Guest post by The NetBSD project has announced two important stability updates for its highly portable operating system.
The NetBSD Project is pleased to announce NetBSD 5.1.5, the fifth security/bugfix update of the NetBSD 5.1 release branch, and NetBSD 5.2.3, the third security/bugfix update of the NetBSD 5.2 release branch. They represent a selected subset of fixes deemed important for security or stability reasons, and if you are running a prior release of either branch, we strongly suggest that you update to one of these releases.
Details on the two updated branches of NetBSD can be found in the release notes for NetBSD 5.1.5 and NetBSD 5.2.3.
Well this is just my opinion but I think NetBSD is the best BSD because it is the only one that properly supports systemd at this point.
As everyone knows the status of systemd support on OpenBSD is still uncertain and it isn’t likely any progress will be made until Theo de Raadt is out of jail (selling 3d printable plastic fire arm models, don’t ask).
Now that Yolanda (born Jordan) Hubbard has made it clear that systemd is too advanced for FreeBSD that OS can’t be considered for serious work either.
Darwin is written in C# so there might be difficulty porting. Looking at the other BSD flavors I think only Emacs is possibly powerful enough to run systemd.
Go NetBSD!
I think there is some ongoing work to run openbsd as a systemd daemon on haiku. Looks stable to me. Maybe that can be extended to support netbsd?
You can’t run systemd on NetBSD. In fact you can’t run it on any *BSD distribution. NetBSD’s init system is called rc.d and has much in common with FreeBSD and OpenBSD init systems: https://www.netbsd.org/docs/guide/en/chap-rc.html
Um, I think he was being sarcastic, especially since almost every recent comment thread has devolved into a systemd rabble rabble.
Bravo! Too bad your sarcasm was lost on some people.
Although it would no doubt be an interesting and ironic point, Darwin is not, in fact, written in a language designed by Microsoft.
Edited 2014-11-23 03:56 UTC
It was a joke! You really missed that which was so obvious?
Edited 2014-11-23 06:01 UTC
A joke??? LOL not… sort of like the GNU/Linux joke – becuase it’s a recursive accronym… it’S funny NOT
What Beerfloat posted was called sarcasm.
What I posted was called a JOKE… maybe you have heard of them?
I suggest we all should cease using any xBSD. It’s not only the BSDs lack proper support for systemd but they are written in ancient C/C++ language while all modern operating systems have large portions written in BASH script or Python which is essential for speed in critical areas like the CPU scheduler, multithreading support or SMP. C was developed in 1969 while latest stable BASH release was made on 5 October 2014!
Adding to that, BSDs are evil. They aren’t free software at all, lacking a proper free software license such as GNU GPL v2.0 or later and are not endorsed by a proven Free Software Guru such as Richard Stallman or Eric S. Raymond.
Use GNU/Linux or GNU/Hurd instead!
Edited 2014-11-24 08:19 UTC
In fact NetBSD seems to be not much more than just a “toy project” of a few developers, and – in fact – it looks it’s going nowhere. Just a hobby for a free time, nothing with a particular goal(s) set. You can read more for example here: http://julipedia.meroh.net/2013/06/self-interview-after-leaving-net…
Some of the concerns jmmv lists are valid, but he never wrote that NetBSD is a toy project of a few developers. That is not the case. There are many people working on it, just that it is mostly a volunteer project, without financial backing of big companies. So people mostly work on what they find interesting. Not what you find interesting, not necessary what most users find interesting.
And the level of trolling in this thread saddens me deeply.
Edited 2014-11-24 21:48 UTC
You just defined “toy project”. Yes, it’s a project, where you work on the things you are interested in, not actually caring about the users, and not even setting any particular goal. Just playing with the “things you find interesting”.
Yes, he didn’t use the term “toy project” – it was my conclusion. And you just confirmed I was right.
Edited 2014-11-24 23:34 UTC
So then Linux is a toy project too and so are all the operating systems based on it.
Wrong – Linux developers care what Linux users say.
These days NetBSD is more of a research operating system. But that does not mean it can’t be used commercially. The term “toy” that you use, suggests that NetBSD is not usable in a real world or commercial environments.
Tell that to Dell (where NetBSD is used on their 10 GBit switches), Internet Initiative Japan or other vendors that are basing their devices or appliances on it with great success.
In the open source projects, it was always very important for developers to work on stuff that is interesting. I would never waste my time on things that are boring. Developing NetBSD is fun. I know, I am a NetBSD developer.
The term “toy” that I use suggests, that NetBSD – as I wrote – is just hobby project of a few persons, nothing with particular goal set. There’s nothing wrong with having hobbies – and I really don’t understand, why are you denying.
Such hobby projects – if you dislike “toy” so much – can be (but they doesn’t have to) useful for some of the users. Dell uses NetBSD? That’s good – still no wonder, that some developers are leaving the board, being disappointed with no particular policy. With no answer to the question: “Actually, where are we going?” (unless your target is creating the software for Dell gigabit switches).
I get your point about goals and direction, but if Dell, a major corporation, doesn’t consider it a toy, then characterizing it as a toy is unfounded.
It’s also pretty much insulting to a 20+ year project with the contributions it has made, so surely you can understand the push-back you are getting, despite saying you don’t understand it.
Maybe indeed I should have write “pet project” instead – well I’ve explained, what I meant. Not that NetBSD is a toy – just it seems rather to be “hobby project” rather than endeavour.
Like Red Hat is caring, while shoving systemd up everyone’s bottoms?
We have 170 active developers. I wouldn’t use the word “few” here.
There might not be any particular “goal” for the whole project, but we have many internals teams or even single developers, that are working on their goals. Some of them might be not interesting for general public, and some of them is extremely valuable research work (think rump).
I am merely correcting your FUD.
Not every one needs a written policy, or a goal sets for the whole project, to continue working on it.
Now as I said, jmmv brought up some valid points. But they don’t reduce the value of the whole project.
The question is: do NetBSD really need to answer such question? The current conservative architecture of NetBSD is perfect for people who want an OS designed with pure UNIX principles. Continuing development along the same lines, while updating the system components, writing new drivers, adding support for more platforms, etc. is perfectly fine with most developers working on NetBSD. Perhaps that’s why we don’t have a “goal” set.
Do most open source projects have such written policies? Developers have to be managed somehow? Even within Linux, most people just work on what they want.
Some Linux features are developed only when companies and funding come into play, because otherwise they would be too boring to work on.
Not sure, what you mean – cannot address that. Anyway: there are plethora of distros out there. If anyone doesn’t like Redhat’s policy, the switch to other distro won’t take that long.
It’s not much considering, that NetBSD doesn’t support as much hardware, as Linux – and, for example, until today couldn’t complete full internalization of its text mode. i18n of graphics mode AFAIK isn’t the work of NetBSD devs.
No, there’s no pressure – but consider an example, please: let’s complete a team, whose goal shall be “to build a car” (just “a car”). One of the team members wants to build an engine, the other the body, next one transmission and brakes etc. etc. Is it possible to create a car this way? Probably. But until they agree, what kind of car they’re actually designing – is it racing car, roadster, truck or van – the result of their common efforts most probably won’t be that impressive. Do you agree?
So far just one thing can be told for sure: NetBSD’s team is developing Unix-like OS. It has a slogan, it can be used everywhere – “even on toaster” – but let’s face it: Linux supports about as many platforms (if not more). And the second thing: yes, NetBSD can be run on many platforms – but not equally good on each selected one. Support for more common ones – like x86 – is better, but for more osbcure ones (Atari 68k?) can be (much) worse (and usually is). Other BSDs also support many platforms – maybe not as many, as NetBSD, but still. Therefore that doesn’t make NetBSD unique, and we still don’t know, what kind of “car” it’s going to be.
It is already almost impossible to avoid systemd and friends on the Linux desktop. It’s a great example of a changed forced by a single company on the whole community. And it is all done while saying “oh we are doing it so that Linux will be better for the user!”.
Now compare that with NetBSD, which is just set on taking conservative 4.4BSD architecture and making it better, instead of reinventing it.
NetBSD is a complete operating system. Linux is a kernel. There is no single Linux distribution that supports as much hardware as NetBSD does.
Support for UTF-8 localised text consoles is lacking indeed, but it seems that no one cared enough about it, considering that you can easily use X11 on most ports. Also users that really need UTF-8 in text mode are using mlterm-fb, so it is possible to achieve.
It’s mostly done by X.org devs, as in Linux distributions. So you could say the same about Linux.
I understand where you’re going, but most NetBSD developers understand that we are building a modern 4.4BSD derivative. Not some new fancy-ultra-hyper OS. If someone do not understand this, then he will be disappointed with NetBSD conservative approach.
Linux kernel yes, Linux distributions no. NetBSD is more like a complete Linux distribution.
If we are comparing to Linux, NetBSD is doing awesome job supporting these obscure architectures. How many drivers Linux gained in last few years for Amiga? 0. NetBSD? Around 15. Not sure about Atari, since I don’t have any MMU-equipped Atari, but I heard it is not doing bad at all. Japanese developers are hacking furiously on NetBSD for Sharp X68000 and Luna 68K. There is no other UNIX currently developed with support for obscure hardware as good as NetBSD’s.
When it comes to base NetBSD OS, it mostly delivers the same features for all supported platforms. Sure, some features don’t have a point on slow or embedded hardware. Some machine dependent APIs are lacking here and there, but taking NetBSD/amiga and NetBSD/amd64, you mostly get the same feature set (as they are build from a single source tree). And naturally, not as many developers have rare machines, so they are usually less tested in real world applications, but we are working hard on delivering the same user experience across all ports.
Now we’ve reached a point, where I’d rather spend my time working on NetBSD (have a few drivers in progress, lol), than continuing this fruitless discussion. What I wrote here you might as well read on NetBSD web site, mailing lists. etc. So I feel like stating the obvious.
Not going to argue further, just a few words:
As a Slackware user I assure you, that it’s perfectly possible. BTW: but not sure actually, whether systemd is a good, or “evil” thing – maybe indeed it’s something worthy? Just didn’t read about it yet too closely.
Oh, “sorry – no bonus”. This is usual mantra of most of xBSD-fans/devs: “Linux is just a kernel” etc. etc. Now show me, please, a working Linux system containing “just a kernel”.
This is very formal, very “artificial” argument (but you aren’t attorney at law, aren’t you?
. Willing to do anything with Linux, you do need a distribution, therefore “Linux is just a kernel” means just: “I would to mock Linux, but I’ve got no better arguments”.
No idea, how could you use “just Linux kernel alone” without at least minimal distribution.
Maybe, but who needs presently NetBSD on Amiga? Does NetBSD allow to use Amiga’s advantages (I mean graphics, sound etc.) the same complete way, as AmigaOS does? If not – it’s better to install NetBSD on older 486 machine, which is more powerful than Amiga, has better support being x86-derivative, and can be obtained even for free. How many people in the whole world use NetBSD on original Amiga hardware (and why, actually) these days?
So you really think, that NetBSD users care more about Amiga support than about full i18n?
What I was trying to express – maybe not that cleverly – in my initial post, was that NetBSD is rather floating, than swimming. Which, of course, doesn’t hurt me at all.
Deleted.
Edited 2014-11-25 17:13 UTC
Glad they are doing deep maintenance releases! 1.6.2 was very solid!