That is what he is getting knighted for, and not for his achievements in amassing a huge amount of wealth.
Well, no. If you RTFA it says that he is being knighted for his "contributions to enterprise" and because "Microsoft software has had a profound impact on the British economy".
This is the kind of crap that makes me sick to be British. I don't normally have a strong opinion on the Crown, but sometimes I wish they'd piss off. The knighthood system could be used to recognise truly great people who had served society beyond the call of duty. Instead, it seems to be used as a kind of archaic Oscars for "important people", basically anybody the government feels like sucking up to. Rejecting honours has actually become a serious problem: there was a story a while ago about the government maintaining a list of people who had been given honours but rejected them, and it's growing all the time.
I'd probably reject it too - I wouldn't want to be associated with a stupid popularity contest, let alone Sir Bill.
dmix isn't a full solution as OSS apps still access the device directly. And if you are thinking of the aoss wrapper, it's very buggy. In fact last time I used it I had to debug it myself.
It's almost certainly the fault of Linux, which has no unified or reliable software mixing solution. Efforts are being made, I think FC4 and Ubuntu Hoary have (different) solutions in place for this. But we'll be plagued by this problem for a long time - really until somebody implements a kmixer.
Thanks for mentioning your real age by the way. For people outside the states it's not immediately obvious what age "9th grade" corresponds to, which is probably why the original poster got it wrong,
... and the ones that are good at theory and not at practice become academics. It cuts both ways: good programmers understand algorithmic optimisation but also very practical details like how compilers optimise code and what branch prediction is.
Algorithmic optimisation is definitely important but it's not the be-all and end all of optimisation. In particular a lot of real world optimisation tasks aren't related to pure runtime speed. They're things like "how do I make my program use less memory" and "how do I make it start quicker" or even "how do I make it feel fast" which has more to do with psychology than maths. A lot of desktop software especially is not CPU bound most of the time, so a zen-like understanding of sorting algorithms will only get you so far.
Everything you say is right, but IMHO the reason end users often associate the word "modular" with "bad" is because it means it's much harder for them to try it out, as each dependency must be manually compiled etc. This is especially true of stuff like E17 which being so bleeding-edge isn't widely packaged.
IMHO you guys could do a lot for the E project by having an rsyncable nightly build tree for Linux/x86, so it's trivial to try the latest code. That way you'd get more testers as well.
Talk is cheap, but the Rasterman doesn't just talk the talk, he walks the walk too. The whole point of what he's written is that the Red Hat "next generation rendering" team are talking about things he's already done.
I can see where he is coming from, but for all the hype the E team generate over their amazing new libraries, how many apps actually use them? As far as I can tell, basically none. I don't know why that is though.
The vast majority of code in MacOS is proprietary just like Windows is. This includes things like IOKit which are pretty low level - remember the BSD license lets you take but not give back (which is basically what happened with FreeBSD)
What makes you think it's not Steve Jobs that encourages and re-inforces this atmosphere? You know how he throws enormous hissy fits when his secrets leak.
The problem is that several years after the avoidable screwup of claiming "we're walking away from home users and the desktop: go use Windows"
Oh please, that is such wank. The guy said that for some people who have apps or hardware not compatible with Linux, they should use Windows instead. Guess what? I agree with him! People who point out the obvious truth that Linux does not work out of the box for most people are not heretics, they're just being rational.
I still can't walk into a random computer store, as far as I know, and buy a $20 boxed DVD with any kind of Red Hat Linux on it. This is bad.
Why is it bad? You can download it for free. Given the huge costs involved with setting up boxed sales I'm not surprised they dropped it. Not many people buy operating systems in boxes.
more importantly, we could always recommend the latest bits to newbies without any download hassle.
If you were doing that you were being irresponsible. The volume of updates that all major distributions push today means it's not feasable to use Linux on dialup and stay secure at the same time.
But incremental breakage is just a lot easier to manage.
Not to most people. Most people want to be able to read a review of the next Fedora, see "it breaks stuff" and avoid it until they iron out the problems. They don't want to do an apt-get upgrade and find a few hours later they've been locked out of their system by a busted PAM upgrade.
There are a ton of people replying to you and saying, "Linux is great because XYZ" or "open source is inherantly superior so it won't matter" or whatever.
The real answer is, we just do not know. Right now Linux is targetted mainly by intelligent blackhat hackers working on their own to penetrate servers with large amounts of resources, eg mirror sites and such. That's quite a different type of threat to mass desktop viruses and worms.
But does it really matter? The fact is, our current IT infrastrucure is a mess. Half the backbone traffic is noise from spam and viruses. If you buy a PC from a shop and plug it directly into the net, within 4 minutes it'll be infected with a virus.
How much worse can it get? Does it even matter that we don't know how well Linux will fare when it's replaced Windows? We have a second chance, and we are learning from the mistakes of the past. That is what matters. You have to hope it'll turn out better in the end. It may end up the same way - a crumbling and diseased infrastructure. But we have to hope that it won't.
There are no known kernel exploits for Mac OS X, there is no known spyware, there are no known viruses, there have been a handful of OS X specific exploits that require the user to run a program (and generally ask you to supply an admin password), and have all been "proof of concepts". The bulk of OS X security updates have been for Open Source/Unix apps, which are all turned off by default, and have never been reported as actually exploited.
That's because the open source apps have all their exploits reported as separate incidents, with incident IDs and so on. Apple (and Microsoft) slipstream security fixes into other patches all the time and just don't report them.
For Microsoft this technique is no longer useful because hackers reverse engineer the patches to determine the security flaws.
HackerDefender lets you "cd" into the directories it hides, it just prevents a directory listing from seeing them. So it's like a secret passage: if you know it's there you can get in, but if you don't, you can't see it.
So how does a Mac with no services, daemons, or ports open get rooted in the first place?
Well, via the usual routes: Safari has had some nasty "broken by design" remote code execution exploits to do with the way appfolders are mounted into the system, there were the help viewer URL exploits, your average IM program can also sometimes be exploited - then what you need is a kernel race or improper argument validation and you have root.
There seem to be a bunch of people who believe that Macs are magically more secure than anything else, despite a several exploits that bear a startling resemblence the things Windows has seen (eg help exploits, desktop integration features being used to run remote code etc etc).
Also, until IBM releases a native Linux client, Notes will continue running under WINE. The development team actually tests on WINE and if Notes doesn't run, they track down why and fix it in Notes.
They do? This is news to me. I did the compatibility work to make Notes 6.5.1 run on Crossover, and I have a nice big database full of bugs they should feel free to take off my shoulders. Given that when I started the thing wouldn't even load up, I find the idea that they make it run on Wine a little hard to believe.
On the other hand, we do have several IBM employees doing regression testing. Whether they do it on their own time or IBMs is a little hard to figure out, but given IBM has multiple personality disorder with respect to Wine neither would surprise me.
It is checking directly, feel free to do a relay trace and observe its behaviour yourself. The check is clear and unambiguous - it does a RegOpenKeyEx on the Wine config key (which Wine itself never does), and if it's there immediately pops up the message box.
I'm actually still pretty sure that is what it is, but then why would Windows have a registry key for Wine? Doesn't that imply that you installed it? So that you can run Windows applications on.. Windows?
That's exactly the point - this key would not ever exist on Windows, so the app checks for it and if found it knows you must be running the validation tool on Linux using Wine. In other words, MS are actively checking for and blocking their competitors (again).
The key point here is that it's product tying - their new AntiSpyware app is a protected download and they are tying two products together, which is a violation of anti-trust law.
That's a poor example because an anti-spyware tool isn't very useful to run on Linux. But nonetheless, they are doing the same thing that got them into trouble with the courts back in the Netscape years.
That's not what it's doing. It's checking directly for Wine, and if it finds it, it's bailing out. That is significantly different to Wine simply not being accurate enough.
It's not broken slashcode, popular though that misconception is. It's a race condition inside Gecko that has since been fixed, but the bugfix isn't in 1.0 - there are other websites broken by the same bug, I've encountered a few.
So, don't bash Slashdot prematurely guys;) It's not their fault.
Yes, maybe. I'm sure they take it a lot more seriously on the server. But, they seem to be pretty blase about the desktop. He has tried Linspire and Xandros? Hardly mainstream desktop Linux distros - how many people do you know that use that? They probably have more people dedicated to analysing them elsewhere but given that he's in charge of selling Windows over Linux you'd think he'd use it on the desktop if only so he could know its weak points.
Huh? How does that work? Isn't all blackmail "arm twisting" and "making threatening comments"?
One definition I quite like is "Blackmail is the offer to refrain from engaging in an act that one has the right to perform." Walter Block & Christopher F. Kent. Blackmail, in MAGILL'S LEGAL GUIDE, 109 (1999). Fits the bill quite nicely here.
Let's see. Gates is saying "Do what I say, or I'll make life painful for you" (by damaging the Danish economy). That sounds very much like "Do what I say, or I'll [release embarassing information|trash your stock|kidnap your kiddies]" or whatever Hollywood cliche-blackmail you want. It's all about making threats (in private, do you think it was Gates who leaked this information?) to get what you want.
PaX is the flawed system here. The PaX guys have some weird philosophy where runtime code generation is viewed as "insecure" and it blocked by default, along with a host of other ridiculous restrctions which quite apart from not enhancing security in a useful manner break things as basic as the Nautilus file manager or OpenGL. The fact that Mono doesn't work with it is the least of their concerns.
Quite how runtime code generation can be considered bad is beyond me: I can "generate code" by typing a shell script into emacs.
Meanwhile exec-shield trades off usability and compatibilty against security, and IMHO gets that tradeoff right. If anybody is now thinking "oooh Mono is insecure!!" then don't, it's only insecure according to the PaX developers extremely unusual and by no means accepted world view.
We provide an NDR/RPC/DCOM implementation, MAPI is most often used for "please open the users mail app with this To address" which is trivial to implement and the rest you mentioned are hardly used by any apps at all. What kind of program needs "workstation management"?
Slashdot Mirror
Well, no. If you RTFA it says that he is being knighted for his "contributions to enterprise" and because "Microsoft software has had a profound impact on the British economy".
This is the kind of crap that makes me sick to be British. I don't normally have a strong opinion on the Crown, but sometimes I wish they'd piss off. The knighthood system could be used to recognise truly great people who had served society beyond the call of duty. Instead, it seems to be used as a kind of archaic Oscars for "important people", basically anybody the government feels like sucking up to. Rejecting honours has actually become a serious problem: there was a story a while ago about the government maintaining a list of people who had been given honours but rejected them, and it's growing all the time.
I'd probably reject it too - I wouldn't want to be associated with a stupid popularity contest, let alone Sir Bill.
dmix isn't a full solution as OSS apps still access the device directly. And if you are thinking of the aoss wrapper, it's very buggy. In fact last time I used it I had to debug it myself.
It's almost certainly the fault of Linux, which has no unified or reliable software mixing solution. Efforts are being made, I think FC4 and Ubuntu Hoary have (different) solutions in place for this. But we'll be plagued by this problem for a long time - really until somebody implements a kmixer.
Thanks for mentioning your real age by the way. For people outside the states it's not immediately obvious what age "9th grade" corresponds to, which is probably why the original poster got it wrong,
Algorithmic optimisation is definitely important but it's not the be-all and end all of optimisation. In particular a lot of real world optimisation tasks aren't related to pure runtime speed. They're things like "how do I make my program use less memory" and "how do I make it start quicker" or even "how do I make it feel fast" which has more to do with psychology than maths. A lot of desktop software especially is not CPU bound most of the time, so a zen-like understanding of sorting algorithms will only get you so far.
IMHO you guys could do a lot for the E project by having an rsyncable nightly build tree for Linux/x86, so it's trivial to try the latest code. That way you'd get more testers as well.
I can see where he is coming from, but for all the hype the E team generate over their amazing new libraries, how many apps actually use them? As far as I can tell, basically none. I don't know why that is though.
The vast majority of code in MacOS is proprietary just like Windows is. This includes things like IOKit which are pretty low level - remember the BSD license lets you take but not give back (which is basically what happened with FreeBSD)
What makes you think it's not Steve Jobs that encourages and re-inforces this atmosphere? You know how he throws enormous hissy fits when his secrets leak.
Oh please, that is such wank. The guy said that for some people who have apps or hardware not compatible with Linux, they should use Windows instead. Guess what? I agree with him! People who point out the obvious truth that Linux does not work out of the box for most people are not heretics, they're just being rational.
I still can't walk into a random computer store, as far as I know, and buy a $20 boxed DVD with any kind of Red Hat Linux on it. This is bad.
Why is it bad? You can download it for free. Given the huge costs involved with setting up boxed sales I'm not surprised they dropped it. Not many people buy operating systems in boxes.
more importantly, we could always recommend the latest bits to newbies without any download hassle.
If you were doing that you were being irresponsible. The volume of updates that all major distributions push today means it's not feasable to use Linux on dialup and stay secure at the same time.
But incremental breakage is just a lot easier to manage.
Not to most people. Most people want to be able to read a review of the next Fedora, see "it breaks stuff" and avoid it until they iron out the problems. They don't want to do an apt-get upgrade and find a few hours later they've been locked out of their system by a busted PAM upgrade.
The answer is, we don't know.
There are a ton of people replying to you and saying, "Linux is great because XYZ" or "open source is inherantly superior so it won't matter" or whatever.
The real answer is, we just do not know. Right now Linux is targetted mainly by intelligent blackhat hackers working on their own to penetrate servers with large amounts of resources, eg mirror sites and such. That's quite a different type of threat to mass desktop viruses and worms.
But does it really matter? The fact is, our current IT infrastrucure is a mess. Half the backbone traffic is noise from spam and viruses. If you buy a PC from a shop and plug it directly into the net, within 4 minutes it'll be infected with a virus.
How much worse can it get? Does it even matter that we don't know how well Linux will fare when it's replaced Windows? We have a second chance, and we are learning from the mistakes of the past. That is what matters. You have to hope it'll turn out better in the end. It may end up the same way - a crumbling and diseased infrastructure. But we have to hope that it won't.
That's because the open source apps have all their exploits reported as separate incidents, with incident IDs and so on. Apple (and Microsoft) slipstream security fixes into other patches all the time and just don't report them.
For Microsoft this technique is no longer useful because hackers reverse engineer the patches to determine the security flaws.
HackerDefender lets you "cd" into the directories it hides, it just prevents a directory listing from seeing them. So it's like a secret passage: if you know it's there you can get in, but if you don't, you can't see it.
Well, via the usual routes: Safari has had some nasty "broken by design" remote code execution exploits to do with the way appfolders are mounted into the system, there were the help viewer URL exploits, your average IM program can also sometimes be exploited - then what you need is a kernel race or improper argument validation and you have root.
There seem to be a bunch of people who believe that Macs are magically more secure than anything else, despite a several exploits that bear a startling resemblence the things Windows has seen (eg help exploits, desktop integration features being used to run remote code etc etc).
They do? This is news to me. I did the compatibility work to make Notes 6.5.1 run on Crossover, and I have a nice big database full of bugs they should feel free to take off my shoulders. Given that when I started the thing wouldn't even load up, I find the idea that they make it run on Wine a little hard to believe.
On the other hand, we do have several IBM employees doing regression testing. Whether they do it on their own time or IBMs is a little hard to figure out, but given IBM has multiple personality disorder with respect to Wine neither would surprise me.
So I guess you can consider yourself flamed ;)
That's exactly the point - this key would not ever exist on Windows, so the app checks for it and if found it knows you must be running the validation tool on Linux using Wine. In other words, MS are actively checking for and blocking their competitors (again).
That's a poor example because an anti-spyware tool isn't very useful to run on Linux. But nonetheless, they are doing the same thing that got them into trouble with the courts back in the Netscape years.
That's not what it's doing. It's checking directly for Wine, and if it finds it, it's bailing out. That is significantly different to Wine simply not being accurate enough.
So, don't bash Slashdot prematurely guys ;) It's not their fault.
Yes, maybe. I'm sure they take it a lot more seriously on the server. But, they seem to be pretty blase about the desktop. He has tried Linspire and Xandros? Hardly mainstream desktop Linux distros - how many people do you know that use that? They probably have more people dedicated to analysing them elsewhere but given that he's in charge of selling Windows over Linux you'd think he'd use it on the desktop if only so he could know its weak points.
One definition I quite like is "Blackmail is the offer to refrain from engaging in an act that one has the right to perform." Walter Block & Christopher F. Kent. Blackmail, in MAGILL'S LEGAL GUIDE, 109 (1999). Fits the bill quite nicely here.
Let's see. Gates is saying "Do what I say, or I'll make life painful for you" (by damaging the Danish economy). That sounds very much like "Do what I say, or I'll [release embarassing information|trash your stock|kidnap your kiddies]" or whatever Hollywood cliche-blackmail you want. It's all about making threats (in private, do you think it was Gates who leaked this information?) to get what you want.
Quite how runtime code generation can be considered bad is beyond me: I can "generate code" by typing a shell script into emacs.
Meanwhile exec-shield trades off usability and compatibilty against security, and IMHO gets that tradeoff right. If anybody is now thinking "oooh Mono is insecure!!" then don't, it's only insecure according to the PaX developers extremely unusual and by no means accepted world view.
Are you similarly disturbed by Office, MSN or Visual Studio, which likewise do not use the XP widgets (and don't even attempt to resemble them) ?
We provide an NDR/RPC/DCOM implementation, MAPI is most often used for "please open the users mail app with this To address" which is trivial to implement and the rest you mentioned are hardly used by any apps at all. What kind of program needs "workstation management"?