Can Linux Distros get rid of bugs on some day? [Archive]

brunovecchi

November 15th, 2008, 03:31 PM

As already said, no unless you want to brin software *and* hardware development to a halt.
There are examples of software that is so mature that is practically bug-free, such as the LaTeX package. But for an entire distro, I guess that's more difficult.

Zootropo

November 15th, 2008, 03:40 PM

There's not such thing as an operating system without bugs.

You can make mistakes with a "Hello world" type of program, so...

eternalnewbee

November 15th, 2008, 03:51 PM

OK. Can I change my mind? (just for the sake of the argument:-#

Sealbhach

November 15th, 2008, 03:56 PM

If everybody stopped making new software, and everybody agreed to use only one distribution of software, it might be possible. But it would not be much fun....:(

.

Skripka

November 15th, 2008, 04:03 PM

If everybody stopped making new software, and everybody agreed to use only one distribution of software, it might be possible. But it would not be much fun....:(

.

....Then "bugs" would instead be called "features" like WGA, UAC, and the gawd-awful Talking Paper-Clip in Office

eternalnewbee

November 15th, 2008, 04:05 PM

If everybody stopped making new software, and everybody agreed to use only one distribution of software, it might be possible. But it would not be much fun..
Fun is in the eyes of the beholder

etdsbastar

November 18th, 2008, 06:24 PM

I think this can be possible one day....

I can tell you the reason why? but please try to find it yourselves for a couple of days...

aeacides

November 18th, 2008, 08:09 PM

Bugs are like secondary objectives in a game.
You learn from bugs you made or from other's mistakes.

But rather than see bugs disappear, it would be cool to see technical difficulties minimized from proprietary drivers ;- p.

aaaantoine

November 18th, 2008, 08:24 PM

The packages that do not need any further changes, the "perfect" packages, would eventually become bug free.

Take the GNU coreutils for example. How often does a command like "cd" get changed?

kernelhaxor

November 18th, 2008, 08:25 PM

Your question is like 'Can the world be perfect (everyone person in the world be good)?' .. Thats just not possible ..
however bugs keep getting discovered and then fixed ..

Moreover, u cannot know whether a software is bug-free? because lets say u did testing for years and years and fixed every bug found but u still cannot call it bug-free because what if the next day a new bug is discovered .. so we never can say a software is bug-free .. there is no way to know

Bölvaður

November 18th, 2008, 08:47 PM

Lets see. If everyone working on it stopped innovating and developing, all the bugs might get fixed before it became obsolete and people stopped using it.

If you stop making anything new, no new bugs will arise, therefore there will always be bugs in linux while other systems might stop having bugs (funny though how there are less bugs in linux compared to quantity of new lines of code in other systems)

cardinals_fan

November 19th, 2008, 01:54 AM

I have never run into a distro-specific bug on Slackware. Ever.

init1

November 19th, 2008, 03:23 AM

Lets see. If everyone working on it stopped innovating and developing, all the bugs might get fixed before it became obsolete and people stopped using it.
That's basically what Debian stable is. It's nice though, since it's very stable and predictable.

etdsbastar

November 20th, 2008, 06:33 PM

:popcorn: Ya, That was too my decision.... It can be a 50-50 position.

etdsbastar

November 30th, 2008, 05:15 PM

That's basically what Debian stable is. It's nice though, since it's very stable and predictable.

are there any other linux distros with stable compatibility.

etdsbastar

December 13th, 2008, 07:29 PM

are there any other linux distros with stable compatibility.

bump

Changturkey

December 13th, 2008, 08:08 PM

bump

Debian, Slackware, uhhhh....

cardinals_fan

December 13th, 2008, 08:26 PM

bump
Again, it depends on what "getting rid of bugs" means to you. I believe that Slackware already has gotten rid of all distro-specific bugs. These occur from faulty packaging and it's a distro's job to purge them. As for upstream bugs, it's not a distro's responsibility. With that said, Slackware and Debian both choose software based on stability.

JohnFH

December 13th, 2008, 08:33 PM

I think this can be possible one day....

I can tell you the reason why? but please try to find it yourselves for a couple of days...

Come on then, tell us, O Smart One. I'd be very interested in your answer.

EdThaSlayer

December 14th, 2008, 05:17 AM

If the developers code an A.I that can do the programming for them, test and debug(trillions of times, ofcourse-only limited by the speed of the hardware), and then share the code for them to double check then possibly all the bugs could be eliminated.

Bachstelze

December 14th, 2008, 05:18 AM

I would surely hope not! See signature.

etdsbastar

January 15th, 2009, 08:03 AM

I would surely hope not! See signature.

How can u be so sure...

samjh

January 15th, 2009, 09:10 AM

It would be extremly difficult and take many years to achieve. Just take a look at Debian's release cycle to see how long it takes to produce a "stable" distro. Even Debian stable releases have many bugs.

Theoretically it is possible. Practically it isn't. There is too much code, too many programs designed and coded by unrelated developers, mashed together into a distro. Furthermore, much of the code is written ad hoc - with no thorough design or analysis of security or reliability. If you fix one bug, you may end up creating a regression on another part of the code, inadvertently creating another bug.

The only practical way to develop bug-free software is to do it from scratch, with a "perfect" design (which is probably impossible), using a programming language which is developed specifically for safe programming (eg. SPARK Ada), and have developers adhere to a very disciplined regime of incremental testing and documentation. Even then, the software will have some bugs, although they will probably be minor ones.

adamlau

January 15th, 2009, 09:44 AM

Never. Too many package to track, too many dependencies.

etdsbastar

January 24th, 2009, 06:42 PM

Never. Too many package to track, too many dependencies.

Lets give you a simple answer on this : Lets prove 1=2 first (You think that it is impossible, let me show you ...):

Let us first take equation Ist as,

a = b = 1 ... (1)

Now from Eq. (1) extracting,

a = b ... (2)

Multiplying both sides by 'a' in Eq. (2),

a2 = ab ... Here (a2) means a to the power 2 (a square)

Now substract both sides by b2 (b square):

a2 - b2 = ab - b2

Putting the formula in the above equation and extracting b from RHS:

(a+b)(a-b) = b(a-b)

(a-b) is common in LHS and RHS hence omited:

(a+b) = b

But, from equation I, the value of a and b is 1, keeping those values:

1+1 = 1

hence,

2 = 1

therefore 1 = 2

NOW TELL ME IS ANYTHING IMPOSSIBLE IN THIS WORLD...

Just go through the above problem throughly...

Eisenwinter

January 24th, 2009, 07:36 PM

Lets give you a simple answer on this : Lets prove 1=2 first (You think that it is impossible, let me show you ...):

...

therefore 1 = 2

NOW TELL ME IS ANYTHING IMPOSSIBLE IN THIS WORLD...

Just go through the above problem throughly...
Dude... what the hell?

smartboyathome

January 24th, 2009, 09:27 PM

Lets give you a simple answer on this : Lets prove 1=2 first (You think that it is impossible, let me show you ...):

Let us first take equation Ist as,

a = b = 1 ... (1)

Now from Eq. (1) extracting,

a = b ... (2)

Multiplying both sides by 'a' in Eq. (2),

a2 = ab ... Here (a2) means a to the power 2 (a square)

Now substract both sides by b2 (b square):

a2 - b2 = ab - b2

Putting the formula in the above equation and extracting b from RHS:

(a+b)(a-b) = b(a-b)

(a-b) is common in LHS and RHS hence omited:

(a+b) = b

But, from equation I, the value of a and b is 1, keeping those values:

1+1 = 1

hence,

2 = 1

therefore 1 = 2

NOW TELL ME IS ANYTHING IMPOSSIBLE IN THIS WORLD...

Just go through the above problem throughly...

That actually can't be true. You are extracting a zero, which means that you are losing that zero, and the equation becomes untrue. I remember seeing this back in (Precalculus or calc 1), and it just gets rid of the zero and makes things no longer equal.

Basically, it is because you are dividing out zero on both sides of the equation. Since mathematicians have declared you cannot under any circumstance divide by zero, what you did doesn't work.

KIAaze

January 24th, 2009, 11:42 PM

YES WE CAN!

edit: Hey, why can't I write in CAPS lock?
re-edit: hehe, tricked it. :p (worked around the "bug")

etdsbastar

January 25th, 2009, 07:55 PM

That actually can't be true. You are extracting a zero, which means that you are losing that zero, and the equation becomes untrue. I remember seeing this back in (Precalculus or calc 1), and it just gets rid of the zero and makes things no longer equal.

Basically, it is because you are dividing out zero on both sides of the equation. Since mathematicians have declared you cannot under any circumstance divide by zero, what you did doesn't work.

I can't get you.... Actually I am confused... will you please tell me clearly...

cardinals_fan

January 25th, 2009, 08:01 PM

I can't get you.... Actually I am confused... will you please tell me clearly...
You say:
Putting the formula in the above equation and extracting b from RHS:

(a+b)(a-b) = b(a-b)

(a-b) is common in LHS and RHS hence omited:
Since a and b both equal one, a-b=0. When you divide both sides by (a-b), you are dividing by zero.

etdsbastar

January 26th, 2009, 01:45 PM

You say:
Since a and b both equal one, a-b=0. When you divide both sides by (a-b), you are dividing by zero.

Dear friend,

Dividing both sides by (a-b) cancels the upper side and the lower side and the result will be one ...

Go to your previous mathematics books and be confirmed again...

I am waiting for the reply...

KIAaze

January 26th, 2009, 02:55 PM

Dear friend,

Dividing both sides by (a-b) cancels the upper side and the lower side and the result will be one ...

Go to your previous mathematics books and be confirmed again...

I am waiting for the reply...

No, he's right.
You go back to your previous mathematics books (or read more "advanced" ones).

This is a very well known problem and everybody should know it. It's basic maths and if I recall correctly, it's taught around the same time as a^2-b^2=(a+b)(a-b). (or you just learn that division by zero is forbidden)

Here:
http://en.wikipedia.org/wiki/Division_by_zero
http://en.wikipedia.org/wiki/Division_by_zero#Fallacies_based_on_division_by_ze ro

Keyper7

January 26th, 2009, 03:01 PM

Dear friend,

Dividing both sides by (a-b) cancels the upper side and the lower side and the result will be one ...

Go to your previous mathematics books and be confirmed again...

I am waiting for the reply...

Wow. It's one thing to make such basic mathematical mistake. But to be this arrogant while doing so, geez... Trolling at its pure form.

Now, to go back on topic: one thing that has not been mentioned here is how hardware also changes. I believe a completely stable, rock-solid distro wouldn't be obsolete just in terms of features, but also in terms of hardware support. After all, hardware changes usually mean at least a little bit of code base changes and possible introduction of new bugs.

etdsbastar

January 26th, 2009, 03:03 PM

No, he's right.
You go back to your previous mathematics books (or read more "advanced" ones).

This is a very well known problem and everybody should know it. It's basic maths and if I recall correctly, it's taught around the same time as a^2-b^2=(a+b)(a-b). (or you just learn that division by zero is forbidden)

Here:
http://en.wikipedia.org/wiki/Division_by_zero
http://en.wikipedia.org/wiki/Division_by_zero#Fallacies_based_on_division_by_ze ro

Ok ... I was wrong and I got the answer.... But, do you know it was not invented by mine but I got it somewhere in Mathematics Tweaking books...

Do you have something like this ... If yes then please tell me friend... Ok...

and Thanks for your replies and suggestions...

wersdaluv

January 26th, 2009, 03:22 PM

https://launchpad.net/bugs/1

KIAaze

January 26th, 2009, 06:19 PM

Wow. It's one thing to make such basic mathematical mistake. But to be this arrogant while doing so, geez... Trolling at its pure form.

Now, to go back on topic: one thing that has not been mentioned here is how hardware also changes. I believe a completely stable, rock-solid distro wouldn't be obsolete just in terms of features, but also in terms of hardware support. After all, hardware changes usually mean at least a little bit of code base changes and possible introduction of new bugs.
Mmh, yes, sorry. I reused his formulation without consideration. :oops:
edit: Oh, you were referring to what he said. ^^'

jespdj

January 27th, 2009, 03:48 PM

Can linux distros get rid of bugs some day? Please tell your answer in detail.
What is this supposed to mean?

Bugs are being fixed all the time in the Linux kernel and all the other software that comes with Ubuntu.

It will never be 100% bug free, because that's simply impossible for such a large software ecosystem.

There is no operating system that is bug free. Windows and Mac OS X also contain bugs, that's why Microsoft and Apple are also constantly providing updates, just like Ubuntu.

tjwoosta

January 27th, 2009, 04:49 PM

bugs are a necessary part of developement

end of story!

etdsbastar

February 20th, 2009, 03:38 AM

bump

aaaantoine

February 20th, 2009, 04:59 PM

Ask the folks who put together Debian 5.0.

Also, here's a C program without any bugs:

int main(){return 0;}

The compiler didn't like void main(){}.

Simian Man

February 20th, 2009, 05:20 PM

Ask the folks who put together Debian 5.0.

Also, here's a C program without any bugs:

int main(){return 0;}

Assuming the C compiler has no bugs :).

The compiler didn't like void main(){}.
That's because it isn't standard C.

etdsbastar

February 23rd, 2009, 12:36 PM

Assuming the C compiler has no bugs :).

That's because it isn't standard C.

what does this mean?

KIAaze

February 23rd, 2009, 10:58 PM

The compiler translates the C source code into machine language.
If the compiler makes errors during the translation, there may be bugs in the resulting executable.

cf: http://en.wikipedia.org/wiki/Compiler

As for the second part, I had to google:
http://users.aber.ac.uk/auj/voidmain.shtml
http://homepages.tesco.net/J.deBoynePollard/FGA/legality-of-void-main.html
http://en.wikipedia.org/wiki/ANSI_C

I haven't found the "official" proof in the ANSI documentation yet, but if somebody knows where it is, let me know. :)

edit: Found it in case anybody is interested:

5.1.2.2.1 Program startup
1 The function called at program startup is named main. The implementation declares no
prototype for this function. It shall be defined with a return type of int and with no
parameters:
int main(void) { /* ... */ }
or with two parameters (referred to here as argc and argv, though any names may be
used, as they are local to the function in which they are declared):
int main(int argc, char *argv[]) { /* ... */ }
or equivalent;9) or in some other implementation-defined manner.

Source: C99 + TC1 + TC2, WG14 N1124, dated 2005-05-06: http://www.open-std.org/JTC1/SC22/WG14/www/docs/n1124.pdf
Available at: http://www.open-std.org/JTC1/SC22/WG14/www/standards.html#9899

I even downloaded the "Technical Corrigendum 3" from the ISO site!

Yay, I just gained one geek level! xD

etdsbastar

March 26th, 2009, 07:44 PM

bump

gjoellee

March 26th, 2009, 08:31 PM

No, it is impossible

etdsbastar

May 30th, 2010, 07:59 PM

can u tell me why coz canonical is improving day by day and bugs are being removed in each version, nothing in this world is stable, if any other reasons then please tell why?

etdsbastar

June 17th, 2010, 04:08 PM

bump...

RiceMonster

June 17th, 2010, 04:17 PM

Please stop bumping threads you made 2 years ago.

aaaantoine

June 17th, 2010, 04:47 PM

can u tell me why coz canonical is improving day by day and bugs are being removed in each version, nothing in this world is stable, if any other reasons then please tell why?

I'll continue to humor you.

While some bugs are being removed, more bugs are being introduced.

Take a look at Launchpad (https://bugs.launchpad.net/ubuntu/+bugs). Are there more bugs today than when you started this thread two years ago, or fewer?

It's been about two years since I looked at Launchpad myself. We were at ~36,000 bug reports at the time. Now the list has exploded to 76,792. Highly improbable that we'll ever fix them all.

True that the number of reported bugs is not necessarily indicative of the number of actual bugs in Ubuntu and its supported software, and I assure you there are many duplicates in that sample size. But the point remains, and aside from this post I believe your question has already been thoroughly answered.

23dornot23d

June 17th, 2010, 05:18 PM

A 100 % bug free system ...... with the complexity that has now been built into every computer and each individual chip and all the code .........

I would find this very hard to believe ..... that somewhere there would be a way to check each item 100 % ...... so a bug ..... as in the first computer bug (http://en.wikipedia.org/wiki/File:H96566k.jpg) recorded ......

A bug can be caused by so many things in a computer .....

Even with a self check done of every item at the start up and a detailed listing given and the system set to stop on every small error ........

The chance of probability would probably say that even with every conceivable thing possible thought of (there is the flaw) - the checks would not be able to include every known possibility of how each item in a computer will interact with each other item.

If it could be proved that one person could be 100 % perfect ...... and that person thought of every conceivable problem ....... and built one computer .......

Then it could be possible ....... anybody know of this person .......

I am sure there may be many like this in our World .......

But are they doing every task that makes a computer work ........

So my feelings on this ....... not possible ..... but it does not hurt trying to get as
close as possible to a bug free system ...... maybe 99.9999:% one day .....

and the music played on ....... :guitar:......... ):P

MasterNetra

June 17th, 2010, 06:20 PM

+1 to no such thing as impossible.

It can be done, but for as long as there are new features or changes to the code, the odds are more bugs will arise. But its can be done.

etdsbastar

June 21st, 2010, 08:06 PM

+1 to no such thing as impossible.

It can be done, but for as long as there are new features or changes to the code, the odds are more bugs will arise. But its can be done.

Great, I can really say that ubuntu is having less bugs than windows machines... specially causing due to viruses.... We are free of viruses till now... since still no 100% active virus for linux systems (ubuntu)