Comparison of Code Quality

[JMJ_coder]

O.K. - this will probably be subjective as all else - but, which *BSD (FreeBSD vs. OpenBSD vs. NetBSD vs. etc.) has the best quality of code in their source base.

For me in this thread, I define code quality as

1) free of bugs and errors
2) no security holes
3) efficiency of design (i.e., CPU and memory usage, etc.)

[BSDfan666]

I can't decide if this is a troll or not.. #1 and #2 are just ridiculous.. #3 will always be speculative.

[TerryP]

Troll no, but probably young.


For my point of view: code should be robust, understandable, and maintainable -> that's what I consider code quality. One of the best bits of advice I ever read amounts to: all large programs will have many bugs, the question is how hard they are to kill [on account of your system].

[vermaden]

Originally Posted by JMJ_coder

O.K. - this will probably be subjective as all else - but, which *BSD (FreeBSD vs. OpenBSD vs. NetBSD vs. etc.) has the best quality of code in their source base.

For me in this thread, I define code quality as

1) free of bugs and errors
2) no security holes
3) efficiency of design (i.e., CPU and memory usage, etc.)

1) NetBSD
2) OpenBSD
3) FreeBSD

Like that

BSDs on the left while Linux on the right side:

[anemos]

Always code as if the guy who ends up maintaining your code will be a violent psychopath who knows where you live

There is a book which is certainly in my todo (|toread) list.
The author is a FreeBSD commiter.

[graudeejs]

Read Unix philosophy

[s2scott]

To be sure,

#1 = netBSD.

I'm conflicted wrt awarding positions 2 vs. 3.

Using the suggested criteria, then

#2 = openBSD
#3 = freeBSD.

I'd like to award points for native-to-their-project's innovation, and the intrinsic risks to the code-quality-criteria in doing so, then,

#2a = dragonflyBSD,
#2b = freeBSD,
#3 = openBSD.

/S

[JMJ_coder]

Originally Posted by BSDfan666

I can't decide if this is a troll or not.. #1 and #2 are just ridiculous.. #3 will always be speculative.

I warned from the start that this would most likely be subjective as all - but I like to get other people's opinions on such topics - I'm quirky like that.

[JMJ_coder]

Originally Posted by anemos

There is a book which is certainly in my todo (|toread) list.
The author is a FreeBSD commiter.

It's on my bookshelf - but I can't read every book I have instantly. It takes a while for books to be processed in my reading queue; especially when work and school takes priority on such reading assignments during the semester.

[JMJ_coder]

Originally Posted by s2scott

To be sure,

#1 = netBSD.

I'm conflicted wrt awarding positions 2 vs. 3.

Using the suggested criteria, then

#2 = openBSD
#3 = freeBSD.

I'd like to award points for native-to-their-project's innovation, and the intrinsic risks to the code-quality-criteria in doing so, then,

#2a = dragonflyBSD,
#2b = freeBSD,
#3 = openBSD.

/S

Thanks. I always had the thought (completely subjective) that NetBSD was the best BSD - codewise. I certainly find it better for me than FreeBSD. I never was able to test it against OpenBSD, because when I was testing them a couple years back, OpenBSD did provide ISO's - something I see they now provide (yipee!). So I am about to install OpenBSD and give it a whirl.

[guerrier-cachalot]

opensolaris ;-)

[DrJ]

OpenSolaris is not a *BSD.

[TerryP]

if memory serves [classic] SunOS was BSD based, [traditional] Solaris is SVR4 based. While BSD began on a AT&T UNIX base, and System V always had a little bit of code lifted from the Berkeley distribution making its way into the family tree.


Occasional pissing contests and legal issues about pedigrees aside, of course.

[DrJ]

That's basically right. Some Berkeley code dribbled into SysVR4, but it is an AT&T product and codebase.

OpenSolaris is a modern SVR4, so it is no *BSD, which was my point.