Rube Goldberg Code

[sebster]

A couple that I saw in some code sent to me:

Why not a simple string constant 'Enable IrDA Device'?

When will the next state be 'IRDA_ENABLE_3V'?

[Hooovahh]

---

@sebster wrote:

When will the next state be 'IRDA_ENABLE_3V'? 

---

I was going to guess it would be 3V if you close the one button dialog using the red X...but I just tested it and even then it is true.  So what is the point of this output?  Data flow?

[Yamaeda]

Nice catches Seb!

Regarding the last one, maybe it was a 2-button dialog earlier?

/Y

[altenbach]

Your task is to take four boolean controls and build their state into a boolean array with four elements. Any of us could program that in 10 seconds, but others do it in different ways. (seen here)

[altenbach]

---

@altenbach wrote:

.... (seen here)

---

Another lead nugget from the same VI:

What is the purpose of building a scalar into an array and then taking the "array max"???? 😮

(Since I am clarivoyant, I can predict that taking the "array min" would probably give the same result! :D)

 

[Hooovahh]

---

altenbach wrote:

(Since I am clarivoyant, I can predict that taking the "array min" would probably give the same result! :D)

 

---

This has to be a case of randomly throwing down nodes and finding ones that don't break the wires when connected.  Then keep wiring stuff together until the result is what is expected, then stop because you are done.

[Yamaeda]

---

@Hooovahh wrote:

This has to be a case of randomly throwing down nodes and finding ones that don't break the wires when connected.  Then keep wiring stuff together until the result is what is expected, then stop because you are done.

---

DNA/Evolution programming? Random mutations to the code until it's better adapted to the evironment/target! 😄

/Y

[yenknip]

---

@Yamaeda wrote:

DNA/Evolution programming? Random mutations to the code until it's better adapted to the evironment/target! 😄

/Y

---

I've used evolutionary computation with AI / machine learning. Throwing the occasional random noise into a learning algorithm has somewhat similar intentions to preventing integral windup in a PID controller

[Hooovahh]

---

@Yamaeda wrote:

DNA/Evolution programming? Random mutations to the code until it's better adapted to the evironment/target! 😄

/Y

---

I proposed a game of random scripting when I was first learned about scripting.  Start with two inputs, something simple like two booleans.  Then have a truth table of what you want.  Then setup a routine that would drop random functions from the palette down which is of the correct data type, until there is a run-able routine.  Then have it run the VI looking for the expected result.  If the result isn't correct, add more random functions until there isn't a broken arrow and then run again.

Theoretically this could produce unexpected ways to get the same result.  For more complex functions this could result in a more efficient way of programming a subroutine than the developer...of course the actual output would probably be a fun experiement in a Rube Goldberg using way more functions than it should need.

[parthabe]

Today's delight! 🙂

Stern Warning! When there is no error from the H/W, you should be extra careful with the way you code. 😉