FreeBSD poised to take desktop away from Linux May 15, 2006Posted by rjdohnert in Opinions.
Many people from my previous post asked me why did I make the statement that FreeBSD can take what limited desktop marketshare Linux has away. Well I made the statement because I believe it can happen. As I posted earlier thats the FreeBSD groups plans and in my opinion those plans seem feasible. There are a few factors that will inhibit Linux adoption in the future as mainstream and if FreeBSD shapes itself up, Linux is in deep trouble.
GPL 3 — My biggest problem is that the GPL 3 is going to restrict developers and content creators from being able to properly protect their assetts through patents and DRM. For initiatives like online music stores, video stores and such which a lot of consumers use, i doubt we will see many developers of those services port to Linux if it goes under the GPL 3 for the fact it takes away their rights to do with their code as they see fit. The BSD license does not have those restrictions in place and so its a non-issue in regards to the BSD community, while Im sure they care about it, its been my experience that they are more aware of the real world and less political group than the Linux community.
While the GPL is considered an Open Source license I dont consider it to be so. The GPL is too political. All you have done when you trade a proprietary license with the GPL is trade one set of restrictions for another set of restrictions. Restriction is not what I think when I hear the words Open Source.
Lack of binary cooperation — 2nd problem, not less important than the first, not more important than the rest. The Linux kernel maintainers and such do not like binary modules, according to a letter received by a live CD distributor its also against the GPL to link the binary ATI and NVidia drivers to the kernel. Also its been said that the maintainer of the USB subsystem on Linux plans to restrict the ability of the stack to load USB modules that are binary and not open source. For developers of wireless lan devices and ISDN gear, thats bad news. This is a case of the Linux developers trying to impose their will upon hardware and software developers, never a good thing to try to stroing arm people you want to actually help you. Oh sure, they have good excuses, like "Oh you cant provide proper support", but alas they are just excuses. the BSD community does not have these types of restrictions in place yet. NVidia produces and supplies drivers for FreeBSD and they work rather well.
Infighting and self cooperation — While yes every community has its fights, the BSD community is no exception, the BSD community tends to get things done rather quickly and they can agree on standard tools. The BSD guys can at least agree on package management tools. Thats one argument, among many, that the Linux community cant seem to agree on. And its a shame.
The feature set of FreeBSD is on par with Linux. Wireless works with FreeBSD, while its a chore to get working its equal with Linux to get wireless working. There is vendor support for common video cards. Now that they have a Java Developer kit and Java Runtime available for BSD there goes the native Java stack for compatibility and Oracle supports FreeBSD, even Microsoft supports FreeBSD to a certain point. Most open source software has BSD ports and MySQL and PostgreSQL have BSD ports. FreeBSD also has a Linux compatibility layer and the fact a lot of the useful OpenSolaris stuff is getting ported to FreeBSD is also a positive sign. As of right now, I consider it to be on an equal playing field.
Will FreeBSD kill Linux? Absolutely not. You will have your die hard Linux and GNU bigots who will stick with Linux and shun BSD. I just personally feel the BSD community if they get their share up will have much more success with hardware and software vendors than the Linux community has done thus far. If you look at the state of Linux hardware while its gotten better its not because of vendor cooperation, its called reverse engineering.
So what does the FreeBSD community need to do to ensure success? Its very simple. Create a decent, easy to use installer. Create easy to use configuration tools. As they said in the original story get plug and play working, get desktop features that users have come to expect working and start cooperating with hardware and software vendors and dont try to force anyone to do anything they dont want to do. In other words dont play politics..