadsheets – Absolute References:
  This was probably the most common error made on the quiz. There were quite a few instances where students failed to recognize the necessity of using absolute references in their solutions. (Questions 2 and 6 in the Spreadsheets section of the quiz.)
• Databases – Nested Queries:
  The final question of the quiz involved nested queries. A few of the students did not use nested queries correctly, or failed to use them at all.
• Version Control – Comments:
  Ah, of course. There was tendency for students to skip the “give a meaningful comment” part when explaining how to commit changes to a repository.

Overall, the students did very well. We did not assign numeric scores, but the majority of the quizzes were perfect or near to it.

Well done, Fall 2010 class!

We have added brief notes on the books in the course’s bibliography—we hope you find them useful.

Over the last two weeks we’ve had short one-on-one conversations with the current students about how the course is going for them, and what they’ve enjoyed and haven’t. We’ve heard from 30 students of the ~48 signed up in the course.   For most students, their feedback was about the first three lectures (Version Control, Databases, Spreadsheets) and to some extent the Python lectures.

Here is a summary of the feedback:

Overall comments about the course:

Generally, the feedback was pretty much all positive.  Students are enjoying the course so far and see the benefit in learning the material.  Some are already using techniques or technologies they’ve learned (mostly, version control), or are referring back to the course material.  Almost everyone that responded liked the short episode screencast format.   One student said that they had dropped out of the course because it was hard to decide in advance whether to committ without better idea of the structure/time commitment involved.

Pacing:

The opinion on pacing is generally mixed.  Some students are finding the pacing fine, at least one  feels we are moving too quickly, and several have found the pace too slow.  A few of those that have said the pace was slow also have pointed out that spreadsheets & databases aren’t applicable to them or that their interest is more on Python so their opinion might change as the course progresses.

Time allocated for course work:

The most common answer was about 1-2 hours a week spent on the course.  Many students (over half) said that they have been busy and haven’t had enough time to really work through the course material. A couple of students have said that they didn’t intend to spend much time on exercises, but instead will just dip into the course if the material seems applicable to them.  Several students mentioned that they find it challenging to do the work at their own pace.

Exercises & Quizes:

Generally students have said the exercises and quizes have been helpful and are good motivators to keep working on the course.   Several students mentioned that the quiz was lengthy but not difficult.

Version control:

Most students that gave feedback about the version control lecture said that they were excited by it and thought it would be useful.   A reoccurring issue that came up was on how to set up an SVN repository for themselves now that they understood how to use one.  They liked the idea that we write up a short HOWTO directed to their sys admins.

Spreadsheets & Databases:

Generally students said they were familiar with spreadsheets and so didn’t find the lectures useful.  Not many students commented on the database lectures, but those that did either said it was easy or irrelevant to their work.   As with SVN, the issue of how to create a database was raised (which we don’t cover in our lectures).

Python:

Almost everyone seems excited about starting Python, but at the time of our check-ins not many people had gotten into the material.  Those that had asked about accessing databases from python, numerical programming (which we are creating a lecture on), and raised the concern that there was a significant jump in difficulty from the spreadsheet lectures to the Python lectures.

Other topics to cover:

We had lots of interest, suggestions and requests for future topics students wanted to see covered in the course.  They include:

• more advanced python programming
• Web programming (under construction)
• Programming in Matlab, and integrating with python
• Working with text (the lecture on regular expressions may helpful)
• Testing
• Shell, Make or generally automating scripts
• Perl
• Algorithms and parallel processing