The current state of NetBSD

[Oko]

Has somebody read NetBSD Foundation 2013 Financial Report. I knew that NetBSD was in trouble for a long time but I didn't realize how bad trouble was until I read the report. They spent less than $30 000 last year and raised only $25 000. I am also having hard time finding anything about impending 7.0 release (please no links to http://www.netbsd.org/changes/changes-7.0.html) upgrade from one version of GCC to anther or upgrading OpenSSL are not significant new features. What is the status of Xen Dom0? It seems that that thing is rotting and it is in the same shape from almost 10 years ago.

[gpatrick]

They've been running with about $250,000 cash on hand for a few years. About what they spend is what they receive in donations, 2013 wasn't any different than 2012 or 2011. I'd assume 2014 will be the same.

[jkl]

Things were even rotting if they had more money. So many projects at a time, so many half-baked developments. Sigh.

But all we can do for now is to try keeping them up.

[fn8t]

I'm looking forward to the 7 release. I'd like to see how much progress is made on evbarm. Support for SMP is recent. Boards like the BananaPi, Cubieboard 2,3, and 4. I was hoping to see the Radxa Rock added to the list, but the Cubie 4 is equivalent.

Out of all the boards being looked at, I wonder which one(s) will get close to full support?

I noticed another device was added to urtwn, ported from FreeBSD.

I haven't noticed any new developments of interest to the "Production Environment World".
It almost seems like the interest isn't there for it.

There has been some work to move pkgsrc over to clanq, but only on select architectures.

Honestly, I don't mind it developing at the speed of molasses. However, it is concerning to think it might not be there one day.

I always hope that the main focus is maintenance to the base system and application of innovations made gently for the retention of that focus.

I have read about developers being displeased, some being very displeased, because the project seems to be trying to sustain a very peculiar set of goals. Some of those goals are very antagonistic towards the ambitious contributor. Wish I had the inside scoop on all of that, perhaps I'd find the reason for peculiarities honorable. It could be just indecision and fragile egg shells binding up the development procedure. Maybe too many folks wanting too many divergent paths, and the only way to prevent massive forking is by letting everything boil down to nothing? Overall, the opinion seems to be that the movers and shakers are dinosaurs. Some people wanna see the project instead thriving with their own new blood.

Its just guessing on my part. I've only heard the word of those that are complaining. The others aren't saying much (that I've encountered anyway).

I sometimes wonder why little easy things aren't done, like trimming gtranscoder and gmencoder from pkgsrc.

[abs]

Originally Posted by Oko

Has somebody read NetBSD Foundation 2013 Financial Report. I knew that NetBSD was in trouble for a long time but I didn't realize how bad trouble was until I read the report. They spent less than $30 000 last year and raised only $25 000. I am also having hard time finding anything about impending 7.0 release (please no links to http://www.netbsd.org/changes/changes-7.0.html) upgrade from one version of GCC to anther or upgrading OpenSSL are not significant new features. What is the status of Xen Dom0. It seems that that thing is rotting and it is in the same shape from almost 10 years ago.

Just curious, what was the other bad news for which this is more?

I think a lot of work is currently going into stabilising the KMS support for netbsd-7 (native resolutions and reasonable performance on current Intel and ATI display hardware), and a chunk more of arm platform support (allwinner mainly), plus cutting more ports across to gcc-4.8 (which is apparently not as simple as you might hope given port specific hardware or boot code which makes assumptions about how code will be compiled which are no longer valid in recent gcc).

I have a NetBSD/Xen DOM0 running nine assorted test VMs (mainly CentOS) on an old i5 server. I upgrade xen when there is a security notice, and otherwise pretty much forget its there.

[Beastie]

I hope it doesn't disappear. I feel it's always a shame when this happens, even when I don't even use the software.

[Oko]

Originally Posted by abs

Just curious, what was the other bad news for which this is more?

1. 2005 Google acquired Android Inc. dashing hopes that technically superior NetBSD will be used as a base for its OS.
2. 2006 Charles Hannum expresses his concerns about the future of NetBSD
3. 2009 Wasabi Systems Inc. (heavy NetBSD user/sponsor) is out of business
4. 2013 Julio Merino one of NetBSD Board of Directors gives up
   http://julipedia.meroh.net/2013/06/s...ng-netbsd.html

Now lets put above facts in prospective knowing that project spends less than $30 000 a year in operational costs (that used to be my NFS Summer support).

If that doesn't look like the ship is sinking I don't know what. Maybe project leaders like Martin Husemann who claims that OpenBSD is not "really UNIX and BSD" see what we other mortals don't see.

[gpatrick]

NetBSD received donations of $150,000 in 2009 and more in 2010 that left them with a final balance which is about what it is today. They spend about what their donations are.

If you think their ship is sinking, then I'd assume you'd feel the same way about OpenBSD which was in worse shape because they had no money to pay their light bill.

Seems OpenBSD likely receives less money than NetBSD.

[rocket357]

http://arstechnica.com/information-t...coin-donation/

$100k in 2014 seems a bit more alive than $150k in 2009 (but I'll give you that we're discussing two projects that both seem dead compared to, say, Linux). I'd argue that OpenBSD code gets used more heavily in other projects than NetBSD code, though (OpenSSH alone makes that statement possible), though that doesn't mean shit when it comes to corporate sponsorship.

[fn8t]

I really like both Net and Open. I have different use scenarios for each.

I can't say anything negative about Open or the authority in its development process, other than attributes that are practical but yet still hindering. Practical negatives are viewed as positive (in my eye). This might be my stance on Theo's so called overbearing control and or general intolerance, if I actually knew personally that he was overbearing or unduly intolerant. If it accomplishes the goal, then it can't be that bad.

While I like Net, I can't say anything more certain about its development process. Different members of the Net group have admitted that the treatment Theo and others have received was an investment in bad karma. I have read a ton of "historical" email archives, and still feel there is some background information missing. What isn't missing is the improper treatment of persons due to some mentioned disagreements that are not likely the sum of the igniting issues.
One area about Net that leaves me questioning, is the member agreement contract. There is a clause about "private information" that seems excessive, but could be completely proper. Guess I don't know enough about anything to make any stone engraved judgments.

None of this actually addresses the concept of decaying operating systems. To some, Plan 9 is good and dead. To others DOS is dead and rotten. If you still use and develop on them, they are plenty alive enough for you. Dead or decaying might just mean not used by the majority or a slower level of maintenance toward satisfying modern expectations. If that is the case, I haven't been using a living operating system in a long time. Yet another concept of dead is, "Its dead to me". This is the case with my regard to Windows and Linux, regardless of their root strength globally.

This thread could go some interesting directions.

[Oko]

Originally Posted by gpatrick

NetBSD received donations of $150,000 in 2009 and more in 2010 that left them with a final balance which is about what it is today. They spend about what their donations are.

That was 5 years ago. Do you know how long is 5 years in Tech industry? What is more troubling is that the money has NOT being spent, which means there is no activity.

Originally Posted by gpatrick

If you think their ship is sinking, then I'd assume you'd feel the same way about OpenBSD which was in worse shape because they had no money to pay their light bill.

Seems OpenBSD likely receives less money than NetBSD.

OpenBSD electric bills alone are over $150 000 so yes OpenBSD has to plea people to get serious about funding and community responded. You can check that the goal of 150K was reached within a week. More importantly the sign of the community vitality is that even increased founding level (I think they ended up with close to 300K) was also spent and they could spend even more if they had money.

Now politics aside OpenBSD is associated with the following significant coding projects:
OpenSSH, OpenBGPD, OpenNTPD, OpenSMTPD, OpenIKED, mandoc, LibreSSL. LibreSSL alone would be enough to make a big splash.

Now tell me one, just one significant NetBSD project in the last five years which was absorbed by other BSDs or OSs. Now I am fully aware of WAPBL and excited that it is part of Bitrig and possibly in the future of OpenBSD. However WAPBL is Wasabi Systems' file system and it has been released to public after Wasabi went out of business in spring of 2009 (that is more than 5 years ago). NPF is vaporware. What else? Porting ZFS and HAMMER went nowhere. Famed portability is reduced to cross compiling everything on amd64. Compare the state of Sparc64 in OpenBSD and NetBSD or something similar. pkgsrc doesn't work as advertised on anything but NetBSD and more testing and less talk would help (DragonFly people bailed out of pkgsrc, Minix seems to stick with it for now). Xen is cool but is it maintained? How does it compare to Jails+Behyve? NetBSD documentation feels like from the previous century. There is nothing new added there in the long, long time. New books are written about OpenBSD. None so far has ever been written about NetBSD.


Nobody here hates NetBSD. I am personally worried about once mighty NetBSD (it was more popular than FreeBSD in mid 90s). NetBSD is very important for the BSD and UNIX ecosystems. Something has to happen to shake that community because otherwise will just further sink into obscurity. They have to come up with something cool to work on and more importantly they have to finish it, test it, and brag about it.

[shep]

They have to come up with something cool to work on and more importantly they have to finish it, test it, and brag about it.

I would add that the project needs to decide what it is about. OpenBSD has a very well defined mission and ALL of their projects are in line with security and code correctness.

As near as I can tell, NetBSD is about running on mutliple architectures. In my opinion, they should prioritize the architectures they wish to support, cut the cruft. and focus on making sure those system run well.

[gpatrick]

I'll admit NetBSD needs an identity, much the same as OpenIndiana. To me, both projects are foundering without adequate direction (OI's project leader resigned so that's a given).

cut the cruft

Agreed. As two examples, either make dtrace and zfs work or remove them.

pkgsrc doesn't work as advertised on anything but NetBSD

I use pkgsrc on SmartOS often since it my platform for my websites and mail server. It works very well and I've never encountered any problems.

NPF is vaporware

Is also inaccurate if you talking about NetBSDs npf firewall, because it was released for production in the 6.x branch. Mindaugas is very receptive to emails and will reply directly. He also fixed a problem https://mail-index.netbsd.org/netbsd...msg032210.html I reported.

[Oko]

Originally Posted by gpatrick

I'll admit NetBSD needs an identity, much the same as OpenIndiana. To me, both projects are foundering without adequate direction (OI's project leader resigned so that's a given).

For starters they have to drop things like Amiga or openly declare themselves archaeological OS. As much as I enjoyed Atari and Amiga in late 80s supporting those platforms makes no practical sense. Now maybe NetBSD wants to support dead hardware platforms. If that is the case that is a very noble goal and they should cut all that nonsense about practically usable OS.

Originally Posted by gpatrick

I use pkgsrc on SmartOS often since it my platform for my websites and mail server. It works very well and I've never encountered any problems.

Played 4-5 years ago on OpenBSD. It was laughable. DragonFlyBSD people really, really tried to make it work and they gave up on pkgsrc. I am not saying that pkgsrc is bad. What I am saying is that either test all advertised platforms or stick with NetBSD. Same goes with their annoying cross building. If you are going to support 50 something platforms then you have to do 50 something native builds and test it. Otherwise NetBSD has no credibility.

Originally Posted by gpatrick

Is also inaccurate if you talking about NetBSDs npf firewall, because it was released for production in the 6.x branch. Mindaugas is very receptive to emails and will reply directly. He also fixed a problem https://mail-index.netbsd.org/netbsd...msg032210.html I reported.

I am very happy to hear that I am wrong!

[ibara]

So I'm going to do something I never do and post to a non-OpenBSD section of the forums. Not that I think NetBSD needs help defending themselves; they're adults and can speak for themselves. But there is one recurring sentiment made in this thread that needs to be challenged and/or halted because it's senseless. I'm going to paint you all with a broad brush. I don't particularly apologize for this.

Complaint: NetBSD should drop smaller platforms/"cruft".
Why? What sense does this make? Should OpenBSD drop their smaller platforms, like OpenBSD/loongson? Think really hard before you answer that question. If OpenBSD took your suggestions then I would not have had the opportunity to join the team. I got my account because I wanted to work on that platform so I did and got rewarded for that work. The OpenBSD/luna88k is essentially a labor of love from a single person. Why take that away?

There's some bizarre folk logic that says "if you take away what I think are the distractions then developers will work on the things I think are important." No. That's 100% wrong. Taking the Loongson platform away doesn't translate to me working on something else. It translates to me, and I am going to stress this, NOT WORKING ON ANYTHING. The things are are important to the developers (i.e. not you, i.e. the only people that actually matter in this discussion) would not be there. And they would have nothing to work on.

So I went and read the NetBSD CVS logs, clearly something none of you bothered to do, and there's real activity going on in the Amiga port. Go look for yourself: NetBSD/amiga. If people want to work on it, great! You don't get to control what other people voluntarily work on. Remember: the equation isn't "they work on amiga or they work on something else" it's "they work on amiga or they don't work."

And yes having those "dead archs" really matters. Because sometimes you find bugs you don't find elsewhere. And sometimes those bugs are bugs that affect all platforms. Yes, NetBSD would do better at finding those bugs if they did more native building but that's their decision to make.

And if someone else wants to work on ZFS or anything else and the project is willing to let them work on it in-tree then it's a sad day for you.

If you don't like the direction NetBSD is headed, do something about it. Yeah, that's difficult. And it takes a lot of hard work and/or money (but sometimes not as much money).

The conference I helped organize and run this past year, NYCBSDCon 2014, was able to raise enough money to donate over $1100 to each of the four main *BSD projects. See our logo on the NetBSD donations page for 2014.

Or maybe donating your time and code? Maybe you're not a kernel hacker, but maybe how about one of you take it upon yourself to clean up pkgsrc on some platform you think is less than ideally supported. A friend of mine has been doing this with pkgsrc on PowerPC Mac OS X.

Seriously, go and do any of those things. Donate money or go donate your time and skills. Then come back and repeat any of that to me again. I dare you. You won't be able to because you'll finally have an appreciation for all those things you don't know and never think about.

I'm done. Don't make me come back to this thread.

[gpatrick]

I don't agree with dropping architectures, but remove what is half-baked like ZFS or fund it so it works. I'm looking forward to the NetBSD 7.0 release.

FWIW NetBSD is my preference between FreeBSD, NetBSD, and OpenBSD.

[darktrym]

I think ZFS on NetBSD is dead, nobody really wanted it. If there are no efforts to take ZFS to the main file system, there is no chance to update this outdated non functional port. You can look on the wiki site, not so much progress for years.
I also see any progress to Lua interfaces. My hope was i can use this for skripting i.e. to replace some shell scripts or make it easier to get control of my accellerator sensor for the hard disk.
Then i don't see much benchmarks how good or bad NetBSD is. When npf is better scaling why is no actual slides/tests to demonstrate this? Maybe it is worth to work with phoronix to get more statistics and attention to NetBSD.
Also there is also a very bad information politic. When i want to know whats going on in NetBSD have to visit cvs logs, news aggreator(not so much NetBSD inside) sites and the changes in the next version(netbsd.org/changes).
I saw a video from AsiaBSD that Net-/OpenBSD has now valgrind. Thanks to japanese NetBSD developer, but is it to much work to send a short message to news/bsd sites like bsdnow? I also see working on networking layers, maybe useful for better smp performance.
All and all many NetBSD relevant information are hard to find. When i want to see which wireless adapter are supported, i have to visited wikipedia. For many changes there i responsible. Also a info of the state of X, which driver are currently supported when come newer X release would be nice. On the OpenBSD plus site i can follow the progress of the project, NetBSD doesnt have an equivalent. Also a specific NetBSD site like OpenBSD undeadly would be nice to keep in touch users and developers. Our community blog is full of dragonfly digest entries and questions from forums with no answers, not very useful from a NetBSD user perspective.

Last i hope we can see progress in 802.11n, uefi. NetBSD isn't in that bad state, it is under heavy active development in different regions, but a little more information what going on would show that the project is there and donation aren't waste.

[Oko]

I feel compel to respond to this well intended post which I believe contain some logical fallacies. None the less I think that the post is very valuable in the sense that it might help us understand how some of the inner circle NetBSD people think.

Originally Posted by ibara

Complaint: NetBSD should drop smaller platforms/"cruft".
Why? What sense does this make? Should OpenBSD drop their smaller platforms, like OpenBSD/loongson? Think really hard before you answer that question. If OpenBSD took your suggestions then I would not have had the opportunity to join the team. I got my account because I wanted to work on that platform so I did and got rewarded for that work. The OpenBSD/luna88k is essentially a labor of love from a single person. Why take that away?

There's some bizarre folk logic that says "if you take away what I think are the distractions then developers will work on the things I think are important." No. That's 100% wrong. Taking the Loongson platform away doesn't translate to me working on something else. It translates to me, and I am going to stress this, NOT WORKING ON ANYTHING. The things are are important to the developers (i.e. not you, i.e. the only people that actually matter in this discussion) would not be there. And they would have nothing to work on.

While I concur that some of your logic here is correct it also contains some logical flows. First one is that I am not aware that NetBSD project is for developers only unlike OpenBSD which publicly states that as one of its principals. Apart of the semantics the fact that Amiga developer might not be compelled to do any work after her/his favorite port is officially dropped that might also translate into 10 more capable ARM or AMD developers who are now sitting on the sidelines due to constrains imposed on the project by supporting hardware with serious technical limitations like Amiga or Atari.

Originally Posted by ibara

So I went and read the NetBSD CVS logs, clearly something none of you bothered to do, and there's real activity going on in the Amiga port. Go look for yourself: NetBSD/amiga. If people want to work on it, great! You don't get to control what other people voluntarily work on. Remember: the equation isn't "they work on amiga or they work on something else" it's "they work on amiga or they don't work."

This might be very well how many core NetBSD developers think. If that is the case NetBSD should be pronounced Archaeological OS. As such it might be very interesting to me when I want to fire my old Atari 1000 but has no relevance for my day job.

Originally Posted by ibara

And yes having those "dead archs" really matters. Because sometimes you find bugs you don't find elsewhere. And sometimes those bugs are bugs that affect all platforms. Yes, NetBSD would do better at finding those bugs if they did more native building but that's their decision to make.

This argument is frequently invoked by OpenBSD developers when somebody is trying to argue against some legacy ports (most recently Sparc). The major fallacy of your argument in my point of view is that unlike OpenBSD, NetBSD folks are actually doing very little native builds. As a member of inner OpenBSD circle you know all too well that no platform can be adapted as official platform until at least two physical machines of that type are not available to developers one of which has to be located in the famous Theo's rack. Are there any images of NetBSD rack that I am not familiar with with machines in all 56 or so different architectures NetBSD is officially supporting? For some of those architectures actual physical machines have never been built IIRC. OpenBSD has never been shy in cutting down legacy platforms for various reasons. Check out the attic. OpenBSD used to support Amiga too. I argue that supporting for old platforms is important as long as native builds are done and no major new features are prevented from being imported into the code base due to limitations of legacy platforms. There would be no faster way to infuriate OpenBSD community than to say that new packet queuing and prioritization can't be imported into the base because of limitations of Luna or Shark platforms. Even with current fairly reasonable policies we saw two recent forks of OpenBSD (defunct AerieBSD and Bitrig).

Originally Posted by ibara

And if someone else wants to work on ZFS or anything else and the project is willing to let them work on it in-tree then it's a sad day for you.

No it is not. I use ZFS to make living and if I can't get it from NetBSD I will get it from somewhere else. Again you are using your OpenBSD mantra "OS by developers for developers and everyone else is just for a good ride". NetBSD is officially not like that and some people would argue that was the main point of contention between Theo and the rest of core when he forked OpenBSD.

Originally Posted by ibara

If you don't like the direction NetBSD is headed, do something about it. Yeah, that's difficult. And it takes a lot of hard work and/or money (but sometimes not as much money).

The conference I helped organize and run this past year, NYCBSDCon 2014, was able to raise enough money to donate over $1100 to each of the four main *BSD projects. See our logo on the NetBSD donations page for 2014.

Or maybe donating your time and code? Maybe you're not a kernel hacker, but maybe how about one of you take it upon yourself to clean up pkgsrc on some platform you think is less than ideally supported. A friend of mine has been doing this with pkgsrc on PowerPC Mac OS X.

Seriously, go and do any of those things. Donate money or go donate your time and skills. Then come back and repeat any of that to me again. I dare you. You won't be able to because you'll finally have an appreciation for all those things you don't know and never think about.

I'm done. Don't make me come back to this thread.

I see no reason for vocabulary of threats. I am not afraid of you or anybody else for that matter. I have been supporting OpenBSD financially for a while like many people who frequent this forum arguably with small but reoccurring donations. It is a tool that I use at work. It is a tool that I like using, but it is a tool, no more no less. I would be happy to support NetBSD and I started this thread being a UNIX lover. At this point I personally see no piece of NetBSD worth of my support. What I am afraid is that my opinion is not isolated.

[pawaan]

All and all many NetBSD relevant information are hard to find

NetBSD News:
http://www.NetBSD.org/changes/rss-netbsd.xml
NetBSD Code Changes:
http://www.NetBSD.org/changes/rss-netbsd-internals.xml
NetBSD Fresh Packages:
http://www.NetBSD.org/changes/rss-netbsd-pkgs.xml
NetBSD Security Advisories:
http://www.NetBSD.org/support/securi...advisories.xml

Also a specific NetBSD site like OpenBSD undeadly would be nice to keep in touch users and developers.

Nice idea , there are a few NetBSD-centric sites though :
http://blog.netbsd.org
http://wiki.netbsd.org
http://mail-index.netbsd.org/
http://www.feyrer.de/NetBSD/bx/blosxom.cgi/index.front
http://pkgsrc.org/
http://pkgsrc.se/
http://pkgsrc-wip.sourceforge.net/
http://netbsdfr.org/
http://www.netbsd.se/

[darktrym]

If i remember right, two/three weeks ago some listeners wanted in bsdnow more news from other BSD like NetBSD and Dragonfly. It is not too had to send them some useful links to give their audience a better look of all BSDs.