FreeBSD development seems lost

ralphbsz said:
And systems research (meaning: operating systems research) died somewhere around 99 or 2000, according to Rob Pike. Strange that he is still employed as a systems researcher, now 25 years later.
Click to expand...
More code has been written since then, true, but what new OS research ideas have come about since 2000? I recommend reading Brinch Hansen's The Evolution of Operating Systems paper as a refresher.

It all seems more of the same piled on top of more and more complexity.

PS: I suspect cy@ was being sarcastic.
 
bakul said:
It all seems more of the same piled on top of more and more complexity.
Click to expand...
I think, the main purpose of the OS is to abstract the machine, write drivers.

We can see the C compiler also as a driver, for making more abstract the CPU machine language.

What BSD created a C compiler?

I real OS is plan9, with a lot of own compilers for many CPUs.
 
It's actually pretty funny that everyone dislikes Wine (for whatever reason) -- and yet Wine was completely designed, coded, tested and built on Unix systems -- mainly *BSD systems -- over many (decades/years), starting far back in the 1990s.

Do you think Linux invented (everything)? It did not, a ton of what Linux is comes from Unix systems.

When you say you don't want something because that's "What they do on Linux" - that's just silly.
 
kpedersen said:
I don't disagree with you about Wine being great tech, but I must admit, OpenBSD not supporting it, keeps the community a little more focused on computing rather than gaming.
Click to expand...

I think that's a fair point -- and a big reason people switch to OpenBSD, Net2, etc.

If that is where you want to go then good luck.
 
kpedersen said:
The world is in a race to the bottom
Click to expand...
Started watching a (terrible) movie, Idiocracy, where a man is put to sleep but accidentally doesn't wake up for 500 years. What happened in all that time is smart people kept waiting for the perfect time to have kids that never happened while dumb people were mating with anything that moved. Hence, the world became stupid over time.

The plot is that, even though he was an average Joe, 500 years later he's considered a genius. A terrible movie so don't bother with it.
 
hruodr said:
Unfortunately the problem of all scripting languages.
Click to expand...
Mostly aggree with it.
But not applicable to /bin/sh. I consider it as a scripting language, rather than command line interface. So would be awk, sed and so on in base.

hruodr said:
Completely unnecessary change for luanatic reasons.
Click to expand...
Another reason is (if I recall correctly) decreasing number of forth programmers in loader guys.
Just my understanding, but if sufficent number of forth programmers pop in constantly, the switch wasn't happen. So needed change to be sustainable.
 
T-Aoki said:
Another reason is (if I recall correctly) decreasing number of forth programmers in loader guys.
Just my understanding, but if sufficent number of forth programmers pop in constantly, the switch wasn't happen. So needed change to be sustainable.
Click to expand...
In my opinion not enough justification.

People that deal with the loader in that way are not clueless users and would quick learn forth for their purpose.
 
category wishful thinking
loader scripting should have been tcl. if cisoo could do it on crappy hardware 1000 years ago its certainly possible now. also tcl should have been in base
 
kpedersen said:
Proponents of PkgBase and other terrible ideas mainly.
Click to expand...
It is not a change that I would have asked for, but PkgBase is working so well for me (so far) that 15-ALPHA has replaced 14.3-RELEASE as my main installation.

The Xfce desktop environment is looking better than ever, and I haven't had any of the recent problems that were affecting me in 14.3, such as missing VLC codecs or a packages update breaking my custom MariaDB-based WikiMedia installation by replacing MariaDB with MySQL.
 
When Judgement Day comes it'll find me busy building the kernel & world. I'll shrug it off saying "wait" and then calmly proceed to run make installworld. That's why I don't care about the end of the world or your favourite Doomsday scenario, religious or secular.
 
rbranco said:
That's why I don't care about the end of the world or your favourite Doomsday scenario, religious or secular.
Click to expand...
Doomsday scenarios aren't what they once were: at one time it looked like the world was close to nuclear war, but society and governments keep adapting. Even when U.S. politics has drifted too much to one extreme or another, it has been followed by a readjustment the other way.

I don't think FreeBSD development is lost - it's just adjusting to more recent expectations.
 
pkgbase concerns me and I may look elsewhere, as I have been using FreeBSD for the last 5 years or so precisely because of the sensible split between base and everything else. A big part of "the BSD way".

I you decide to look elsewhere, it's best to just do it - railing impotently against a project, on their forums usually achieve little. We are after all "along for the ride" and we can move on elsewhere whenever we choose.
 
blackhole said:
I you decide to look elsewhere, it's best to just do it - railing impotently against a project, on their forums usually achieve little. We are after all "along for the ride" and we can move on elsewhere whenever we choose.
Click to expand...
Yep. One thing that has always boggled my mind was people that announce they are leaving a forum.
Why do that? Just ask the mods to delete your account/login and move on.

If one doesn't like the direction a project is moving, why not get more involved (or at least try to) and change the direction or maybe get a better understanding of why it's moving in that direction.

I'm on the fence about pkgbase at the moment; I can see the usefulness, but I think it may take a few cycles to get it to stabilize. It sounds like it's optional in 15.0, if it remains optional through 15 and enough people use it to shake it out, then maybe mandatory in 16 works.
How long did it take for people to be comfortable and decide 5.x was stable?
 
Useradd said:
It is not a change that I would have asked for, but PkgBase is working so well for me (so far) that 15-ALPHA has replaced 14.3-RELEASE as my main installation.
Click to expand...
The issue with PkgBase is not if the code works currently (its just extracting files from many archives rather than one archive). The issue is that it is the first step in destroying FreeBSD's deterministic base install and turning it into a random Linux-style ricers playground.

gotnull said:
Real question, how is this going as a file server with FFS?
How would you compare it to UFS we have on FreeBSD?
Don't you miss bhyve(vmm seems less capable), jails or ZFS?
Click to expand...
Actually FFS is solid as hell. It is slower though but for my use-case (I am not an enterprise), it is easily good enough.

vmm and chroot are also good enough for me. I only allocate one core to my VMs anyway and my use of a jail is mainly building software / administrative rather than security, so chroot will suffice.

For me personally, OpenBSD is an easy compromise.

CShell said:
I think that's a fair point -- and a big reason people switch to OpenBSD, Net2, etc.
If that is where you want to go then good luck.
Click to expand...
It will be fine for around a decade, then I am sure the crap will follow me just like it did with FreeBSD from Linux :/
 
By now I think most of the fuss about pkgbase is a storm in a teacup. The seperation between ports and base is drawn on the file system level and dependencies, it is not that important where the files come from. I like tracking STABLE, for example, but being able to get llvm from a package instead of building it again and again sounds like a good idea. Having all the packages for base build from areas of the base source tree takes care of always having a running system. As long as there are no dependencies from these to some packages who are build from ports, we keep the seperation of state and church system and applications. Rust in base is a hot topic right now, but it is also a strawman for adding code written in something not C. Complaining that perl is gone (iirc since a looong time) at the same time is a bit hypocritical don't you think?
 
Crivens said:
By now I think most of the fuss about pkgbase is a storm in a teacup. The seperation between ports and base is drawn on the file system level and dependencies, it is not that important where the files come from.
Click to expand...
Worst of it is that whiners are acting like this is so sudden and unexpected change. C'mon! It has been in the works for years and years now, can't even remember when I heard about plans for it for the first time, during 12, 11?
 
kpedersen said:
The issue is that it is the first step in destroying FreeBSD's deterministic base install and turning it into a random Linux-style ricers playground.
Click to expand...
My experience with freebsd-update is that it is indeterministic & lame, takes hours and one can end with a broken system.
I upgraded debian many times, really good experience, works without troubles and is not an endless process.
 
rbranco said:
When Judgement Day comes it'll find me busy building the kernel & world. I'll shrug it off saying "wait" and then calmly proceed to run make installworld. That's why I don't care about the end of the world or your favourite Doomsday scenario, religious or secular.
Click to expand...
End Times, Judgment Day, Al-Qiyāma or not, don't rush it:
etcupdate -p; cd /usr/src; make installworld; etcupdate -B
😉
 
You must log in or register to reply here.
Back
Top