I use grouping and love it. I thin it's a better option than "tabbed" browsing (tabbed browsing is trying to solve a bigger UI problem within a specific application IMHO). I use grouping on laptops or desktops - I fail to see the difference. My laptop screen has more real estate than my 19" LCD (1400x1050 vs 1280x1024 respectively). Now, you may not like grouping, and that's your opinion, but that doesn't make it useless.
As far as a patent, yes, software patents are rediculous, but to blame the player, blame the game.
No OS should ever crash for any software-induced reason, ever
I disagree. Device level drivers need a certian level of performance, and it's still too expensive to provide enough saftey nets around them to completely prevent them from foobaring your OS.
I'm running Windows XP on VMWare in Linux. Linux doesn't crash, it keeps on chugging along fine, but Windows XP in the vmware session is what reboots. its not a hardware problem, otherwise it would kill linux too
It is most definitely a "hardware" problem - it's a virtual hardware problem. You are incorrectly assuming that VMWare doesn't have any bugs with emulating your hardware and running Windows.
I'm not claiming to know your specific situation, but the only BSOD situations that I've run accross are bad hardware (eg bad memory). One specific situation was with ASUS's poor handling of DDR400 memory when using dual channel mode. The box BSOD'd all the time. Dude was swearing at MS left and right. I said, "no way, Windows is rock solid (he laughed, I'm serious), fix your hardware". Needless to say, he built his own box. Maybe you've covered a sev 1 bug in win2k3, but I'd investigate other options before coming to that conclusion.
3. Stability.// most uptimes in Linux are measured in months and years rather than days and weeks (with exceptions, of course), and the GUI being a completely separate component from the kernel helps this greatly
The GUI being completely seperate is the biggest reason why Linux, without the GUI, is so stable. I'll admit that it's been while since I've used Linux with a GUI, but when comparing it to Windows 2000 a few years ago, my RedHat box was extremely unstable. I know that 2-3 years can yield a lot of improvement, and I know that many/. experts claim great stability, but for the average user, I don't think that Linux's reputation for stability on the desktop (read: GUI) is with merit.
Look at Windows 2003. They don't have the same usability req's as XP, so it's easier to secure. And it IS secure. It's not bulletproof, it's not OpenBSD, but how many serious exploits have made it into the wild, especially when compared to competing OS's? Windows XP SP2 looks to be a huge improvement - we'll just have to see. Either way, it seems they have a handle on it, without having to tear everything down and start again.
1) This is a trojan. While IE could be improved to help prevent this, this type of trojan can be used with any browser (albeit with a bit more social engineering effort would be required with most other browsers).
2) Yes, XP SP2 is a magic fix. I've seen the dialog screens for BHO's and the like. They're rediculously obvious. Furthermore, I believe that MS is _finally_ sandboxing this stuff (I remember reading it somewhere, but I can't verify). Finally, SP2's super aggressive firewall would detect that an unauthorized application was trying to send data via port 80 (or any port, for that matter) and warn the user. SP2 isn't bullet proof, but MS has put a LOT of resources into it to help minimize it's embarassing history. From what I've seen it looks promising, and hopefully my firewall will stop reporting so much NIMDA etc. traffic.
Microsoft are in a hugely powerful position to control other.Net implementations by changing things arbitrarily.
You have to remember that has a huge installbase of.NET and while they can add new features, making breaking changes will have a negative impact on their customers.
You need to broaden search for.NET, not C#. A lot of.NET jobs are VB.NET, and a small percentage account for COBOL and FORTRAN.NET as well (a lot of mainframe migration projects use those languages).
The registry--nay, the entire concept of the registry, is an abomination.
While not the best solution for the goal, one of the important features of the registry that.conf files lack is that of per user, or per group configuration. Microsoft has done a good job with xml based config for programs such as.NET and IIS that don't require per user configuration. I'm not sure if Longhorn does away with the registry for newer apps (I'm sure it still has it for legacy support), but it'll be interesting to see what they end up doing (I should get off my butt and just install the Longhorn disks 2ft. from me!).
I've never needed to edit the registry to install hardware. And I've installed some pretty off-the-wall hardware in my time (Windows user since Win95). Don't get me wrong, the registry is a mess (much better in NT based OS's), but it's not as bad as you make it out to be. And my WinXP on 3 different machines (one home-built, 2 laptops) runs just fine, no crashes, no reinstalls. It's not a perfect OS, but they've gotten a lot better of late.
Looking at the source is not an "at a glance" task. Depending on the problem, the programmer may need days or even weeks to understand the problem. Even at todays declined rates, that would cost thousands of dollars to the end user just to fix one problem.
Sure, the community can support the software, but if there's no commercial support (ie Redhat), than it's too risky for most people.
1.) Just how much exactly is Microsoft afraid of Linux?
Most people on/. keep asking the wrong question. The question is, "Just how much exaclty is Microsoft afraid of Apple". Apple is the one making huge headway on the desktop, not any Linux distro. I'm a developer who mainly works with MS technologies (C#, MSSQL Server, etc.) and it's amazing how many MS developers (even MS employees) have Powerbooks at home. The hardware is awesome, the OS solid, it's unix, and it has better office productivity software. Linux is an issue on the server side, but most "Linux wins" are wins against other unices, not server or desktop rollouts against Windows.
Longhorn October PDC bits run slow but usable on 2Ghz w/512MB. By the time it's optimized for hardware (the latest builds I've seen mostly are) it will run faster. Beta bits will run even faster. And obviously the production bits will be even faster yet. I'm confident that it will run fine on todays hardware as long as you disable a few of the unnecessary eyecandy bits.
And who cares how much software costs in relation to hardware? Software is the reason you buy hardware. Just one game is 33-40% of a console's cost. If Longhorn has the features that people want, than spending the OEM $25-$40 (depending on OEM and version) on even a $399 value machine is totally worth it.
Apple's DRM is one of the most flexible. However, that doesn't mean it's flexible enough. I've barely even spent $100 on the iTunes store and the main reason is because I want to play my music on devices other than my iPod. I need to convert to MP3 to play on other devices. At least with WMA's DRM I can still use most all devices (except for my iPod!) since MS worked hard to put WMA on everything but your toaster. If Apple doesn't want me cracking their DRM, they better get their technology on a lot more devices.
No, commodity infrastructure software is no exception to that. That softwarwe does come at a price. At the price of thousands of volunteer developers who, for whatever reason, like working for free and donating their time and talents to for-profit organizations.
(Please no replies on how IBM and Apple pay developers to work on Linux and BSD respectively. This illustrates the exception and not the rule for most major OSS packages)
But this has nothing to do with the quality of healthcare. I know a lot of people who believe in active euthanasia for certain medical situations. Heck, there's been a lot of cases in the US where the same has happened except that people PAID for it with their own money! The only thing that's stopped people in the US is the law.
Yet countries with social healthcare cover more of the elderly and other "non productive citzens" better than we do. Sure, we have the best technology (for some treatments) for the privilaged few. The tyranny of democracy is checked because 51% of Congress can never vote anything into existance. Most votes require 2/3rds majority. Our system is a representative republic, so theoretically more politically savvy people are doing the voting on many issues. Our system is nowhere near perfect, and I could critique it for hours, but a social healthplan does not mean the death of elders and retards.
Oh, and Milton Friedman's Nobel Prize was awarded via a democratic vote.
Not in the near future. It's too bug prone, insecure, and difficult to manage in larger codebases. Finally, the time to market with C is too costly for the enterprise.
Will C# get used for "enterprise apps".
It allready does, and we're talking mission critical Fortune 500. Google is your friend, use it before making rash statements. I'd much rather use.NET or J2EE for an enterprise level applications. Both.NET and J2EE support lower level programming (Managed C++/PInvoke and JNDI respectively) for the rare instances that direct hardware access is needed.
Yes, C/C++ will be around for a long time, but new applications will continue to use higher level platforms as those platforms mature.
I think that you're missing the point of this letter. The author is mainly critiquing the concept of an individual working on an OSS project for free. You're right, Google is profiting handsomly, and in part thanks to programmers who contributed to Linux and other OSS projects. Are those programmers making a dime off of Google's success? Other than a couple at Red Hat or maybe IBM, the answer is no.
My question the programmer addressed in this letter is, why on earth would you work for a for-profit organization without requiring compensation? Companies choosing to open up protocols or source code makes sense in certain scenarios. This is however very different then a programmer giving up their own personal time, which in my opinion just deteriorates our pofessional value as a whole.
Slashdot Mirror
but to blame the player, blame the game
:).
Don't blame the player, blame the game.
I'll hit preview next time
I use grouping and love it. I thin it's a better option than "tabbed" browsing (tabbed browsing is trying to solve a bigger UI problem within a specific application IMHO). I use grouping on laptops or desktops - I fail to see the difference. My laptop screen has more real estate than my 19" LCD (1400x1050 vs 1280x1024 respectively). Now, you may not like grouping, and that's your opinion, but that doesn't make it useless.
As far as a patent, yes, software patents are rediculous, but to blame the player, blame the game.
No OS should ever crash for any software-induced reason, ever
I disagree. Device level drivers need a certian level of performance, and it's still too expensive to provide enough saftey nets around them to completely prevent them from foobaring your OS.
I'm running Windows XP on VMWare in Linux. Linux doesn't crash, it keeps on chugging along fine, but Windows XP in the vmware session is what reboots. its not a hardware problem, otherwise it would kill linux too
It is most definitely a "hardware" problem - it's a virtual hardware problem. You are incorrectly assuming that VMWare doesn't have any bugs with emulating your hardware and running Windows.
I'm not claiming to know your specific situation, but the only BSOD situations that I've run accross are bad hardware (eg bad memory). One specific situation was with ASUS's poor handling of DDR400 memory when using dual channel mode. The box BSOD'd all the time. Dude was swearing at MS left and right. I said, "no way, Windows is rock solid (he laughed, I'm serious), fix your hardware". Needless to say, he built his own box. Maybe you've covered a sev 1 bug in win2k3, but I'd investigate other options before coming to that conclusion.
3. Stability. // most uptimes in Linux are measured in months and years rather than days and weeks (with exceptions, of course), and the GUI being a completely separate component from the kernel helps this greatly
/. experts claim great stability, but for the average user, I don't think that Linux's reputation for stability on the desktop (read: GUI) is with merit.
The GUI being completely seperate is the biggest reason why Linux, without the GUI, is so stable. I'll admit that it's been while since I've used Linux with a GUI, but when comparing it to Windows 2000 a few years ago, my RedHat box was extremely unstable. I know that 2-3 years can yield a lot of improvement, and I know that many
Tear everything down and start again.
Look at Windows 2003. They don't have the same usability req's as XP, so it's easier to secure. And it IS secure. It's not bulletproof, it's not OpenBSD, but how many serious exploits have made it into the wild, especially when compared to competing OS's? Windows XP SP2 looks to be a huge improvement - we'll just have to see. Either way, it seems they have a handle on it, without having to tear everything down and start again.
1) This is a trojan. While IE could be improved to help prevent this, this type of trojan can be used with any browser (albeit with a bit more social engineering effort would be required with most other browsers).
2) Yes, XP SP2 is a magic fix. I've seen the dialog screens for BHO's and the like. They're rediculously obvious. Furthermore, I believe that MS is _finally_ sandboxing this stuff (I remember reading it somewhere, but I can't verify). Finally, SP2's super aggressive firewall would detect that an unauthorized application was trying to send data via port 80 (or any port, for that matter) and warn the user. SP2 isn't bullet proof, but MS has put a LOT of resources into it to help minimize it's embarassing history. From what I've seen it looks promising, and hopefully my firewall will stop reporting so much NIMDA etc. traffic.
Microsoft are in a hugely powerful position to control other .Net implementations by changing things arbitrarily.
.NET and while they can add new features, making breaking changes will have a negative impact on their customers.
You have to remember that has a huge installbase of
You need to broaden search for .NET, not C#. A lot of .NET jobs are VB.NET, and a small percentage account for COBOL and FORTRAN .NET as well (a lot of mainframe migration projects use those languages).
The registry--nay, the entire concept of the registry, is an abomination.
.conf files lack is that of per user, or per group configuration. Microsoft has done a good job with xml based config for programs such as .NET and IIS that don't require per user configuration. I'm not sure if Longhorn does away with the registry for newer apps (I'm sure it still has it for legacy support), but it'll be interesting to see what they end up doing (I should get off my butt and just install the Longhorn disks 2ft. from me!).
While not the best solution for the goal, one of the important features of the registry that
I've never needed to edit the registry to install hardware. And I've installed some pretty off-the-wall hardware in my time (Windows user since Win95). Don't get me wrong, the registry is a mess (much better in NT based OS's), but it's not as bad as you make it out to be. And my WinXP on 3 different machines (one home-built, 2 laptops) runs just fine, no crashes, no reinstalls. It's not a perfect OS, but they've gotten a lot better of late.
Looking at the source is not an "at a glance" task. Depending on the problem, the programmer may need days or even weeks to understand the problem. Even at todays declined rates, that would cost thousands of dollars to the end user just to fix one problem.
Sure, the community can support the software, but if there's no commercial support (ie Redhat), than it's too risky for most people.
1.) Just how much exactly is Microsoft afraid of Linux?
/. keep asking the wrong question. The question is, "Just how much exaclty is Microsoft afraid of Apple". Apple is the one making huge headway on the desktop, not any Linux distro. I'm a developer who mainly works with MS technologies (C#, MSSQL Server, etc.) and it's amazing how many MS developers (even MS employees) have Powerbooks at home. The hardware is awesome, the OS solid, it's unix, and it has better office productivity software. Linux is an issue on the server side, but most "Linux wins" are wins against other unices, not server or desktop rollouts against Windows.
Most people on
Longhorn October PDC bits run slow but usable on 2Ghz w/512MB. By the time it's optimized for hardware (the latest builds I've seen mostly are) it will run faster. Beta bits will run even faster. And obviously the production bits will be even faster yet. I'm confident that it will run fine on todays hardware as long as you disable a few of the unnecessary eyecandy bits.
And who cares how much software costs in relation to hardware? Software is the reason you buy hardware. Just one game is 33-40% of a console's cost. If Longhorn has the features that people want, than spending the OEM $25-$40 (depending on OEM and version) on even a $399 value machine is totally worth it.
Try to develop a for-profit operating system to compete with Windows and you'll get crushed.
Apple seems to be doing quite nicely.
Apple's DRM is one of the most flexible. However, that doesn't mean it's flexible enough. I've barely even spent $100 on the iTunes store and the main reason is because I want to play my music on devices other than my iPod. I need to convert to MP3 to play on other devices. At least with WMA's DRM I can still use most all devices (except for my iPod!) since MS worked hard to put WMA on everything but your toaster. If Apple doesn't want me cracking their DRM, they better get their technology on a lot more devices.
No, commodity infrastructure software is no exception to that. That softwarwe does come at a price. At the price of thousands of volunteer developers who, for whatever reason, like working for free and donating their time and talents to for-profit organizations.
(Please no replies on how IBM and Apple pay developers to work on Linux and BSD respectively. This illustrates the exception and not the rule for most major OSS packages)
But this has nothing to do with the quality of healthcare. I know a lot of people who believe in active euthanasia for certain medical situations. Heck, there's been a lot of cases in the US where the same has happened except that people PAID for it with their own money! The only thing that's stopped people in the US is the law.
"Only a fool thinks price and value are the same." - Antonio Machado
That still doesn't negate the fact that value generally comes at a price.
Yet countries with social healthcare cover more of the elderly and other "non productive citzens" better than we do. Sure, we have the best technology (for some treatments) for the privilaged few. The tyranny of democracy is checked because 51% of Congress can never vote anything into existance. Most votes require 2/3rds majority. Our system is a representative republic, so theoretically more politically savvy people are doing the voting on many issues. Our system is nowhere near perfect, and I could critique it for hours, but a social healthplan does not mean the death of elders and retards.
Oh, and Milton Friedman's Nobel Prize was awarded via a democratic vote.
Will enterprise apps be developed with C,
.NET or J2EE for an enterprise level applications. Both .NET and J2EE support lower level programming (Managed C++/PInvoke and JNDI respectively) for the rare instances that direct hardware access is needed.
Not in the near future. It's too bug prone, insecure, and difficult to manage in larger codebases. Finally, the time to market with C is too costly for the enterprise.
Will C# get used for "enterprise apps".
It allready does, and we're talking mission critical Fortune 500. Google is your friend, use it before making rash statements. I'd much rather use
Yes, C/C++ will be around for a long time, but new applications will continue to use higher level platforms as those platforms mature.
I think that you're missing the point of this letter. The author is mainly critiquing the concept of an individual working on an OSS project for free. You're right, Google is profiting handsomly, and in part thanks to programmers who contributed to Linux and other OSS projects. Are those programmers making a dime off of Google's success? Other than a couple at Red Hat or maybe IBM, the answer is no.
My question the programmer addressed in this letter is, why on earth would you work for a for-profit organization without requiring compensation? Companies choosing to open up protocols or source code makes sense in certain scenarios. This is however very different then a programmer giving up their own personal time, which in my opinion just deteriorates our pofessional value as a whole.
It says that PC makers were restricted (by Microsoft's action) from including programs like RealPlayer, if they wanted to.
Which is why my Sony Vaio had RealOne preinstalled?
Why does real suck? ... I can turn off most of the spy features.
Case in point.