All applications will have bugs

>All applications will have bugs

Do we have proof that it's impossible to write a bug-free application or is it just human nature that's in the way?

Other urls found in this thread:

ece.cmu.edu/~ganger/712.fall02/papers/p761-thompson.pdf
np.reddit.com/r/OutOfTheLoop/comments/4bwshx/why_are_people_so_mean_to_neil_degrasse_tyson_on/d1daa05/?context=4
en.wikipedia.org/wiki/Halting_problem
twitter.com/SFWRedditGifs

>proof

>He doesn't know what a proof is
Found the code monkey

>WOOD U LIKE SURBAY? :DDDddd

Seems like a stupid thing to ask, hello world doesn't have any bugs.

The compiled binary could still have bags due to bad compilers.

I'll put a bag over ur mum's head m8

it IS possible, but it's way too hard to bother. we just fix the major bugs.

>hello world doesn't have any bugs
>proof by contradiction

Wow, I'm recommending you for the next Turing award.

>it's impossible to write a bug-free application
That's a matter of definition

ece.cmu.edu/~ganger/712.fall02/papers/p761-thompson.pdf

Found the Engineer.

An application that does exactly as intended? Surely we have conventions and protocols to describe our intentions accurately?

>tfw no test suite before pushing to production

...

>He thinks he can catch all issues in unit tests

OP asked for at least ONE bug-free application. You are saying a compiler COULD be poorly written, which means that at least one decently written compiler will do the job.
So yes, basically this thread is complete bait and op should get fucked by an octopus for not posting in /sqt/

This is an unsolvable NP problem.

OP asked for proof, which doesn't exist even for your hello world application. Enjoy your Meme Engineering degree.

Even if we had unhackeable systems, the 8th layer is still vulnerable to attacks like rubber hose cryptography or formal-undeinable brute force attacks.

guy who who took screenshot, what OS are you using?

neil is the biggest fucking joke, even most of reddit hates him now.

It's hard to even prove a binary corresponds with the source code.

sure it is, just compile it again and compare the two binaries

Go back to your slum, Pajeet.

wait, really? [citation needed]

oh wow, remember when the US created a shitstorm to help Sony hide the fact they treated employees like shit

>just create unhackable systems lmao
American "scientists".

>compiler COULD be poorly written, which means that at least one decently written compiler will do the job

The bug-free status of the compiler is also relevant, moron. If you're gonna assume the hypothetical bug-free status of the compiler, you might as well do the same with the code.

Public opinion of him on reddit shifted when one user told a tale of paying him $85,000 to lecture at their university and him being a huge asshole.

>Been waiting for some flak directed at him for this for years. My student club spent a year organizing and lobbying him to come to our college (including raising his 50k speaking fee) and I spent the whole day with him. He was insufferable and a bully.

np.reddit.com/r/OutOfTheLoop/comments/4bwshx/why_are_people_so_mean_to_neil_degrasse_tyson_on/d1daa05/?context=4

Fuck this ignorant cocksucker

That won't work if you use a different compiler or different compiler settings and you still can't prove that the source code looked exactly like that (for example if things got optimized by the compiler).

>He unironically believes in deterministic binaries

Looks like OS X font rendering to me.

>unsolvable
>NP
(((You)))

God you two are fucking idiots.

>He thinks all NP problems have a solution in finite time

His relative fame has gone to his head and he's completely delusional now.

The way he's been carrying on for the past year or so has made his shitty self-serving sagan-science peacocking apparent even to plebs.

if the definition of "bug" is determined by human perception, yes it's impossible

if the definition of "bug" is limited to technical implementation and execution, then maybe

I think it will be bug free if we develop a universal bug detection software.

create unhackable systems? that's a pretty good idea, i wonder why nobody has thought of this before

He is still very relevant in the scientific community. He works with NASA to help them discover new planets. He's also the director of a museum. The man is a workaholic because he loves science.

Yo holmes you just activated my trap card

How do we determine if the universal bug detection software is bug free

I think we should make a list of bugs and check if it has them. Once we do it right just once, we can fix all bugs in other applications.

chances are he's hiking the price to avoid wasting time talking to college kids in the middle of nowhere

also, like many public intellectuals that come from STEM fields (like Dawkins), they think Humanities are superfluous, the guy from the comment thread you linked should have expected for him to shit on philosophy majors

Tyson is just a discount Sagan, no one should expect much of him

By making another bug detection software for the bug detection software.

For someone who's so active and prolific he's highly irresponsible. The way he represents science makes it seem myopic and closed minded.

Defends his shitty opinions even against the criticism of experts in fields not his own.

His is typical showman bullshit. Highly visible with little tangible value.

I asked because of the font rendering, looks very nice indeed

Alright you wore me down, I'm tired now. Go read Godel, Escher Bach and leave me alone.

by investing on a marketing campaign bullshing people into believing it, than selling it

Apple did it with "apple doesnt have malware"

>a scientist shitting on a philosopher
where did we go wrong?

He is one of the most smartest people on the planet, moron. Ofcourse he's gonna defend his opinions when they're correct.

Some braindead moron on Facebook who subscribes to Natural News isn't an expert and their opinions doesn't matter against an Astrophysicist. Neil is on the cutting edge of science unlike you morons who won't amount to anything in life.

for increasing complexity the chance of being completely free of bugs approaches zero. that's all there is to it.

>He works with NASA to help them discover new planets.
Doing what? And what papers or even actual scientific work has he done in the last 10-15 years? He's reaching Kaku-ness levels of spokesman with the same amount of bill bye self-delusion.

Pop science and science journalism.
We tried to make science cool and accessible and we just ended up empowering cunty self-centered science types.

Nice bait

Don't feed the gnolls

>He is one of the most smartest people on the planet, moron. Ofcourse he's gonna defend his opinions when they're correct
According to bloody who? He's output has been overshadowed by a bloody musician working on his free time, and he has made literally 0 contributions of merit other than letting people stick their hands up his ass to use as a spokeperson.

Well, currently it's very hard to make software unhackable but you can make quality software that is almost unhackable. Look at Truecrypt. It was "unhackable" for a while but later it was hacked. That's sorta proof. That shit was pretty fucking strong.

Now everyone knows, how will we corner the market on bug free software if everyone knows about this top secret algorithm. You fool.

You can only prove that a piece of code does what you designed it to do. You can't prove that it won't do extra shit.

Lmao'ing at people in this thread, you fuckfaces are so fucking clueless it's hilarious.
> it's impossible to write a bug-free application
We have no proof that such a thing is possible but we KNOW that it's impossible to write a
program which can check if a program is bug-free, it's an indecidable problem.
hints for smart people : Universal Turing machine, reduction, decidability

Any CS graduates MUST be able to solve that, otherwise you went to a shitty school

>inb4 muh test framework
This is the poor man solution to a complexe problem, sadly this is the only thing we have now.

You should learn about lambda-calcul (ie functional languages mostly) if you want to see bug-free programs as there is no side effect and the underlying mathematical structures are known for decades.

>Do we have proof that it's impossible to write a bug-free application or is it just human nature that's in the way?
yes
en.wikipedia.org/wiki/Halting_problem

Shitty analogy time.
Yeah a strong-enough bridge is great, because we don't have an indestructible material to build all bridges out of. This is why engineering exists to begin with, to find the intersect of good-enough and feasible.

However we're not arguing about it being worth doing, we're arguing about whether or not it's possible to make something indestructible. Which you can't.

> muh halting problem
Go suck a dick you pretentious fuck, most of the people in this thread know what's up they're all just being assholes

The fact that you have to soap box proves you're more ignorant and cunty than the rest of us.

I pity you, go have yourself a sad wank and stop shitting up the thread

open source is literally the only way to get close "unhackableness"

the more people readthe code, bigger the chance someone will spot a vulnerability

Disagree, if I can read the source code it's easier for me to find bugs/vulnerabilities. I don't have to report them before exploiting them.

Though you are right in that certain bugs are discovered and fixed through OSS contributions that wouldn't have been sorted or even found otherwise. Wonder what the real world statistical split is on the two scenarios though.

The sel4 microkernel was formally proven to be bug free

Just write programs in lisp with formal proofs.
Fucking easy.

No, that's basically the halting problem

>god is perfect
>god instructed terry to create templeos
>therefore templeos is perfect
>even if templeos crashes it's still completely perfect because god works in mysterious ways

>this tweet is real

man this guy is insufferable

>>Do we have proof that it's impossible to write a bug-free application or is it just human nature that's in the way?

look up the halting problem

>people don't know you can mathematically prove the correctness of a program
Haha oh wow

Think about it this way: if you can log into it, it can be "hacked". Even if hacked just means phishing or holding a gun to your head or physically putting a keylogger on your computer.

Hacking is just being creative, there's no ultimate protection

IT'S BEST TO TEST

Human nature but also the fact that in most projects you don't actually write large parts of the code nowadays and just rely on various libraries. Not only can the library itself have actual bugs, but the way you use it may also introduce hard to notice and hard to reproduce bugs because you may be subtly misusing it since you have no knowledge of how it actually does what it does and its developer has no knowledge of exactly what use case you have.

It's hard to know exactly what happens even when just using C and the standard lib, when you get to shit with VMs, various runtime environments, interpreters or various "application servers" you have way more shit that executes code you haven't written and most likely never touched. So yeah it's kind of inevitable in practice for most reasonably complex applications.

welcome to 2 years ago

> cyber hacking

I fucking hate Neil DeGrasse Tyson so fucking much.