Julia Scientific Programming

I was not satisfied with the course, mostly because I felt that it taught a lot of syntax without giving good explanations of the logic underlying it. Put differently, in contrast with other coding courses I have taken before (including the fantastic Matlab course offered by Coursera), I do not feel that this course taught me about computers and programming in general. Instead, it only presented me with a set of useful tools which are offered in Julia.

I must emphasize that there is a big difference between the first 3 weeks of the course, which were not good, and the fourth week, which was fine. During the first 3 weeks, the explanations given in the lectures were opaque. Moreover, much of the (short) lecture time was spent on things which are not substantial, like recurrent attempts to fit a curve or models from biology (SIR). At the same time, important basic notions, like types, were given to short of a treatment when compared to their importance and complexity. The quizzes in the first 3 weeks did not offer a good practice for the core material taught in class (I was disappointed to see, that one graded quiz even included a question about properties of the Ebola virus, which is really not a part of the core material in a programming course).

The course had not provided enough practice material. The lecturers highly advised us to "try those tools on our own projects", which is a good advice but is by no means supplements for sufficient practice opportunities which I would have expected to find in the course itself.

Last, the course should not be offered as a first course in programming, as critical basic notions, such as loops or logical operators, are hardly explained.

With all due respect, this is the worst course I followed on Coursera.

- First of all week 1 is extremely basic. I would think that most people learning Julia are coming from Python ann/or data-science background with programming knowledge. Yet the week 1 feels like for people with no programming experience.

- But then week 2 goes directly into plotting! Plotting is the last phase of any data analysis. Before you need to manipulate your data and there is a lot to learn about ad Julia's Array seems very different than Python or Numpy's arrays but is absolutely not covered.

- Worse, week 3 has nothing to do with Julia. I appreciate having a real world example but most of the lectures are explaining epidemiology SIR models and teach nothing more about Julia. In terms of learning Julia, that entire week can be skipped.

- Week 4, which should have been before week 2 is more interesting going into DataFrames and data manipulation. But again, I think there's a lot to teach first about basic Julia data types like Arrays, Vectors, Sets...

An example. In week 4 there is a code doing the following on DataFrames dataA = data[data[:Treatment] .== "A", :] But never in the course is it explained that the period, when prefixing a function or operator is used to indicate broadcasting (performing the operation element wise).

In short, the course fails on teaching any of the important specificities about Julia.

And on top of that, as other mentions, it is outdated.

Thanks to the entire team of Julia at the University of Cape Town.

Realmente has been very important for me to finish the course.

A month ago, at the beginning of the course, I was not in the ideal situation to continue it:

1- First handicap and not minor: I do not know the English language at all.

2-t he old macbook-apple stopped working,

3- I do not have any kind of regulated studies.

And despite everything, I managed to finish the course at 4 weeks!!!

thanks to:

online translators
the little Raspberry Pi 3 with Linux
the great JuliaBox
and above all, the magnificent code containing your Notebooks
to my precarious knowledge of basic, assembler 8-bits and interest in Turing machines
Only by studying the code (I do not know English to follow the videos) I managed to pass. I sincerely believe that he says a lot about the Julia team at that University.

What I am sure of, is that I will continue codifying, the interest that this course has awakened in me, I hope to continue it.

I'm euphoric!

J U L I A, eres grande !!

Team Julia of UCT You are great !!

from Spain, A big hug to everyone

and see you soon...

Firstly, the positive. Both lecturers are very clear, give good examples and I think overall provide a very useful and fun introduction to Julia. The range of topics covered in this course provides a nice foundation for further exploration into what Julia can do.

As a note for future improvement. I really believe the course should update its materials to the current 1.0 version of Julia. I understand that this is not a simple task but there were numerous occasions where the current implementation of Julia is very different (in for example array creation etc.). Bringing the course up-to-date to the 1.0 version I think will be worth it as this should be a more stable releases for the years to come.

Nevertheless, thanks for creating this course I enjoyed it and learned a lot!

Sincerely,

Daan Rademaker