Post a Comment
The unfortunate fact is that testers are a very very critical asset to any kernel development process.
With the rise of other free unix-like (and BSD derived) kernels and strong competition between them, the "marketplace" is looking decidedly rosy if you're a tester, and quite sour if you're a sub-par kernel development process.
Now it is all well and good to have unstable development cycles, however the more unstable it is, the less testers it will get and the less testers it gets can easily mean the less stable it gets.
Also, it doesn't help if the unstable cycle is so long that the production quality release becomes obsolete and lacking new features or hardware support.
My comment might seem off topic at first glance, but one of the many little things that I like about DragonFly is the fact that debug kernels are installed by default, even in releases. While I'm sure that is the case for FreeBSD betas and release candidates, I'm sure that it has proven to be a bonus to the DragonFly devs to have had debug kernels being the default for relases as well, when something goes wrong.
Well telling the truth they first said it was released on middle of august then they decided to avoid to put dates on their schedule.
We are in the middle/end of september and we have beta-5, is it a record ?
Frankly I understand that a BETA could have bugs, but a beta-5 that crashes, at least if that it's true, is not a good sign of the quality.
The fact that it is delayed so much is not a good sign, imho.
I still prefer netbsd.
If there is anyone that beta4 crashed for right away (or whenever it was), chances are that it is on a very particular setup.
Of course that doesn't excuse there being bugs in the code, but I haven't seen anyone crash beta4 really.
It is delayed so much because they want to ensure quality of the actual release. Unfortunately, not everything can be fixed if you want to actually release it.
I don't think it is good to constantly delay it, but I don't see any other choice either.
I like FreeBSD and all, I don't see anything wrong with useing it. But theres people out there that is NOT going to install something that thet can't understand what it's telling them yo do upon instllation. Now if it had a GUI/point&clik I would say they would be more BSD user's.. cuz the pub wants something that will work.. and WE know BSD works right? So it's not going to go any where unless it gets a GUI upon installing.
Hack
Great to see the development taking there time with this release.
I have been using 6 since before the first beta and I have found it to be extremely stable with the only exception being my USB Hdd which makes the system flaky. Other than that its been great.
Keep up the great work guys 
I'm seriously thinking of switching to FreeBSD 6.x or Ubuntu in a few months.
I'm growing tired of Debian. I don't know why they don't "fire" the volunteer maintainers that don't do anything.
There should be an *automated* system that posts a "hall of shame" for Debian package maintainers that fail to maintain their packages. And anyone that persistently makes this list should have their maintainer status removed so that someone else can take over without a fight.
Those quite different choices, though I can't say FreeBSD is not great on a desktop - it is!
Maintainers - that is certainly one of the strong points of FreeBSD (although I guess this is true of Ubuntu as well). If you go with pkg_add -r (and forgo using ports), than you'll get _slightly_ outdated packages, though I think the package cluster builds packages once a month or so from recent ports. If you go with ports, www.freshports.org is the place to look first. Yes, there are broken packages, but at least they are tagged as such, and periodically ports with unresponsive maintainers are purged. Also, broken often means that it won't work on one supported branch only out of the box. For instance, openoffice.org-2.0-devel is marked as broken, but works for 5.x and 6.x. It is marked broken because if you run 4.x, you need to update rtld before you can build it.
Good luck, either way.
Frankly I understand that a BETA could have bugs, but a beta-5 that crashes, at least if that it's true, is not a good sign of the quality.
Here I have a *very* common Dell system [Dimension P2400] and it runs like a charm. So why do you think that your experience reflects BETA5? It's reflects BETA5 on *your* system. Says nothing! ;-)
Now if it had a GUI/point&clik I would say they would be more BSD user's.. cuz the pub wants something that will work..
Why the heck would FreeBSD wants that public? I think FreeBSD developers make things like they want them to be. If other people want to use it.. fine! If they don't want to.. fine as well! It has nothing to do with business but with quality software.
Well telling the truth they first said it was released on middle of august then they decided to avoid to put dates on their schedule.
We are in the middle/end of september and we have beta-5, is it a record ?
Frankly I understand that a BETA could have bugs, but a beta-5 that crashes, at least if that it's true, is not a good sign of the quality.
I'll tell you what. You get 12-15 people together, take a full operating system, rework the internals by rewriting major sections of nearly every part of the system, then you see how fast you can get the bugs out.
Or, here's a better idea. How about if you learn the difference between a BETA and a Release Candidate. It's BETA because it has bugs. It's not that "Beta could have bugs.", it's that "Beta has bugs. Lots of bugs." It's beta because they are getting help working the bugs out, Oaf.
If it was "Relaese Candidate 5" and there were major bugs causing lots of crashes that wolud be different, but it's a bloody Beta you cheese eating wanker.
"beta version
A pre-shipping release of hardware or software that has gone through alpha test. A beta version of software is supposed to be very close to the final product, but, in practice, it is more a way of getting users to test the software in the first place under real conditions. Given the complexity and ambiguous standards in the PC industry, it is impossible to duplicate the myriad of configurations that exist in the real world."
Chances are the software now works flawlessly under most conditions on all the systems of every developer on the project, and apparently works perfectly for most people here. It's still in beta because people like you have one random system component that completely kills everything, and they are trying to take the extra time to iron that problem out so that people like you with weird peripherals can actually use the system the way you might want to, rather than the way the rest of use see fit.




