Well, special purpose ok, but lean? After seeing the CPU specifications (3 symmetric cores, each with 2threads and running at 3,2Ghz each with plenty of registers) I thought that's pretty high end hardware. May be it's cripped in some other way that I can't see?
It sure would be quite fun running Linux on this box for the power and form factor - all the power to the hackers!
People might be better off with the Coral Cache link - http://www.mi70k.com.nyud.net:8090/video.htm. (The site is hosted in India and they most probably had no idea they would hit slashdot:)
The project was called Mission Impossible 70K (MI70K) and there are some live videos detailing his flight (Flash required) on this site - http://www.mi70k.com/video.htm. The site also carries information about Mr. Singhania and his some other records.
I use bicycle when possible, I don't employ slaves, I do charity in my capacity to help starving people with food, and I will protest when my country wreaks havoc on another country.
And yes, I feed even the anonymous trolls - they might be starved. So in short I do everything in my capacity without bull s**tting, I guess if everyone did that, the world will be a better place (Slashdot included.)
...that good men do nothing. I guess good people in Nigeria did not do anything to stop the scammers even as they watched the whole country's credibility go down the drain. What you can do now, in my opinion, will have to be at a very large scale involving the government , the police and common man in Nigeria. Given the volume of Nigerian scammers it would have to be history's biggest revolt yet if it intends to regain the whole country's credibility.
Good luck.
Yep. The only remaining pain with x64 is 32bit PCI devices which still can't access RAM beyond the 4Gb boundary even when on a full blown 64bit OS. AMD sort of makes it better by providing a hardware IOMMU which does the bouncing in hardware (it still is limited - People doing graphics intensive stuff for example, still run into scarce IOMMU memory) so it's fast. Intel EMT64 doesn't have this in H/W - so there is a software layer required to do this which is obviously slower.
For an OS which is continually evolving and was not designed with a lot many future developments in mind, it is very natural to say no to the stable binary API/ABI concept for drivers. But as it matures and there is no longer a need to fix interfaces to support some out-of-world functionality, the driver interfaces are automatically going to be stabilized. (Unless kernel folks decide they get bored with having one function name for more than a year or that they want to keep driver writers continuosly on their toes - all of which is unlikely.)
API/ABI compatibility obviously has it's own pros and cons - some times it's impossible to break things, take Windows for example. The world is going with LP64 model for 64 bit machines but Windows developers had to stick with LLP64 just because they made some design mistakes and now they cannot break the tons of applications. (See http://blogs.msdn.com/oldnewthing/archive/2005/01/ 31/363790.aspx).
Linux on the other hand can afford to break and fix things until the time where binary and out-of-tree drivers grow to out number the in-tree stuff. By that time I guess there will be a very less need to break things such as driver interfaces and the like.
And I think the mad rush to put everything in the official kernel tree is not a good idea from maintenance and complexity stand point. So if and when the Linux ABI/API stabilizes that will be a good thing for out-of-tree kernel drivers and Linux itself.
Absolutely true, it just changes the perception and wastes virtual address space. They could separate the rt.jar into smaller pieces though - console, gui, networking, corba orb,.... they all could be separate jars mapped on demand. This has an added benefit of more available address space - server side apps need it to work better with the 32bit JVM's address space limits.
OP's comment is based on the fact that Java is usually deployed on those big iron Sun boxes (8-way minimum, 16-way, 32-way etc. with 32Gb RAM etc. - you get the picture.) Any darn thing runs reasonably fast on these machines. And when things execute in-memory and are JIT'ed there is no reason to believe Java code will execute slower than say C/C++ code. Plus lots of design patterns have grown to make up where Java performance sucks - Use huge thread pools to avoid creating threads on demand, cache as much as you can, pool objects to avoid doing new on big objects etc. Java/J2EE makes it possible to write robust server side applications quickly without having to deal with low level details. People pay a big price for this by having to buy those big boxes on the server side. It could be argued that smart C/C++ programmers can build something more reliable/scalable with less resources - but matter of the fact is that world is full of lazy "high-level" "abstract" "mediocre" programmers.
Destop is a different matter altogether - people still have machines with mere 256Mb RAM with one pale CPU - C/C++ rocks there. Java sucks badly on the desktop - Graphics/Drawing is slow, startup time is horrible, simple apps can require boat load of memory and on top of this the UI looks nasty.
Just a data point - Microsoft's VC# IDE requires 29Mb - it's written in C/C++ and embeds the C# VM. Netbeans and Eclipse - last I checked they needed atleast 100Mb!
All you said is true - but what prevents Microsoft from playing catch up? They are in a better position - money wise, market share wise, and technology wise, to do all the things Google is thought of doing. They have an OS, an office suite which is already ubiquitous and enough money to buy/build anything which matters and they don't already have. They recently started Windows OneCare which includes antivirus, online backups etc. - How long it would take for them to offer online Office on a pay per use basis?
Nothing Google is doing can kill Microsoft - they can only be forced to play catch up and they do that very well. They will continue to earn all the moolah till they can and when they can no longer, they still can afford to play the competitors' own game a little late and still can beat them at it.
In my experience MS cares a lot about backwards compatibility. So if you used standard C++ code (may be with some anti-standards tweaks to get the VC++ 6.0 compiler work - it's a shitty compiler when it comes to C++ standards compliance) you should be fine. The APIs are still there, albeit some are now deprecated like the unsafe string handling functions - Your code will still compile but with warnings.
Needless to say your projects will be converted automatically.
And you might want to clean up the code of any VC6 compiler specific kludges as the VS 2005 C++ compiler is nearly perfect when it comes to C++ standards compliance.
If you are using any 3rd party libraries compiled with VC++ 6.0 you might need to get an updated version compiled for VS2005. I found it problematic mixing C++ libraries compiled by VC6 and VC7, for example.
In our computer lab we had a very un-friendly admin. We used to hate him like anything. Our revenge was to screw up the Win 95 boxes assigned to us. They were the so called protected ones - hooked on to the Novell netware server, had no floppy / cd drives and no internet access. But being running Win95 we were easily able to achieve our goal of hosing the OS. We had to put in efforts only until we figured out the following, after which OS destruction was automagic! -
As I said the boxes didn't use to have floppy/cd drives and so re-installation of OS was very problematic. The net admin figured out a 'slick' way to deal with the situation -
a. Boot 95 to command prompt b. Enable network and CD to the novell netware share which hosted the Win 95 CDROM c. His theory being that reformatting destroys the hard drive - fdisk it. (Actual reason being that he won't be able to access the 95 CDROM share after the format) Remove all partitions and recreate them as-is. d. Run win95 setup from the netware share. e. Profit!
The machines generally remained on unless a software install required a reboot, in which case the partition table was re-read on boot and things were screwed up, once again.
The admin used to feel better by cursing at the software installation program which hosed the OS!!
He never found out why this was happening and we never bothered to tell him!!
Coz Dvorak misses the fundamental fact - Linux's desktop market share is too inisignificant to make a dent into Linux's overall market share - majority of which is on server side. AND Linux on Servers is an area where neither MS nor Apple are in a good position to compete - MS due to licensing/cost issues and Apple due to licensing, hardware support and scalability issues.
I actually tried to install and work around the failures on FC3, Gentoo and Debian. Did not work. I hadn't tried RHEL3 so I shouldn't have commented on that but I thought RHEL3 == FC3, but maybe not.
Netscape Directory used to have a GUI (Java Swing based, and so it used to suck but..) for performing routine management tasks (searching, setting up replication, adding/deleting/modifying the attributes etc.). So you can hope that the Fedora DS which is based on Netscape will have atleast some GUI tools to make an admin's life easier. Some one should probably write a Softerra LDAP Administrator alike tool integrated with the Fedora Directory server and it will be an ideal OSS directory server to use. And last time I checked Netscape Directory server was fairly easy to install.
[Better/Corrected version for English loving/.ers;)]
Your ignorance doesn't make it a fact - Ever heard of Anjuta IDE for GNOME? Others may not be opening a new OSS project but I know there are many working for the OSS cause not necessarily for money. (I contribute small things in my spare time if you needed proof.)
Just because of your ignorance doesn't make it a fact - Ever heard of Anjuta IDE for GNOME? Other may not be opening a new OSS project but I know there are many working for the OSS cause not necessarily for money. (I contribute small things in my spare time if you needed an proof.)
Windows error reporting does report kernel crashes back to Microsoft. If your machine crashed, on reboot it will identify which module (Driver, Kernel core, etc..) caused the crash and send the dump back to Microsoft server. If it was a driver from say, ATI, ATI people are able to see that data.
Linux may never quite reach that level of sophistication due to inherent diversity and non-uniformity of things. (Some distro adds a patch which main line doesn't have - there is no way to know automatically to send the crash due to this patch back to the Distro - etc. You get the point..)
I just bought the 2004 edition:) Not that Linux has problems running on it but would have been fun asking MS support about the problem that happens with dynamic CPU frequency change when running Linux - The CPU stays in Max power mode once you run a Linux VM under Virtual PC - It never goes back to normal on-demand operation unless you reboot.
I fail to understand a) Where can India-China collaborate and b)how exactly and who will benefit from this collaboration. Given the fact that both are populous countries with abundant and cheap labour, India doesn't benefit from cheap Chinese goods - We can and do manufacture them and that's best for our economy. Same might most likely be true with China. And as far as both countries competing for hardware and software business coming from the USA goes - In that situation there isn't a need for collaboration. At least I fail to imagine.
Educational and research level collaboration requires at least some commonality, common interests and plus some mutual trust.
So what's left to collaborate?
Don't worry - Given India and China's stark historical, political and cultural differences, it will never happen and even if it does, it will not succeed in practice. There has to be at least a little commonality somewhere in order for things of such mangnitude to materialize.
I run FC3 on x86_64 and find it quite okay after fixing certain nuances by hand. (Random hangs with kudzu, 2 minute hang if USB storage drive is attached while booting etc.)
Bad part is that Fedora as a distro doesn't get enough attention from the developers at Redhat. With limited developers, they obviously have time to fix only things which affect their customers. 10 out of the 10 bugs I submitted (they are serious ones) never got attention despite people confirming them.
I sincerely think Microsoft is turning all of its employees into trolls. If you take a peek at some of the MSDN blogs, you will see what I mean - They are engaged in drawing baseless conclusions and spreading FUD at any and all costs. Kind of shows how much they are hurt by Windows being insecure and Linux eating there market share.
Its high time we stopped feeding the trolls and giving them undeserved publicity. I mean I would not mind a fair comparison on sound basis but this is complete BS.
Slashdot Mirror
onto lean and special-purpose hardware?
Well, special purpose ok, but lean? After seeing the CPU specifications (3 symmetric cores, each with 2threads and running at 3,2Ghz each with plenty of registers) I thought that's pretty high end hardware. May be it's cripped in some other way that I can't see? It sure would be quite fun running Linux on this box for the power and form factor - all the power to the hackers!
People might be better off with the Coral Cache link - http://www.mi70k.com.nyud.net:8090/video.htm. (The site is hosted in India and they most probably had no idea they would hit slashdot :)
The project was called Mission Impossible 70K (MI70K) and there are some live videos detailing his flight (Flash required) on this site - http://www.mi70k.com/video.htm. The site also carries information about Mr. Singhania and his some other records.
I use bicycle when possible, I don't employ slaves, I do charity in my capacity to help starving people with food, and I will protest when my country wreaks havoc on another country. And yes, I feed even the anonymous trolls - they might be starved. So in short I do everything in my capacity without bull s**tting, I guess if everyone did that, the world will be a better place (Slashdot included.)
...that good men do nothing. I guess good people in Nigeria did not do anything to stop the scammers even as they watched the whole country's credibility go down the drain. What you can do now, in my opinion, will have to be at a very large scale involving the government , the police and common man in Nigeria. Given the volume of Nigerian scammers it would have to be history's biggest revolt yet if it intends to regain the whole country's credibility. Good luck.
Yep. The only remaining pain with x64 is 32bit PCI devices which still can't access RAM beyond the 4Gb boundary even when on a full blown 64bit OS. AMD sort of makes it better by providing a hardware IOMMU which does the bouncing in hardware (it still is limited - People doing graphics intensive stuff for example, still run into scarce IOMMU memory) so it's fast. Intel EMT64 doesn't have this in H/W - so there is a software layer required to do this which is obviously slower.
Parent has been wrongfully modded flamebait - PLEASE MOD UP.
For an OS which is continually evolving and was not designed with a lot many future developments in mind, it is very natural to say no to the stable binary API/ABI concept for drivers. But as it matures and there is no longer a need to fix interfaces to support some out-of-world functionality, the driver interfaces are automatically going to be stabilized. (Unless kernel folks decide they get bored with having one function name for more than a year or that they want to keep driver writers continuosly on their toes - all of which is unlikely.)
/ 31/363790.aspx).
API/ABI compatibility obviously has it's own pros and cons - some times it's impossible to break things, take Windows for example. The world is going with LP64 model for 64 bit machines but Windows developers had to stick with LLP64 just because they made some design mistakes and now they cannot break the tons of applications. (See http://blogs.msdn.com/oldnewthing/archive/2005/01
Linux on the other hand can afford to break and fix things until the time where binary and out-of-tree drivers grow to out number the in-tree stuff. By that time I guess there will be a very less need to break things such as driver interfaces and the like.
And I think the mad rush to put everything in the official kernel tree is not a good idea from maintenance and complexity stand point. So if and when the Linux ABI/API stabilizes that will be a good thing for out-of-tree kernel drivers and Linux itself.
Absolutely true, it just changes the perception and wastes virtual address space. They could separate the rt.jar into smaller pieces though - console, gui, networking, corba orb, .... they all could be separate jars mapped on demand. This has an added benefit of more available address space - server side apps need it to work better with the 32bit JVM's address space limits.
OP's comment is based on the fact that Java is usually deployed on those big iron Sun boxes (8-way minimum, 16-way, 32-way etc. with 32Gb RAM etc. - you get the picture.) Any darn thing runs reasonably fast on these machines. And when things execute in-memory and are JIT'ed there is no reason to believe Java code will execute slower than say C/C++ code. Plus lots of design patterns have grown to make up where Java performance sucks - Use huge thread pools to avoid creating threads on demand, cache as much as you can, pool objects to avoid doing new on big objects etc. Java/J2EE makes it possible to write robust server side applications quickly without having to deal with low level details. People pay a big price for this by having to buy those big boxes on the server side. It could be argued that smart C/C++ programmers can build something more reliable/scalable with less resources - but matter of the fact is that world is full of lazy "high-level" "abstract" "mediocre" programmers. Destop is a different matter altogether - people still have machines with mere 256Mb RAM with one pale CPU - C/C++ rocks there. Java sucks badly on the desktop - Graphics/Drawing is slow, startup time is horrible, simple apps can require boat load of memory and on top of this the UI looks nasty. Just a data point - Microsoft's VC# IDE requires 29Mb - it's written in C/C++ and embeds the C# VM. Netbeans and Eclipse - last I checked they needed atleast 100Mb!
All you said is true - but what prevents Microsoft from playing catch up? They are in a better position - money wise, market share wise, and technology wise, to do all the things Google is thought of doing. They have an OS, an office suite which is already ubiquitous and enough money to buy/build anything which matters and they don't already have. They recently started Windows OneCare which includes antivirus, online backups etc. - How long it would take for them to offer online Office on a pay per use basis? Nothing Google is doing can kill Microsoft - they can only be forced to play catch up and they do that very well. They will continue to earn all the moolah till they can and when they can no longer, they still can afford to play the competitors' own game a little late and still can beat them at it.
In my experience MS cares a lot about backwards compatibility. So if you used standard C++ code (may be with some anti-standards tweaks to get the VC++ 6.0 compiler work - it's a shitty compiler when it comes to C++ standards compliance) you should be fine. The APIs are still there, albeit some are now deprecated like the unsafe string handling functions - Your code will still compile but with warnings.
Needless to say your projects will be converted automatically.
And you might want to clean up the code of any VC6 compiler specific kludges as the VS 2005 C++ compiler is nearly perfect when it comes to C++ standards compliance.
If you are using any 3rd party libraries compiled with VC++ 6.0 you might need to get an updated version compiled for VS2005. I found it problematic mixing C++ libraries compiled by VC6 and VC7, for example.
In our computer lab we had a very un-friendly admin. We used to hate him like anything. Our revenge was to screw up the Win 95 boxes assigned to us. They were the so called protected ones - hooked on to the Novell netware server, had no floppy / cd drives and no internet access. But being running Win95 we were easily able to achieve our goal of hosing the OS. We had to put in efforts only until we figured out the following, after which OS destruction was automagic! -
As I said the boxes didn't use to have floppy/cd drives and so re-installation of OS was very problematic. The net admin figured out a 'slick' way to deal with the situation -
a. Boot 95 to command prompt
b. Enable network and CD to the novell netware share which hosted the Win 95 CDROM
c. His theory being that reformatting destroys the hard drive - fdisk it. (Actual reason being that he won't be able to access the 95 CDROM share after the format) Remove all partitions and recreate them as-is.
d. Run win95 setup from the netware share.
e. Profit!
The machines generally remained on unless a software install required a reboot, in which case the partition table was re-read on boot and things were screwed up, once again.
The admin used to feel better by cursing at the software installation program which hosed the OS!!
He never found out why this was happening and we never bothered to tell him!!
Coz Dvorak misses the fundamental fact - Linux's desktop market share is too inisignificant to make a dent into Linux's overall market share - majority of which is on server side. AND Linux on Servers is an area where neither MS nor Apple are in a good position to compete - MS due to licensing/cost issues and Apple due to licensing, hardware support and scalability issues.
I actually tried to install and work around the failures on FC3, Gentoo and Debian. Did not work. I hadn't tried RHEL3 so I shouldn't have commented on that but I thought RHEL3 == FC3, but maybe not.
Except that Sun JSDS does not work on anything other than RedHat 7.2 or RHEL2 which is not what most people use today.
Netscape Directory used to have a GUI (Java Swing based, and so it used to suck but..) for performing routine management tasks (searching, setting up replication, adding/deleting/modifying the attributes etc.). So you can hope that the Fedora DS which is based on Netscape will have atleast some GUI tools to make an admin's life easier. Some one should probably write a Softerra LDAP Administrator alike tool integrated with the Fedora Directory server and it will be an ideal OSS directory server to use. And last time I checked Netscape Directory server was fairly easy to install.
[Better/Corrected version for English loving /.ers ;)]
Your ignorance doesn't make it a fact - Ever heard of Anjuta IDE for GNOME? Others may not be opening a new OSS project but I know there are many working for the OSS cause not necessarily for money. (I contribute small things in my spare time if you needed proof.)
Just because of your ignorance doesn't make it a fact - Ever heard of Anjuta IDE for GNOME? Other may not be opening a new OSS project but I know there are many working for the OSS cause not necessarily for money. (I contribute small things in my spare time if you needed an proof.)
Windows error reporting does report kernel crashes back to Microsoft. If your machine crashed, on reboot it will identify which module (Driver, Kernel core, etc..) caused the crash and send the dump back to Microsoft server. If it was a driver from say, ATI, ATI people are able to see that data. Linux may never quite reach that level of sophistication due to inherent diversity and non-uniformity of things. (Some distro adds a patch which main line doesn't have - there is no way to know automatically to send the crash due to this patch back to the Distro - etc. You get the point..)
I just bought the 2004 edition :) Not that Linux has problems running on it but would have been fun asking MS support about the problem that happens with dynamic CPU frequency change when running Linux - The CPU stays in Max power mode once you run a Linux VM under Virtual PC - It never goes back to normal on-demand operation unless you reboot.
I fail to understand a) Where can India-China collaborate and b)how exactly and who will benefit from this collaboration. Given the fact that both are populous countries with abundant and cheap labour, India doesn't benefit from cheap Chinese goods - We can and do manufacture them and that's best for our economy. Same might most likely be true with China. And as far as both countries competing for hardware and software business coming from the USA goes - In that situation there isn't a need for collaboration. At least I fail to imagine. Educational and research level collaboration requires at least some commonality, common interests and plus some mutual trust. So what's left to collaborate?
Don't worry - Given India and China's stark historical, political and cultural differences, it will never happen and even if it does, it will not succeed in practice. There has to be at least a little commonality somewhere in order for things of such mangnitude to materialize.
I run FC3 on x86_64 and find it quite okay after fixing certain nuances by hand. (Random hangs with kudzu, 2 minute hang if USB storage drive is attached while booting etc.) Bad part is that Fedora as a distro doesn't get enough attention from the developers at Redhat. With limited developers, they obviously have time to fix only things which affect their customers. 10 out of the 10 bugs I submitted (they are serious ones) never got attention despite people confirming them.
I sincerely think Microsoft is turning all of its employees into trolls. If you take a peek at some of the MSDN blogs, you will see what I mean - They are engaged in drawing baseless conclusions and spreading FUD at any and all costs. Kind of shows how much they are hurt by Windows being insecure and Linux eating there market share. Its high time we stopped feeding the trolls and giving them undeserved publicity. I mean I would not mind a fair comparison on sound basis but this is complete BS.