Can you do a bsd-only app list?

N

Nicushor

Guest
I understand BSD and Linux too seem stronger kinda visually to make the freedom and open source message louder for now, but do you think BSD and Linux will split ideologically, will it's differences make a stronger identity for both of them?
 
This is a reddit-like question but I'll give my thoughts anyway. Linux split ideologically years ago and is no longer a Unix-like system like FreeBSD is. Linux came from a college kid's dream while FreeBSD is a direct descendant of ATT UNIX. (Yeah, I like to get my digs in.)

While Linux main goal in life is to become the next Windows, FreeBSD's main goal is technical excellence.
 
The OP's paragraph doesn't really match the title very well. Which is really the question? If it's the title, then that question has been asked a few times on the forum, and the answer is usually "Just about any app can be compiled on any platform, so there are very very few BSD-only apps".

But, a more informative narrative might be "which apps were first developed on one of the BSD operating systems, and perhaps distributed in that channel for a while before being ported to other systems?" For instance, Postgres has BSD (or at least Unix) history in it. So - such apps may very well perform a little better on the first (native) platform that existed for the app.
 
Linux very definitely does not have a single ideology. There is no single entity setting the goals for Linux. Linus Torvalds alone can't do it, RedHat or Suse alone can't do it, the various commercial development centers run by companies such as Oracle, IBM, HP, ... can't do it.

Furthermore, to claim that "Linux wants to become the next Windows" is silly. Today, Linux has 100% market share on supercomputers (every single one on the Top500 list). Windows never had that, nor did any one one OS before Linux. On the other extreme, Linux has a very high market share (probably 70%) on small embedded systems that are used for IoT and industrial control; matter-of-fact, running anything other than Linux on a small device like a Raspberry Pi or Beaglebone is outright difficult (it's doable, I ran FreeBSD on a Pi for a few weeks, but gave up). Again, this is a market where Linux is cleaning up fragmentation.

In the middle, on the desktop (which today is mostly laptops), the market share of Linux is abysmally small, about 3% or so; the bulk goes to Windows, with roughly 10% being MacOS. I think on desktops, *BSD is barely even measurable, about 10 or 100 times smaller than Linux, fraction of a percent. Linux has been technically capable of taking over the desktop market for a long time now: I started running Xwindows on Linux on a 386-40 in about 1995, when installing X meant an extra 30 floppies of the SLS distribution. It has been over 20 years, and the market share of Linux on the desktop can still be counted on the fingers of one hand, even if you work in the sawmill. The prediction that Linux will become the next Windows seems far off.
 
In the middle, on the desktop (which today is mostly laptops), the market share of Linux is abysmally small, about 3% or so;
It's even less than that. Numbers range from 0.8% to 1.5% depending on which article you read.
 
to claim that "Linux wants to become the next Windows" is silly.
I'm only quoting what I read on Linux forums and articles over the decades. "Linux is the next desktop!", they'll scream. Linus Torvalds states that when Windows does X, it means Linux wins. (I forgot what X was.)
Today, Linux has 100% market share on supercomputers (every single one on the Top500 list). Windows never had that, nor did any one one OS before Linux.
Not true. I've seen Windows on there and I think I saw FreeBSD, too. Only one or two but they were there. That other OSes aren't on that list doesn't mean much. Data and computer scientists use what they and their "friends" all use cause they all read the same papers and tutorials.

Why Windows wins the desktop and Linux wins other things is the same reason Intel won the processor wars. All the above are technically inferior to other products but big money and business deals won out for whatever reason they were. IBM wanted Intel and IBM was the big name in desktop computers at the time despite far superior Motorola and Natinoal Semiconductor processors. IBM wanted Windows so Windows won. And so on.

And so on includes all the kiddies and their games, fireworks and flashing lights to keep them entertained. In the meantime, those with real technical knowledge and the freedom to choose superiority tink away on awesome machines that don't run Windows.
 
I'm only quoting what I read on Linux forums and articles over the decades. "Linux is the next desktop!", they'll scream. Linus Torvalds states that when Windows does X, it means Linux wins.
That may be what they say. The reality is different. Linux has been the next desktop for 20 years. In those 20 years, it has managed to increase its market share on the desktop from zero to (as SirDice corrected me) 1.5%. Windows has done "X", and Linux did not win.

(About the Top500 list of supercomputers)
Not true. I've seen Windows on there and I think I saw FreeBSD, too. Only one or two but they were there.

Look at top500.org. The current list (November 2017) is 100% Linux. The previous one (June 2017) was 99.6% Linux, and had two AIX machines. The last time a Windows system was on the list was 2015, and the last time a *BSD System in 2013.

Data and computer scientists use what they and their "friends" all use cause they all read the same papers and tutorials.

You really think that purchasing decisions on computers that cost from $10M to $500M are made based on tutorials? Sorry, that's nonsense. These systems are far too expensive for making emotional decisions; the suppliers and customers of these machines are ruthless and efficient in determining the best technology.

The question of why Windows continues to have such a giant market share on the desktop is interesting and complicated. 30 years ago, when IBM introduced the PC (and made an existing fragmented industry socially acceptable), there were few alternatives for a low-cost desktop OS, and Windows won by default, for lack of competition. But beginning in the early 90s, there were other operating system choices. I bought my first PC in 93 or 94, and immediately ran Linux on it: 386BSD was not useful (it had no Xwindows, and no way to get there with the Jolitzes blocking the road), and a BSDi license cost about $1000. By 95 I was running Linux with X and a GUI at home, and could do real productive work on it (which at the time required graphical tools for data analysis). Yet, in spite of there being alternatives to Windows on the desktop for over 20 years, it continues to succeed. Why? Because those alternatives are not viable for the vast majority of users. The reasons for that are complex, but the single biggest component is ease of use of Windows and MacOS: You pick up your laptop, preinstalled with the OS, from either the store or your corporate IT department, and from then on things are very easy if you remain on the same OS.
 
You really think that purchasing decisions on computers that cost from $10M to $500M are made based on tutorials? Sorry, that's nonsense. These systems are far too expensive for making emotional decisions; the suppliers and customers of these machines are ruthless and efficient in determining the best technology.

Completely agree.
 
That may be what they say. The reality is different.

I know that but it doesn't change their goals.
Look at top500.org.
I was replying to you saying Windows and other OSes were never on the list but you actually said they never had 100% ownership of the list. I mis-read.
You really think that purchasing decisions on computers that cost from $10M to $500M are made based on tutorials? Sorry, that's nonsense.
I didn't say that. I said they all read the same papers, meaning research papers, but I probably shouldn't have said tutorials though research papers do explain how they did things which is what I meant by that.

I can understand why average users choose Windows but I will never understand why professionals choose Windows for professional use. Seeing the Microsoft logo in the background behind the TV weatherman drives me nuts. Every time the Windows POS at one of my restaurants has issues, I want to tear my hair out.
 
One reason may be that there may be a Windows program that does what they need that one just installs and makes work.
Even if it's not set it and forget it, various software will be created by companies that charge a lot of money, provide training and support, and so on.

Whereas with opensource, whether BSD or Linux, paid support is often limited, or one is expected to RTFM. To many of us, the computer and system software are ends in themselves, but to the majority, and there's nothing wrong this--I'd rather my doctor spent his spare study time figuring out to fix me--they are simply a means to an end, and usually Windows is an easier means. I suspect (not researched at all on my part) that most medical, dental, legal and so on software, is written for Windows, and there are opensource equivalents, they'll require a lot more research and work to have them function.

Obviously a generalization, and I don't know how true it is for more academic sciences, but I know that every time I've seen a computer in a medical or legal office, it's been running Windows.
 
I'm not talking end users. I'm talking the originators, the creators of the software. What knowledgeable, experienced programmer would choose Windows over any other operating system when asked for their expert advice?

I'll use my restaurants as an example, again. We're a large chain and every cash register, maybe over a hundred thousand throughout the world, is a Windows computer. The user interface, the cash register, is custom made by some outfit hired by the main office but there are tech support people for us and it's all they work on. It's one screw up after another and updates to Windows or the POS software is met with dread. There are times when we have daily unknown issues, slow downs and we're still told that the best fix for any problem is to reboot the computer and it is wise for us to reboot every night or first thing in the morning.

So the question is not why Weatherman Bob is using software that runs on Windows but why did the software company create it to run on Windows and not Unix/Linux/BSD? In our case, as I said, the interface for us is not Windows at all except to reboot the system. Tech support even gets in remotely when they have to.

I'm betting I'd save a hundred dollars on the OS alone for each system and not need to upgrade my hardware as often if I was running a proper unit running FreeBSD. In fact, I am positive it would be true.
 
I'm talking the originators, the creators of the software. What knowledgeable, experienced programmer would choose Windows over any other operating system when asked for their expert advice?
One who wants to get paid? Because with 50x more Windows computer than Linux computers on the desk top, the chance of getting a paycheck for Windows software is significantly higher. Obviously, this argument only applies to desktop software.

I'll use my restaurants as an example, again. We're a large chain and every cash register, maybe over a hundred thousand throughout the world, is a Windows computer.

I agree. Why anyone would use Windows for an embedded system is hard to understand. Your cashiers don't use the register as a general-purpose computer, they only want to ring up sales and do the associated functions. They don't care whether the underlying OS is Windows, Linux, BSD, VxWorks, FreeDOS, or cp/m, as long as the machine functions. I can come up with at best semi-plausible theories to explain Windows on these machine. By the way, the same goes for display kiosks that are used in places like airports: why they boot Windows, just to run a display application that immediately takes over the screen makes no sense. I don't play in those markets (neither as a user nor as a developer), so I can't change that.

BUT: What's good for embedded systems is not necessarily good for IoT, supercomputers, or the desktop. I think people continue to use Windows (and to some extent MacOS) on the desktop because they are sane and rational: it is, for most people (*), a more productive and efficient way to get computing done. (*) Footnote: most people are not computer programmers, software engineers, or computer scientists.
 
One who wants to get paid? Because with 50x more Windows computer than Linux computers on the desk top
You missed my point sort of. We're talking of custom software, written from scratch in a professional environment where the operating system does not matter. The weatherman example is one I am close to and I can attest they bought those computers new to run the software but it could have been any kind of computer. They bought Windows computers cause the software was Windows. If they were told they needed FreeBSD on those systems, they would have bought FreeBSD systems. Thus the question we both can't answer, why is the software on Windows when professional programmers are given the choice for professionals who will happily use what is given to them?
 
If it is for an embedded environment, where a computer is used for a single purpose with canned software, and not as a general-purpose machine, then I agree. For such a use, Windows doesn't make sense to me. Your weatherman example seems to be such a case: they probably have a dedicated computer to run those monitors behind the weatherman.

A different situation is software that is intended to be used on general-purpose machines. For example, my wife (who is very computer literate, but not employed in the computer field) has a Windows laptop. She runs a wide variety of software on it, from the generic (Outlook for mail, Word/Excel/Powerpoint for office stuff), to some rather specific things (hardware-specific lab interfacing programs, specialized data and image processing software, expensive packages that are very tailored to her profession). Why are those written for Windows (and in some cases also available for Mac)? Because the vast majority of general-purpose workstations (desktops and laptops) use Windows, with a sizable minority MacOS. These are programs that people buy and intend to use, with many others, on their workstations, not dedicated hardware.

Actually, many years ago I interviewed at a company that made dedicated systems (ocean sonar data analysis), and used only Windows. Their justification for using only Windows was that it is much easier to find programmers with Windows experience than with Unix experience, and staffing was their biggest problem. That argument seems somewhat sensible, but only somewhat. This was in the time before Java and Python were popular, and when POSIX compliance was still ways off, and where real world C and C++ code was still highly platform dependent.
 
Back
Top