FreeBSD 5.0 Available 372
Vegard writes "Although not yet officially announced, the 5.0 version of FreeBSD is beginning to appear on the FreeBSD FTP site and mirrors world wide." Congrats to the developers. Update: 01/19 17:44 GMT by T : Some more detail -- Dan writes "Scott Long of FreeBSD Release Engineering team has officially announced the availability of FreeBSD 5.0 release. Improvements include second generation UFS filesystem, GEOM, the extensible and flexible storage framework, DEVFS, the device virtual filesystem, Bluetooth, ACPI, CardBus, IEEE 1394 and many more! FreeBSD is also available on 64-bit sparc64 and ia64 platforms."
Oh, hooray (Score:5, Insightful)
Nice linking (Score:5, Insightful)
Also by using the mirror list, our US friends wouldn't have to download from a server in Denmark, but maybe a local one instead. Oh, well I guess that's just me, but I really think that in the lengthly, time consuming screening process of each article, someone would show a bit of responibility, knowing the effects, posting a article with links have.
Re:My review of FreeBSD 5.0 (Score:5, Insightful)
NTFS is intentionally underdocumented, so most attempts to support it in other OS's have been mostly reverse-engineering attempts. You could sign an NDA, but probably wouldn't be able to write free code with that information. Do not blame FreeBSD for not supporting undocumented features of another OS.
If you have an example of any non-Microsoft OS that can install on NTFS, please prove me wrong!
Hasnt this happened before (Score:5, Insightful)
Announcement. (Score:4, Insightful)
Uh, maybe there's a reason? Like they want to finish pushing everything out to the mirrors?
--saint
FreeBSD 5.0 NOT released (Score:5, Insightful)
Murray Stokely writes "We have gone over this for the past 2 releases now. I thought I had made it clear that you were not to publish information about FreeBSD being released until you saw a signed PGP message from one of the release engineers. Are you trying to help the spread of trojanned copies of FreeBSD? The release is not ready yet, and will not be until the front page of FreeBSD.org is updated and a PGP signed announcement message is posted to announce@FreeBSD.org."
Unless the rules have changed, slashdot screwed up again.
Early annoucements (Score:5, Insightful)
Despite being idiotic, this behaviour is really harmful. FreeBSD takes care to let their mirrors prepare for the traffic peak when a new version is released. The early "announcements" on slashdot of course mean that the people managing the mirrors - voluntarily, people not only FreeBSD but lots of free software projects depend on - don't have this time to prepare, and might get major problems, which in turn might mean that they decide not to support FreeBSD and other projects by providing bandwidth for free any more.
Unless this is some funky plan of VA Software or whatever their name is this week to push SourceForge, it would be really nice if slashdot could just stop damaging the Free Software infrastructure.
Keeping it a secret. (Score:3, Insightful)
Re:A few mirros (Score:5, Insightful)
And, of course, instead of downloading ISO images, consider using CVSup to save time and bandwidth. Or at least don't download all ISOs - you don't need all packages, and installing the stuff you want from the network works without any problems.
Mirroring with peer2peer networks (Score:4, Insightful)
Re:Early annoucements (Score:2, Insightful)
Sure, the maintainers could restrict access until everything is ready, then announce the release and open up the servers. But why should they have to?
Jesus people, it only takes a little common courtesy to wait until the announcement is made. Is it really that important to scoop even the project's own site?
Re:My review of FreeBSD 5.0 (Score:5, Insightful)
Re:Oh yeah ! (Score:5, Insightful)
This is a valid point that I haven't yet seen addressed: which is better for the average user, 4.7 or 5.0?
There have been many, many changes to the code in 5.0, and there are bound to be more than a few bugs. If you're running a site that can have zero downtime, and you don't have redundant servers, don't bother switching to 5.0, it's simply not ready yet.
If you're a home user, don't mind a few make buildworld, make buildkernel, make installkernel, make installworld sequences, upgrade. There's enough new that you'll enjoy it, and there's enough stability that you probably won't notice the infrequent bugs.
If you're asking yourself "Why should I upgrade when everyone says there's going to be bugs?", the answer is simple: the bugs can't be found without testers, so everyone on the team needs your help to find them quickly. If you encounter a bug, file a PR, and maybe even try publicizing it on a mailing list. Letting the developers know that bugs exist is the first step in getting bug-free code.
Re:Nice linking (Score:2, Insightful)
MOD PARENT DOWN (Score:5, Insightful)
I mean, I expect this from one of the junior "editors", but Cmdr Taco? Come on.
Re:Release Notes (Score:5, Insightful)
The BSD license is a beautiful thing. Software that carries the BSD license can really, seriously, no-shit change the world for the better.
Re:Keeping it a secret. (Score:3, Insightful)
Thank you for that.
I just have one thing to add here;
I've been reading all these comments from the BSD crowd here in awe. I mean, all this hostility over... what? An announcement that linked to a PGP-signed release announcement. The ISOs are on the servers. The time to rejoice is nigh! But no rejoicing from this crowd. No "Awesome new features ... I can't wait to test this on my home rig ... " postings; just adolescent whining.
Seriously folks; you respect the FreeBSD development team, right? You respect their programming talents and their combined decades of computer, operating system, and networking experience, right? Do you really take them to be this naive? Would you really have us believe that they would roll release-grade, Version 5.0 (no RC-*) CD images and make them public when they weren't ready? Do you really think they'll be at all SURPRISED when people start to notice, download, and tell all their friends about this release? Don't you think they have a solid, stable (FreeBSD) FTP server pumping out these requests, properly configured with reasonable user/transfer limits in place, and QoS on their upstream bandwidth? If you're that unsure of FreeBSD's ability to handle high loads - why are you downloading it?
It was inevietable that this would find its way to Slashdot. That's how Slashdot works. It's been seen time and time again. KDE, GNOME, Linux Kernel, XFree86, [Open|Star]Office, or any other project of significant magnitude (and interest) - the release files are made publically available, someone notices and the Slashdot editors respond to the influx of "It's here! It's here!" submissions. As a result, Slashdot is very often the first place to find out about new software updates. Is this really 'news' to anyone?
Sure, they could link to the mirrors, but not doing so isn't by any means a conspiracy, it may be poor taste, but it's the same taste that links directly to kernel.org when a new Linux kernel is released. It's been pointed out to me more times than I can count that Slashdot readers are "IT professionals" - so stop talking about being professional and act like it. Download reaponsibly; use a mirror.
I'll download a mini-ISO later, when the tide has ebbed, and install it at my leisure.
</RANT>
Re:Nice linking (Score:3, Insightful)
As for Ext3, it's been improved, not hacked. And counting its age it's already reliable. Compare to UFS back to the same age.
As for scaling, I doubt that IBM made a mistake choosing Linux as a replacement for AIX. Otherwise, why IBM did not do the same or similar step as Apple did? The answer is simple: IBM doesn't trust to non-scalable design of BSD.
Recent news from SGI (Linux on new SGI servers) just proves it.
Re:GEOM sounds interesting (Score:1, Insightful)
*BSD already has a virtual filesystem layer in the kernel, but it's not accessible from userland.
JJ