Can software replace a programmer?

[zidane1strife]

"In ten years we will laugh about the fact that we still had to write code at the beginning of the 21st century. It is about time to automate the software industry,"

http://www.tomshardware.com/hardnews/20040607_140114.html

What will you do, the day after tomorrow?

 

[London Geezer]

What will you do, the day after tomorrow?

Cover up or freeze to death.

No really, lots of jobs would go down the drain, but thankfully this won't happen too soon, and i believe there will always be the need to "tweak" the code here and there, in the end nothing created by man will ever be perfect, therefore i don't expect automatised code will also be perfect. Errors are like viruses. There will always be a little "thing" going around and around, until it has the opportunity to "come up" and mess things up. Look at Windows. And any program ever created.

 

[cristic]

I don't think we will see that in ten years from now... not until SkyNet is up and running!

 

[Simon F]

Can software replace a programmer?

Maybe, maybe not, but I'm sure it could replace a fanb' or two in the console forum.

 

[K.I.L.E.R]

Programmers write software for specific set of hardware specs.
IE: DirectX coding requires a Direct X capable card that supports feature A, B, etc...

Does this software write for the hardware specs? Or is it just just a very detailed code generator?

The more simpler programs are to write the more complex our software becomes and the more possibilities of bugs and hardware/software incompatibilities.

I don't think any machine could replace a coder.

 

[Hanners]

IF programmer=human THEN killwithlaser=true

 

[nutball]

Can software replace a programmer?

Maybe, maybe not, but I'm sure it could replace a fanb' or two in the console forum.

while (1)  {
 rant();
}

 

[K.I.L.E.R]

Nutball, that's a fucking aweful coding practice.

Replace '1' with 'true'.

 

[Simon F]

KILER, it is perfectly normal coding practice. There is no ambiguity.

Your suggestion, OTOH, gives:

main.c:...: `true' undeclared (first use this function)

 

[bloodbob]

If we program code creators so smart they can create better code creater be prepaired for judgement day!

 

[evil]

Nutball, that's a fucking aweful coding practice.

Replace '1' with 'true'.

And then it's still wrong!

loop:
rant();
goto loop;


A machine would know that you don't write 'while' when you're really doing a 'goto'.

/me ducks

 

[K.I.L.E.R]

KILER, it is perfectly normal coding practice. There is no ambiguity.

Your suggestion, OTOH, gives:

main.c:...: `true' undeclared (first use this function)

Not in Java or C#.

 

[London Geezer]

^^ I think the above few posts sum up pretty well why there will never be perfect code.

 

[K.I.L.E.R]

Common sense suggests you should avoid infinite loops anyway.

 

[Crisidelm]

At times common sense fails....

 

[K.I.L.E.R]

At times common sense fails....

I don't think so.
It's just that common sense is uncommon.

 

[nutball]

KILER, it is perfectly normal coding practice. There is no ambiguity.

Your suggestion, OTOH, gives:

main.c:...: `true' undeclared (first use this function)

Not in Java or C#.

What's Java?

 

[nutball]

Nutball, that's a fucking aweful coding practice.

Replace '1' with 'true'.

And then it's still wrong!

loop:
rant();
goto loop;


A machine would know that you don't write 'while' when you're really doing a 'goto'.

/me ducks

Ugh! I haven't used a GOTO in fifteen years, not even when programming in FORTRAN!

 

[WhiningKhan]

I don't think any machine could replace a coder.

A coder? Of course it can.
Any coder? Possibly.
An engineer? Not likely.

Coding is just current implementation method, transforming and evolving to something else in time. The essence of technology is in design, not in implementation.

 

[K.I.L.E.R]

We'll see if those code generators work as good as John Carmack when it comes to making graphics engines.

After all, a machine can't be better than a god.