If you have Microsoft Office you probably have a version of Visual Basic already available to you. I'd suggest starting by learning how to write functions and macros in Excel. You'll find this particularly easy if you start by learning how to record macros, as you can then inspect the code afterwards. There's also a great deal of satisfaction in writing your first function in Excel. Drop me a line if you want to know more (though my response time may be slow, as I don't log on here much). Good luck.

I call bullshit on this one.

Yes.

Generally I don't have to do this very much, and if I do it tends to be in contexts where someone else has already written a number in hex, so I just use the same case that's already been used. If I have to do it on my own I generally use upper case, because that's what I was taught at school.

*** Waves to the Brexit voters. ***

Absolutely. Agile methods are the curse of modern business. In the right place they can be useful, but when large-scale mission-critical systems are being developed using agile methods something is severely wrong.

Possibly. I think it would be very difficult to persuade senior management in many commercial firms to adopt this approach without a paradigm shift in the way corporations make their decisions. Imagine the conversation in the board room:

"The CTO advocates that we switch to this new system that is 99.9999% reliable. There will never be any hardware failures, but about 0.000001 of our bank balance calculations will be wrong."

"Sounds good. What's the reliability of the existing system?"

"Well, we aim for 100% accuracy in our bank balance calculations, but we have occasional hardware problems that prevent us from reporting any bank balances at all. This happens about once in 100,000 transactions."

"So you're offering us a choice between a system that goes wrong one time in a million, or a system that doesn't tell people anything at all one time in one hundred thousand?"

"Yes."

"We'll stick with the existing system. Failing to deliver any information is a better outcome than delivering incorrect information."

And this is before the legal officer steps in with questions about legal liability!

A central problem with this approach is that, at least in the form presented here, it doesn't guarantee accuracy. There will always be a probability (presumably low, but nevertheless non-zero) of error. Whilst this may be useful in some areas it's not obvious to me that it's always a desirable approach. How would you feel about a system that mostly got your bank balance correct, but occasionally got it wrong? And how about a system that, for example, monitors some mission-critical component? Would you feel happy with a system that "mostly" got things right when it comes to your heart monitor, or "mostly" spotted problematic objects in Spaceguard?

I don't deny that this is an interesting way of building on ideas in the cellular automaton tradition, but I'm not yet convinced that it can be exploited in a real world problem.

Could you provide a link to "the whole SJW thing", please. Thanks in advance!

If you enjoy maths at school, you might try working on the Collatz Conjecture. This should be within your ability level right now, or very soon. Of course the conjecture is well studied, but the nice thing about it is that there are almost certainly areas where an amateur might find something new. In particular, the controlling parameters are 2, 3 and 1 (divide by 2, or multiply by 3 and add 1). Once you've worked out how to deliver some basic answers, try changing the controlling parameters and see what happens, or add more simple steps to the basic process and see what happens then. It won't take long till you're in unknown territory, and have a chance of finding something that no-one else has ever seen.

Ordinary R user here. This made me laugh.

Very sorry to hear this. RIP.

Very sorry to hear this. I remember learning BNF (well, teaching myself, really) at age 14 from a bunch of manuals for an ICL1903A: it was one of my first attempts to seriously understand the nature of computers and programming.

Have you considered writing some code that would do the procrastinating for you?

Yes. This role should be taken by an experienced assistant project manager, and full documentation should be a standard project deliverable.

I'm sure that's true. I suppose I tend to be one of those people for whom correctness is more important than timeliness, but I certainly see the point you're making.

I don't know. I'm not sure whether I'm a mathematician or a system architect or an economist or an information theorist or an econometrician - I work at a rather curious intersection of all of these areas - but I do find that almost anything can be considered from a "functional" point of view. However, I do this by being very relaxed about what I consider to be the inputs and outputs of functions; in particular, I'm very happy to take the entire universe at some point in time to be the input to a function, a position that tends to boggle the minds of some of the people I talk to.

It's also worth noting that I'm much less worried about inefficiencies than most people. I tend to take the view that it's better for systems to match reality correctly than be efficient. I recognise that my views are not widely shared, though.

Agreed. Personally I'd prefer to see more movement from Haskell towards the "top right" of his diagram, but I recognise that there is a lot of movement from the "top left" families of languages. There's still a lot of work to be done, though.

Really not true... Java's earliest forms date back to mid-1991, while the first website dates to December 1990. Further, the Internet has been around since 1969, and widespread information processing has been available at a corporate level since the 1950s. Java has been around for a while, but it wasn't nearly as significant in the early stages of the www as you're suggesting.

I usually do two things after "Hello, World". First, I check that I can perform simple mathematical operations. Second, I check that I can open files, read from them and write to them.

There's no point in getting involved in projects with this kind of immature nonsense. Get involved in something where the other participants are adults. There's a basic rule that many actors follow, and it can readily be applied to other areas: never work with children or animals.

Don't overload yourself too much - Haskell will still be there when you've got less pressure. For now, spend most of your energy on your exams, because they really do matter (I sound like your Dad, don't I?). Good luck with your studies!

The best way to learn Haskell (and probably any other programming language) is to take a small step forward each day. A little regular practice every day will help you to learn to think "functionally". What I did was to make up little programming exercises for myself, so I was continually trying out Haskell in different little situations. Good luck!

There were some Haskell-related queries the other day, so I thought it might be useful to post this classic text to this subverse. The book is very readable, and takes you through the basics of Haskell and functional programming very gently. Strongly recommended for anyone who wants to see what all the fuss is about. Just download the Haskell platform, and away you go!

EDIT: Damn, I can't believe this made the front page... Guess Voat really is different from Reddit! TEN MINUTES LATER: Off the front page again, but it was fun while it lasted.

Haskell. If languages could enter beauty contests Haskell would be Miss Vermont.