when it happens, it promptly apologizes and offers you to open the windows you were browsing at the moment of the crash.
Actually this is one thing that really pisses me off about Opera. No matter how many times I tell it "start with no pages, and don't ask me again", it always gives me that dialog after a crash. Is there any way to get it to stop doing that.
There is an exception for libraries that are a normal part of the system, but since the drivers in question are not normally distributed with the OS, I don't think that that exception applies.
Huh? They are being distributed with the OS, and every OS I have ever seen has shipped with graphics drivers.
It's silly duplication of effort. Why not make frontpage better for this sort of thing - isn't that what it's there for? Certainly I think this is a feature that would be better elsewhere than in a word processor.
If it was actually organised in 3D it could be very useful. Have links appear just below the page, so you can see them without following them completely. Categorise wikipedia in 3D. However, I suspect it will just be used for gimmicky eyecandy, in which case, yes, it will suck.
As a software developer whose experience goes back more than 40 years, to the Stanford Time-Sharing System on the DEC PDP-1, I can assure you that the only way to keep the kernel API from changing is to kill the project.
I refer you to solaris. Still very much alive, and backward compatible as far as version 5.
But that doesn't have much to do with microkernels.
It makes it a lot easier. If most things are in userspace processes you only need to worry about specific handling of a small amount of code.
As for other drivers failing, well, I'm not so sure it is really that big of a problem to be talking about radical OS redesigns. Especially one that, traditionally, has significant performance penalties. I think most people would rather have a system that runs fast all the time vs. one that can handle that rare event when some faulty code in their sound driver writes to the wrong bit of RAM or whatever.
I don't. I came to linux because of the promise of stability, no more random crashes. I have much more CPU power than I need.
By not giving the user permission to do this. When there is essentially no boundary between system and user programs, your access control system that stops the bluetooth drivers giving a local root (real example from linux) can equally well stop processes that have no need to from opening sockets.
Say what you will about the monolithic kernel, but the fact that one bad driver can crash the whole works tends to make people work much harder to create solid drivers that don't crash.
And they fail. What do you think of the statement "Say what you will about C, but the fact that one missing () can give attackers root tends to make people work much harder to create solid code that doesn't have holes."
In Andrew Tanenbaum's world, a driver developer can write a driver, and not even realize the thing is being restarted every 5 minutes because of some bug. This sort of thing could even get into a shipping product, with who knows what security and performance implications.
I don't think it's any more or less likely to have a security issue if it has a particular bug. As for performance, if you don't notice the problem, then it isn't a problem.
On a well-designed system it is absolutely realistic to be able to restart it, especially since so much kernel code is drivers for non-essential hardware. If my sound card drivers crash, just block on any reads/writes until they're restarted. Same for my network card, cd drives, even hard disk should be alright for short periods. Heck, there is no reason this shouldn't work for losing a cpu on an SMP system, or a stick of ram provided the core kernel parts (scheduler etc.) weren't on that one.
I hate US foreign policy and think those who attacked Iraq belong in jail. However, there is a difference here in that there is a fundamental human right, there in the UN charter, to freedom of speech. And I would support foreign countries working around my own or US law if there were human rights being violated - all I can think of right now is releasing those who are being detained without trial, but I'm sure there are other examples.
No, I am saying that the completely standard hardware is more likey to be maintained by volunteers than the less common nonstandard hardware. That's just a fact of Free Software.
But if there are working drivers for nonstandard hardware, they shouldn't be arbitrarily broken.
What SATA chipset/mobo are you using? I have had to flash my stupid motherboard 3 times now with updates (MSI K8N mobo, I think choosing this mobo may have been a mistake) because of various things not working but it was always the fault of the motherboard so you might want to consider that also. Overall SATA works great. Just last night I upgraded my smartmontools to support SATA and now I can monitor my hard drive for signs of impending failure. My HD has 6132 hours of powered up time and has remapped one bad sector. Pretty cool to be able to get that kind of useful diagnostic info out of it.
VIA chipset, optronix motherboard. Nothing wrong with it afaict, I'm just saying that's the one thing that's not 100% bog-standard PC.
How does having everything in the kernel make it hard to maintain separate drivers?
It doesn't. The kernel devs' refusal to provide a stable kernel API does.
It does not hurt the reliability of your kernel to have drivers in the source tree that you are not actually running.
Agreed, but a stable kernel should not ship with drivers that don't work. External drivers make it easier to find the source of a problem, for starters.
So far 2.6 has been just as solid for me as previous kernel versions but I try really hard to avoid using bizarro hardware and drivers that probably do not get much testing, and rightly so.
Are you saying it doesn't matter if anything other than completely standard hardware fails to work? Because that's not an attitude I like, and if that's the approach it's taking now, I will be looking at moving from linux as soon as feasible. Besides, it's not just that - I have a standard PC with the only even slightly unusual thing being SATA. 2.6 has consistently failed to keep it up more than a week, whether by actual locking up, leaking memory to the point of unusability, or suddenly ceasing to drive a random piece of hardware.
I think we need to distinguish between bugs in the core kernel (code that everyone runs) and bugs in drivers. The vast majority of the Linux kernel code is drivers.
Yes, and that's because the kernel devs insist on having everything in the kernel and deliberately make it difficult to maintain separate drivers. They are reaping what they've sown in this regard.
As I said, get a new bank. They are obviously not particularly security concious. I am willing to bet that the last few digits of all of your bank accounts are different. Why would the bank show anything more than they need? Do their ATMs print the account number on the receipt?
Well, yeah. If they didn't, I can't see it being that useful as a receipt - proof that money came out of or went into that account.
I've been using it at least 7 years. All that quote says they're doing is adding extensions for new features - which is the only way to support up to date features, ODF has extensibility as a big marketing point too - and how many features that weren't in word 6 do you normally use?
Just out of curiosity, when did BA stop checking passports at the airport for US flights?
I'm pretty sure they didn't when I flew in 2003 and after that. You get generic passport control on the way to the plane but you've already got your boarding pass at this point.
Now, if BA had to collect the information through the web, why on earth would they have to make it accessible and easily changable?
So you can see if you've entered it wrong, and change it if it changes.
Does your bank show your account number on the website when you log in? If they do, change your bank
Yeah, they show it, and so did the last one. How else would I identify my different accounts?
The Geneva Convention doesn't state how long you have to wait to get your tribunal
I'm sure there's "entitled to a speedy decision by a competent tribunal" in there. If you want to treat them as unlawful combatants, you have to give then the tribunal as soon as reasonably practical.
in other words, they could have you supply the information when you buy the ticket, ship it accross, and promptly remove it.
The whole point of the online service is not having to queue up, you can simply collect your tickets from the machine and go. If your passport had to be checked at the airport, that would make the service essentially useless. So the new US policy gave them a choice between removing a popular service that paying customers liked and storing that information on those servers. Really it's no surprise what they chose, and I can't say I blame them.
A data format specification is only useful if the applications that implement it do so correctly. What the parent poster was trying to convey was how little respected the RTF specfication is, both in the FOSS community and by Microsoft. RTF, as a document specification, is almost as useful as CSV is for spreadsheets: it's a guideline, but the devil is in the details,
Really? Because I have *never* seen an RTF look different in different programs, something I see all the time with doc and even sometimes with odt.
The automatic ToC is a good example. The actual contents of the ToC are retained. (The "data"), the tabular formatting of the ToC is retained (The "metadata"), but the the fact that that table is an automatic ToC is gone (the "meta-meta-data"?).
IIRC it remained a TOC, and had gone in and out of kword. But this could still be an example of the embedded comments thing if kword keeps them in there (I have no idea if this is the case or not)
So use a windows installer unpacker. Seriously, if you're going to be a dick about it I can say "WTF is this ungodly format called PDF that your supposedly open specification is in" (the pdf spec is only available as a pdf, and guess what I have to do to read those? Yep, run a random executable, and it won't work on my alpha either).
Slashdot Mirror
Actually this is one thing that really pisses me off about Opera. No matter how many times I tell it "start with no pages, and don't ask me again", it always gives me that dialog after a crash. Is there any way to get it to stop doing that.
Huh? They are being distributed with the OS, and every OS I have ever seen has shipped with graphics drivers.
It's silly duplication of effort. Why not make frontpage better for this sort of thing - isn't that what it's there for? Certainly I think this is a feature that would be better elsewhere than in a word processor.
OMG!!! That was, like, totally the best!!
If it was actually organised in 3D it could be very useful. Have links appear just below the page, so you can see them without following them completely. Categorise wikipedia in 3D. However, I suspect it will just be used for gimmicky eyecandy, in which case, yes, it will suck.
I refer you to solaris. Still very much alive, and backward compatible as far as version 5.
It makes it a lot easier. If most things are in userspace processes you only need to worry about specific handling of a small amount of code.
As for other drivers failing, well, I'm not so sure it is really that big of a problem to be talking about radical OS redesigns. Especially one that, traditionally, has significant performance penalties. I think most people would rather have a system that runs fast all the time vs. one that can handle that rare event when some faulty code in their sound driver writes to the wrong bit of RAM or whatever.
I don't. I came to linux because of the promise of stability, no more random crashes. I have much more CPU power than I need.
You're new here, aren't you
By not giving the user permission to do this. When there is essentially no boundary between system and user programs, your access control system that stops the bluetooth drivers giving a local root (real example from linux) can equally well stop processes that have no need to from opening sockets.
And they fail. What do you think of the statement "Say what you will about C, but the fact that one missing () can give attackers root tends to make people work much harder to create solid code that doesn't have holes."
In Andrew Tanenbaum's world, a driver developer can write a driver, and not even realize the thing is being restarted every 5 minutes because of some bug. This sort of thing could even get into a shipping product, with who knows what security and performance implications.
I don't think it's any more or less likely to have a security issue if it has a particular bug. As for performance, if you don't notice the problem, then it isn't a problem.
On a well-designed system it is absolutely realistic to be able to restart it, especially since so much kernel code is drivers for non-essential hardware. If my sound card drivers crash, just block on any reads/writes until they're restarted. Same for my network card, cd drives, even hard disk should be alright for short periods. Heck, there is no reason this shouldn't work for losing a cpu on an SMP system, or a stick of ram provided the core kernel parts (scheduler etc.) weren't on that one.
I hate US foreign policy and think those who attacked Iraq belong in jail. However, there is a difference here in that there is a fundamental human right, there in the UN charter, to freedom of speech. And I would support foreign countries working around my own or US law if there were human rights being violated - all I can think of right now is releasing those who are being detained without trial, but I'm sure there are other examples.
That'd be fair enough if the comment were actually funny
WTF is this modded up for? It's learning the same as anything else.
But if there are working drivers for nonstandard hardware, they shouldn't be arbitrarily broken.
What SATA chipset/mobo are you using? I have had to flash my stupid motherboard 3 times now with updates (MSI K8N mobo, I think choosing this mobo may have been a mistake) because of various things not working but it was always the fault of the motherboard so you might want to consider that also. Overall SATA works great. Just last night I upgraded my smartmontools to support SATA and now I can monitor my hard drive for signs of impending failure. My HD has 6132 hours of powered up time and has remapped one bad sector. Pretty cool to be able to get that kind of useful diagnostic info out of it.
VIA chipset, optronix motherboard. Nothing wrong with it afaict, I'm just saying that's the one thing that's not 100% bog-standard PC.
How does having everything in the kernel make it hard to maintain separate drivers?
It doesn't. The kernel devs' refusal to provide a stable kernel API does.
It does not hurt the reliability of your kernel to have drivers in the source tree that you are not actually running.
Agreed, but a stable kernel should not ship with drivers that don't work. External drivers make it easier to find the source of a problem, for starters.
Perhaps their client is rigged to not work if they aren't making money off the ads? Wouldn't surprise me.
Are you saying it doesn't matter if anything other than completely standard hardware fails to work? Because that's not an attitude I like, and if that's the approach it's taking now, I will be looking at moving from linux as soon as feasible. Besides, it's not just that - I have a standard PC with the only even slightly unusual thing being SATA. 2.6 has consistently failed to keep it up more than a week, whether by actual locking up, leaking memory to the point of unusability, or suddenly ceasing to drive a random piece of hardware.
I think we need to distinguish between bugs in the core kernel (code that everyone runs) and bugs in drivers. The vast majority of the Linux kernel code is drivers.
Yes, and that's because the kernel devs insist on having everything in the kernel and deliberately make it difficult to maintain separate drivers. They are reaping what they've sown in this regard.
Well, yeah. If they didn't, I can't see it being that useful as a receipt - proof that money came out of or went into that account.
I've been using it at least 7 years. All that quote says they're doing is adding extensions for new features - which is the only way to support up to date features, ODF has extensibility as a big marketing point too - and how many features that weren't in word 6 do you normally use?
I'm pretty sure they didn't when I flew in 2003 and after that. You get generic passport control on the way to the plane but you've already got your boarding pass at this point.
Now, if BA had to collect the information through the web, why on earth would they have to make it accessible and easily changable?
So you can see if you've entered it wrong, and change it if it changes.
Does your bank show your account number on the website when you log in? If they do, change your bank
Yeah, they show it, and so did the last one. How else would I identify my different accounts?
I'm sure there's "entitled to a speedy decision by a competent tribunal" in there. If you want to treat them as unlawful combatants, you have to give then the tribunal as soon as reasonably practical.
The whole point of the online service is not having to queue up, you can simply collect your tickets from the machine and go. If your passport had to be checked at the airport, that would make the service essentially useless. So the new US policy gave them a choice between removing a popular service that paying customers liked and storing that information on those servers. Really it's no surprise what they chose, and I can't say I blame them.
Really? Because I have *never* seen an RTF look different in different programs, something I see all the time with doc and even sometimes with odt.
IIRC it remained a TOC, and had gone in and out of kword. But this could still be an example of the embedded comments thing if kword keeps them in there (I have no idea if this is the case or not)
So use a windows installer unpacker. Seriously, if you're going to be a dick about it I can say "WTF is this ungodly format called PDF that your supposedly open specification is in" (the pdf spec is only available as a pdf, and guess what I have to do to read those? Yep, run a random executable, and it won't work on my alpha either).