Excuse me DemoCoder!

[K.I.L.E.R]

Output is irrelevent under some circumstances.
In scientific computing you aren't looking for outputs, you are looking at the process.
Outputs can be used to assess the performance of a certain model but there are other ways to do this and the most unique being validity convergence by Reiner's law.

I dare anyone to say this isn't the case.

If a programmer starts writing code without thinking about what he wants output first, then he is doing something wrong.

 

[RussSchultz]

Output is irrelevent under some circumstances.
In scientific computing you aren't looking for outputs, you are looking at the process.
Outputs can be used to assess the performance of a certain model but there are other ways to do this and the most unique being validity convergence by Reiner's law.

I dare anyone to say this isn't the case.

You are taking the terminology 'output' too literally.

Your validity convergence is an output. A properly functioning calculator is an output. Measurement of algorithmic complexity in real world units is an output. The point of the statement was that if you don't know what you're aiming at when you begin, your project is already in serious trouble. That's software engineering 101.

 

[K.I.L.E.R]

Validity convergence has nothing to do with output.
It has to do with calculating a process and not giving any outputs.
I think it's based on the laws of quantum mechanics or something.

You are taking the terminology 'output' too literally.

Your validity convergence is an output. A properly functioning calculator is an output. Measurement of algorithmic complexity in real world units is an output. The point of the statement was that if you don't know what you're aiming at when you begin, your project is already in serious trouble. That's software engineering 101.

 

[RussSchultz]

Validity convergence has nothing to do with output.
It has to do with calculating a process and not giving any outputs.
I think it's based on the laws of quantum mechanics or something.

Again, you're taking the terminology 'output' too literally.

 

[K.I.L.E.R]

There are things that have absolutely no outputs in the world.
What do you call an output?

Again, you're taking the terminology 'output' too literally.

 

[Captain Chickenpants]

There are things that have absolutely no outputs in the world.

Nothing that you need to compute.

 

[K.I.L.E.R]

That's beside the point.

Nothing that you need to compute.

 

[WhiningKhan]

Never mind, I can't read.

 

[RussSchultz]

There are things that have absolutely no outputs in the world.

Then applying computation to those things seem particularly pointless, doesn't it?

What do you call an output?

Some information you want out of it, and/or some desired functionality.

Game behavior, is an 'output'.
Statistical analysis is an 'output'.
Efficiency measurements is an 'output'.

The statement you originally quoted is more precisely worded: "If a programmer starts writing code without a specific goal in mind, then he is doing something wrong."

Nothing extends schedules more than not knowing what or why you're making something.

 

[Blazkowicz]

There are things that have absolutely no outputs in the world.

the only such thing I can think of is /dev/null

 

[AlNom]

output = result

or... a purpose of a process
...maybe.

 

[K.I.L.E.R]

That makes sense.
Assuming that's actually considered an output.

Some information you want out of it, and/or some desired functionality.

 

[Captain Chickenpants]

That's beside the point.

Nope.
If you have no expected output then there is no point in doing the computation. A program that has no outputs is utterly pointless.
Even if you are examining some process the purpose of the process is to transform some input and generate an output.

If you disagree I suggest that you present an example where this isn't the case.

CC

 

[K.I.L.E.R]

System->in("Hello world"):
{
do| as::executed |
}

It takes an input and does nothing else.

Nope.
If you have no expected output then there is no point in doing the computation. A program that has no outputs is utterly pointless.
Even if you are examining some process the purpose of the process is to transform some input and generate an output.

If you disagree I suggest that you present an example where this isn't the case.

CC

 

[Gubbi]

It takes an input and does nothing else.

And is utterly pointless.

Cheers

 

[K.I.L.E.R]

The point I am trying to make isn't whether it is useless or not but that it's possible to create a program that processes random data and does nothing with it.

And is utterly pointless.

Cheers

 

[RussSchultz]

Some of us use our powers for good. Others for evil.

 

[K.I.L.E.R]

Huh?

Some of us use our powers for good. Others for evil.

 

[AlNom]

Sounded like a Star Wars quotation.

 

[DemoCoder]

Im traveling in south east asia/indochina right now, so my time is limited, but Russ explained it sufficiently. Don't start wacking away at building something before you figure out what it is you wish to build.

If you're doing something for research, fine, but your program must still produce some result (even if the result is as simple as 1-bit or halt/no-halt), otherwise, you won't know the result, and thus the exercise is pointless as research since you gained zero knowledge.

Output does not mean returning an 'answer'