Type II Error Program

[Hovers...@aol.com]

Hi everyone,

 

Yates, Moore, Starnes (authors of the The Practice of Statistics) included a great program for the 84 to calculate the probability of a Type II error.  Does anyone know if there is a program out there to do the same thing on the nspire?  I know the nspire has the switchable keyboard, but I would love to have a program for the nspire operating system.

 

Thanks,

Heather Overstreet

Staunton River HS

Moneta, VA

---

Feeling the pinch at the grocery store? Make dinner for $10 or less.

[Nelson Sousa]

I think there isn't (unless someone allready built it). The question
should be "can such a program be built on TI-Nspire?". Well, the
answer is: it depends. (although in this case I believe it's perfectly
possible to adapt the algorithm).


You can't, in general, translate a program written for the TI-84 Plus
to use directly on TI-Nspire.

Not only are the languages substantially different (TI-Nspire's is
more like the programming language on TI-89 Titanium or Voyage 200),
but also the input of a program on TI-Nspire must be given at the
beginning of the program, as arguments passed when calling the
program.

This means that it's not just a matter of finding the equivalent
statements on TI-Nspire, as in some cases, there are no such
equivalent statements.


However...


I'm sure it's possible to adapt or redesign (almost) all of TI-84+
programs to be used on TI-Nspire. The reason for this is that
TI-Nspire's programming language is actually more powerfull and more
flexible than the language of TI-84 Plus (with string manipulation,
possibility of calling subroutines, defining functions, etc.). The
greatest limitation is the inability to use graphical input and
output, but this also can be overcome, in most cases, with 1%
inspiration and 99% transpiration, using geometrical constructions, a
spreadsheet and a few carefully designed functions.

The "TI gaming community" strongly disagrees with this and there are a
lot of users and developers complaining that TI-Nspire is impaired on
what concerns programming. (actually, they're correct about the
existence of limitations, they are wrong about the scope and relevance
of such limitations).


In this particular case, the program will need some input from the
user that on a TI-84 Plus is asked with Input and Prompt statements.
There are no such statements on TI-Nspire, but you can redesign in the
following way:

TI-84 Plus:
Program Test
Input "A=",A
Input "B=",B
Input "C=",C
...

TI-Nspire:
test(a,b,c)
Prgm
...
EndPrgm

Instead of being prompted to enter data while the program is running,
the user enters the data as arguments when the program is called.

The rest of the program is most likely a sequence of computations that
can (somewhat) easily be translated into TI-84 Plus.

The output on a TI-Nspire program must be done using the Disp instruction only.


Hope this helps.
Let me know if you need any further help.


Nelson

[hovers...@aol.com]

Thank you for the response.  I am quite proficient at programming on the 84; however, I have been a little intimidated about trying on the nspire.  I do have a background with programming in Visual Basic (although I am very rusty).  Could you possibly give me some basics or direct me to where I could find programming instructions on the nspire?

Thank you so much!
Heather

I think there isn't (unless someone allready built it). The question
should be "can such a program be built on TI-Nspire?". Well, the
answer is: it depends. (although in this case I believe it's perfectly
possible to adapt the algorithm).


You can't, in general, translate a program written for the TI-84 Plus
to use directly on TI-Nspire.

Not only are the languages substantially different (TI-Nspire's is
more like the programming language on TI-89 Titanium or Voyage 200),
but also the input of a program on TI-Nspire must be given at the
beginning of the program, as arguments passed when calling the
program.

This means that it's not just a matter of finding the equivalent
statements on TI-Nspire, as in some cases, there are no such
equivalent statements.


However...


I'm sure it's possible to adapt or redesign (almost) all of TI-84+
programs to be used on TI-Nspire. The reason for this is that
TI-Nspire's programming language is actually more powerfull and more
flexible than the language of TI-84 Plus (with string manipulation,
possibility of calling subroutines, defining functions, etc.). The
greatest limitation is the inability to use graphical input and
output, but this also can be overcome, in most cases, with 1%
inspiration and 99% transpiration, using geometrical constructions, a
spreadsheet and a few carefully designed functions.

The "TI gaming community" strongly disagrees with this and there are a
lot of users and developers complaining that TI-Nspire is impaired on
what concerns programming. (actually, they're correct about the
existence of limitations, they are wrong about the scope and relevance
of such limitations).


In this particular case, the program will need some input from the
user that on a TI-84 Plus is asked with Input and Prompt statements.
There are no such statements on TI-Nspire, but you can redesign in the
following way:

TI-84 Plus:
Program Test
Input "A=",A
Input "B=",B
Input "C=",C
.....

TI-Nspire:
test(a,b,c)
Prgm
.....
EndPrgm

Instead of being prompted to enter data while the program is running,
the user enters the data as arguments when the program is called.

The rest of the program is most likely a sequence of computations that
can (somewhat) easily be translated into TI-84 Plus.

The output on a TI-Nspire program must be done using the Disp instruction only.


Hope this helps.
Let me know if you need any further help.


Nelson



On Thu, Mar 19, 2009 at 00:19,  <Hovers...@aol.com> wrote:
> Hi everyone,
>
> Yates, Moore, Starnes (authors of the The Practice of Statistics) included a
> great program for the 84 to calculate the probability of a Type II error.
> Does anyone know if there is a program out there to do the same thing on the
> nspire?  I know the nspire has the switchable keyboard, but I would love to
> have a program for the nspire operating system.
>
> Thanks,
> Heather Overstreet
> Staunton River HS
> Moneta, VA
> ________________________________
> Feeling the pinch at the grocery store? Make dinner for $10 or less.
> >
>

---

Live traffic, local info, maps, directions and more with the NEW MapQuest Toolbar. Get it now!

[Pat Flynn]

John Hanna has an excellent primer on programming the TI-Nspire at his
web site. http://www.johnhanna.us/TI-nspire.htm



On Mar 19, 9:57 am, hoverstree...@aol.com wrote:
> Thank you for the response.? I am quite proficient at programming on the 84; however, I have been a little intimidated about trying on the nspire.? I do have a background with programming in Visual Basic (although I am very rusty).? Could you possibly give me some basics or direct me to where I could find programming instructions on the nspire?

> TI-Nspire:
> test(a,b,c)
> Prgm
> ....

> EndPrgm
>
> Instead of being prompted to enter data while th
> e program is running,
> the user enters the data as arguments when the program is called.
>
> The rest of the program is most likely a sequence of computations that
> can (somewhat) easily be translated into TI-84 Plus.
>
> The output on a TI-Nspire program must be done using the Disp instruction only.
>
> Hope this helps.
> Let me know if you need any further help.
>
> Nelson
>

> On Thu, Mar 19, 2009 at 00:19,  <Hoverstree...@aol.com> wrote:
> > Hi everyone,
>
> > Yates, Moore, Starnes (authors of the The Practice of Statistics) included a
> > great program for the 84 to calculate the probability of a Type II error.
> > Does anyone know if there is a program out there to do the same thing on the

> > nspire?? I know the nspire has the switchable keyboard, but I would love to

> > have a program for the nspire operating system.
>
> > Thanks,
> > Heather Overstreet
> > Staunton River HS
> > Moneta, VA
> > ________________________________

> > Feeling the pinch at the grocery store? Make dinner for $10 or less.- Hide quoted text -
>
> - Show quoted text -

[Hovers...@aol.com]

Thank you for the direction.  His instructions are really useful. 

 

I looked further into the Type II error program and realize that it acts more like a function.  The 84 program, TYPE2, displays screens for the user to select what kind of test they are using, (normal, t, proportion), and then more.  This is very much like the statistical functionality of the nspire when selecting a specific test or operation, etc, and windows are displayed for certain input.  So I guess what I am asking is, is it possible to create such a function (and I know you can create a function), with windows, etc, and how would I go about doing this?

 

Thanks,

Heather

[Nelson Sousa]

not with windows to select parameters.

You must specify all input when you're calling the function.

What you can do, however, using an idea given by Piman for another
issue, is creating a spreadsheet to assist the user.


Here's an example:

Suppose you have an algorithm that uses 4 or 5 different arguments.
And suppose that some of these arguments are not numbers, but strings
of text (such as "normal", "t", "proportion"), etc.

So, create a spreadsheet and on column A write down what the
parameters are; on column B write their values.

Now, go to another column, preferably one "hidden" far out right, say
Z and define:

z1=b1
z2=b2
...
zn=bn

For each of the cells on Z store them in variables with suggestive names.

Now write a function that doesn't take any input, but instead uses
those variables as parameters (the text variables will of course be
part of a cycle where you have if string="normal" then ... endif).

As the function returns it's value to any application, you can:
- call it on another spreadsheet cell
- call it from a calculator page and use the result in other calculations.

On the above I reccommend storing the variables in column Z to avoid
accidental deletion. The user only edits column B, so by storing the
cells of column B in variables there's the risk of the user
accidentaly deleting a variable definition and messing with the
results of the function.


See attached file for an example (in this case, the algorithm is
simply the 2nd degree equation). Column D is where variables are
stored; move it further to the right as a precaution.

Page 1.2 as the function itself (pretty trivial in this case), and a
calculator page where you can also call the function.


I hope this helps.

Nelson

[Joe]

Nelson said: "also the input of a program on TI-Nspire must be given

at the
beginning of the program, as arguments passed when calling the
program."

Is TI working on fixing this situation so that a dialog between the
user and program can occur while the program is running? Not allowing
that is a very grave limitation.

[Eagle-Man]

Not to say that I don't also want input, but is it really that grave a
limitation? For how many mathematical programs/functions do you not
know all the information beforehand? How many of the programs
presented here would benefit or be drastically improved with the
ability to input in the middle of the program/function.

Again, I'd love to see some input for non-mathematical reasons, but I
wouldn't be mad at TI for not including it.

You know what I would like to see? Classes.

--Eric F.

[Rex Boggs]

From: "Eagle-Man" <eagl...@duetsoftware.net>

> You know what I would like to see? Classes.

Eric, what do you mean by 'classes'?

Cheers

Rex

[Joe]

Eric, for me and probably many others, it is grave because I can not
create the kind of educational programs for the nspire that I did for
the 84 series. I have been waiting for TI do do something about this
problem for quite a while now and I am loosing hope. Do you see any
indication at all that TI is going to fix this problem? Also, what
advantage is there to severely limiting programability this way?
Shouldn't a newer product be at least as good as a previous product?
Obviously the programs presented here have to be of a type that don't
require inputs while the program is running so that point is mot.

> >> - Show quoted text -- Hide quoted text -

[Hovers...@aol.com]

One suggestion I would have is for TI to maybe incorporate a version of Visual Basic programming.  It seems that the programming function on the nspire is somewhat loosely familiar to the Visual Basic course I took in 2000.  Either way, I would love to be able to create windows for user input(since this is what the nspire already has as a wonderful piece of its functionality).  This programming capability would yet again show that the functionality of the nspire is cuts above all other handhelds.

 

Just some thoughts and ideas,

 

Heather Overstreet 

[Nelson Sousa]

Here's my two cents on this matter:

There have been, since ever, huge criticisms about TI-Nspire's
programming abilities. The developer's community turned it's back on
TI-Nspire (although the true reason is purely because they can't
program games in assembly like they did on the TI-84 Plus - just
browse ticalc.org's archives and count the files in each folder), and
a lot of users have been complaining. This is a fact, and I'm not here
to judge anyone's opinion on the matter. There have been criticisms.

In fact there's loss of flexibility (do not confuse flexibility with
power, TI-Nspire has much MORE programming power than the TI-84 Plus,
but less flexibility) on programing input and output. While on a TI-84
Plus you can add intermediate input, repeated input and menus and have
your output either as text or as graphics, a program on TI-Nspire
accepts only input as arguments and only displays output as text. This
is also a fact.

We have, however, functions. They are limited when compared to
programs, as they cannot create or edit global variables, can't have
try-endtry blocks, can't call programs and can only return an
expression (numbers, lists, strings, matrices, etc.), not control more
complex output. However, there are no functions on TI-84 Plus, there
isn't either a try-endtry block (which is an outstanding instruction
when dealing with user input, that allows one to trap errors much
easier!) and there's almost no string manipulation. In TI-Nspire we
have all the string manipulation allready present on TI-89 Titanium
and Voyage 200 (although I'd like a few more functions that a guy gets
used to).

Functions can be called from other applications (spreadsheet, for
example), and can be used to give you output updated in realtime,
something that was impossible on TI-84 Plus (at least without some
serious programming). Also their results can then be imported to a G&G
page and transfered to an axis to create points that depend of the
function's results.

We can, using functions, have geometrical objects like points,
measurements or sliders to define input of a function and have their
results stored on a spreadsheet cell and inserted back to a G&G page.
This allows one to create both graphical input and output.

Also we can have a spreadsheet page define input and output, again,
using a function.

Using a program one's limited to the fact that a program must be
called from the calculator app, eliminating the real time updates.

If you look at the kind of content I've been creating you can see that
there's a lot you can do in TI-Nspire. What you can't do is develop
the same kind of contents in the same way as you did on the TI-84
Plus! But that's a completely different problem. "How you develop it"
is a different question than "can it be developed".

As for "why are things as they are, if users are complaining", here's
my guess (it's merely a guess, I have no insiders info whatsoever):
programming for a multi-page document structure on TI-Nspire is a
totally different beast than creating the (rather simple using
computer standards) programming language of a "classical calculator".
On TI-84 Plus the command "DispGraph" displays the Graph Screen. The
ONLY graph screen available on the TI-84 Plus. On TI-Nspire that
instruction can mean (a) display the next available graph app,
whatever that is, (b) display the first graph app or (c) create a new
graph app and prepare to draw stuff on it. Among other meanings I
haven't thought of! Being able to add 10 graphical windows in the same
document is great, but a program needs to know which one do you mean
when you ask it to display a graph. So, you'll need to set some sort
of "document model", where you can define a page or application ID,
and a non-cumbersome language to allow you to reand and write to each
of these objects, given their ID. It's not an easy task, and just by
looking at the current Document Object Model of the Internet pages, a
standard a lot (thousands of people from all quadrants of the
industry) of people took years developing, and the way some browsers
(one should read Internet Explorer) messes things up, it's not
something you want to jump in loosely.


But the situation is the one we have now, there are differences, some
are limiting, some others can be circumvented, but in total I'm of the
opinion that:
- TI-Nspire's programming language is BY FAR more powerfull than the
one on TI-84 Plus, although LESS flexible; its power is similar to the
programming language of TI-89/Voyage 200 and the like.
- The flexibility can be, in many cases, circumvented by using
original solutions. In fact, if one asks a question about development
in this forum alone an answer will probably pop up in a matter of
hours.
- In other cases, some luminous idea is required and nobody got it
yet, so things can be, at the time, impossible.

Example of things that are limiting and annoying and cause complaints:

Problem: need a dialog box to input data, as there are lots of parameters
Solution: use a spreadsheet with the parameter's names on column A and
input on column B. Use a function instead of a program (if possible)
to display the results automatically on another spreadsheet cell.

Problem: but my input is graphical!
Solution: then, define it geometrically! If they're numbers, use
sliders or pure text with numbers in them; store that in variables;
call those variables on a spreadsheet. Define a function with the
algorithm part of the program you want to build. Call it from the
spreadsheet with the input you have allready stored in variables or in
the spreadsheet cells. Store the function's result in another
variable.

Problem: I also want my program to draw stuff.
Solution: Store the result on a variable and call it to the G&G page
by either: creating a point and linking its coordinates to the
variable or calculating an expression "x", using that variable as
parameter "x".

Problem: ah, but my program outputs text.
Solution: on G&G use when statements that return either a adequately
chosen number or undef; Use those results as point coordinates. Label
the point with the text you want.

Problem: but there are a LOT of text messages there.
Solution: so what? Define a LOT of points! I said it's possible, not
that it's easy ;) The periodic table uses 118 such messages and I
lived through it!!! Twice, as I did it in English and in Portuguese!

Problem: but I want free text. text can be anything
Solution: ok, now you got me. Tell me exactly what you're trying to
do, I'll think about it. Either I'll find a solution, or I won't.


So, to go back to the concerns manifested here:
- I have no idea whether a different paradigm is being prepared by TI
(you'll have to ask them; although I wouldn't keep my hopes up if the
project is ongoing but on an early stage, they probably won't tell
you)
- I find it very difficult to believe that something (other than
games!!!) possible on TI-84 Plus is simply impossible on TI-Nspire. If
I believed that I would have never created the sort of content I've
been developing.
- In some cases, there will be a loss of performance. But that's a
different beast. Even in games, performance is the limiting factor,
not power! All that I need to create games (even real time games)
exists allready and you can "see it" in some of my constructions (if
you look closely enough!). The only problem here is performance as you
can't just add 2000 points to a G&G page and hope it runs fast.


Believe it or not, I'm currently facing performance issues, not design
issues, on a electronic circuit simulator, that would have a function
generator and an oscilloscope with trigger and realtime display! And
it's not the "how can this be built" that is bothering me, that's
taken care of! It's the 1 to 2 seconds (on the computer software, not
the handheld!) it takes to update the oscilloscope screen everytime
something changes that's holding me back. (Oh, and there's not 1
single line of programming in a signal generator or an oscilloscope;
well, maybe a few small functions, but only to try to gain some
performance).

My other ongoing projects include a battleship game (with fleet
checking, score keeping and victory detection) and, just a proof of
concept, a car game. With a track and some obstacles. But those
projects take time, and they're not a top priority for me right now.


Nelson

[Eagle-Man]

I mean Classes in the C++ sense of the word. The ability to create a
new data structure which contains properties and functions.

For example, create a Class called "Student." Each "Student" contains
the properties gender, name, age, student number, and a class/grade
matrix (x by 2). Then you could create a list of Students and do
something like this:

high := 0
For i, 1, i+1
if sList[i].age > high then
high := sList[i].age
EndIf
EndFor

(finds oldest student)

This could be faked through global variables, but I'd prefer to have
some level of invisibility so the user doesn't need to know exactly
how it works, and also of security so that the user doesn't play
around with the values that might break the program.

Nelson: Great posts. I share many of your thoughts on the issue. And
it also got me thinking. The Nspire uses dialog boxes to get input
for certain functions (and creating programs and settings, etc), so
why can't they add an input dialog option? That wouldn't seem too
bad. It's still not necessary, though.

Also, in response to Heather, Windows are harder to make than you
think. In Visual Basic it seems pretty simple, but the underlying
code that goes on in the background to make the windows appear is very
complex.

Immovable windows with predefined dropdown menus and buttons are hard
enough, so I doubt that they will make user-defined windows. Even if
they did, the screen is too small to make use of more than one window
at a time. Splitting the screen is an example of that (more than one
split gets pretty crowded).

Speaking of which, it's already like we have windows anyway. Each app
is a new window. We can draw on them (G&G), calculate (Calc), write
(Notes), and even spreadsheet (D&S). What more could you ask for?
And like Nelson said, most things can be faked.

--Eric F.

--- Original Message ---

>

[Eagle-Man]

Joe, what kinds of programs are you talking about? Could you give an
example? Because I'm hardpressed to find a mathematical example
(since that's what the calc is for) where you do not know all the
values needed when you call the function.

I do not know if TI is planning on including this functionality at any
time, but as I stated before, I don't believe it's a "problem," as you
put it. It's just different. Like Nelson said, there are probably
ways to work around the input issue depending on your use.

And I would say that the Nspire IS at least as good as the 84+. In
fact, I'd say it's hugely better. The fact that we can save files,
work on multiple problems at once, use libraries, draw interactive
graphs, measure lengths and angles, etc. makes the Nspire much better
(not to mention the resolution).

Perhaps TI didn't include the functionality because they also had a
hard time thinking of mathematical examples where this would be
needed. Maybe they would look into it if some good examples were
brought to their attention.

In closing, let me say that I, too, would like to see input - in fact,
I requested it to one of the TI personnel during our local T3
institute (Marc might remember my list of suggestions), and this was
even before the TI was out - but while it's a nice feature, I don't
see it as necessary by any means and wouldn't mind if it never came out.

I look forward to hearing your comments.

[Joe]

Nelson, I appreciate your well thought out post. It is obvious that
you are a very skilled professional programer and your message has
caused me to view nspire in a different light. The designers of
nspire have apparently raised calculator proformance to new levels in
some areas. For example the abililty to store and work multiple
problems at once. In the process they have also raised programming to
more powerful but also more difficult levels that are workable for
professionals like you but not for people like me.

I am not a professional programmer nor a simi-professional programmer
nor even an advanced amatuer. I am just a guy that is good at math,
good at teaching math, and was able to write useful programs for the
84 after an hour or so of study. I am also *extremely* busy teaching
and taking care of family and household responsibilities. So if (and
this seems be true) I have to devote a large amount of time to
obtaining semiprofessional programing skills in order to create
programs for nspire similiar to those I have for the 84, that just
isn't going to happen. Not for me and probably not for a lot of other
people, dispite the admirable job this site is doing to promote nspire
programing.

For people like me, without input, output, and pause statements like
the 84 has, nspire might as well be a non-programmable calculator. I
was surprized that TI initually allowed nspire programing to be so
difficult and I have hoped that would change for the better, but
apparently it will not. Again, I want to thank you for your well
thought out post. As I said, it has caused me to change the way I
view nspire. I had thought that eventually TI would provide input,
output, and pause statements for nspire. Now I think it more likely
that TI assumes an increase in the customers programming skills will
eventually make that unnecessary.

> ...
>
> read more »- Hide quoted text -

[Joe]

Hi Eric, I just finished a reply to Nelson's post. In it I indicated
that I consider him to be a professinal programer and based on your
posts, I consider you to be in that league also. While nspire
programing may be suitable for professionals like you guys, the
simplified programing of the type used by the 84 is more suitable for
the larger number of users like me.

As for the example you asked for, consider a program that presents a
problem and asks the studet to choose from among the multiple choice
answers and enter that choice. When the student enters the correct
answer he/she is then presented with a different or more difficult
problem. When the student enters an incorrect choice he/she is
presented with the choice of another similiar problem or the page in
the text that covers the math principle involved. After several
incorrect choices, the program terminates with the message that the
student should review pages such and such in the text and then restart
the program. When the student completes the program, he/she is
presented with the message "Well done. You have mastered that
material." Students seem to prefer this type of program compared to
simply doing a test.

[Eagle-Man]

Self testing. Now there's an application I hadn't thought of. While
not ideal, here is a possible solution to your situation.

While not able to make this in a program, you could use a Notes page
with several questions on it.

If you have the Nspire Teacher Edition software, then you can choose
Insert->Question which will help you create a test document.

1) Insert->Question->Custom Choice
2) Select Teacher Tool Palette->Question Properties
3) Change Document Type to Self Test
4) Type questions and answers
5) Check correct answer on the left
6) Split page vertically
7) Insert Notes page
8) Type page and question numbers if they get it wrong
9) Repeat Steps 1, 4-9 as necessary
10) Save and distribute to class.

As students finish each question, they can press Menu->2:Check Answer
and it will tell them if the answer is right or wrong. If it is
wrong, then they have the option to try again or show the correct
answer.

Unfortunately, there isn't a way to limit the number of retries, and
this method requires a bit more honesty in what they choose and which
questions they need to do, but it's a pretty good way of doing things.

See the attached files.

[Nelson Sousa]

Joe:

I'm not a computer programmer. Not a professional one, not even a
semi-professional one. I'm a Physics graduate and I went to grad
school to study Mathematics (differential geometry). My first contact
with programming of any sort was on a Casio scientific calculator
(step-by-step programming) in middle grades and then on the TI-85 in
high school (I didn't even own a computer until I was 17!). Although I
can do small stuff on a computer, I'm actually more skilled at
programming TI calculators.

I agree with you that TI-Nspire takes time to master. Of course it
does, it's power is almost endless! However, programming for the
TI-Nspire, although more structured, isn't more difficult than for the
TI-84. It's in fact easier for the non professional programmer as
there are a few aids that allow you to fix syntax errors or structured
errors on a program in a much easier way. It is, of course, more time
consuming, as you allready know how to program TI-84 Plus and haven't
learned about TI-Nspire yet. But that will always happen anytime
something significant changes!

You don't need to be a full-time Nspire developer to create beautiful
things. You have always the group where you can ask anything and
someone might be able to answer.

TI-Nspire is much different than the TI-84 in a lot of ways, but, as I
like to say when I'm introducing TI-Nspire to a group of teachers (I'm
a full-time trainer and consultant in educational technology), we (all
of us: TI, teachers, trainers, developers,...) can't pretend anymore
that the world didn't evolve! Available technology is impressive and
at maths class we were still using a calculator platform that was 15
years old (TI-84 Plus Silver Edition is just a TI-82 with some
improvement). At some point evolution is required or else TI would be
wiped out by the competition from PDAs, smartphones, netbooks,
mathematics software, etc. That's what TI-Nspire tries to do, use the
technological developments available and get them in reach of math
teachers.

I don't know whether TI will add more flexible input or output. IT
seems to me that they started looking at other priorities first: a
good geometry application, application interoperability, stat graphs,
etc., and programming was left a bit behind as there was no time to
develop everything at once. It doesn't seem to me that TI will leave
things as they are in the long run. But we'll have to wait and see.



As for the example you mention: that's a tough one! I'd suggest having
a program and a geometry page in a split screen. As answers are
probably multiple choice, my first atempt would be something like
this:

1. Have a left/right split screen with G&G and calculator.
2. On the G&G page Create a set of points, as many as the questions
you want, and type in the questions as labels of those points. If more
than 1 line is needed, use a second point created as a translation of
the 1st point by a fixed vector (so that both points move together).
Store the coordinates on variables xq1. yq1, xq2, yq2, etc.
3. Do the same with your possible answers. store the coordinates as
xa1, ya1, xb1, yb1, xc1, yc1, etc.
4. As you create more and more questions and answers, move the points
away from the screen, so that they're not filling the screen
completely.
5. Add a slider that can take an integer value from 1 to 4; minimize
it so as to have only the up and down arrows.
6. Create a matrix with 1 line for each question. The first column is
the correct answer, the second column is the question that follows
this one if the student got the answer correctly and the third one is
the question to follow if the answer is incorrect.
7. Add a spreadsheet and set-up a data capture from the variable
you're controling with the slider.
8. Now, build the program! The initial data is: number of current
question, last captured value from the slider. The last value of the
spreadsheet tells you which option the student chose; compare with the
answer on the first column of that matrix, and see if it's correct or
not. Store the answer on a answer list and keep adding student's
answers.
8a) in each run of the program, a large number is stored in the
variables of the previous question, making it disappear. Using the
question number from that matrix, store adequate values in the
variables refering to the next question, making the points apprear on
screen. This is where the power of the language comes into place: if
variable question_num is 5, for example, a number, you can store
something in xq5 by using #(xq&string(question_num)) - see indirection
string on the reference manual, it's a very powerfull tool.

In the end, all question branches should end with "next question is
zero", so that the program knows when to stop doing anything.

This is just a sketch, it will work, although it might not be
spectacular. But it does the job...

Although it may look really nasty to build (and it takes quite some
time), the structure underneath it is somewhat similar to the one I'd
use on a program for the TI-84 Plus to do the same thing.

Hope this helps.


Nelson

[Joe]

Nelson, I appreciate very much the time and effort you have taken to
explain nspire programing, and I appreciate what a difficult time you
must have convincing people to learn the complicated programing
schemes that nspire requires. I respectfully submit however that the
need for such difficult programing is not the natural result of
inevitable technological progress, but an example of technology gone
awry. To be more specific, to go from the easy to use input and
output programing statements that the TI-84 uses, to the complicated
use of spread sheets, sliders, etc., is (in my opinion) not progress.
I believe that progress is the other way around. It is going from the
complicated programing that nspire requires (at this time), to easy to
use input and output statements.

I think that to transition from the simple direct programing available
with the TI-84, to the complicated programing scheme you have
explained (and I appreciate your effort) tends to make nspire somewhat
undesirable and I hope that TI will soon fix this condition. As a new
product nspire inevitably has problems but ignoring them, as if to
wish them away, is not the answer. I have to wonder if people are
thinking twice about equipping a whole room full of nspires with the
result that they then have to spend a lot of time and effort dealing
with its user unfriendly (in my opinion) programmability.

That is the truth as I see it and now I want to move on to some thing
else. Again, I thank you for explaining nspire programing. I am
impressed with your programing ability. I don't completely understand
all that you said, but I will print out a copy and I will study it.

Best Regards,

[Nelson Sousa]

Oh, I think you got it the wrong way!

When I say programming on TI-Nspire isn't more difficult, I'm not
refering to the techniques needed to create input/output using other
tools. That's a different thing, that's not programming, it's
Nspiring. ;)

I mean the programming language itself. That is, it's easier to read
and understand code written in TI-Nspire, as it's more similar to
English. Instead of having 4 or 5 End statements in a row, you have a
EndIf, EndFor, EndWhile, etc.

Also, most of the instructions are given in a more natural way and the
code is presented more nicely. It's more powerfull, and a lot more
functions are available, but that's the natural course of evolution.

The input/output issue is in fact a complicated issue. On one hand,
developing such a object oriented language in a bug-free way that can
be at the sime time used by teachers and students and powerfull and
complete enough to please the programmer community can be a herculean
task.

On the other hand, for most (almost all) mathematical programming
teachers need, other tools can be developed. The need math teachers
have to develop interactive content and that usually required some
programming skills can actually be done without any programming.

The input/output limitation is a limitation and a serious restriction
of the language's flexibility, that's not in dispute. My goal when I
describe what the alternative would look like, isn't as much convince
you that Nspire is better in every way, or simpler in every way, but
to show that in spite of the apparent (and in some cases real)
difficulties, one can achieve the same type of results. Only using
different techniques. It goes more to say that things are different on
TI-Nspire as oposed to things are impossible on TI-Nspire.


Nelson

[Marc Garneau]

I've been a bit swamped lately, hence I haven't been able to join in this worthwhile conversation.  All I'd like to say is that I am one who has done a ton of TI-84 programming in the past and depended heavily on it for enhancing the learning experience in my classroom.  With the TI-Nspire, I haven't felt the same need to rely on programming - I've been able to develop activities that have been far more effective just using the built-in structure of the Nspire applications, and I get to 'keep' what I've developed because of the document capability.  Only occasionally do I now feel the need to program in the more particular sense, and when I do I've found what's there to be adequate.  That being said, would I like to have built in functionality for input dialog boxes and menus?  Yes, absolutely, and I hope it comes one day.  But overall, the Nspire for me has been a HUGE step forward, not backwards.

Marc Garneau

[Andy Kemp]

Just to add to the debate, I have to agree with Marc and Nelson.  I have done a fair bit of programming on various platforms including the 84 and an old Casio calculator (as well as some VB, PHP etc) however I have as of yet not felt the need to do any "programming" on the Nspire...

I agree some form of input/output would be a nice addition but now that I can create documents and share these, and do sophisticated geometry and link that to spreadsheets and graphs without needing to do any programming at all is what makes the Nspire platform more powerful than the the 84 ever was...

That said I am sure there are things that the 84 could do, and things you could program the 84's to do that are not easy to do on the TI-Nspire but I am confident that most of those features will come in time...  and until then you could always use the 84 faceplate for those occasions when you feel the 84 would be a simpler solution and the Nspire faceplate for those other occasions...

That was certainly what I intended to do, that said I haven't put my 84 faceplate back in for 4-5 months now...

However if there is anyone from TI listening (and I'm sure there is!) I was thinking the kind of solution I would like to see to this input output issue would be a 'inputbox' type command like that in VBA which could just pop-up a dialogue box where the user could enter a value/string, but could also be used to produce a kind of pop-up window which could display a message with an OK button to close.  A solution along these lines would then work (in theory) within any of the page types as it would just pop over the content much like the menus do...

Do persevere with the Nspire, it may not be perfect but it is much more usable for most things you want/need to do in a classroom than the 84 ever was...  Also don't be put off but the level of complexity in some of the documents shown on here, they are very complicated and time-consuming to produce but would have been equally (if not more-so) to produce on the 84, this isn't because the Nspire is complicated it is because the things people are trying to achieve here are complicated!!

Cheers

Andy

[Eagle-Man]

"However, programming for the TI-Nspire, although more structured,
isn't more difficult than for the TI-84. It's in fact easier for the
non professional programmer as there are a few aids that allow you to
fix syntax errors or structured errors on a program in a much easier
way."

Also, the fact that you can type in function and instructions names
into the editor makes it much easier to program. You don't have to
search through endless menus to find the command you're looking for,
you just type it. Also, should you need to search, the Catalogue has
several sorting methods which makes it easier to search.

"I respectfully submit however that the need for such difficult
programing is not the natural result of inevitable technological
progress, but an example of technology gone awry."

Joe, I disagree. As things like computers get more powerful, they
become more and more complicated to program. THIS is the nature of
technology. As you have guessed, I am a programmer. Not
"professional," but I obtained a Computer Science degree in 2005, and
I can say with out a doubt, that programming for computers has become
much, much more complicated in the last 10, even in the last 5 years.

For an extreme example, look at Basic. Now open up a webpage that
contains javascript. It's really different, and more complicated, but
it can do more. And I could provide you with examples of C++ Win32
programming for windows. To open a single window requires about a
page and a half of code.

I respectfully submit to you that the more "difficult" programming is
a) necessary, and b) not that different. The TI-Nspire has a lot of
new features that are on the cutting edge for graphing calculators and
to control that, a more complicated language is needed. However, the
commands on the 83 is like a subset of the commands on the Nspire.
While some of the names of commands are different, I'd say about 90%
of the programs on the 83 (excluding games) can be easily translated
into Nspire programs without much effort.

Please don't be so quick to disregard the capabilities of the Nspire.
Yes, it's new and different, but it's not a step back, as you say. If
you give it a shot, you'll see that for the most part, it's very
similar to the 83 Basic you're used to. You don't even need to learn
about the new functions or commands. And like Nelson said, there's a
whole group of people here that can help you with programs.

If TI truly did simplify the programming capabilities, now that would
be a step back. We would lose much of the control that they've given
us in this new calc, such as the ability to detect errors and handle
them in different ways.

I also noticed that you didn't mention the file I attached to my
previous email. Check it out. That took me less than 5 minutes to do
on my computer and required no programming (just the Teacher Edition
software). I think you might be pleased.

"and until then you could always use the 84 faceplate for those
occasions when you feel the 84 would be a simpler solution and the
Nspire faceplate for those other occasions..."

That's another option. They include an 84 faceplate for a reason.

"I was thinking the kind of solution I would like to see to this input
output issue would be a 'inputbox' type command like that in VBA which
could just pop-up a dialogue box where the user could enter a
value/string, but could also be used to produce a kind of pop-up
window which could display a message with
an OK button to close."

Andy: I'm sure that this is what TI is considering, but it's much
easier said than done. Some of the ability is obviously there, but to
be able to make it user-customizable and possibly use different
buttons; to take the value and put it into a variable of the user's
creation (it doesn't exist before, like the A-Z of the 84); to know
how to interpret the input and make sure it's of the type the user
wants; these are the hard parts, and it will probably take some time
to figure it out. Give them time.

[Andy Kemp]

Sorry didn't mean to imply that it would be easy to add a useful/
flexible input output system... Was just the designer in me doing a
little dreaming!!

I have managed to work round this need so far and am very pleased with
my TI-Nspire!!

Sent from my iPhone

> ----~--~----~------------~-------~--~----~
> To post to this group, send email to tins...@googlegroups.com
> To unsubscribe send email to tinspire+u...@googlegroups.com
> For more options, visit this group at
> http://groups.google.com.au/group/tinspire?hl=en-GB?hl=en-GB
> The tns documents shared by group members are archived at
> http://lafacroft.com/archive/nspire.php
> -~----------~----~----~----~------~----~------~--~---
>
> -~---
>
>
>

[hovers...@aol.com]

Wow, little did I know what a discussion my question would bring about.  Concerning "90% of the 84 programs can be rewritten for nspire", I have attached the Type II error program I mentioned.  I have not been able to wrap my head around it.  Would anyone be interested in helping me rewrite it, if it can be done?

Thank you so much for the wonderful posts and I have become even more interested in programming on the nspire.

Thanks,
Heather

-----Original Message-----
From: Andy Kemp <an...@1kemp.co.uk>
To: tins...@googlegroups.com <tins...@googlegroups.com>
Sent: Tue, 24 Mar 2009 4:41 am
Subject: [tinspire] Re: Type II Error Program

Sorry didn't mean to imply that it would be easy to add a useful/ 
flexible input output system... Was just the designer in me doing a  
little dreaming!!

I have managed to work round this need so far and am very pleased with  
my TI-Nspire!!

Sent from my iPhone

On 24 Mar 2009, at 04:27, Eagle-Man <eagl...@duetsoftware.net> wrote:

>
> "However, programming for the TI-Nspire, although more structured,
> isn't more difficult than for the TI-84. It's in fact easier for the
> non professional programmer as there are a few aids that allow you to
> fix syntax errors or structured errors on a program in a much easier
> way."
>
> Also, the fact that you can type in function and instructions names
> into the editor makes it much easier to program.  You don't have to
> search through endless menus to find the command you're looking for,
> you just type it.  Also, should you need to search, the Catalogue has
> several sorting methods which makes it easier to search.
>
> "I respectfully submit however that the need for such difficult
> programing is not the natural result of inevitable technological
> progress, but an example of technology gone awry."
>
> Joe, I disagree.  As things like computers get more powerful, they
> become more and more complicated to program.  THIS is the nature of
> technology.  As you have guessed, I am a programmer.  Not
> "professional," but I obtained a Computer Science degree in 2005, and
> I can say with out a doubt, that programming for computers has become
> much, much more complicated in the last 10, even in the last 5 years.
>
> For an extreme example, look at Basic.  Now open up a webpage that
> contains javascript.  It's really different, and more complicated, but
> it can do more.  And I could provide you with examples of C++ Win32
> programming for windows.  To open a single window requires about a
> page and a half of code.
>
> I respectfully submit to you that the more "difficult" programming is
> a) necessary, and b) not that different.  The TI-Nspire has a lot of
> new features that are on the cutting edge for graphing calculators and
> to control that, a more complicated language is needed.  However, the
> commands on the 83 is like a subset of the commands on the Nspire.
> While some of the names of commands are different, I'd say about 90%
> of the programs on the 83 (excluding games) can be easily translated
> into Nspire programs without much effort.
>
> Please don't be so quick to disregard the capabilities of the Nspire.
> Yes, it's new and different, but it's not a step back, as you say.  If
> you give it a shot, you'll see that for the most part, it's very
> similar to the 83 Basic you're used to.  You don't even need to learn
> about the new functions or commands.  And like Nelson said, there's a
> whole group of people here that can help you with programs.
>
> If TI truly did simplify the programming capabilities, now that would
> be a step back.  We would lose much of the control that they've given
> us in this new calc, such as the ability to detect errors and handle
> them in different ways.
>
> I also noticed that you didn't mention the file I attached to my
> previous email.  Check it out.  That took me less than 5 minutes to do
> on my computer and required no programming (just the Teacher Edition
> software).  I think you might be pleased.
>
> "and until then you could always use the 84 faceplate for those
> occasions when you feel the 84 would be a simpler solution and the
> Nspire faceplate for those other occasions..."
>
> That's another option.  They include an 84 faceplate for a reason.
>
> "I was thinking the kind of solution I would like to see to this input
> output issue would be a 'inputbox' type command like that in VBA which
> could just pop-up a dialogue box where the user could enter a
> value/string, but could also be used to produce a kind of pop-up
> window which could display a message with
> an OK button to close."
>
> Andy: I'm sure that this is what TI is considering, but it's much
> easier said than done.  Some of the ability is obviously there, but to
> be able to make it user-customizable and possibly use different
> buttons; to take the value and put it into a variable of the user's
> creation (it doesn't exist before, like the A-Z of the 84); to know
> how to interpret the input and make sure it's of the type the user
> wants; these are the hard parts, and it will probably take some time
> to figure it out.  Give them time.
>
>   --Eric F.
>
> --- Original Message ---
>
>>
>> Joe:
>>
>> I'm not a computer programmer. Not a professional one, not even a
>> semi-professional one. I'm a Physics graduate and I went to grad
>> school to study Mathematics (differential geometry). My first contact
>> with programming of any sort was on a Casio scientific calculator
>> (step-by-step programming) in middle grades and then on the TI-85 in
>> high school (I didn't even own a computer until I was 17!).  
>> Although I
>> can do small stuff on a computer, I'm actually more skilled at
>> programming TI calculators.
>>
>> I agree with you that TI-Nspire takes time to master. Of course it
>> does, it's power is almost endless! However, programming for the
>> TI-Nspire, although more structured, isn't more difficult than for  
>> the
>> TI-84. It's in fact easier for the non professional programmer as
>> there are a few aids that allow you to fix syntax errors or  
>> structured
>> errors on a program in a much easier way. It is, of course, more time
>> consuming, as you allready know how to program TI-84 Plus and haven't
>> learned about TI-Nspire yet. But that will always happen anytime
>> something significant changes!
>>
>> You don't need to be a full-time Nspire developer to create beautiful
>> things. You have always the group where you can ask anything and
>> someone might be able to answer.
>>
>> TI-Nspire is much different than the TI-84 in a lot of ways, but,  
>> as I
>> like to say when I'm introducing TI-Nspire to a group of teachers  
>> (I'm
>> a full-time trainer and consultant in educational technology), we  
>> (all
>> of us: TI, teachers, trainers, developers,...) can't pretend anymore
>> that the world didn't evolve! Available technology is impressive and
>> at maths class we were still using a calculator platform that was 15
>> years old (TI-84 Plus Silver Edition is just a TI-82 with some
>> improvement). At some point evolution is required or else TI would be
>> wiped out by the competition from PDAs, smartphones, netbooks,
>> mathematics software, etc. That's what TI-Nspire tries to do, use the
>> technological developments available and get them in reach of math
>> teachers.
>>
>> I don't know whether TI will add more flexible input or output. IT
>> seems to me that they started looking at other priorities first: a
>> good geometry application, application interoperability, stat graphs,
>> etc., and programming was left a bit behind as there was no time to
>> develop everything at once. It doesn't seem to me that TI will leave
>> things as they are in the long run. But we'll have to wait and see.
>>
>>
>>
>> As for the example you mention: that's a tough one! I'd suggest  
>> having
>> a program and a geometry page in a split screen. As answers are
>> probably multiple choice, my first atempt would be something like
>> this:
>>
>> 1. Have a left/right split screen with G&G and calculator.
>> 2. On the G&G page Create a set of points, as many as the questions
>> you want, and type in the questions as labels of those points. If  
>> more
>> than 1 line is needed, use a second point created as a translation of
>> the 1st point by a fixed vector (so that both points move together).
>> Store the coordinates on variables xq1. yq1, xq2, yq2, etc.
>> 3. Do the same with your possible answers. store the coordinates as
>> xa1, ya1, xb1, yb1, xc1, yc1, etc.
>> 4. As you create more and more questions and answers, move the points
>> away from the screen, so that they're not filling the screen
>> completely.
>> 5. Add a slider that can take an integer value from 1 to 4; minimize
>> it so as to have only the up and down arrows.
>> 6. Create a matrix with 1 line for each question. The first column is
>> the correct answer, the second column is the question that follows
>> this one if the student got the answer correctly and the third one is
>> the question to follow if the answer is incorrect.
>> 7. Add a spreadsheet and set-up a data capture from the variable
>> you're controling with the slider.
>> 8. Now, build the program! The initial data is: number of current
>> question, last captured value from the slider. The last value of the
>> spreadsheet tells you which option the student chose; compare with  
>> the
>> answer on the first column of that matrix, and see if it's correct or
>> not. Store the answer on a answer list and keep adding student's
>> answers.
>> 8a) in each run of the program, a large number is stored in the
>> variables of the previous question, making it disappear. Using the
>> question number from that matrix, store adequate values in the
>> variables refering to the next question, making the points apprear on
>> screen. This is where the power of the language comes into place: if
>> variable question_num is 5, for example, a number, you can store
>> something in xq5 by using #(xq&string(question_num)) - see  
>> indirection
>> string on the reference manual, it's a very powerfull tool.
>>
>> In the end, all question branches should end with "next question is
>> zero", so that the program knows when to stop doing anything.
>>
>> This is just a sketch, it will work, although it might not be
>> spectacular. But it does the job...
>>
>> Although it may look really nasty to build (and it takes quite some
>> time), the structure underneath it is somewhat similar to the one I'd
>> use on a program for the TI-84 Plus to do the same thing.
>>
>> Hope this helps.
>>
>>
>> Nelson
>>
>>
>>
>> On Mon, Mar 23, 2009 at 01:54, Joe <wjb...@yahoo.com> wrote:
>>>
>>> Nelson, I appreciate your well thought out post.  It is obvious that
>>> you are a very skilled professional programer and your message has
>>> caused me to view nspire in a different light.  The designers of
>>> nspire have apparently raised calculator proformance to new levels  
>>> in
>>> some areas.  For example the abililty to store and work multiple
>>> problems at once.  In the process they have also raised  
>>> programming to
>>> more powerful but also more difficult levels that are workable for
>>> professionals like you but not for people like me.
>>>
>>> I am not a professional programmer nor a simi-professional  
>>> programmer
>>> nor even an advanced amatuer.  I am just a guy that is good at math,
>>> good at teaching math, and was able to write useful programs for the
>>> 84 after an hour or so of study.  I am also *extremely* busy  
>>> teaching
>>> and taking care of family and household responsibilities.  So if  
>>> (and
>>> this seems be true) I have to devote a large amount of time to
>>> obtaining semiprofessional programing skills in order to create
>>> programs for nspire similiar to those I have for the 84, that just
>>> isn't going to happen.  Not for me and probably not for a lot of  
>>> other
>>> people, dispite the admirable job this site is doing to promote  
>>> nspire
>>> programing.
>>>
>>> For people like me, without input, output, and pause statements like
>>> the 84 has, nspire might as well be a non-programmable  
>>> calculator.  I
>>> was surprized that TI initually allowed nspire programing to be so
>>> difficult and I have hoped that would change for the better, but
>>> apparently it will not.  Again, I want to thank you for your well
>>> thought out post.  As I said, it has caused me to change the way I
>>> view nspire.  I had thought that eventually TI would provide input,
>>> output, and pause statements for nspire.  Now I think it more likely
>>> that TI assumes an increase in the customers programming skills will
>>> eventually make that unnecessary.
>>>
>>> On Mar 22, 2:07 pm, Nelson Sousa <nso...@gmail.com> wrote:
>>>> Here's my two cents on this matter:
>>>>
>>>> There have been, since ever, huge criticisms about TI-Nspire's
>>>> programming abilities. The developer's community turned it's back  
>>>> on
>>>> TI-Nspire (although the true reason is purely because they can't
>>>> program games in assembly like they did on the TI-84 Plus - just
>>>> browse ticalc.org's archives and count the files in each folder),  
>>>> and
>>>> a lot of users have been complaining. This is a fact, and I'm not  
>>>> here
>>>> to judge anyone's opinion on the matter. There have been  
>>>> criticisms.
>>>>
>>>> In fact there's loss of flexibility (do not confuse flexibility  
>>>> with
>>>> power, TI-Nspire has much MORE programming power than the TI-84  
>>>> Plus,
>>>> but less flexibility) on programing input and output. While on a  
>>>> TI-84
>>>> Plus you can add intermediate input, repeated input and menus and  
>>>> have
>>>> your output either as text or as graphics, a program on TI-Nspire
>>>> accepts only input as arguments and only displays output as text.  
>>>> This
>>>> is also a fact.
>>>>
>>>> We have, however, functions. They are limited when compared to
>>>> programs, as they cannot create or edit global variables, can't  
>>>> have
>>>> try-endtry blocks, can't call programs and can only return an
>>>> expression (numbers, lists, strings, matrices, etc.), not control  
>>>> more
>>>> complex output. However, there are no functions on TI-84 Plus,  
>>>> there
>>>> isn't either a try-endtry block (which is an outstanding  
>>>> instruction
>>>> when dealing with user input, that allows one to trap errors much
>>>> easier!) and there's almost no string manipulation. In TI-Nspire we
>>>> have all the string manipulation allready present on TI-89 Titanium
>>>> and Voyage 200 (although I'd like a few more functions that a guy  
>>>> gets
>>>> used to).
>>>>
>>>> Functions can be called from other applications (spreadsheet, for
>>>> example), and can be used to give you output updated in realtime,
>>>> something that was impossible on TI-84 Plus (at least without some
>>>> serious programming). Also their results can then be imported to  
>>>> a G&G
>>>> page and transfered to an axis to create points that depend of the
>>>> function's results.
>>>>
>>>> We can, using functions, have geometrical objects like points,
>>>> measurements or sliders to define input of a function and have  
>>>> their
>>>> results stored on a spreadsheet cell and inserted back to a G&G  
>>>> page.
>>>> This allows one to create both graphical input and output.
>>>>
>>>> Also we can have a spreadsheet page define input and output, again,
>>>> using a function.
>>>>
>>>> Using a program one's limited to the fact that a program must be
>>>> called from the calculator app, eliminating the real time updates.
>>>>
>>>> If you look at the kind of content I've been creating you can see  
>>>> that
>>>> there's a lot you can do in TI-Nspire.  What you can't do is  
>>>> develop
>>>> the same kind of contents in the same way as you did on the TI-84
>>>> Plus! But that's a completely different problem. "How you develop  
>>>> it"
>>>> is a different question than "can it be developed".
>>>>
>>>> As for "why are things as they are, if users are complaining",  
>>>> here's
>>>> my guess (it's merely a guess, I have no insiders info whatsoever):
>>>> programming for a multi-page document structure on TI-Nspire is a
>>>> totally different beast than creating the (rather simple using
>>>> computer standards) programming language of a "classical  
>>>> calculator".
>>>> On TI-84 Plus the command "DispGraph" displays the Graph Screen.  
>>>> The
>>>> ONLY graph screen available on the TI-84 Plus. On TI-Nspire that
>>>> instruction can mean (a) display the next available graph app,
>>>> whatever that is, (b) display the first graph app or (c) create a  
>>>> new
>>>> graph app and prepare to draw stuff on it. Among other meanings I
>>>> haven't thought of! Being able to add 10 graphical windows in the  
>>>> same
>>>> document is great, but a program needs to know which one do you  
>>>> mean
>>>> when you ask it to display a graph. So, you'll need to set some  
>>>> sort
>>>> of "document model", where you can define a page or application ID,
>>>> and a non-cumbersome language to allow you to reand and write to  
>>>> each
>>>> of these objects, given their ID. It's not an easy task, and just  
>>>> by
>>>> looking at the current Document Object Model of the Internet  
>>>> pages, a
>>>> standard a lot (thousands of people from all quadrants of the
>>>> industry) of people took years developing, and the way some  
>>>> browsers
>>>> (one should read Internet Explorer) messes things up, it's not
>>>> something you want to jump in loosely.
>>>>
>>>> But the situation is the one we have now, there are differences,  
>>>> some
>>>> are limiting, some others can be circumvented, but in total I'm  
>>>> of the
>>>> opinion that:
>>>> - TI-Nspire's programming language is BY FAR more powerfull than  
>>>> the
>>>> one on TI-84 Plus, although LESS flexible; its power is similar  
>>>> to the
>>>> programming language of TI-89/Voyage 200 and the like.
>>>> - The flexibility can be, in many cases, circumvented by using
>>>> original solutions. In fact, if one asks a question about  
>>>> development
>>>> in this forum alone an answer will probably pop up in a matter of
>>>> hours.
>>>> - In other cases, some luminous idea is required and nobody got it
>>>> yet, so things can be, at the time, impossible.
>>>>
>>>> Example of things that are limiting and annoying and cause  
>>>> complaints:
>>>>
>>>> Problem: need a dialog box to input data, as there are lots of  
>>>> parameters
>>>> Solution: use a spreadsheet with the parameter's names on column  
>>>> A and
>>>> input on column B. Use a function instead of a program (if  
>>>> possible)
>>>> to display the results automatically on another spreadsheet cell.
>>>>
>>>> Problem: but my input is graphical!
>>>> Solution: then, define it geometrically! If they're numbers, use
>>>> sliders or pure text with numbers in them; store that in variables;
>>>> call those variables on a spreadsheet. Define a function with the
>>>> algorithm part of the program you want to build. Call it from the
>>>> spreadsheet with the input you have allready stored in variables  
>>>> or in
>>>> the spreadsheet cells. Store the function's result in another
>>>> variable.
>>>>
>>>> Problem: I also want my program to draw stuff.
>>>> Solution: Store the result on a variable and call it to the G&G  
>>>> page
>>>> by either: creating a point and linking its coordinates to the
>>>> variable or calculating an expression "x", using that variable as
>>>> parameter "x".
>>>>
>>>> Problem: ah, but my program outputs text.
>>>> Solution: on G&G use when statements that return either a  
>>>> adequately
>>>> chosen number or undef; Use those results as point coordinates.  
>>>> Label
>>>> the point with the text you want.
>>>>
>>>> Problem: but there are a LOT of text messages there.
>>>> Solution: so what? Define a LOT of points! I said it's possible,  
>>>> not
>>>> that it's easy ;) The periodic table uses 118 such messages and I
>>>> lived through it!!! Twice, as I did it in English and in  
>>>> Portuguese!
>>>>
>>>> Problem: but I want free text. text can be anything
>>>> Solution: ok, now you got me. Tell me exactly what you're trying to
>>>> do, I'll think about it. Either I'll find a solution, or I won't.
>>>>
>>>> So, to go back to the concerns manifested here:
>>>> - I have no idea whether a different paradigm is being prepared  
>>>> by TI
>>>> (you'll have to ask them; although I wouldn't keep my hopes up if  
>>>> the
>>>> project is ongoing but on an early stage, they probably won't tell
>>>> you)
>>>> - I find it very difficult to believe that something (other than
>>>> games!!!) possible on TI-84 Plus is simply impossible on TI- 
>>>> Nspire. If
>>>> I believed that I would have never created the sort of content I've
>>>> been developing.
>>>> - In some cases, there will be a loss of performance. But that's a
>>>> different beast. Even in games, performance is the limiting factor,
>>>> not power! All that I need to create games (even real time games)
>>>> exists allready and you can "see it" in some of my constructions  
>>>> (if
>>>> you look closely enough!). The only problem here is performance  
>>>> as you
>>>> can't just add 2000 points to a G&G page and hope it runs fast.
>>>>
>>>> Believe it or not, I'm currently facing performance issues, not  
>>>> design
>>>> issues, on a electronic circuit simulator, that would have a  
>>>> function
>>>> generator and an oscilloscope with trigger and realtime display!  
>>>> And
>>>> it's not the "how can this be built" that is bothering me, that's
>>>> taken care of! It's the 1 to 2 seconds (on the computer software,  
>>>> not
>>>> the handheld!) it takes to update the oscilloscope screen everytime
>>>> something changes that's holding me back. (Oh, and there's not 1
>>>> single line of programming in a signal generator or an  
>>>> oscilloscope;
>>>> well, maybe a few small functions, but only to try to gain some
>>>> performance).
>>>>
>>>> My other ongoing projects include a battleship game (with fleet
>>>> checking, score keeping and victory detection) and, just a proof of
>>>> concept, a car game. With a track and some obstacles. But those
>>>> projects take time, and they're not a top priority for me right  
>>>> now.
>>>>
>>>> Nelson
>>>>
>>>>
>>>>
>>>> On Sun, Mar 22, 2009 at 13:24, Joe <wjb...@yahoo.com> wrote:
>>>>
>>>>> Eric, for me and probably many others, it is grave because I can  
>>>>> not
>>>>> create the kind of educational programs for the nspire that I  
>>>>> did for
>>>>> the 84 series.  I have been waiting for TI do do something about  
>>>>> this
>>>>> problem for quite a while now and I am loosing hope.  Do you see  
>>>>> any
>>>>> indication at all that TI is going to fix this problem?  Also,  
>>>>> what
>>>>> advantage is there to severely limiting programability this way?
>>>>> Shouldn't a newer product be at least as good as a previous  
>>>>> product?
>>>>> Obviously the programs presented here have to be of a type that  
>>>>> don't
>>>>> require inputs while the program is running so that point is mot.
>>>>
>>>>> On Mar 22, 12:53 am, Eagle-Man <eagle-...@duetsoftware.net> wrote:
>>>>>> Not to say that I don't also want input, but is it really that  
>>>>>> grave a
>>>>>> limitation?  For how many mathematical programs/functions do  
>>>>>> you not
>>>>>> know all the information beforehand?  How many of the programs
>>>>>> presented here would benefit or be drastically improved with the
>>>>>> ability to input in the middle of the program/function.
>>>>
>>>>>> Again, I'd love to see some input for non-mathematical reasons,  
>>>>>> but I
>>>>>> wouldn't be mad at TI for not including it.
>>>>
>>>>>> You know what I would like to see?  Classes.
>>>>
>>>>>>    --Eric F.
>>>>
>>>>>> --- Original Message ---
>>>>
>>>>>>> Nelson said: "also the input of a program on TI-Nspire must be  
>>>>>>> given
>>>>
>>>> ...
>>>>
>>>> read more »- Hide quoted text -
>>>>
>>>> - Show quoted text -
>>>>
>>>
>>
>>>
>>
>
>
>
> >
> ----~--~----~------------~-------~--~----~
> To post to this group, send email to tins...@googlegroups.com

[Nelson Sousa]

Hi!

I've taken a look at the program and yes, it can be done.
And it will look nice!

It's somewhat complex, so give me a couple of days (reading 84 Plus
code is boring).

Nelson

[ido...@comcast.net]

While certainly not a program, the file I have attached is what I am showing my students to help them understand (hopefully) Type I error, Type II error, and Power.  It is a first run, so I only built it to work if the specific alternative is greater than mu_naught. 

 

You can change the settings on the sliders to be in a range specific to the problem you are working.  At this time, also type in the value for each variable in each slider.  This is enough to see the new calculation of Type II Error and Power.  However, if you want to see the corresponding picture, change the Window Settings on the graph so the x-values match a decent range for the mu values (naught and specific alternative) in your problem.

 

Enjoy, but please offer suggestions to make it better...

 

Kara

 

 

----- Original Message -----
From: "Nelson Sousa" <nso...@gmail.com>
To: tins...@googlegroups.com

[arjun gill]

Not sure if anyone is still looking for a program for type 2 error, but I made one for the TI Nspire. If you want the files just reply with your emails and I'll send them, or I coud post them to this group if you want.