Slashdot Mirror
Why is BSD Not As Popular As Linux?
hill writes "An article over on Economic Times explains why BSD is as not as popular as Linux. Both use an open-source model, but Linux demands the user community to disclose modifications on its source code, while BSD allows its users to make proprietary changes. The current size of the BSD community is estimated at 2 million, with Linux being around 10 million. This is definately worth the read for anyone interested in comparing the two operating systems. " I'm sure we have a few opinions on the subject.
First, we should look at it this way: there are 12 million people that use a free unix-like operating system. Most if not all opensource applications run equally well on both. One (Linux) is an implementation from scratch, the other (Free/Net/OpenBSD) has royal blood as it is the direct descendant of 4.4BSD which itself descends from Unix. This should keep happy both the new army of coders that like to toy with new concepts, and the traditionalists for whom 30 year old code doesn't mean outdated, but proven and stable. Both points of view can be defended I think. We therefore have 12 million users and users-developers of free unixish applications, that's great and was absolutely unthinkable 10 years ago!
:). Same for NIC drivers usually (hello, donald becker, do comment weird things :-). But the linux kernel is full of good and new ideas.
:) Long and happy life to all the linux and free/open/netbsd hackers, be it kernel or office applications writers :)
As for the technical side, I keep having to look at both the linux and freebsd kernels as part of my work; they are good references. Both have very good parts. I have to say that usually, the solution adopted by FreeBSD is simpler and a lot more commented/documented (take the bogomips case for example; people are starting to wonder what will happen if the cpu speed changes at runtime, how to detect and recalculate it, etc; freebsd spins simply by looking at changes in the hardware clock counter. simpler
So we need both if we want to keep the high standards we are used to have in the free unices now. That was my original point
There is no billion dollar IPO backing the hype about BSD.
The hype will come when:
1) There is a billion dollar BSD IPO.
2) When the BSD community starts explaining the biggest advantage of the BSD licence to Multinational corporations. That advantage is, you can choose to HIDE your own source code if you wish. (Get them to at least start supporting OpenSource. Once they find its not as bad as Microsoft says, they will keep coming back for more. Like drugs...the first hit, we'll give ya free.)
3) Some cleaver BSDers (Hi Pat!) start whispering in Wall Streets ears "Feel that you mised out on the Linux IPO frenzy? Take heart, here is BSD...the next big IPO launchpad. It runs Linux binaries, its OpenSourced, AND the licencing difference over Linux doesn't cause the heads of the lawyers in your IP departments to spin about."
When the first IPO of BSD is successful, then you will see the people who use Linux instead of the word OpenSource, refer to BSD as OpenSource...and Linux also. And, the more OpenSource is out there, the better for BSD, Linux, Apache, Sendmail, vi,
NO CARRIER
If it was said on slashdot, it MUST be true!
1. Time. BSD was held back by the ATT lawsuit and Linux already had so much mindshare when that was over.
2. The BSD license doesn't enforce the quid-pro-quo. This is a real sticking point for me personally. When I put a lot of work into something, I like to be a partner in a free software development, not someone's unpaid employee dupe. But I feel like a dupe when somebody takes that work private, makes proprietary modifications to my work and doesn't return their modifications to me or the other free software authors who gave him our work.
Unfortunately, history shows that without a license requirement the return of code doesn't happen. Most of the workstation Unix systems are BSD-derived (although these days there is more System V in there) and all of their X servers are derived from software under a very similar license to the BSD. Try to get the source code for those systems. Sun only released its modifications to the BSD system recently, 10 years late, and then under a license that would not allow their reincorporation into the BSD system as free software! Most other workstation manufacturers didn't bother to release source at all.
So, I am more likely to put work into a GPL project. It is possible to take the BSD system and GPL it. The new BSD license and the GPL are compatible, and you can GPL all new work that you do, and in general establish a GPL source thread. But that would annoy a lot of the long-time BSD folks.
Thanks
Bruce
Bruce Perens.
This applies somewhat to users and to an extreme with developers. As a user, a question revealing that you don't know UNIX, not just *BSD, is enough to have you shouted out the door. As a developer, unless you're a 20 year BSD veteran, suggest an idea or ask where you can begin to help and you should be prepared to be stomped on. Hard and repeatedly. Largely by many of the project principals.
Review some of Matt Dillon's contributions to FreeBSD in the mailing lists. He's repeatedly helped to pull large portions of FreeBSD up to and even past their Linux equivalents. Then consider the rationale behind the community's treatment of him.
A similar type of treatment resulted in the split of NetBSD and OpenBSD. Again, reviewing their mailing list archives shows that this kind of childish animosity and cliquish cult behavior abounds.
To the contrary, it takes all of five minutes to find something to do for Linux and to find a mentor who will help you find your way to the in crowd the first few times you've got a core-level contribution to make. They give you the benefit of the doubt as a new contributor, reviewing and considering your contribution, not your credentials or your ability/willingness to pose as a BSD veteran long enough to be heard.
Frankly, it's surprising that this group exists outside of acedemia at all.
As self-fufilling prophecies go, this is another one. BSD continues to be less known, because it is less known. Over half of those same college undergrads I knew in computer science and engineering got hands on experience with Linux before they graduated, myself included.
BSD continued to languish in the realms of unknown software.
Many of the undergrads went out into the work force and are now doing jobs where they can at least provide knowledgable input about Linux. Many of them went to find jobs specifically where they could work on Linux systems. There was no similarly large pool of individuals who knew BSD amongst the dozens of fellow students I knew, including the systems operators (I was one) for our UNIX systems, or much in the faculty. Perhaps a few people seemed knowledgable about BSD, but they didn't talk about it much, because people knew more about and were already interested in Linux.
For the most part, colleges provide the ground where our next generations of individuals in the computer industry learn UNIX-based OS's and determine what technologies they will bring to their initial workplaces. If BSD is as absent from most colleges as it was from mine, BSD won't catch on, because many of the people who would use it will not know about it.
B. Elgin
B. Elgin
"Read at your own risk; feel free to ignore."
The first free version of BSD (Networking Release 2) was distributed in June 1991, but got tied up in lawsuits from 1992 to 1994. By the time that was cleared up, early versions of Linux were already available.
--
"But, Mulder, the new millennium doesn't begin until January 2001."
send all spam to theotherwhitemeat@ropine.com
Yes, I fully agree that *BSD has numerous merits and that this is something that could easially be flamebait. However, Linux is just a bit ahead of the game (most likely due to the ATT crap), and it has caught the media attention. Its one thing when something comes out of a university, but when "the young finnish student created his own operating system because he didn't like what was out there" grabs peoples' attention, it seems to be a more heartwarming story.
Ok, now let's look at this part about the "heartwarming". Yes, we as techies like to look at things for their technical merit, not their popularity. As I said above, they are both quite good and nitpicking is justified, but almost pointless. Wall Street knows about BSD, but they just don't really care. There are no Red Hats or VA Linux companies for *BSD, and Microsoft doesn't acknowledge *BSD (from what I've seen; tell me otherwise, please). I remember a few months back reading in the Wall Street Journal an article on how "If you thought Linux was the underdog, BSD is underground". People have read about it. They don't care.
I guess I could rant about this for a while, and I'm sure people will flame and argue with this. The point I'm getting to is that Linux just has the head start on the public eye, and it is simply the center of a lot of attention. *BSD may be better than Linux. Linux may be better than *BSD. However, they're both quite good and certainly much better than that software from Redmond. RedHat and VA have both brought Linux to the public attention. I fear, however, that if there is a 'Red Hat BSD', it will just confuse people; it could turn out to be a good thing, but it could also just bring us back to the 80s when there were 20 different platforms and little in the way of 'cross platform' standards.
Rant, rant, rant. I better stop before I talk in circles
--
Never hit your grandmother with a shovel, for it leaves a bad impression on her mind...
Initially Linux is decipherable with IRC help, mailing lists and on-line docs. Whereas, BSD takes some previous understanding and the man pages on BSD assume Unix know-how. The LDP HOWTOs are written for the uninitiated and that is a major reason why Linux appeals to people more.
This whole Unix rebirth is very new. So people new to Unix will choose Linux first. Once they realize there is something objectively more mature for advanced purposes, they may consider a switch.
I started out learning Unix by trying out the various Linuxes. Now I've settled on OpenBSD b/c security a huge issue for my business. And my level of security must be high. That is not so for other people. While I'm a huge proponent of security and privacy I feel most people can be by with their Windows computers if they have a good firewall/ip-masquerading gateway installed that runs either Debian (most secure Linux), FreeBSD or OpenBSD. With the growth of the home LAN, such a configuration is a no-brainer and you can install it on some relic of a PC that you thought could only have been used as a door stop.
If people want to try a more stable desktop system; I usually will configure a system with KDE and FreeBSD or Debian for them. In terms of application capability they are about the same so it boils down to the person's taste in licencing features.
But for someone who wants to go it alone and install and learn as one goes, I recommend something easy to install like Corel Linux or Caldera Linux (no not Red Hat which I recommend for the corporate environment).
If a company came along that made a BSD easy to install and use it would be a truly awesome product; that is what Darwin and MacOS X is all about and they are awesome but expensive. If you have the money for Apple's new OS, the advantages for using a BSD based system speak for themselves after you've used them for a while. Unix gurus don't need convincing. They either only run BSD because it's 'real' Unix or they only run Linux because the GPL is preferrable. The arguments about Linux having more applications and better hardware support are, of course, silly because if that is the basis for an argument then we'd all be using Windows instead.
The bottom line for new users is documentation that's easy to access and meant for them and an install process that people perceive as easy (i.e., it has a GUI). Linux has it and BSD doesn't.
The article contends that Linux is not as sophisticated as BSD. While I agree that certain features of BSD might be more advanced (e.g. from a brief chat with one of the NetBSD folks, the UVM sounds cool), Linux is braving uncharted water in a number of previously shunned areas (I was stunned to find, for example, that I can choose to enable a kernel-based static http server in my Linux kernel as of 2.3.x). This willingness to break with UNIX tradition is what sets Linux apart, and frankly is the reason that many of us like it.
I also like BSD (I was a huge fan of 4.2, back when Ultrix was 4.2 with the serial numbers filed off). BSD has a tradition of stability and innovation that is hard to match, and look forward to a world where BSD and Linux are equal participants in the operating system development community. But can we stop pretending that one OS is "better" than another, and focus on which OS is right for a given task/environment?
It's a small world and it smells funny; I'd buy another if it wasn't for the money; Take back what I paid (SoM)
And all of these things will run, mostly without modification, on BSD.
Who cares which kernel is used! That's a small (but very important) part of the whole picture. The important thing is that we are rapidly developing more and more user-space stuff that will run on any modern UNIXy platform -- whether its Linux, FreeBSD, or the Hurd.
Linux's success helps to insure BSD's long term viability. Don't forget it. From some stuff I've seen, I gather that the core *BSD teams are well aware of it.
-- Slashdot sucks.
3. Linux supported IDE long before *BSD did, and these were the inexpensive drives that the masses had.
To that, I would add, now that Linux is way ahead,
4. Linux is way ahead in market share, and just like the internal combustion engine, the incandescent bulb, Microsoft OSes, and other less than optimal technologies, once a product is entrenched with sales networks, R&D networks, customer support networks, etc., it takes a vast improvement in the underlying technology to overturn the economic advantage of sticking with the status quo.
Any open sourced kernel offered just that sort of advantage over windows (to developers) and with the internet paradigm shift to vastly increase the numbers of servers, open source (especially of unix) on the commodity platform offered a compelling enough functionality jump to create the new market/new standard. But since Linux won the initial sprint, expect it to continue its hegemony. The BSDs have a chance of gaining share by being very compatible and utilizing highly transferable skills, so all is not lost for them, but things are often the way they are for many small good reasons, not by random chance, nor for one reason.
If moderation extended to articles I'd mark this one down Score -1: Flamebait. Not necessarily because it's inflammatory in and of itself (although the last comment was practically asking the soapboxers to come out of the woodwork), but because we've all seen this ground hashed over again, again, and again: "BSD license sucks! Disinfect the GPV! BSD==Proprietary! GPL==Commie Facists! BSD users are elitist jerks! Linux users are clueless idiots! BSD is k00l! Linux is k00l!" /. to try for a little more discretion in posting articles and to try to cut a little of the hype and bullbaiting. Not that the odds are in favor of this occuring..
I've browsed the first few comments and found that, unsurprisingly, they say nothing that hasn't already been repeated ad nauseum. I'd like to ask
Luck,
Daniel
Hurry up and jump on the individualist bandwagon!
Oh, you mean the discussion that includes
i.e., that Linux later accepted the change:
As for "others can be found elsewhere", please give references - perhaps they're also not bad implementations.