The first tuples example has too many concepts

[PaulBone]

The first tuples example has:

wtf = (z, wtf)

Which introduces recursive data (infinite data) at the same time as it shows infinite typew (which are ilegal). Before we see this example we need to understand what:

ones = 1 : ones

So that we understand infinite data first.

Also contrast this with:

let q = 3 in q + q 

Where q appears twice (on the LHS of = and in the 'in' expression)

And

let q = 3 in let q = "troz" in q ++ q

Now there are two qs, one shadows the other. Which q is used in the innermost expression and in a slightly outer expression may need to be show. The distrinction between all these situations should be made at some point in the tutorial.

Thanks.

[sordina]

True, but it's difficult to introduce lists before tuples. Have a look at the dependency graph in the TOC to see what I mean.

[PaulBone]

Hrm, you may be right. And the problem might be that earlier I introduced Sarah to let, and how let q = 3 in let q = "troz" in q ++ q behaves. Which may have confused her when she saw wtf = (z, wtf]. The similarity being that q/wtf appear on both sides of something.

I don't actually use tuples much in practice. But other Haskell users probably do, I would usually introduce data types first, and tuples only incidentally when you want to return multiple things. But I'm not certain if that's "best".

[PaulBone]

I didn't click the table of contents and notice the graph before. Thanks for doing that. Dependencies between concepts is very important and something that's often forgotten.

[sordina]

I wish there was a simple way to attack things like this, but it's quite time consuming to get right. I decided to put tuples early since they work so nicely as a precursor to several upcoming topics, but could probably be left out somehow if I spent a bit more time factoring the concepts.