Future of FreeBSD

Status
Not open for further replies.
B

BSDAppentic3

Guest


#26
So curious that you said that...Linux is actually spreading in an innumerable amount of distros, while BSD stays in a few.
And the two are still functioning. The substantial difference it's how each one works.
 

drhowarddrfine

Son of Beastie

Thanks: 872
Messages: 2,724

#29
As I said somewhere, FreeBSD is not an application developer. Ask the OpenSim people why their code doesn't run on FreeBSD. FreeBSD has no influence in this.
Why anyone wants Windows code to work on BSD is beyond me. And i don't care what anyone says or thinks about the usefulness of Mono and .NET. It's origins are for the Microsoft world, not BSD/Linux/Unix.

I understand Mono's usefulness in porting over Microsoft software in some cases but most talk of its use is not that. Mono is an inefficient infection in most cases and, if you noticed, Microsoft more and more is integrating Unix style software into Windows to the point where I see them dropping Windows core and running a Unix system. We'll see.
 

vermaden

Son of Beastie

Thanks: 1,042
Messages: 2,694

#30
Careful: Isilon and EMC are really two different companies. Isilon did all the engineering and development when they were a free-standing company, before being bought by EMC. And even today, the corporate parent (Dell-)EMC does not take much direct technical influence over its Isilon division. The story with NetWorker is similar: it was a product of Legato, fully developed, before the company was bought by EMC. So Legato is not identical to EMC, even if it is owned by them.
I will start by saying that I do not want to convince You to my point of view.

Generally that is how Enterprises work, EMC bought Isilon, Data Domain, Legato (and renamed to Networker) then it was bought by DELL and now the rumors are that DELL EMC would be bought by VMware ... these are often financial optimization or to omit the taxes, HP bought DEC and Compaq, recently HPE (after split into HP and HPE) sold their software division to Micro Focus (which also owns SUSE) ... it never ends.

Going back to Isilon, EMC bought it 8 years ago, for me as EMC user it does not matter if it aquired some company or it developed it themselves, they have ALL REQUIRED INFORMATION to provide at least basic client for FreeBSD as they are using FreeBSD and even commiting to/developing the FreeBSD itself, so not providing a EMC Netowrker client for FreeBSD by EMC is a hypocrisy for me, that is how I interpret hypocrisy, you may interpret hypocrisy otherwise and I do not have problem with that.

One also has to remember that Legato NetWoker is at its core pretty old. I looked it up, and they still have full client support for Irix and Tru64, and for HP-UX on PA-RISC, which are operating systems that have not been sold new in many years. I know that Legato was a big company already 20 years ago; I used to see their building somewhere here in Silicon valley already long ago. Isilon is comparatively young, a product of the scalout-storage wave of the early 2000s. Neither company is located anywhere near EMC central (which has always been on the east coast).
Yes and no. EMC Networker 8.x and earlier codebase is old. In Networker 9.x most critical components are rewritten, some solutions migrated from file based database like solutions to PostgreSQL or SQLite databases. The very core of Networker 9.x has been rewritten, even Netowrker 8.2.3 and later uses that new core, Networker 8.2.2 and earlier uses the old 'Legato' core. Thus, after all these rewirites all Backup Policies has to be reimplemented when migrating from Networker 8.x to Netowrker 9.x, it cannot be just upgraded.
 

krisb

New Member

Thanks: 4
Messages: 8

#31
Today I came through a very wired topic on YouTube. A guy was saying that the future of BSD os is in danger as we have lesser developers compared to Linux and also the code quality is poor compared to Linux distros. He also stated that when a bug is reported in Linux it is fixed within few days or weeks but BSD has many serious bugs over than a decade. And also due to permissive BSD licence it is loosing it's popularity. Here is the website which has posted an article on this subject.

https://www.csoonline.com/article/3...dying-some-security-researchers-think-so.html

Now my question is if it is real or if becomes true in near future then what will happen to the diehard BSD users or common BSD fans like us? Even what will happen to those company who are using BSD for their business(WhatsApp/Netflix) or mostly to mac os as it has Darwin(based on FreeBSD) in it's core? I know that I/anyone can switch to Linux anytime but end of the day BSD is like home
:Dbuhahaha...
One famous quote comes to my mind: “The reports of my death are greatly exaggerated.”
 

Spartrekus

Well-Known Member

Thanks: 45
Messages: 303

#32
Today I came through a very wired topic on YouTube. A guy was saying that the future of BSD os is in danger as we have lesser developers compared to Linux and also the code quality is poor compared to Linux distros. He also stated that when a bug is reported in Linux it is fixed within few days or weeks but BSD has many serious bugs over than a decade. And also due to permissive BSD licence it is loosing it's popularity. Here is the website which has posted an article on this subject.

https://www.csoonline.com/article/3...dying-some-security-researchers-think-so.html

Now my question is if it is real or if becomes true in near future then what will happen to the diehard BSD users or common BSD fans like us? Even what will happen to those company who are using BSD for their business(WhatsApp/Netflix) or mostly to mac os as it has Darwin(based on FreeBSD) in it's core? I know that I/anyone can switch to Linux anytime but end of the day BSD is like home
FreeBSD is not in danger, at all.

Some number of excellent softwares in Linux have been derived from FreeBSD.

A must to read really: https://www.over-yonder.net/~fullermd/rants/bsd4linux/01 Nice quote, isn't it?
BSD is what you get when a bunch of Unix hackers sit down to try to port a Unix system to the PC. Linux is what you get when a bunch of PC hackers sit down and try to write a Unix system for the PC.
BSD doesn't use GNU ls or GNU libc, it uses BSD's ls and BSD's libc,
I've never in my life deployed a BSD system running just what's in base, and I probably never will.
Me not! I do, my all applications are made using only a C compiler!! all softwares on my servers are on the base, which is clean, and coming from the C compiler only. Clean and reliable. There is no junk addon, unreliable, in my system at all, but it takes 3-10 minutes to compile. If there is a error or bug, I can fix it.

Now, it's true that most Linux users install binary packages, and most BSD users install by building from source.
This leads to a lot of differences. In a very real sense, BSD systems are constantly developed; I can always update my system to the absolute latest code, irrespective of "releases". In Linux, that doesn't really have as much meaning, because the release process is very different. I think the most appropriate verb for a Linux release is "assembled". A Linux release is assembled from version A.B of this program, plus version C.D of this program, plus version E.F of this program... all together with version X.Y.Z of the Linux kernel. In BSD, however, since the pieces are all developed together, the verb "cut" makes a lot more sense; a release is "cut" at a certain time.
True and true:
BSD tends to always shy away from hackish solutions when there's even a hint of a proper solution in the wings. The theory is that it's far easier to wait for the clean answer, than to integrate the dirty answer now, for several reasons. For one thing, if you integrate the dirty answer, that reduces the incentive to implement a better one. For another, once you dirty up the architecture to integrate something it'll never get cleaned up again.
However, (-) for FreeBSD, GCC is still being standard and begin used.
 

Oko

Daemon

Thanks: 764
Messages: 1,620

#33
1/ Reasons:
The best softwares and core of Linux comes from FreeBSD.
Personally I read "the best" in your post as beauty is in the eye of the beholder. Apart of IPTables which is bastardized child of FreeBSD's IPFW there is nothing in Linux I can think of which comes from the FreeBSD. Actually there is almost nothing in FreeBSD which comes from FreeBSD in the past 10 years and that is why some (hopefully well-meaning) people are getting a bit concern about the future of this OS. All that being said I prefer to work with FreeBSD any time day or night over the Linux.

I couldn't find any facts in your post/rant. Everyone is entitled to her/his opinion but we should refrain from presenting it as facts.
 
Last edited:

shepper

Aspiring Daemon

Thanks: 240
Messages: 713

#34
Actually there is almost nothing in FreeBSD which comes from FreeBSD in the past 10 years ... .
1+

In my view, apart from the kernels, the main distinguishing feature of the BSD's is the core userland.

OpenBSD has distinguished itself by being critical of some of the userland applications and writing their own:
OpenSSH, LibreSSL, OpenNTP, OpenSMTP, PF. I am not aware of any core userland utilities that have written or are being currently maintained by FreeBSD.
 

Oko

Daemon

Thanks: 764
Messages: 1,620

#35
1+

In my view, apart from the kernels, the main distinguishing feature of the BSD's is the core userland.

OpenBSD has distinguished itself by being critical of some of the userland applications and writing their own:
OpenSSH, LibreSSL, OpenNTP, OpenSMTP, PF. I am not aware of any core userland utilities that have written or are being currently maintained by FreeBSD.
IPFW since you listed PF although PF is the default firewall of OS X, Solaris (before untimely death) so it is really different thing. Junoos (customized FreeBSD) is using IPFilter IIRC. The last thing which came out of FreeBSD camp was ULE scheduler

http://web.cs.ucdavis.edu/~roper/ecs150/ULE.pdf

Other most exciting things came out of Solaris (ZFS, DTrace). However Jail (Indigenous FreeBSD technology) and in particular VNET are not nearly as usable as Solaris zones and crossbow. I don't want to offend Bhyve developers but I think they are too late for the party. For me personally Bhyve is too little too late.

IIRC 10-15 years ago FreeBSD guys decided to stop any userland work. Until fairly recently even many of the basic UNIX filters were strait GNU utilities. This is no longer the case.
 

bryn1u

Well-Known Member

Thanks: 9
Messages: 336

#36
I recommend HardenedBSD wich is of course based on FreeBSD.
https://hardenedbsd.org/content/easy-feature-comparison
There are so many security features included many security flags in packages. Linux is a bag of exploits. One by one.

I hope that HBSD will have a great future. Functionallity from FreeBSD and great security from HardenedBSD. Maybe poeple can see this as a product which has "all" what servers need.
 

Spartrekus

Well-Known Member

Thanks: 45
Messages: 303

#37
I recommend HardenedBSD wich is of course based on FreeBSD.
https://hardenedbsd.org/content/easy-feature-comparison
There are so many security features included many security flags in packages. Linux is a bag of exploits. One by one.

I hope that HBSD will have a great future. Functionallity from FreeBSD and great security from HardenedBSD. Maybe poeple can see this as a product which has "all" what servers need.
https://hardenedbsd.org/content/easy-feature-comparison
it looks like a distro ;) indeed it is much "hardened".
 

lebarondemerde

Daemon

Thanks: 691
Messages: 1,564

#39
HardenedBSD is a FreeBSD fork aimed to add and test security features and later bring those new features back to FreeBSD. The 'bring back' part does not seem to be really happening (don't ask me why).

So, it is more like a research OS, specially if you pay attention to they RELEASE model. OPNsense is built on top of HardenedBSD.
 

bryn1u

Well-Known Member

Thanks: 9
Messages: 336

#40
HardenedBSD is a FreeBSD fork aimed to add and test security features and later bring those new features back to FreeBSD. The 'bring back' part does not seem to be really happening (don't ask me why).

So, it is more like a research OS, specially if you pay attention to they RELEASE model. OPNsense is built on top of HardenedBSD.
The problem is add those features from HardendBSD to FreeBSD. It is not that easy. That's way HardenedBSD had to split and go own way. There are many solutions like hbsd-update, secadm etc. HardenedBSD is compiled with strong flags. Ports are compiled additionaly with SafeStack and will with CFI harden. Freebsd dosen't have it. I was reading Shawn Web (HardnedBSD creater)statment about ASLR in FreeBD and he said that implementation was sooo long and sooo hard. It was really complicated. This the one of the main reason why HardenedBSD has chosen own way. And Shawn say it was probably unavoidable. So now, you know why.
 

Oko

Daemon

Thanks: 764
Messages: 1,620

#41
HardenedBSD is a FreeBSD fork aimed to add and test security features and later bring those new features back to FreeBSD. The 'bring back' part does not seem to be really happening (don't ask me why).
In my understanding fork has happened out of Shawn Web's frustration to get his ASLR into the FreeBSD proper. (Disclaimer Shawn and I had a lunch once and we didn't talk about politics at all. Whatever you read here are my own words). The same forces which prevented ASLR code in FreeBSD proper are holding the breaks on 'bring back' part. I don't know enough about FreeBSD internal politics nor I care for that matter but from outside it looks like ASLR and other security enhancements Shawn is championing serve no economic interest to the parties whose economic interests are served by FreeBSD. Sorry in this day and age FreeBSD altruism story is about as real as Santa Claus.
 

ShelLuser

Son of Beastie

Thanks: 1,415
Messages: 3,073

#42
Just be sure to always keep in mind that security isn't a product you "just" install, but an ongoing process. Using one OS over the other is never any guarantee that your setup will also be actually "safer", that's merely a ruse created by people in order to cater to their own agendas.
 

Datapanic

Well-Known Member

Thanks: 140
Messages: 260

#43
As I said somewhere, FreeBSD is not an application developer. Ask the OpenSim people why their code doesn't run on FreeBSD. FreeBSD has no influence in this.

Why anyone wants Windows code to work on BSD is beyond me. And i don't care what anyone says or thinks about the usefulness of Mono and .NET. It's origins are for the Microsoft world, not BSD/Linux/Unix.

I understand Mono's usefulness in porting over Microsoft software in some cases but most talk of its use is not that. Mono is an inefficient infection in most cases and, if you noticed, Microsoft more and more is integrating Unix style software into Windows to the point where I see them dropping Windows core and running a Unix system. We'll see.
I don't know, but I want portability for any program to run on any OS. I don't give a donkey's hind excretes about the politics behind it.

On a lighter note, I have gotten OpenSim 0.9.0.0 release Standalone and Grid to run on FreeBSD 11.1-p8 with 144 Regions and over 20,000 assets with Percona SQL 5.7 server... Emphasize, ALL ON FREEBSD. That's no small accomplishment since the latest docs on how to do it are for FreeBSD 6.something and obviously don't work any more.

To answer what the Doc said "FreeBSD is not an application developer" Not. It's whatever you can make it to be and to address the topic of this thread, it certainly has a future and for the doubters, go on over to Linux.
 

gnath

Member

Thanks: 18
Messages: 80

#44
Recently I came across the page https://vez.mrsk.me/freebsd-defaults.txt, specially port & pkg part of it. Also memory management is an important part for gurus. It is true that onion approach for security is good but lack of "how to harden freebsd" is not also good. This is my little understanding. (daemon_computing 'core team.d' is not LISTENING , joke)
 

scottro

Daemon

Thanks: 422
Messages: 1,242

#45
Distrowatch (and I always consider links that give you no clue of where they're going to be a bit clickbaity, though one can hover the mouse over the link and see), will be the first to tell you that those numbers aren't to be taken seriously. As it says somewhere on Fedora's site, anyone making claims for open source numbers is either mistaken, lying, or trying to sell you something.

That being said, yeah, Linux is a lot more popular and you'll have trouble getting a FreeBSD job--generally, you have to get a Linux one and try to get them to make a change if you see a valid reason. However, it's even easier to get a Windows job. If popularity is the true indicator of quality, then Windows wins.
 

drhowarddrfine

Son of Beastie

Thanks: 872
Messages: 2,724

#46
To answer what the Doc said "FreeBSD is not an application developer" Not. It's whatever you can make it to be
You are not addressing what I said. By "FreeBSD", I mean the developers of the operating system. If one has problems with application software, looking to FreeBSD to fix those woes is the wrong place to be looking.
 

Spartrekus

Well-Known Member

Thanks: 45
Messages: 303

#47
HardenedBSD is a FreeBSD fork aimed to add and test security features and later bring those new features back to FreeBSD. The 'bring back' part does not seem to be really happening (don't ask me why).

So, it is more like a research OS, specially if you pay attention to they RELEASE model. OPNsense is built on top of HardenedBSD.
In other words, might it be good for FreeBSD for having a fork?
 

Chipper8827

Member

Thanks: 5
Messages: 33

#48
HPUnix,AIX Solaris (rip} osx and IRIX are all directly derived from Kin Thompson and Dennis Richie's original reimplementation of the dinosaur named multix as well as Linux Torvald's Linux and the whole FOSS community.
 

Oko

Daemon

Thanks: 764
Messages: 1,620

#49
In other words, might it be good for FreeBSD for having a fork?
I respectfully disagree with you. FreeBSD has being already forked once in June 2003 by Matthew Dillon. As much as I love DragonFly and Matt as a programmer and a person the fork has never gained as much traction as originally thought. Quite on the contrary it seems that it peaked 2004-2006 and right now you can hear crickets on their mailing list. Bruised FreeBSD egos never allowed any of DragonFly technical innovations back into FreeBSD proper. So it is double loss for FreeBSD (core developer + many new technologies). With all due respect Shawn Web is not in the same league as Matt Dillon. Probably the only person in FreeBSD camp who was as charismatic as Matt is Poul-Henning Kamp. Other people that come to mind when I speak about Matt are Fabrice Billard and Theo de Raadt.

I think Shawn is beating a dead horse. FreeBSD was never truly about security. That is neither bad nor good. It is just the fact that people who work on FreeBSD have different focus. That is one of the beautiful things about BSDs. Each one its four flavors is a truly a separate operating system which has a very narrow focus.

Out of 500 of something Linux distros on Distro Watch 350 are Ubuntu with custom wall paper, 100 are Debian derivatives and everybody uses the kernel developed by Red Hat. There are less than 10 truly interesting Linux distro and that is counting turn key appliances (Clonezilla, GParted, Parted Magic, SystemRescueCD, Proxmox, ThinStation, OSMC). FreeBSD alone has four highly popular and very unique appliances pfSense, OPNsense, FreeNAS, NAS4Free far more interesting than Linux counterparts ( IPFire, Sophos UTM, Untangle NG Firewall, OpenMediaVault).
 

ralphbsz

Daemon

Thanks: 711
Messages: 1,210

#50
HPUnix,AIX Solaris (rip} osx and IRIX are all directly derived from Kin Thompson and Dennis Richie's original reimplementation of the dinosaur named multix as well as Linux Torvald's Linux and the whole FOSS community.
It's Ken, not Kin. In the bay area, you can meet him occasionally. Unfortunately, Dennis has become swapped out, permanently. Very sad.

Apple's OS X is not derived from Unix, nor is Linux or the BSDs that are available today, in the following sense: they contain no single line that was written by Dennis and Ken, and are not subject to the original Unix copyright. One could say that Linux is a complete re-implementation of Unix; the history of BSD is much more complex (it contained Unix source code for a while, but it was removed again).

It's probably a mischaracterization to claim that Unix is a re-implementation of Multics. Clearly, it was inspired by Multics, including its name being a pun on the name. But to a large extent, it did things deliberately different, learning from the mistakes of Multics. And the goal of the Unix project was not to implement a new Multics (nor was it to drive a new phototypesetter, which was a feeble bureaucratic excuse to get a budget for it). Most likely one could say that the operating systems research group at Bell Labs (meaning: Dennis and Ken) needed a way to do research, and having lost access to the Multics project, they created their own little research OS. About a decade later, when the group got bored with that, they abandoned Unix (which by then had become a success), and worked on Plan 9.
 
Status
Not open for further replies.
Top