Slashdot Mirror
FreeBSD Upcoming Release Dates
"The following constitutes the FreeBSD provisional release schedule into mid-2002. The release engineer and the FreeBSD project have always committed to a "best effort" where these dates are concerned and will continue to do so, but all dates are also subject to change without notice given prevailing conditions in the source tree.
2001-04-20: FreeBSD 4.3 released
--
2001-08-20: FreeBSD 4.4 release date
2001-11-11: FreeBSD 5.0 release date [EARLY ACCESS]
2001-12-15: FreeBSD 4.5 release date
2002-03-15: FreeBSD 5.1 release date [GENERAL ACCESS]
2002-04-20: FreeBSD 4.6 release date
2002-07-15: FreeBSD 5.2 release date [BEGIN -STABLE]
The dates for 5.x are particularly SWAGish in nature and I'm sure that a lot of people in -current will scream and jump up and down about a November release date, but all of us here also know that when we wait for a dot-zero to be "ready". It turns into a multi-year debacle, during which time *nobody* is pleased, and we finally end up rushing something out the door at the last minute anyway except that it's rushed because it's overdue and everyone's upset, not because we had an aggressive deadline. 5.0 certainly won't be perfect, nor will it achive all the goals we'd originally envisioned for it, but then that's been true every single time we've done this and somehow we've survived. :)
I'm also proposing that we branch fairly late in 5.x so that we have quite a bit of time to finish a few works in progress, but we still have some time to talk about that. The 4.x schedule should be fairly accurate as it stands since no significant issues are forseen there.
- Jordan"
The one thing that 5.0-RELEASE and future 5.x releases will bring is a completely revamped kernel and SMP support. Instead of having one giant central lock, the new SMP code will introducing a scheduler lock and per-CPU idle processes. The SMP project page for FreeBSD can be found at http://people.freebsd.org/~jasone/smp/. It looks very promising... mostly when combined with a dually Athlon (or even a dually Alpha) machine :)
FreeBSD's C2 security certification is horrible also. Even NT can do better than it! For systems that need any level of security FreeBSD is not the answer.
First - NT certification was based on a machine that did NOT have a network card. That was NT 3.51.
Second - To get C2 certfication, someone has to PAY over 20,000 dollars. As you MUST have knowledge of someone how paid, will you share it? Oh wait.... YOU ARE A TROLL who has no facts to backup your claims.
Feel free to post again when you have some facts, k?
If it was said on slashdot, it MUST be true!
I'm sure you heard about FreeBSD's merge with BSDI. FreeBSD has very close ties with BSDI now and outwardly thats a postive thing, but it really isn't. I'm not one to preach doom by association, but I'm afraid FreeBSD has doomed itself by the move. BSDI is best known for its archaic OS and high cost of ownership.
This is irrelevant since FreeBSD is picking from BSD/OS not the other way around. IOW, FreeBSD is not dependent on BSD/OS. They have no dependence on BSDi which is no more; Wind River has a portion of it now.
FreeBSD's C2 security certification is horrible also. Even NT can do better than it!
C2 certification is a joke. The U.S. Government does not even seem to follow it since they use NT 4 in many places without it conforming to C2.
Actual memory bandwidth performance is a fraction of all of Sun's offerings...
Memory bandwidth is a hardware item. Sun's architecture is much faster than x86. OS's have little control over the design of the memory bus.
the multiprocessor support is a joke since it has a poorly implemented semaphore locking mechanism.
This has been acknowledged and is why SMPng is being worked on.
But even then the open source nature of FreeBSD will cripple it - all open source code does is allow poor quality code be used in mission critical systems.
You are now comparing FreeBSD development style against Linux. You obviously have never done your "fair share of testing on the VM scheme" as you stated. If you had, you would know how FreeBSD is developed. Thankfully, it is nothing like Linux, but you should already know that.
Recent tests done by SysAdmin magazine have *BSD coming in last place in terms of performance when paired against Linux and Solaris.
Without even basic tuning for FreeBSD? They did not even have SoftUpdates active. Sorry. The test was flawed.
I think its time to let FreeBSD be merged out of use and let operating systems such as Solaris, NT, or even Linux take its place.
With the incredible growth FreeBSD has been experiencing, your opinion is a joke.