FreeBSD to rethink target audience?

[oliverh]

No, quite the opposite. If they come along and say "here, we provide you with a desktop variant of FreeBSD", they cannot lean back and go "OK, FreeBSD devs, now you give our users what they want". After all, those users who *are* interested in running FreeBSD on modern notebooks and with cheap hardware are *right there in the DesktopBSD/PC-BSD support channels* and can contribute, and if only through giving feedback on what they want and testing of modifications.

As somebody else said, there's few FBSD developers, and they're busy as it is. The very least the PC-BSD and DesktopBSD people can do is contribute to the FreeBSD code base, making it more desktop/notebook-friendly. I don't know if they do, but I sure know they should. Otherwise they're just leeches.

For example all the manpower that went into the pointless PBI infrastructure could as well have been used to implement a graphical ports/packages manager and contribute it back to FreeBSD. This way everyone would have benefitted. As it is, only PC-BSD users "benefit", and how starting a parallel software universe is a benefit to anyone is still up for discussion. For all I know they could be distributing trojans with all applications, because I cannot verify those proprietary binary files one way or the other.



Yes, me too. I also was delighted when DesktopBSD and PC-BSD started their work, but the lack of manpower shows. I'm not contributing to them myself, so I won't complain, but I observe that given more love, they could really give Ubuntu a serious beating.

Do you think so? I was for some years an active member of the DesktopBSD-Team - a couple of people. There is no money, there is a shortage of time etc.pp. and there is almost no interest by the media to see another alternative on the desktop. It's different with PCBSD, they have some money but it's not comparable to the ressources of Ubuntu. So some love will not help at all.

 

[susanth]

Let the full focus of FreeBSD be on Servers (Like@present). Especially on SMP capability, Networking, Security, Speed ...etc that are related to servers.

I fear, Satisfying the Desktop n Lap top users will reduce the power and strength that FreeBSD Enjoy @ present.

After all FreeBSD is a POWER OS which have the "Power to Serve" for all Power User.

 

[bsddaemon]

I fear, Satisfying the Desktop n Lap top users will reduce the power and strength that FreeBSD Enjoy @ present.

I think its a valid point, in this case, popularity would spoil quality

 

[thortos]

I was for some years an active member of the DesktopBSD-Team - a couple of people. There is no money, there is a shortage of time etc.pp. and there is almost no interest by the media to see another alternative on the desktop.

The media shouldn't bother you; they don't know anything about FreeBSD, yet Netcraft still lists it a lot among the longest-lasting sites.

I just fear as long as Linux has such strong momentum, you can't score with a FreeBSD based desktop distro unless it's absolutely amazing from the layman's point of view. You might enjoy the fact that I always thought that *if* one "distro" can do it, then it's DesktopBSD, because it aims to make FreeBSD usable, not create something similar to all those Linux distros.

It's different with PCBSD, they have some money but it's not comparable to the ressources of Ubuntu. So some love will not help at all.

And what does that money buy them? They've got no credibility at all because they insist on using their proprietary way of installing software instead of using the awesome mechanisms already in place.

With "love" I actually meant manpower. If DesktopBSD could only get 1% of the people working on Ubuntu to work on DesktopBSD instead, that would make a huge difference. However, we probably won't see this until Linux keepts getting raped badly over all those superfluous kernel modules it has in the default configuration. Then the security of a well-architected system environment such as FreeBSD will gain mindshare.

Yeah, one can dream. OTOH, I like the FreeBSD comunity as it is, small and full of hackers in the original meaning of the term.

 

[overmind]

I noticed lately in the news that Canonical is porting Ubuntu to ARM, and that ARM notebooks will be produced soon (maybe are already on the market).

It would be a good idea to have FreeBSD on ARM laptops as a Workstation distro.

 

[graudeejs]

uh, he, i dream for long time to try some architecture other than x86, but i won't buy other arch, if FreeBSD ain't supported on it. as there is no OS other than FreeBSD for me

I love FreeBSD.

and if it stays as good as it is, i'm ready to buy real server just to use FreeBSD as my desktop (i'm dam serious)

 

[none]

The media shouldn't bother you; they don't know anything about FreeBSD, yet Netcraft still lists it a lot among the longest-lasting sites.

I just fear as long as Linux has such strong momentum, you can't score with a FreeBSD based desktop distro unless it's absolutely amazing from the layman's point of view. You might enjoy the fact that I always thought that *if* one "distro" can do it, then it's DesktopBSD, because it aims to make FreeBSD usable, not create something similar to all those Linux distros.



And what does that money buy them? They've got no credibility at all because they insist on using their proprietary way of installing software instead of using the awesome mechanisms already in place.

With "love" I actually meant manpower. If DesktopBSD could only get 1% of the people working on Ubuntu to work on DesktopBSD instead, that would make a huge difference. However, we probably won't see this until Linux keepts getting raped badly over all those superfluous kernel modules it has in the default configuration. Then the security of a well-architected system environment such as FreeBSD will gain mindshare.

I just used FreeBSD as is, no Desktop one. How Can I help them ? I'll see the site and begin to use (at least try :] ). I'm no developer so my help is limited.

Yeah, one can dream. OTOH, I like the FreeBSD comunity as it is, small and full of hackers in the original meaning of the term.

it really feels great this way

none

ps: just saw that DesktopBSD is KDE based ... is there a gnome sibling ?! I like most Windowmaker, but between KDE and Gnome I prefer the latter.

 

[snes-addict]

There is nothing wrong with the FreeBSD Project focusing on servers. Looking at the recent improvements that the team has made, we see lots of talk about SMP and the new scheduler. Performance improvements such as these are absolutely vital for FreeBSD to remain a top contender for the best server OS, and with many desktop users on multi-processor machines, they also happen to benefit the desktop as well.

Also, one of the aspects of FreeBSD which I have always liked has been the cleanliness of the system on an initial install: the included software is simple and doesn't bloat. Seeing as how desktop users typically have everything from word processors to web browsers to video games, the addition of extra software to meet those requirements would make the FreeBSD cumbersome and particularly unusable for the server and cluster guys.

FreeBSD is already the best OS; perfect for servers, and (with the help of ports) is a great choice for the desktop.

 

[Oko]

I don't mean to be offensive or anything but if you refer to this
You will find a whooping 77% running LINUX
I mean like would they run their ultra expensive super computer on a lousy OS?

No offense is taken but there is something very fishy about
your link and the list. Last time I checked RedHat the same as Cent OS or SuSE were all Linux-es. Why are they listed separately
on your list from the Linux share of 77%? If you add all the Linux-es then according to your list Linux-es run 95% of all supper computers. I have used supper computers twice in my life and I can assure you none of them run Linux. Just those two are
0.4% market share of the top 500 supper computers which is according to your list is entire market share of Solaris.

Lets be realistic. All those lists are non-sense to begin with.
Probably 80% of supper computers are not even listed on any lists as they are used for Weapons design or for other classified purposes. Probably half of those are illegally smuggled into the countries in which they are used.

Let me give you one more realistic link. It is from 2005 and talks about domination of MS Windows on the server market.

http://news.cnet.com/Windows-bumps-Unix-as-top-server-OS/2100-1016_3-6041804.html

One has to be very careful about reading the list and understand its meaning. List is completely accurate in its claim that more than 50% of sold servers are sold with pre-installed Windows.
That leaves old poor Unix with less than 50% of market share.

what the table is not telling you is how many of those servers which are sold with pre-installed with Windows still run Windows. One think is sure. Even if all those server are installed with Linux the Linux market share will not be 77%. I have a hunch that the server sold with pre-installed Unix still run pre-installed Unix. I know I would not remove AIX or Solaris from my servers.

To make long story short. It is nice to live in the fantasy land. In my fantasy land 100% of computers run OpenBSD. I know I do run OpenBSD on
all 8 computers I have. In reality, actually I have met only handful of people in person who run OpenBSD.

Please do not put me now another list with the domain names and OS which they run. Bank of America where I bank has only one domain name and thousands of SUN servers. I on the another hand have 3 domain names and have OpenBSD on all three of them. Does it mean that the OpenBSD market share is bigger than Solaris

 

[Weinter]

How about this
The reason thy do not group under linux is all is probably because the group Red Hat as a separate OS because i filtered under OS instead of OS Family the link above should be better

If you don't like it how about i rephrase it Top 10
1)IBM Road Runner (Linux)
2)Cray Jaguar (CNL)Compute Node Linux
3)Pleiades (SLES10 + SGI ProPack 5) SUSE Linux 10
4)BlueGene/L (CNK/SLES 9) SUSE Linux 9
5)Blue Gene/P Solution (CNK/SLES 9) SUSE Linux 9
All the rest also Linuxes

The only Non-Linux top 10 is
10)Dawning 5000A (Windows HPC 2008)

Doesn't this statistic mean anything?
Also even if there is more powerful supercomputers out there the fact that they place Linux on a PETAFLOPS capable machine means something

Don't worry i never trusted anything from CNET they are commercial of course they promote Windows (Rich Rich)
But if you look at netcraft it is different

Longest uptime
1 tgpsubmit.persiankitty.com 327 1603 1648 1066 FreeBSD Apache/1.3.9(Unix) PHP/3.0.11 ViaNet Communications
It is a porn site what a publicity =P

Jokes aside although FreeBSD are top among longest uptime more major companies are using Linux
Google is using Linux
I don't know about you but i think the Linux hype is driving the usage up
So who says marketshare don't matter
Marketshare -> Hype -> Free publicity -> More Usage

 

[graudeejs]

adds.... man, adds...
probably the only reason i'v ever hear of FreeBSD is Gnetoo/FreeBSD project.
I didn't want to try some eta-beta which was hard to set up, so i tried the real thing.
If i never hear of Gentoo/FreeBSD.... id's still be on linux

 

[susanth]

Yes; I join your view

 

[ninjaslim]

Yahoo uses FreeBSD for everything. If anything, they should do some advocacy.

 

[Weinter]

Yahoo uses FreeBSD for everything. If anything, they should do some advocacy.

Yea Netcraft shows yahoo almost runs totally on FreeBSD

but that is the only Commercial Giant using FreeBSD i can think of

 

[susanth]

Yeah, one can dream. OTOH, I like the FreeBSD comunity as it is, small and full of hackers in the original meaning of the term.

Yes; ME too join your view about FreeBSD

 

[Mel_Flynn]

Longest uptime
1 tgpsubmit.persiankitty.com 327 1603 1648 1066 FreeBSD Apache/1.3.9(Unix) PHP/3.0.11 ViaNet Communications
It is a porn site what a publicity =P

No, it's a security disaster. Uptime actually means you didn't patch, it doesn't have any real relation to stability.

The only way FreeBSD would make a good desktop/laptop OS, is if kernel parts are split using a KERNCONF tuneable. Especially within the scheduler.
Server workloads are different from desktops. You want network throughput, forking and pipes to work as fast it can and interrupts from human interface devices are secondary. On the desktop you value responsiveness of the mouse and keyboard over disk IO.

Right now, the biggest problem FreeBSD is facing, is hardware support (mostly ACPI, sata) for "off-the-shelf" servers. This is the core audience, because FreeBSD has been made big through "fast serving with standard hardware".

I'd rather have them focus on that, then add more kludges for desktops/laptops. It's nice if it works, but it shouldn't have focus. Since they're personal computers, you can decide for yourself what you are going to buy and users should learn that it works that way, not the other way around ("I bought this el cheapo hardware, why FreeBSD not work?").
For servers, the decision to buy what hardware depends on many factors, not just cost and is usually not a single person's decision, so the broader the support, the better it will be for FreeBSD and a side-effect of fixes in that area, I'm sure some desktop hardware will start working as well.

 

[bsddaemon]

Right now, the biggest problem FreeBSD is facing, is hardware support (mostly ACPI, sata) for "off-the-shelf" servers. This is the core audience, because FreeBSD has been made big through "fast serving with standard hardware".

I beg to differ. I think 64bit and the virtualisation field would have higher priority, afaik, FreeBSD doesnt support virtualisation applications that can access real hardware. That is also reason why ISP, hosting companies...choose Linux over FreeBSD

 

[DutchDaemon]

No, it's a security disaster. Uptime actually means you didn't patch, it doesn't have any real relation to stability.

Hear hear. I wish more people would get this into their thick skulls! I may be at the other end of the spectrum (I rebuild my STABLE operating systems every month, just for the fun of it), but just leaving a machine up for uptime or treating it as a measurement of quality is madness. All it says is: "It didn't just crash". Well, since XP, even Windows hasn't done that anymore.

 

[DutchDaemon]

Yea Netcraft shows yahoo almost runs totally on FreeBSD

but that is the only Commercial Giant using FreeBSD i can think of

Yahoo! hosts FreeBSD

Name: http://www.freebsd.org
Address: 69.147.83.33

$ whois 69.147.83.33

OrgName: Yahoo
OrgID: YHOO
Address: 701 First Ave
City: Sunnyvale
StateProv: CA
PostalCode: 94089
Country: US

 

[Mel_Flynn]

I beg to differ. I think 64bit and the virtualisation field would have higher priority, afaik, FreeBSD doesnt support virtualisation applications that can access real hardware. That is also reason why ISP, hosting companies...choose Linux over FreeBSD

64 bit, is hardware, though I'm not aware of any true 64-bit related problems, other then x11/nvidia-driver.

Virtualisation won't work anyway, if the machine can't boot, or can only access SATA drives in IDE mode, or when ACPI configures a controller in a way that doesn't make sense to FreeBSD.

Personally, I don't find virtualisation anything more then hype, that allows people to run propriety software on open systems and as a result, operating systems all adopt the same sometimes broken interfaces, rather then do things in a way that's the best for the OS in question.
But as with many hypes, it takes a few years for people to adopt the next hype.

 

[oliverh]

Guys, FreeBSD is what you make out of it!

Exactly it's not one of the dying dinosaurs of the 80s - the so-called server os, and it's not the shiny desktop os full of glimmer for some bucks. It just serves _me_ and does a good job on _my_ desktop and server ;-)

I don't think there is anything different between a server and a non-server machine. If there has started to be -- and by that I mean that we secure our desktop machines less -- then we have got a serious problem. We must try better then.

Theo de Raadt, http://ezine.daemonnews.org/200603/theo_interview.html

 

[oliverh]

Yahoo uses FreeBSD for everything. If anything, they should do some advocacy.

They are supporting FreeBSD with servers etc. since years. Nokia,Juniper, etc. pp. are using FreeBSD too and they are supporting it with code and sponsorship. But who cares about Yahoo anymore? It's a dying giant from the dot-com bubble. For most companies Linux is a strategy, so advertising something BSD would be more or less futile. Most server environments are heterogenous, so you will see even OpenBSD once in a while e.g. as firewall between lots of Suns. Most of the time i don't see something like "Linux only" in professional environments, but I see it in advertising.

 

[Barnie]

If FreeBSD is not a desktop OS, then they should delete most or all desktop-related applications from the three CD-ROMs that every user can download. So it's clear for everyone that it is a server OS only. But if the distribution has every piece of desktop and multimedia software as standard, you must think it's also a desktop OS.

This thread is a little bit painful because it looks like the freebsd doesn't know what it want to be. The commitment is... Wischi-waschi (inexact).

 

[Barnie]

Please let me say something about supported hardware: I don't think that a good desktop OS must support every available hardware. Of course it is important to support categories of hardware e.g. a couple of webcams, a couple of video cards etc.

If I know that I can use a webcam and the freebsd-documentation says which model I must buy, then it's OK! I can go to the shop and buy this special model which works in FreeBSD. I don't need to run every webcam model on the planet.

Notebook: I don't own a notebook but I know from user forums that it runs on some models.

Just look the Apple Mac: they support only special mainboards - there own Mac-PC-mainboards! And not more. But is MacOS X not a desktop OS? It is! If you want use MacOS X then you have only 2 or 3 noteboooks as a choice. Less than FreeBSD supports!


But what would make FreeBSD a better desktop OS? I'm a desktop user and the most unhappy thing is, that I must setup to much to use the OS as a typical desktop OS. It's not the quality! It's only the work that I must do. It would be much easier if I can say in the SysInstall "X-User Desktop installation"... and then I can mount Data-CDs as a user. Of course I can do this with FreeBSD, but I must setup to much. This could be automated by installation. This thread is a nice example:
http://forums.freebsd.org/showthread.php?t=409

I think that FreeBSD has all components, tools, drivers and quality for desktop use! It's only the last step that is missing... setup/configuration for desktops.

 

[Mel_Flynn]

If I know that I can use a webcam and the freebsd-documentation says which model I must buy, then it's OK! I can go to the shop and buy this special model which works in FreeBSD. I don't need to run every webcam model on the planet.

You have the right mindset. Plenty of users don't.
Example

I think that FreeBSD has all components, tools, drivers and quality for desktop use! It's only the last step that is missing... setup/configuration for desktops.

This is why putting the effort in PC-BSD is a good idea. This thread is about changing focus of FreeBSD itself and I don't think it's good for the project. Let the PC-BSD developers continue their good work. They have users that will give them feedback and if
base or kernel needs patching, they can put together patches that end up in FreeBSD.

FYI: I'm running RELENG_6 on a Dell laptop, RELENG_7 on a newer HP Pavillion, have an old Gateway desktop that runs servers and compiles ports and a gateway-firewall-wireless AP-multimedia station with tv-out and uber bass speakers that mostly plays Happy Feet for the kids.
So it's not like I don't know what can be done, I just don't think developer focus should change or even that world-domination should be a goal.