For all the supposed work that is going into Mozilla, it had better be good... I am still running Netscape 4.08, and I have been waiting and waiting and waiting for something better. Netscape 4.08 (and even 4.51, for that matter) still seem notoriously unstable to me... Unlike most Linux apps, I have never been able to get Netscape to run for more than a few hours without the thing segfaulting or just hanging up on me. Any idea when the official version of this Mozilla beast is scheduled to come out, anyone?
Enough said... I would recommend ALSA... I had similar problems... Under OSS/Free, I could not get it to play sound on my GUS PnP without the thing spewing out tons and tons of DMA/IRQ errors, even though my DMA/IRQs were okay. IMHO, OSS/Free is what I would call crippleware, masquerading as open source... ALSA rules.
Am I the only one who wonders about the future of this beast?? I mean, I used to be an OS/2 user, and Windows compatibility is one of the possible reasons that no native OS/2 apps were ever developed... I hope the same does not happen to Linux. In any case, I have been compiling every new release of WINE that comes out almost from the first day I started using Linux... And I must say, my problem with WINE is that, while it does look promising, I have yet to see a *SINGLE* Windows application of mine work with it... Every single one crashes and burns. I think WINE still has longer to go than is reasonable at this point before it becomes truly usable... Why don't they just give up on it? It might be better for the Linux community in the longrun, too.
Oh no... Not another OSS platform... I mean, the OSS APIs sound good, but if the drivers they will be doing for Be are anything like the Linux OSS/Free drivers, they sure will stink.
I just wanted to let those of you who are so stuck up about the superiorities of Linux security to know that this problem exists under Linux, too. Go into the/tmp directory, and you can look at the contents of all the nsform* files that are left there after submitting forms. And BTW, if you are thinking file permissions will solve this problem under Linux and not under Windows, demonstrating superior security in Linux, think again... NT has Unix-like file permissions, too, so this solution could be implemented on both platforms. And I speak as a former NT user, who dumped it in favor of Linux. If you use NT, just right-click on a file in the explorer, and click on the "Security" tab... You will see what I mean. Don't get me wrong: I am a Linux user all the way... I do think there are reasons to say that Linux is greatly superior to NT... I know firsthand what some of these reasons are. All I am saying is, don't start making smart-ass claims about things when you haven't a clue what you are talking about... Make sure you know what you are talking about before you post.
A couple of things... First, I do see KDE & QT 1.42 in the current Raw Hide tree. Granted, it might not get included in Red Hat 6.0, but I think it likely, especially since QT 2.0 will be supposedly Open Source (leaving aside any debates on the nature of the QPL)... Perhaps Red Hat is experimenting with it for its own purposes to get ready for it? In any case, they have not said anything publicly about their complete reversal of direction on this issue, after stating that they were completely opposed to it... So who knows??
As to kernel 2.2... Red Hat has DEADLINES to meet on releasing the new versions. Notice that 5.2 shipped with a pre-version of the 2.0.36 kernel (package hence named as 2.0.36-0.7), and 5.1 shipped with a 2.0.34-0.7. I noticed also that the Raw Hide trees that were around before the release of 5.2 included the most recent 2.1 kernel they could get their hands on at the time... When 5.2 finally came out, it included the most recent stable kernel they could get their hands on... The pre-2.0.36 version they had. I suspect that Red Hat was getting fed up waiting for Alan Cox to release the final 2.0.36, hence the pre-version in the release, just so they could meet the release deadline. At this point though, with 2.2 now in the pre stage, I suspect they will release the latest pre-2.2 they can get their hands on, if the final release is not yet available. That is my best guess, based on my experience in the past.
I *REALLY* hope CL decides to support ALSA, rather than the OSS/Free drivers shipped with the standard Linux kernel... I am hoping perhaps this will convince Linus/Alan to dump the pathetic OSS/Free drivers they currently use, and replace them with ALSA... For all the good things I could say about the drivers that come with the standard Linux kernel, like the (generally) outstanding SCSI/IDE/Parallel port drivers it has, I think OSS/Free is probably the only exception I can find... In my case at least, I found ALSA to be vastly more stable and feature-rich than OSS/Free... I tried and tried and tried to get OSS/Free to work with my GUS PnP, but the most I got in the end was sound that kept skipping, and the driver spewing out tons of DMA/IRQ error messages to my terminal, even though I had them setup CORRECTLY. When you consider this, plus the fact that OSS/Free requires RAM just to do DA on the AMD Interwave-based PnP card (which is absolutely pathetic, IMHO, since ALSA does DA without the need for RAM), I think this says that all around, OSS/Free *STINKS*. Needless to say, the last straw with me came when I was told by one of the OSS/Free developers that they "might consider" PnP support, but that I should pay for OSS/Professional if I really wanted to get this feature (since 4Front-Tech, Inc. sponsers the development of OSS/Free)... That was the last straw as far as I was concerned... They can say that OSS/Free is GPL'd if they wish, but IMHO, it is meaningless if all they can tell me is that I must pay for a proprietary binary-only version just to get something half-decent. That is NOT my idea of open-source... That is what I call CRIPPLEWARE, masquerading as open source. Bah. Let's hope CL makes the right decision.
Slashdot Mirror
For all the supposed work that is going into Mozilla, it had better be good... I am still running Netscape 4.08, and I have been waiting and waiting and waiting for something better. Netscape 4.08 (and even 4.51, for that matter) still seem notoriously unstable to me... Unlike most Linux apps, I have never been able to get Netscape to run for more than a few hours without the thing segfaulting or just hanging up on me. Any idea when the official version of this Mozilla beast is scheduled to come out, anyone?
Enough said... I would recommend ALSA... I had similar problems... Under OSS/Free, I could not get it to play sound on my GUS PnP without the thing spewing out tons and tons of DMA/IRQ errors, even though my DMA/IRQs were okay. IMHO, OSS/Free is what I would call crippleware, masquerading as open source... ALSA rules.
Am I the only one who wonders about the future of this beast?? I mean, I used to be an OS/2 user, and Windows compatibility is one of the possible reasons that no native OS/2 apps were ever developed... I hope the same does not happen to Linux. In any case, I have been compiling every new release of WINE that comes out almost from the first day I started using Linux... And I must say, my problem with WINE is that, while it does look promising, I have yet to see a *SINGLE* Windows application of mine work with it... Every single one crashes and burns. I think WINE still has longer to go than is reasonable at this point before it becomes truly usable... Why don't they just give up on it? It might be better for the Linux community in the longrun, too.
Great... KDE 1.1 is out... Now what? Also, what about QT 2.0, and the QPL?? (I still think Troll should GPL QT.)
Oh no... Not another OSS platform... I mean, the OSS APIs sound good, but if the drivers they will be doing for Be are anything like the Linux OSS/Free drivers, they sure will stink.
I just wanted to let those of you who are so stuck up about the superiorities of Linux security to know that this problem exists under Linux, too. Go into the /tmp directory, and you can look at the contents of all the nsform* files that are left there after submitting forms. And BTW, if you are thinking file permissions will solve this problem under Linux and not under Windows, demonstrating superior security in Linux, think again... NT has Unix-like file permissions, too, so this solution could be implemented on both platforms. And I speak as a former NT user, who dumped it in favor of Linux. If you use NT, just right-click on a file in the explorer, and click on the "Security" tab... You will see what I mean. Don't get me wrong: I am a Linux user all the way... I do think there are reasons to say that Linux is greatly superior to NT... I know firsthand what some of these reasons are. All I am saying is, don't start making smart-ass claims about things when you haven't a clue what you are talking about... Make sure you know what you are talking about before you post.
A couple of things... First, I do see KDE & QT 1.42 in the current Raw Hide tree. Granted, it might not get included in Red Hat 6.0, but I think it likely, especially since QT 2.0 will be supposedly Open Source (leaving aside any debates on the nature of the QPL)... Perhaps Red Hat is experimenting with it for its own purposes to get ready for it? In any case, they have not said anything publicly about their complete reversal of direction on this issue, after stating that they were completely opposed to it... So who knows??
As to kernel 2.2... Red Hat has DEADLINES to meet on releasing the new versions. Notice that 5.2 shipped with a pre-version of the 2.0.36 kernel (package hence named as 2.0.36-0.7), and 5.1 shipped with a 2.0.34-0.7. I noticed also that the Raw Hide trees that were around before the release of 5.2 included the most recent 2.1 kernel they could get their hands on at the time... When 5.2 finally came out, it included the most recent stable kernel they could get their hands on... The pre-2.0.36 version they had. I suspect that Red Hat was getting fed up waiting for Alan Cox to release the final 2.0.36, hence the pre-version in the release, just so they could meet the release deadline. At this point though, with 2.2 now in the pre stage, I suspect they will release the latest pre-2.2 they can get their hands on, if the final release is not yet available. That is my best guess, based on my experience in the past.
I *REALLY* hope CL decides to support ALSA, rather than the OSS/Free drivers shipped with the standard Linux kernel... I am hoping perhaps this will convince Linus/Alan to dump the pathetic OSS/Free drivers they currently use, and replace them with ALSA... For all the good things I could say about the drivers that come with the standard Linux kernel, like the (generally) outstanding SCSI/IDE/Parallel port drivers it has, I think OSS/Free is probably the only exception I can find... In my case at least, I found ALSA to be vastly more stable and feature-rich than OSS/Free... I tried and tried and tried to get OSS/Free to work with my GUS PnP, but the most I got in the end was sound that kept skipping, and the driver spewing out tons of DMA/IRQ error messages to my terminal, even though I had them setup CORRECTLY. When you consider this, plus the fact that OSS/Free requires RAM just to do DA on the AMD Interwave-based PnP card (which is absolutely pathetic, IMHO, since ALSA does DA without the need for RAM), I think this says that all around, OSS/Free *STINKS*. Needless to say, the last straw with me came when I was told by one of the OSS/Free developers that they "might consider" PnP support, but that I should pay for OSS/Professional if I really wanted to get this feature (since 4Front-Tech, Inc. sponsers the development of OSS/Free)... That was the last straw as far as I was concerned... They can say that OSS/Free is GPL'd if they wish, but IMHO, it is meaningless if all they can tell me is that I must pay for a proprietary binary-only version just to get something half-decent. That is NOT my idea of open-source... That is what I call CRIPPLEWARE, masquerading as open source. Bah. Let's hope CL makes the right decision.