I am not very experienced with matters regarding the linux kernel, but it seems some what surprising that IT doesn't build in cygwin. Assuming you have the proper verson of GCC and all the correct header files what would stop it from compiling?
Is it safe to assume that 2.6.1 is stable and secure to use, or would it be more reasonable to wait until it enters more main stream usage and possible problems are exposed?
Have there been any articles or reviews comparing the performance of various kernel versions? I'd find it interesting how much progress has been made in areas such as network throughput, disk access, etc. I guess its possible there isn't any more room for progress in some areas, I would find that interesting too.
600 million. Yeah just hold on a sec while I pull that out of my ass =)
Maybe NASA should be turned into or replaced by a global organization so some of this cost can be spread out a bit. I also wonder if they could cut costs by giving tasks to universities or developing some of the required software in an open community manner.
I also wonder how much thought they put into the servicability of their craft. If certain parts require replacement on a given interval it would make sense to place them (if possible) in location which they could be easily replaced.
I'm sure if the design was rather modular you could even perform routine maintainace with expendable unmanned service vehicals.
I have written perl code that runs on scheduled intervals that will gather, compress and upload my data securely to a remote server. The script filters out unnessasry files such as object files, other archives and executables. I also have the ability to incremental backups based on file dates. Using this script I can perform a full (minus the filtered file types) backup of directory tree containing roughly 20GB of files into a file that is ~50MB.
I also manually run the script after I write a piece of code or perform a certain task that I absolutely do not care to do again. It runs in the background and I don't even know its there.
Look. I work for a company. I don't run the company. This company more or less tells me what to do. I don't have the power to tell the company I work for to release everything as open source. I am well aware of the benefits of open source developement.
As a software developer the GPL scares me. Personally I try to stay far away from using any GPL code because I am affraid I will not properly credit it or in some other way infringe on the licence agreement. For instance, I don't have a clue whether or not its ok to use a dynamically linked GPL library with a closed source app. Honestly I don't even really know how I would find answers to this short of posting here or hiring a lawyer.
As a result I probably end up writing a lot more code from scratch than I probably need to. I usually avoid all types of "open source" since im affraid ill accidently do something wrong.
One thing that I would like to see is developers changing the license on some of their code to a less restrictive license once the code becomes very trivial. That is once the code has reached a point where there are MANY other implimentations, and development has stopped, it would seem to me that the GPL would do more harm than good.
I am getting quite tired of these silly comparision between programming languages. Obviously, depending on the requirements of the task you need to complete a certain programming language may prove to be more useful than another. Obviously you aren't going to want to write an interrupt service routine or other such time critical platform dependant piece of code in java. But for something that needs to be platform independant java may be a viable solution.
So... pick a language, then do your job. repeat.
Obviously you aren't going to drive an SUV in a Formula 1 race, but maybe if you need to transport 7 people and tow a 20 foot long boat up to the mountains, it might be a good choice.
Im a programmer for a provider of automation equipment for electric utilities. One thing that many may not realize is that it actually requires power to operate much of the power grid. Once the power is gone things like substation batterys quickly die out. Once they die generators have to be brought in to supply things like breakers with energy to operate. There are thousands of substations and other portions of the grid that take a long time to bring back.
In addition some power companys have switched to completely "high tech" systems in which power has to be present to operate physical equipment and power to operate things like fiber ethernet infastructure. In other words some power companies do not have a means to control equipment in anyway other than over a network which requires power to operate.
It could be argued that a power grid is much more difficult to maintain than a data network due to the fact that the service which it provides isnt required to provide the service that it provides =). A router can go down but it can always be replaced and power and network hookups will be waiting for it.
Due to these factors power grids are very vulnerable to the domino effect.
An operating system could be designed to selectively allow remote hosts control over portions of the system.
For instance an operating system could allow any host to kill the processes that had open sockets to that host. You could even narrow that down to allowing a host to kill any process which had an open socket not originated by the host. This would disable remote hosts from killing servers that they initiate connections to but would allow processes to be killed that originated connections to remote hosts.
At the operating system or even hardware level you could also engineer a system in which a remote host could request a stop on all traffic targeted at it. If this could be done securely people that were being targeted by another machine could request the operating system to stop all processes from directing traffic its direction.
I find that NuMega (Compuware)'s suite of debugging tools to be much better than anything else that I have used. BoundsChecker especially is an extremely useful tool for find bugs in C/C++ code.
With a traditional debugger finding memory leaks, overflows, and other errors that may not show symptoms imidately can be a lot of work.
I think that it is asinine to believe that such tools haven't improved the debugging process in the last 30 years.
I like static buffer sizes. Many times dynamically sizing buffers are inefficient and can cause memory fragmentation (maybe leaking). You JUST NEED TO USE strncpy instead of strcpy! Dynamic buffers can also cause problems if their size is unrestricted. I wouldn't be surprised if there was a large amount of software out there that uses dynamic buffers which allow you to run a system out of memory..
When I make backups that I intend to keep for a long time I try to include text based specifications for the file formats that I am backing up. Obviously this isnt possible with some formats but with things like MP3/JPEG/MPEG it is. So if worst comes to worst I can write a jpeg viewer in 65 years to look at my anchient digital camera pics.
I believe that wasting CPU cycles that could otherwise be used to advance different fields of science is ridiculous. The SETI at home client is ran by people at thousands of companys and institutions around the world none of which (that I can find) have reported any legitimate security problems regarding the client. Quite frankly I find the argument that "you arent supposed to run what your company doesnt want you to" completely ridiculous. If I know a program you wish to run is safe (or pretty friggin certain at least) and doesnt effect your work productivity why should you let the advancement of science fall to the way side! With all these morons opening their binary executable files they get from complete strangers in MS Outlook do you think the main security risks to a system is through SETI???
To the argument of all the people whining about SETI at home not being open source I have only one thing to say. SETI is already having enough problems with people cracking the client to send back fake result so they can moronically increase their score. Just imagine how many results would taint this scientific research project if they allowed people to download the source code so every script kiddy could have his/her way with it. Believe it or not there ARE legitimate reasons for keep things closed source in some instances. If the SETI team engineers a method of verifying all results without wasting too much cpu power (not bloodly likely) they probably would release the source.
Clearly the here problem resides in the implimentation of the client/server design of these applications (games). For instance if you had a website with some information you didnt want people to see unless they entered a password, simply making the text and background colors the same wouldnt be a very secure way of hiding it.
Similarly if you dont want information to be misused in a game you simply dont send cordinant information of enemy players that arent detected as being viewable or very close to viewable to the client (obviously you would need to find balance between security and performance as not having information early enough could break prediction code...)
I dont know a whole lot of the details of the protocols in question here but at work we are charged with hacking protocols on a daily basis to emulate propriatary programs and devices. If there isnt already detailed information available about these protocols it wouldnt be that hard to learn their secrets. Once you did that you could write sophisticated programs that would peek in on these packets before they got to their final destination and displayed this cheat information in a user friend format. (Infact I have heard of cheat proxy servers that do stuff similar to this.)
Bottom line is the graphics drivers cant draw what isnt there in the first place. Dont send this info to the client. And definately dont render it to the screen and hope it will get covered up by a few polygons! If making this secure would inhibit performance of the games simply being able to turn this protection on and off (hopefully on the server side) would let users decide for them selves what was more important: speed or fair compitition.
I wonder if it would be possible to devise some sort of system for cheating in online games like Everquest or Online gambling sites based on predicting pseudo randomly generated numbers. Not that I would ever do such a thing...
Doesn't the Dreamcast run on Windows CE? Doesn't windows ce have a hardware transparent network layer and tcp/ip stack ?!?!
Why would a developer side step an already existing transparent network layer in favor of something that required applications to provide support for multiple piece of network hardware when they shouldn't have to provide any such support in the first place !?!?
Maybe I am missing something because this just doesnt make any sense to me.
I mean common. Are these submissions for patents even read before they are approved? I'm guessing some government workers are getting lazy and just stamping every application that hits their desk.
Slashdot Mirror
I run mandrake 9.1 under windows XP in VMWARE at work. I have a fast machine with tons of RAM so its just like having a dedicated linux box.
I am not very experienced with matters regarding the linux kernel, but it seems some what surprising that IT doesn't build in cygwin. Assuming you have the proper verson of GCC and all the correct header files what would stop it from compiling?
Is it safe to assume that 2.6.1 is stable and secure to use, or would it be more reasonable to wait until it enters more main stream usage and possible problems are exposed?
Have there been any articles or reviews comparing the performance of various kernel versions? I'd find it interesting how much progress has been made in areas such as network throughput, disk access, etc. I guess its possible there isn't any more room for progress in some areas, I would find that interesting too.
600 million. Yeah just hold on a sec while I pull that out of my ass =)
Maybe NASA should be turned into or replaced by a global organization so some of this cost can be spread out a bit. I also wonder if they could cut costs by giving tasks to universities or developing some of the required software in an open community manner.
I also wonder how much thought they put into the servicability of their craft. If certain parts require replacement on a given interval it would make sense to place them (if possible) in location which they could be easily replaced.
I'm sure if the design was rather modular you could even perform routine maintainace with expendable unmanned service vehicals.
I have written perl code that runs on scheduled intervals that will gather, compress and upload my data securely to a remote server. The script filters out unnessasry files such as object files, other archives and executables. I also have the ability to incremental backups based on file dates. Using this script I can perform a full (minus the filtered file types) backup of directory tree containing roughly 20GB of files into a file that is ~50MB.
I also manually run the script after I write a piece of code or perform a certain task that I absolutely do not care to do again. It runs in the background and I don't even know its there.
I have a feeling the sign may read "2600".
Look. I work for a company. I don't run the company. This company more or less tells me what to do. I don't have the power to tell the company I work for to release everything as open source. I am well aware of the benefits of open source developement.
As a software developer the GPL scares me. Personally I try to stay far away from using any GPL code because I am affraid I will not properly credit it or in some other way infringe on the licence agreement. For instance, I don't have a clue whether or not its ok to use a dynamically linked GPL library with a closed source app. Honestly I don't even really know how I would find answers to this short of posting here or hiring a lawyer.
As a result I probably end up writing a lot more code from scratch than I probably need to. I usually avoid all types of "open source" since im affraid ill accidently do something wrong.
One thing that I would like to see is developers changing the license on some of their code to a less restrictive license once the code becomes very trivial. That is once the code has reached a point where there are MANY other implimentations, and development has stopped, it would seem to me that the GPL would do more harm than good.
Thats just my take.
I am getting quite tired of these silly comparision between programming languages. Obviously, depending on the requirements of the task you need to complete a certain programming language may prove to be more useful than another. Obviously you aren't going to want to write an interrupt service routine or other such time critical platform dependant piece of code in java. But for something that needs to be platform independant java may be a viable solution.
So... pick a language, then do your job. repeat.
Obviously you aren't going to drive an SUV in a Formula 1 race, but maybe if you need to transport 7 people and tow a 20 foot long boat up to the mountains, it might be a good choice.
Im a programmer for a provider of automation equipment for electric utilities. One thing that many may not realize is that it actually requires power to operate much of the power grid. Once the power is gone things like substation batterys quickly die out. Once they die generators have to be brought in to supply things like breakers with energy to operate. There are thousands of substations and other portions of the grid that take a long time to bring back.
In addition some power companys have switched to completely "high tech" systems in which power has to be present to operate physical equipment and power to operate things like fiber ethernet infastructure. In other words some power companies do not have a means to control equipment in anyway other than over a network which requires power to operate.
It could be argued that a power grid is much more difficult to maintain than a data network due to the fact that the service which it provides isnt required to provide the service that it provides =). A router can go down but it can always be replaced and power and network hookups will be waiting for it.
Due to these factors power grids are very vulnerable to the domino effect.
An operating system could be designed to selectively allow remote hosts control over portions of the system.
For instance an operating system could allow any host to kill the processes that had open sockets to that host. You could even narrow that down to allowing a host to kill any process which had an open socket not originated by the host. This would disable remote hosts from killing servers that they initiate connections to but would allow processes to be killed that originated connections to remote hosts.
At the operating system or even hardware level you could also engineer a system in which a remote host could request a stop on all traffic targeted at it. If this could be done securely people that were being targeted by another machine could request the operating system to stop all processes from directing traffic its direction.
I find that NuMega (Compuware)'s suite of debugging tools to be much better than anything else that I have used. BoundsChecker especially is an extremely useful tool for find bugs in C/C++ code.
With a traditional debugger finding memory leaks, overflows, and other errors that may not show symptoms imidately can be a lot of work.
I think that it is asinine to believe that such tools haven't improved the debugging process in the last 30 years.
I like static buffer sizes. Many times dynamically sizing buffers are inefficient and can cause memory fragmentation (maybe leaking). You JUST NEED TO USE strncpy instead of strcpy! Dynamic buffers can also cause problems if their size is unrestricted. I wouldn't be surprised if there was a large amount of software out there that uses dynamic buffers which allow you to run a system out of memory..
When I make backups that I intend to keep for a long time I try to include text based specifications for the file formats that I am backing up. Obviously this isnt possible with some formats but with things like MP3/JPEG/MPEG it is. So if worst comes to worst I can write a jpeg viewer in 65 years to look at my anchient digital camera pics.
I believe that wasting CPU cycles that could otherwise be used to advance different fields of science is ridiculous. The SETI at home client is ran by people at thousands of companys and institutions around the world none of which (that I can find) have reported any legitimate security problems regarding the client. Quite frankly I find the argument that "you arent supposed to run what your company doesnt want you to" completely ridiculous. If I know a program you wish to run is safe (or pretty friggin certain at least) and doesnt effect your work productivity why should you let the advancement of science fall to the way side! With all these morons opening their binary executable files they get from complete strangers in MS Outlook do you think the main security risks to a system is through SETI??? To the argument of all the people whining about SETI at home not being open source I have only one thing to say. SETI is already having enough problems with people cracking the client to send back fake result so they can moronically increase their score. Just imagine how many results would taint this scientific research project if they allowed people to download the source code so every script kiddy could have his/her way with it. Believe it or not there ARE legitimate reasons for keep things closed source in some instances. If the SETI team engineers a method of verifying all results without wasting too much cpu power (not bloodly likely) they probably would release the source.
Clearly the here problem resides in the implimentation of the client/server design of these applications (games). For instance if you had a website with some information you didnt want people to see unless they entered a password, simply making the text and background colors the same wouldnt be a very secure way of hiding it. Similarly if you dont want information to be misused in a game you simply dont send cordinant information of enemy players that arent detected as being viewable or very close to viewable to the client (obviously you would need to find balance between security and performance as not having information early enough could break prediction code...) I dont know a whole lot of the details of the protocols in question here but at work we are charged with hacking protocols on a daily basis to emulate propriatary programs and devices. If there isnt already detailed information available about these protocols it wouldnt be that hard to learn their secrets. Once you did that you could write sophisticated programs that would peek in on these packets before they got to their final destination and displayed this cheat information in a user friend format. (Infact I have heard of cheat proxy servers that do stuff similar to this.) Bottom line is the graphics drivers cant draw what isnt there in the first place. Dont send this info to the client. And definately dont render it to the screen and hope it will get covered up by a few polygons! If making this secure would inhibit performance of the games simply being able to turn this protection on and off (hopefully on the server side) would let users decide for them selves what was more important: speed or fair compitition.
I wonder if it would be possible to devise some sort of system for cheating in online games like Everquest or Online gambling sites based on predicting pseudo randomly generated numbers. Not that I would ever do such a thing...
Doesn't the Dreamcast run on Windows CE? Doesn't windows ce have a hardware transparent network layer and tcp/ip stack ?!?! Why would a developer side step an already existing transparent network layer in favor of something that required applications to provide support for multiple piece of network hardware when they shouldn't have to provide any such support in the first place !?!? Maybe I am missing something because this just doesnt make any sense to me.
I mean common. Are these submissions for patents even read before they are approved? I'm guessing some government workers are getting lazy and just stamping every application that hits their desk.