Possible error in TI Basic Starter Pack

Forum Attachments

Possible error in TI Basic Starter Kit

Bunyamin11 27 Nov 2015 23:57

Sorry is this isn't the best place to post this. In the exercises for chapter 2 of the TI Basic Starter Kit, Review question 8 is presented thusly:

What is the value of this conditional — A/B and C≥D — when A=52, B=3, C=0, and D=2?

I think the author intended A and B to be connected by a test (such as =, >, etc.) rather than the division operator, since the result of a division is not a truth value, but a numeric value. So if we modified the question to fit this idea, one possible question would be:

What is the value of this conditional - A<B and ((C>D) or (C=D)) when A=52, B=3, C=0, and D=2?

The answer would be "False".

I'm sure this was just a simple typo. I'm enjoying the course so far.

Last edited on 28 Nov 2015 00:18 by Bunyamin11 Show more

Re: Possible error in TI Basic Starter Kit

Michael2_3B 28 Nov 2015 01:02

No, I believe that is what the author actually meant. Though the author could have made a typo, I believe it is correct as it is.
.
Since A/B isn't connected to an operator, it actually gets interpreted as a Boolean value. If A/B returns a value of 0, it will be interpreted as a Boolean value of False. In any other case, however, A/B will be interpreted as True because it returned a number other than 0. So in this case, here is an example of how it is evaluated:

A/B and C≥D   //original conditional.

52/3 and 0≥2   //plugging the numbers in.

17.3333333 and False   //evaluating 52/3, and evaluating that 0 is NOT greater than or equal to 2 - so it becomes False

True and False   //17.3333333 became True because it was a number other than 0.

False   //because the first value AND the second value were not both True, it becomes False.

Re: Possible error in TI Basic Starter Kit

Bunyamin11 28 Nov 2015 01:18

Hmmmmm…sorry Michael, but that is incorrect. Division is not a boolean operation, and therefore is not true or false, and therefore cannot be part of a boolean expression. If TI BASIC allows any value other than 0 to be interpreted as True, then that is a flaw in the language.

If the members here can't understand this simple distinction, then I am obviously in the wrong place.

Last edited on 28 Nov 2015 01:21 by Bunyamin11 Show more

Re: Possible error in TI Basic Starter Kit

Silver Phantom 28 Nov 2015 01:44

What michael is saying is that the number produced by the division is considered a boolean value. In a traditional logic test, 1 is true and 0 is false. However in TI-Basic, all numbers in the set (-∞,0)u(0,∞) is considered true and 0 is false.

Its a bit of a weird quirk in this language, but it actually simplifies a lot of programs.
It can change

If A≠0

into

If A

Of course this is only a 2 byte difference, but it does imply a lot of optimizations for other programs, such as those using matrices, lists, strings, etc.

---

The Silver Phantom welcomes you

Re: Possible error in TI Basic Starter Kit

Michael2_3B 28 Nov 2015 01:54

Yes, you are right in that division is not a boolean operation. However, TI-Basic is a bit unique in that it will interpret it as true or false when used in programming. For example, when you have an if statement:

If A/B and C≥D
Then

//code

End

The code inside that block will NOT get run because the conditional gets evaluated to False like I showed in my previous post.

Don't ask me why it's like that at all. It is weird, and unique, but it is helpful in optimizing programs a bit. For example, you can check to see if values exist, and if so, run some code. Anyways, that is one quirk of this language.

Perhaps someone else could elaborate a bit better.

Re: Possible error in TI Basic Starter Kit

Mewcancraft 28 Nov 2015 16:04

I think it's related to the fact that reals can't have a value of nil in TI-Basic. Even using Delvar sets a real's value to 0 rather than setting it to nil. If this wasn't the case, it would mean that if we were to use a real variable like A in a Boolean operator like this:

:If A

it would translate to "If A has an actual value". However, since nil doesn't exist for reals, we need an alternative, and the most logical one would be 0.
This means however that by running the Boolean expression above, it doesn't translate to "If A has an actual value", but rather to "If A is something other than 0", which explains you guys' confusion.

But that's only my theory, there could also be a different explanation.

Re: Possible error in TI Basic Starter Kit

kg583 28 Nov 2015 17:25

Another possible reason this might be the case with TI-Basic is just the nature of calculator programming. If conditionals could only interpret values of 0 or 1 for variables, then there would have to be some error called when the variables take on other values. This not only makes coding harder for Texas Instruments, but also for us programmers who have to ensure the variables used in conditionals never take a value other than 0 or 1. It's easy to avoid this on most computer languages, but that's only because you have near unlimited variables to use (you can just make a variable called isTrue or something). TI-Basic only has 27 real variables and a handful of others to use in our programs. And, like Silver Phantom mentioned, not having to include the operators saves space (something the calculators are very limited in).

---

The solution to a complex problem is often a simple answer.

Re: Possible error in TI Basic Starter Kit

Bunyamin11 28 Nov 2015 17:45

Just thinking out loud, this is not directed at anyone in particular. In most languages, variables have to be initialized before they can be used, and often the default value after initialization is zero, at least for numeric variables.

If the value of zero were the same as nil, then zero could be used by an existential operator in predicate logic…but of course, this makes no sense…the origin point on a number line "exists".

Blah blah blah

Re: Possible error in TI Basic Starter Kit

kg583 28 Nov 2015 18:11

Like Mewcancraft said, TI-Basic doesn't have a nil value. Even using DelVar doesn't require the variable to be initialized; the first time you read it it will be zero. The majority of languages don't have this quirk, but TI-Basic is kind of special like that. Because all of the possible variables (reals, finance, windows, etc.) already exist on the calculator and you can't make more (not counting custom lists, which are a completely separate data type), there is no point to have a nil value for something that already exists.

TI-Basic Boolean logic was created to work around this fact and it actually is quite useful. This isn't a flaw in the language, but rather a necessary work-around for an innate problem with calculators that computers don't have to deal with.

---

The solution to a complex problem is often a simple answer.