Posts on The Art of Computing

Student Comments From the 2016 COMP1030 Class

Enthusiastic Hugo User — Thu, 05 Jan 2017 13:10:06 AEDT

The 2016 edition of The Art of Computing has concluded. So what did students say in the Official ANU Student Survey ?

What were the most notable strengths of the course?

All subject matter covered in the course was interesting, and never once boring. Being able to explore the fundamentals of programming was a real joy to learn, and the way in which we learned concepts then applied them to real life situations in our labs was just fantastic. Really enjoyed every second of this course!
Content was interesting and covered relatively well.
It covered a wide variety of topics.
The diversity of the content, and the topics are very interesting - particularly the guest lectures. The lab assignment provided by this course is really interesting, since Professor Xie makes her researches in machine learning, some topics she gives us are from this area, which are really interesting. I cannot imagine that machine learning can be such close to our life, it is fantastic. At the beginning of this course, I couldn’t have understood why this course teach us such a simple programme language, SNAP!, which is not powerful enough like Java, Python, etc. But in the end of this semester, I finally understand. This language is very suitable to us who have never programmed before. And the thinking of programme is conveyed to me in a simple way by using “SNAP!” . In the future programme, I can understand other language very fast since I have got the thinking of programming and computing. It is really a good course.
The tutors were all really helpful and were clear in their explanations when the lecture material wasn’t quite straightforward.
Very interesting assignment

What suggestions for improvement would you like to make?

The learning curve really was steep around the middle of the semester as was explained. I would personally prefer just a gradual climb in difficulty.
Having the entire course be 90% dependent on weekly assignments don’t really give the students a lot of incentive to improve their skills, wherein getting mediocre marks early on due to minor errors can affect motivation and drive once the course material gradually becomes more difficult.
Lecture recording was very poor. If I attended another lecture at the same time, the lecture recording is relatively useless because I can’t see what the lecturer is doing.
I think there is nothing wrong in this course.

Student Comments From the 2015 Class

Enthusiastic Hugo User — Wed, 12 Aug 2015 13:10:06 +1100

The 2015 edition of The Art of Computing has concluded. So what did our students say?

From the Official ANU Student Survey

What were the most notable strengths of the course?

Challenging lab assignments which expanded upon the basics taught in the lectures. The labs were also of a good length and assistance was provided throughout. The lectures followed a clearly defined structure, which was helpful for revision.
Engaging, entertaining and knowledgeable lecturer. Snap is a great medium for teaching abstract concepts which might be held back until second or third year due to semantic problems. Interesting practical assignments which can easily be seen to be useful.
This course was incredible.

The foremost strength was the collection of topics covered. I walk away feeling I have been given real, working literacy in an area totally foreign to my normal studies. Additionally, the integration with interdisciplinary application was well done. Pascal is a top-tier lecturer, and brought the course to life.

The assessment schedule encouraged equal dedication to the many different levels of computational problem solving.

I have actively recommended this course to all of my friends. Please keep this course available to students! Especially ones (like myself) outside of the Computer Science degree.
A comprehensive and fun introduction to computing. i’d never done anything with computers before starting and i could mostly keep up with my classmates that had previous experience and were taking other computing classes. the tutors were good at explaining. … very interesting assignments. overall an interesting, fun, challenging (in a good way) course.

We also conducted an unofficial anonymous survey with a few more detailed questions, here is what we heard:

Briefly remind us why you wanted to take Art of Computing

I'd never done any computing before but was interested as I use computers so often and am probably just going to use them more and more throughout my degree and career.

I wanted to obtain some measure of literacy in the field of computing and computational problem solving. As an Arts/Law student very few opportunities are available.

I thought it would be a more relevant course to take, rather than the discovering engineering course, which was my other option.

Did you learn what you expected to learn?

YES! And more. The syllabus was great.

I didn't know what to expect but I learnt a lot. I particularly enjoyed learning about computational thinking and applying that to other disciplines.

I learned more than I expected to learn. This was a really great class.

What advice would you give to a future student of Art of Computing (or yourself 3 months ago)?

Don't underestimate the value of the course, simply because it starts with a visual programming language - you will learn a lot of advanced content.

Keep asking questions until you understand, even if it feels like you're repeating yourself.

Do you think what you learned in Art of Computing will help your in the future (in ANU and beyond), if so, how?

Yes - I think having some experience with such a wide range of algorithms and applications of computer science, and well as knowing more about circuits and binary, will enable me to learn more about these areas at a faster pace than if they were new concepts.

Absolutely. It has given me some measure of functional literacy of computational methods and systems.

I think it's taught me a different way of thinking through problems and given me a deeper appreciation to the complexity of computers and programs I use everyday.

Would you now consider taking another course in computer science? If so, do you know which ones?

I'm planning on doing a degree in computer science, with a focus on artificial intelligence.

Probably not right now but that's more to do with wanting to other things more, I'm not opposed to computing, the craft of computing looks interesting.

This class was incredible - I very much hope it remains available for students outside of the CECS world.

Thank you for running this course, I feel very lucky to have had this opportunity

Welcome, and 2014 computing breakthroughs

Enthusiastic Hugo User — Mon, 10 Aug 2015 23:12:01 +1100

Welcome to The Art of Computing!

As part of a general overview to the course, let’s discuss a piece of recent computing news – a summary of breakthroughs in artificial intelligence by MIT Technology Review – in the context of what you will learn this semester.

First and foremost, the article focused on a few particular breakthroughs in machine learning, “a scientific discipline that explores the construction and study of algorithms that can learn from data”. The particular breakthroughs are in understanding unstructured content, such as text and images. For thousands of years, written communication has been the means for ideas and knowledge to transcend space and time. It is also been known that human vision accounts for 70% of the data flowing into the central nerous system^[1]. Automated understanding of pictures and langauge has been long-standing challenges in artificial intelligence, that they have been dubbed “AI-complete” – and making strides in such difficult problem is why these breakthroughs are exciting.

In week 4 of this class, you will learn a method to perform sentiment analysis, i.e., automatically deciding the (positive or negative) emotion expressed by a piece of text. In week6 of this class, you will learn the principles of machine learning, as well as one particular algoirthm called decision trees.

In addition, the article mentioned emerging applications in analyzing genetic databases, and choosing personalized treatment plans. In week 5 of this class, we will cover the basics of analyzing genome sequences, and in week 6, the computer learns to diagnose a disease from medical test results.

Last but not least, the article mentioned new design of computer chips that mimic human brain functions. In weeks 12-15, we will remove the ‘magic’ behind the physical complexity of a computing for you, by teaching you how to build one from scratch : )

Finally, we hope this discussion has given you a fair understanding about how different parts of the class, and the idea of computational thinking, connects to everyday life and the forefront of computing research.

For those who are still curious, here are some pointers to related work in and around ANU for you to explore further.

Research on the methods and applications of machine learning.
Resarch on optimization, decision making and planning.
Research on computer Vision and Robotics.
Analyzing text in legal claims and new machine learning methods for text representations.
Machine learning for genome and medical data.
Machine learning for social media and online photos and video.
Other cool applications of AI and machine learning, including traffic and transportation, looking over water pipes, predicting solar power, or designing energy systems, increasing social resilience, optimizing logistics and supply chains.

We wish you a fun and rewarding semester 1!

^{[1] Encyclopedia of Optical Engineering, Volume 2, edited by Ronald G. Driggers}

Running Snap! on your laptop?

Enthusiastic Hugo User — Mon, 19 Jan 2015 11:13:58 AEDT

Q: “Can I run Snap! on my personal computer, or some other computer outside of the lab?”
A: Yes of course! – and we’re including another tip for saving and loading your programs.

Simply point your browser (the latest version of Chrome preferred) to a Snap! instance, such as https://artofcomputing.cecs.anu.edu.au/snap/snap.html and you are good to go!

What you are looking at is a blank Snap! instance, you may wonder how to load/save your work in Snap!, or look at pre-existing programs, as what we did in the warm-up excercise.

Open the Snap! environment via this link, it will automatically load a simple program into the middle gray space. Notice that the ‘program’ consists of a series of blocks colored differently based on their functions. Press the gree flag on its top, or on the upper right-hand-corner of the your browser window to start the program. (With your computer speaker on!), see/hear what happens .

Now open this link, excute the program as above. Can you see/hear how this three-part canon is built from the part above?

The first way is to load the program from a file saved on your computer in so-called xml format. If you click on the paper icon at the upper left hand side of the Snap! window, you will see a menu item that says “Import …” (see picture below). Click on it, and you will be presented with a file chooser. Choose your local xml file, wait, and voila! - the program appears in your workspace.

Now with test your setup with the Canon in the warm-up programs.

Save the program xml somewhere to your computer, e.g. the desktop. http://snap.berkeley.edu/snapsource/demo/kanon.xml.
Load this program via the “Import … ” menu, test that it runs.
Change something in the program, and then user “export project … ” to save your work to disk.

The second way is by encoding the URL that contain the program - i.e., the keyword “#open:” followed by an ULR to the xml file. For example, as was done in the warm-up excercise: http://artofcomputing.cecs.anu.edu.au/snap/snap.html#open:http://snap.berkeley.edu/snapsource/demo/kanon.xml.

A first look at Snap!

Enthusiastic Hugo User — Thu, 01 Jan 2015 11:13:58 AEDT

Snap! is is a visual, drag-and-drop programming language. This post aims to give you a bit of its background, and a few pointers to explore.

Designed in UC Berkeley, Snap! is an extension of the visual programming tool Scratch. It does not need specialized hardware or complicated installation, and will run within any modern browser – we recommend a recent version of Chrome.

If you are new to programming, let’s take a sneak peek at the programming environment together, and excute a first snap! program:

Open the Snap! environment via this link, it will automatically load a simple program into the middle gray space. Notice that the ‘program’ consists of a series of blocks colored differently based on their functions. Press the gree flag on its top, or on the upper right-hand-corner of the your browser window to start the program. (With your computer speaker on!), see/hear what happens .
Now open this link, excute the program as above. Can you see/hear how this three-part canon is built from the part above?

If you are an expereinced programmer, you may want to take a look under-the-hood and see how Snap! is implmented on github.

For the even more adventurous, you may want to think about the one or two questions below, and help us improve Snap!

How to add a new block, such as interpreting a string of text (e.g. “March 21, 2012”) as a date?
How do you usually make a program run faster? Can we speed up Snap?

Posts on The Art of Computing

Student Comments From the 2016 COMP1030 Class

What were the most notable strengths of the course?

What suggestions for improvement would you like to make?

Student Comments From the 2015 Class

What were the most notable strengths of the course?

Briefly remind us why you wanted to take Art of Computing

Did you learn what you expected to learn?

What advice would you give to a future student of Art of Computing (or yourself 3 months ago)?

Do you think what you learned in Art of Computing will help your in the future (in ANU and beyond), if so, how?

Would you now consider taking another course in computer science? If so, do you know which ones?

Any other thoughts you’d like to share about this class?

Welcome, and 2014 computing breakthroughs

Running Snap! on your laptop?

A first look at Snap!