Why FreeBSD Is Liking LLDB For Debugging

[Akka]

Depending on the used libraries, yes.

While we're speaking of POSIX: Why are some Linux distributions (like Gentoo) not POSIX-compatible? Why doesn't the Linux community care about standards? BSD does.



FreeBSD runs Linux applications, Linux doesn't run FreeBSD applications. One more "pro" for FreeBSD then.

Freebsd compatibility hardly give anything to a linux dist....

[mrugiero]

You must be the most ignorant bsd fanboy ever...

Have a look here for a starter: http://en.wikipedia.org/wiki/List_of...sed_on_FreeBSD

How many of the commercial products in this list have their code/modifications available for review?

Another famous example is the PS4, running modified FreeBSD. Where can we review this modifications?

So there you have your user freedom restricted quite clearly.

Even though I prefer Linux, you are wrong here. It isn't FreeBSD restricting your freedom. If you use FreeBSD, you will find the source code. It is the company's fork which is closed source and restricts your freedom.
I had some things to argue to both sides of the argument, but I think I'll stay out of the flamewar.

Aaaah the Linux fanbois still hallucinating that the reason Linux is at 1% is because Microsoft is a monopoly. Did it ever occur to you that maybe Microsoft got where it is because for most people it fits their needs better than any Linux distro?

Yes. Still, most people's needs are pretty basic, so the first roughly usable thing they have on the machine will fit their needs, including a Linux distro or FreeBSD, as long as it comes pre-installed (most people don't know how to install Windows, anyway, and it always was just clicking next a few times, at the most).

As an aside-- market share is really not as important as some people think. It is another cherry picked statistic. You can 'prove' just about anything by selecting the right statistics. For example, the latest release of Ubuntu has a smaller market share than Windows 98. Therefore Ubuntu's latest release is worse than Windows 98. The fact of the matter is, you cannot quantise how 'good' an operating system is. You can't say, for example, Windows has 10 goods, Ubuntu has 13, and FreeBSD has 20. Everything is relative, and depends on what context it is being used in.

Market share is usually not used to try to determine which OS is "good" or "bad", but it is used for companies to decide whether it is viable to support a platform or it is a waste of money. It is also relevant probability-wise to get your bugs fixed and your feature requests approved: the bigger the market share, the bigger the chance someone can reproduce your bug and someone with programming skills use your platform (then, you should get him/her to want the same feature as you, but hey, if your proposal isn't insane or hugely specific, chances are he/she will be interested).

[nightmarex]

... and while the Linsux kids keep stumblin' upon broken workflows, FreeBSD kicks ass.

Lol, all I could think of is this...

" Jets suck, Nicks suck Yankees suck.... Krypton sucks."

Eh, who needs debugging anyway, real developers write flawless code in text editors.

[DrYak]

Someone here claimed market share to be proving anything. Not me.

Yeah, but two times in this thread your answers were about pointing Windows market share and asking if that's a proof of quality.

I'm point out that windows has only a big market share, only if you're very specifically defining the market you're considering.
As "the OS running specifically on the home desktop/laptop and on the few corporate servers in medium-/bigger- enterprises."
Then yes, windows has big market share.
But using the same tactics you could also argue that Contiki has near market monopoly as an operating system, at least if you choose the reference market wisely (such as measurement sensor hardware, and experience on legacy 8bits hardware).

If you take a step back, and look widely at the whole range of CPU above micro-controller (i.e.: any hardware in the class that can run Windows, Linux, all BSD, etc.) windows' position seems very marginal. At both ends of the scale (from supercluster all the way down to embed), Unixes (such as Linux, BSD, etc.) are clearly holding most of the market share.
And with the current highly technical modern society, the most commonly deployed CPUs aren't desktops.
There's maybe about 1 or 2 desktops per familly in occidental society, whereas in the same home, there are probably a dozen of other gadgets powered by similar "above micro-controller"-class of CPUs, none of which run Windows.

so I'm disagreeing with your use of "Windows" and it's market share as an answer for "does more widespread means better quality ?".
Or maybe not. I disagree that Windows has any relevant global market share as of today.
Given that fact, Windows could be taken as one more data-point for the argument "does more widespread means better quality ?": it has probably big problems stepping out of its very restricted market share because *it is* actually a piece of crap.

Interestingly, Linux only runs on my smartphone. Reality can be harsh, huh?

Is you home router powered by WinCE? Are WinCE and Windows XPe running on all the other piece of technology around you?
Chance are that "only on my smartphone" means "as opposed to my desktop which runs Windows", not "as opposed to the 15 other CPU-powered gizmo which I encounter in my everyday life although I barely notice them".
Most of them will be running Linux, BSD derivatives, QNX, or other.

Again, back to my main topic in my post:
- its not faire to consider that Windows has market share of 99%, when you're specifically looking in a fery narrow sub-group of the whole class of hardware able to run such a scale of OS.

Hence the security holes.

Huh ? The various open-source clones of Unix (Linux, *BSDs, etc.) are among the most secure OS.
Yup, indeed, lots of security holes are *FOUND*. But also subsequently *fixed*. With big efforts undergoing to introduce better security technologies and processes (deterministic build for critical software, various forms of protetion, such as SElinux and AppArmor for Linux, FreeBSD being touted for all the effort going into security, etc.)

Windows gets much frequently hosed than any thing else. Some argues that this is due to bigger market share. I don't agree: As a simple exemple, take a subset of the market such as "servers".
Linux is probably having the biggest market share in this class.
If people were right that "biggest player" is the only relevant criterium to whether a OS will be targeted by malware, media would be a constant string of news about viruses spreading like wildfire among Linux servers. Yet this doesn't happen. Microsoft Windows Server + IIS is the typical stack that is regularily targeted and hit by worms such as Red Code, etc.

Worms targeting home router are rather rare even if that market is probably more than 90% of Linux embed.

Over all, Unix derivative, specially the opensource ones where lots of people can audit the code (Linux and *BSDs) have much better security track than windows.

Hence the dozens of new distributions per month, incompatible with each other.

Huh ?

That would be the same as seing that bitcoins will never catch because there's a whole zoo of alt coins. Yeah, completely not missing the point...

Most of the time when "Linux" is referred to, what is meant is one of the "big names" linux. (Red Hat, Ubunti, SuSE, etc.)
There about 4 or 5 distro which cover most of what people designate by linux.

Yes, there are lot of very small specific linux distro that pigeon-hole on a very specifc group of target users (I do use SystemRescueCD a lot, but I doubt that this is relevant for any regular user not into admin tasks), or regularily new experimental disto showing up to explore some new design and stuff.
But these aren't much relevant to general public. These are only to fill specific needs of very small subgroups of users.

In short, if your software works for the 4-5 big names, it will work for most persons out there, and it will very likely work on at least a dozen of other smaller relatives.

As an exemple: Steam was initially released by valve as a Ubuntu-only beta.
that doesn't prevent lots of enthousiats of trying actually managing to run it on other distros. Within a week or two, I have personally witnessed it being available for Arch, Gentoo, and Suse (not as a hack where you have to copy-paste commands from a dozen of blog post. simply as "select it from a given 3rd party repository").

Hence the claim of some Debian developers that Ubuntu never gives anything back.

The main complain is that for most things, Ubuntu does do *anything*. It justs repackages Debian stuff.

RedHat/Fedora and Suse, on the other hand, have developpers on they payroll that actively contribute to code for the whole ecosystem to enjoy. As en exemple, one of the first Radeon drivers was started by Suse employee. Redhat and Suse has actual developper who write code generally useful for everyone and is contributed back.
Canonical mostly only develops projects for their own use, which aren't useful for anyone else (Unity being the best example).

The situation is slightly different in the BSD world, where the licensing allow anyone to grab and run away with it. Thus company have incentive NOT to publish and contribute back what they've done, even if it is things that would massively benefit all BSD users at the same time, because that would disclose advantage that any competitor could grab for themselves.

So what you wanted to say is that only Windows has kernel panics ("bluescreens")? I see.

Until now, bluescreen and XPe desktops (and other typical quircks such as Windows' dialog asking about DST changes) is the most frequent type of failure mode I've seen on public display screens (as opposed to only once a firmware error that Linux PXE couldn't be booted).
Which would tend to indicate two things:
- That public screens are one of the few market subset where Windows XP-e still has a foot hold (my initial idea when posting it) as opposed to almost every other market.
- That the difference in stability is such that only Windows XP-e is seen crashing (wasn't my idea, but it could be possible. That the case with ATMs: there's a whole zoo of OSes running on them, even including OS/2 descendant. But photos of blue-screens are the only critical fails I've seen. Thank fully, the banks I use here around seem not to be showing any blue screen, probably not the Windows XP-e running kind).

While we're speaking of POSIX: Why are some Linux distributions (like Gentoo) not POSIX-compatible? Why doesn't the Linux community care about standards? BSD does.

I dunno. Most of the unixish software I need compiles as well on gentoo as on any other Unix derivative (other Linux distros, BSD, OS X). Either by hand or (much more frequently with Gentoo) in one of their portage overlays.
That's good enough for me, and that's good enough for a lot of people.

[Cthulhux]

two times in this thread your answers were about pointing Windows market share and asking if that's a proof of quality.

This might loosely be related to "BSD has a low market share lulz", coming from, lulz myself, Linux users. You seem to miss the point that my hint on the market share was an ironic answer to that. So much text, so little to say... (so I'd better skip the first part of your posting, just to stay on topic).

I do know market share is irrelevant. Those don't.

Is you home router powered by WinCE? Are WinCE and Windows XPe running on all the other piece of technology around you?

No, as there's more than just Linsux and Windows.

- its not faire to consider that Windows has market share of 99%, when you're specifically looking in a fery narrow sub-group of the whole class of hardware able to run such a scale of OS.

Precisely. This applies to every single operating system.

Huh ? The various open-source clones of Unix (Linux, *BSDs, etc.) are among the most secure OS.

Not sure about that, especially considering Linux to be one of the main targets for malware developers these days. (And they're good at that. Android as the most widely spread Linux derivative is one of the most insecure mobile operating systems available.)

Yup, indeed, lots of security holes are *FOUND*. But also subsequently *fixed*.

Weeks, sometimes years after *exploits have been released*, mostly. Asterisks don't *verify your statements*.

various forms of protetion, such as SElinux and AppArmor for Linux

Additional "hardening" by the NSA. Funny, isn't it? Anyway, most Linux holes are directly related to Linux being a rag rug, gluing various third-party software together. This is not the case with FreeBSD, Mac OS X or even Windows.

Windows gets much frequently hosed than any thing else.

Any sources on that?

Linux is probably having the biggest market share in this class.

Any sources on that?

If people were right that "biggest player" is the only relevant criterium to whether a OS will be targeted by malware, media would be a constant string of news about viruses spreading like wildfire among Linux servers.

Current IT news cover a wide variety of Android and Linux worms spreading for months now. How about you check your facts before making false statements just because of your beliefs?

Over all, Unix derivative, specially the opensource ones where lots of people can audit the code (Linux and *BSDs) have much better security track than windows.

Another statement unproven. I doubt you could.

Most of the time when "Linux" is referred to, what is meant is one of the "big names" linux. (Red Hat, Ubunti, SuSE, etc.)
There about 4 or 5 distro which cover most of what people designate by linux.

Wrong again. And don't forget that this is not a constant. The "top 4" are regularly rotated and replaced. Check the list in 12 months, in 24 months, ...

"The Linux world" is fragmented into the thousands, and everyone can see that. Some call it diversity, I call it a problem.

In short, if your software works for the 4-5 big names, it will work for most persons out there, and it will very likely work on at least a dozen of other smaller relatives.

... assuming the "4-5 big names" will always stay the same and version numbers are not relevant.

Most Ubuntu software I have stumbled upon has different versions for different Ubuntus. Seems like Ubuntu 13.x is not even compatible to 12.x. "Very likely", huh?

The main complain is that for most things, Ubuntu does do *anything*. It justs repackages Debian stuff.

... and fixes things and adds features and doesn't care about letting Debian know of that. They take Debian packages and claim they're theirs. "Grab and run away" or what did you call it? (Interestingly, the Ubuntu website doesn't even mention Linux more than once or twice. Canonical doesn't seem to be interested in being a part of "the community". Typical Linux arrogance.)

Until now, bluescreen and XPe desktops (and other typical quircks such as Windows' dialog asking about DST changes) is the most frequent type of failure mode I've seen on public display screens

This could be related to Windows being a big player and no one using Linux. Blaming Windows for having kernel panics in case of hardware/driver failures is a bit weird though.

That's good enough for me, and that's good enough for a lot of people.

You and "a lot of people" don't care about standards? Now what did you want to pretend?

[xeekei]

You're kindly invited to review the first postings in this thread, the license discussion has been brought up by Linux fanboys.



Simply ignoring the "BSD Sucks Dicks" - or whatever name they currently chose - Linux fan trolls doesn't validate this statement either. I can't remember any BSD user called "Linsux LOL" here. Can you?

So much about "acting like children".

I know there are Linux using trolls that bash BSD too, and they are annoying. However among BSD fans, Linux bashers seem to be the norm. I am trying to get a foot into that world, but get repulsed in every forum, especially the FreeBSD forums.

The first thing you did coming into this thread was saying things like "Linsux" and crap like that. Even though the article have nothing to do with Linux. It's about FreeBSD and LLDB. I've heard good things about LLDB, licences aside. I do not know why the FreeBSD devs won't touch the GPLv3, propably because they want FreeBSD to still be attractive to corporations that want to make modifications without releasing them. GCC is supposed to be messy code, bash that if you want. Linux however, is not made by the same devs.

Oh, and stop requesting sources from people if you won't provide them yourself. Show us these security holes, and issues with "glued together code" you're speaking of. In fact, let us see the proof that BSD is better once and for all, I'm tired of always reading that, but never actually seeing the proof.

[Sergio]

I know there are Linux using trolls that bash BSD too, and they are annoying. However among BSD fans, Linux bashers seem to be the norm. I am trying to get a foot into that world, but get repulsed in every forum, especially the FreeBSD forums.

The first thing you did coming into this thread was saying things like "Linsux" and crap like that. Even though the article have nothing to do with Linux. It's about FreeBSD and LLDB. I've heard good things about LLDB, licences aside. I do not know why the FreeBSD devs won't touch the GPLv3, propably because they want FreeBSD to still be attractive to corporations that want to make modifications without releasing them. GCC is supposed to be messy code, bash that if you want. Linux however, is not made by the same devs.

Oh, and stop requesting sources from people if you won't provide them yourself. Show us these security holes, and issues with "glued together code" you're speaking of. In fact, let us see the proof that BSD is better once and for all, I'm tired of always reading that, but never actually seeing the proof.

Hi, xeekei.
Please don't listen to the Linux/BSD bashers; try the various systems for yourself and choose only as a consequence of your experience with the system.
I like BSD very much for various reasons: their rich history going back to UC Berkeley and DARPA, their notion of freedom and (obviously) their technical merits, which includes lots of innovations over time.

I think all that arguments that say BSD is "technically" superior goes down to the following: Linux, in their urge to keep up with new hardware, and being backed by so many corporations (and interests) are most likely to introduce problems that translate into an unstable system, and perhaps various security flaws. I think this is undeniable and inherent to the incredible pace at which Linux is being developed. Perhaps this can effectively translate to FreeBSD being more stable than Linux, although I wouldn't say it is technically superior (perhaps this was true in the past).

[Cthulhux]

I know there are Linux using trolls that bash BSD too, and they are annoying. However among BSD fans, Linux bashers seem to be the norm.

Most BSD users know Linux better than Linux users know BSD, so they may be right to do so.

The first thing you did coming into this thread was saying things like "Linsux" and crap like that.

How is that relevant for the subject?

I do not know why the FreeBSD devs won't touch the GPLv3

Because the GPLv3 is still denying freedom.

Oh, and stop requesting sources from people if you won't provide them yourself.

Making false statements requires a good source. My statements are not false.

Show us these security holes

Here's some, here's more, and that's just the tip of it.

and issues with "glued together code" you're speaking of.

This one is a good example.

let us see the proof that BSD is better once and for all

There is no "better" operating system. At least, BSD is Linux done right. Even Linus would have used it if it had been free in 1991.

[ceage]

Because the GPLv3 is still denying freedom.

No it's not. It's denying certain specific freedoms while granting certain other specific freedoms, and the BSD licenses are no different in that regard.

[name99]

No, you miss the point about how that doesn't matter. Why does Apple not release their modifications to BSD openly? Because if they did, someone could just take it, make their own proprietary MacOS, do improvements to the code, and start competing with Apple with an improved MacOS where all the improvements would be closed to Apple.

Apple DOES release plenty of their mods to BSD.
CUPS is basically Apple. Everything related to mDNS is Apple. GCD is Apple. launched is Apple.

What exactly are the widely relevant to the outside world mods to BSD that you think Apple should release and which they haven't?