It's about time that FPS sooting games evolve.
on
Carmack Speaks
·
· Score: 3
Imagine entering a room to see three people. One seated, the two others standing over the seated person. The walls are covered in book shelves. The standing person closest to you pulls a gun and dives behind a couch as he shoots at you. The other standing person pulls a gun and starts to aim it at the seated person.
Think about a FPS game where you have much wider environment to work in, and much more complex tasks to perform. Solve puzzles, fly a chopper, etc..
What OS did you start on, when you first touched a PC?
First "home" computer: UCSD P-System on a Sage II hardware. I then moved up to UNIX System V.2 on a Stride 440. I didn't have an IBM compatible PC till the 90's and then only because I could install Linux on it. With the exception of work I don't use MS Windows and even then it's only because they make me. I'm much happier with any UNIX workstation on my desk than I am with a Windows PC.
Data from the Netcraft Web Server Survey. Of a total of 13,106,190 servers surveyed about 21% are Microsoft based, or 2,742,931 servers (actual March count). Figuring an average of 2 minutes to login and delete each file at an average pay+overhead rate of $50 an hour for the web admins deleting the.dll you get a cost of about $4,571,552 just to delete the files.
MicroSoft dosen't know about the little wires one embeds in security glass to keep it from falling apart when broken. Instead they ship pre broken plain window pain glass with their windows.
The more I think about this the more I feel it is down right criminal to stick a backdoor like this into code. This can lead to massive problems for both individuals and businesses that have data stollen. Look at the trouble that was caused to the credit card companies with the stollen credit card number lists from web servers. Add on top of that the fraudlent charges to peoples credit cards. Deliberate backdoors like this and others make it all that much easier for a cracker or script kiddy to break into a system. Who knows if this was the exploit used in any of the previous security breaches. I'd bet that some used it. maby not all, but some.
Ponder this: Would you accept a security system for your house if you knew it could be bypassed by anybody with a standard code?
Forget zip disks. They draw way to much power and are way to delicate. Use Compactflash cards. You can jumper them and they behave just like an IDE HD so you can connect them to the IDE interface on the doughter board. You can get them up to 192 Mbytes now. No moving parts to wear out. No heads to crash. Dust has no effect on them. They may cost more per MB, but they are so much more robust and draw so much less power.
As for adding a pair of DACs. It shouldn't be all that hard. I'd recomend looking at using a FPGA to implement DMA transfer and addressing for the DACs. They could be put on their own card along with a headphone amplifyer circuit.
I've always wanted a encryptor/decryptor board that is a huge FPGA or similar with a PCI busmaster interface. It would be better if it could hold a large number of differnt keys that the OS then can control the setting and use of. The key space would be write only memory so they can't be read back by any process. Only written to. The board would operate in burst mode when takling on the PCI bus. It would transfer in a block of memory, encrypt it, then burst mode transfer it out. It's operation would be much like a DMA controller, but with the encryption added in. By using a FPGA one can change the programming of the board as new methods come out. One would need a program protect jumper, but that is easy to do.
In Minneapolis it's unfortunatly overcast... This really bummed me out as the proton levels were some of the highest I've seen in quite awhile. On NASA's Space Weather site they posted an alert. Proton levels were up over 20 per cc when they are usually down in the 1-5 range. The solar wind was also up much higher than normal. As I'm posting this the protons/cc is only 1.2. Not very good for Aurora.
I couldn't get through when using www.slashdot.org, but when I tried slashdot.org it worked. There seams to be alot of name server dropouts on the net lately. You guys need to keep your name servers code up to date. The Bind that comes with RedHat 6.1 on the CD has a known root exploit that is actively being exploited. You need to upgrade to BIND 8.2.2 patchlevel 5.
If I was to tell you that you can buy a gun in a gun store, then obviously I have committed no crime. But if I was to leave a loaded gun on the sidewalk outside a school, then a case could be made that I am at least partly responsible for the consequences of someone picking up and using that gun.
There is a very important difference here. In your analogy you were the one to place the "gun"/"code" there, not someone else. 2600 is not the one who is publishing the DeCSS code, but only the one pointing to where it can be found.
Compiled binaries can be looked at as a translation of the original work. In this light they could be protected as free speach. This would also protect code from a code generator. The input code is free speach, so the output code would also be.
As an alternate: We need to come up with a general purpose program that knows about many different encryption methods and file formats, and how to get around them. As part of this program it should have the ability to deal with new methods and file formats with either loadable modules or easily extensible source. It should have automated tools for dealing with all known encryption methods, and all known wrappers to those encryption methods.
The program would be a general purpose tool. It would contain generic tools for analysing and decrypting encrypted data in files, streams, etc. Once a specific format is determined a template could be built to automate the decryption process for that format. The point is the tool is generic, and not specific to any one system. The templates allow the transfer of knowlege of how to deal with specific systems.
The path forward is clear to me. We need to come up with a general purpose program that knows about many different encryption methods and file formats, and how to get around them. As part of this program it should have the ability to deal with new methods and file formats with either loadable modules or easily extensible source. It should have automated tools for dealing with all known encryption methods, and all know wrappers to those encryption methods.
I just remembered that at one of the former places I worked they used an intermediary system to allow some really old systems to be served by their main SAN array. Doing something like this may be a possibility to handle systems unsupported by a NAS system. I'd first check to see if it was possible, but it's worth a try. The intermediary system served an old network file system format to the old boxes while actually getting the files via NFS from the SAN system. It did slow things down some, but it worked.
This may end up being the solution to stay under budget, but you do lose alot going this way. One thing is well integrated automated backups. Some of the NAS and SAN products have very good backup abillities built in. Even on an hour by hour basis. Great for those absent minded users who delete critical reports just before they are due.
If you decide to roll your own a few things to remember:
If you use mirrored RAID, place each image on separate SCSI controllers. This way a SCSI controller can die and the system can continue to work.
Lots of RAM for disk caching.
A mirrored/stripped RAID setup will have lower CPU overhead than RAID 5, but if your CPU is very fast this may not be a factor.
How you stripe your disks is important.
Don't skimp on RAM volume. Buy lots of it. you can skimp some on speed but only if you are doing it to get more RAM for the same cost. Your RAM speed isn't the bottleneck, your disk controllers, harddisk, and network speeds are.
For any mirrired or RAID setup buy a few extra disks for replacements, then mount them in the chasis. Hook one disk per SCSI controller up to power and the SCSI chain. It will be your "hot" backup disk. If possible have it not spin up. note you will need to have software to tell it to spin up when it is needed.
DLT tapes are your friend, and place them on their own controller. They are fast, and hold large amounts. Even at 600MB/minute it still takes a long time to backup 100GBytes. 2:51 assuming you get full transfer speed on the tape. It would be better to plan 6 hours per backup and a robot tape switcher from the start.
If you find out backups take longer than you can have the system down. Consider a third set of disks that can be a backup set where you do a copy to them each night, then backup durring the day.
I hope this helps some. In the long run I feel you will likely be happier with a NAS or SAN system.
This is extreemly important. I have so many tools on my resume because I was always been willing to say, shure I'll try, learn, and do it. It sure helps that I bothered to learn the fundamental computer science theories and am willing to learn more.
My own technical book collection is now a couple of shelves larger than the 4 foot wide floor to ceiling book case I originally allocated for it. Sure some of the books are now old and dusty, but then I've been collecting for 2 decades. The dustier ones will soon be cataloged evicted to boxes till I get a larger place. Sure it's expensive to maintain your own persomal technical library, but I've found it indespensible. I use books as refferences nearly daily. I consider it much more important to know how to research and look up something than it is to know it off the top of the head. More often than not the memory isn't quite all there, but you know that printed page ratains what it was told very faithfully. It's also much easier to hand to somebody else so you can off load the task and get on to some more important task.
Some of us fully under stand Microsoft's browser war tatics, but refuse to lower ourselves to their level. The point of your resume is to show how good you are at doing your job. If was looking at your site to determine if you were a suitable canidate for a web coder or admin I would reject you immediatly. For the plain and simple reason you don't show you know how to communicate to the general web user population. Further more it shows that you place confrontation above communication. Both traits are bad in an employee.
What other format is there? Oh, okay...I'll use WordPad...
Plain text or simple HTML are very good universal formats. If somebody can't deal with either of them, are you sure you want to work for them?
I once maintained my resume in MS Word format, but decided that it was a waste maintaining a Windows system at home just so I could maintain my resume. I now have my resume posted on my web site. I keep the HTML in it and my web site simple as it is there for information transfer, not glitz. If someone needs it in text format I will transfer it over to text buy having Lynx translate it. The resume is currently in flux as I'm updating it for a new job search. I'm looking for a place that will allow me to only work 30 hours a week preferably in web security or programming.
Slashdot Mirror
Imagine entering a room to see three people. One seated, the two others standing over the seated person. The walls are covered in book shelves. The standing person closest to you pulls a gun and dives behind a couch as he shoots at you. The other standing person pulls a gun and starts to aim it at the seated person.
Think about a FPS game where you have much wider environment to work in, and much more complex tasks to perform. Solve puzzles, fly a chopper, etc..
First "home" computer: UCSD P-System on a Sage II hardware. I then moved up to UNIX System V.2 on a Stride 440. I didn't have an IBM compatible PC till the 90's and then only because I could install Linux on it. With the exception of work I don't use MS Windows and even then it's only because they make me. I'm much happier with any UNIX workstation on my desk than I am with a Windows PC.
Oops, grabbed the count from Developers list rather than Servers list, but the total cost is roughly the same...
Some fast and lose numbers...
Data from the Netcraft Web Server Survey. Of a total of 13,106,190 servers surveyed about 21% are Microsoft based, or 2,742,931 servers (actual March count). Figuring an average of 2 minutes to login and delete each file at an average pay+overhead rate of $50 an hour for the web admins deleting the .dll you get a cost of about $4,571,552 just to delete the files.
MicroSoft dosen't know about the little wires one embeds in security glass to keep it from falling apart when broken. Instead they ship pre broken plain window pain glass with their windows.
The more I think about this the more I feel it is down right criminal to stick a backdoor like this into code. This can lead to massive problems for both individuals and businesses that have data stollen. Look at the trouble that was caused to the credit card companies with the stollen credit card number lists from web servers. Add on top of that the fraudlent charges to peoples credit cards. Deliberate backdoors like this and others make it all that much easier for a cracker or script kiddy to break into a system. Who knows if this was the exploit used in any of the previous security breaches. I'd bet that some used it. maby not all, but some.
Ponder this: Would you accept a security system for your house if you knew it could be bypassed by anybody with a standard code?
All code should have code reviews. Where were their code reviews? Who missed this? Did somebody let it go buy?
I'm not surprised so many IIS sites have been hacked. I'm wondering what other gems are in their web server.
Before you think that this problem dosen't effect you, consider the web sites you frequent and buy from.
Forget zip disks. They draw way to much power and are way to delicate. Use Compactflash cards. You can jumper them and they behave just like an IDE HD so you can connect them to the IDE interface on the doughter board. You can get them up to 192 Mbytes now. No moving parts to wear out. No heads to crash. Dust has no effect on them. They may cost more per MB, but they are so much more robust and draw so much less power.
As for adding a pair of DACs. It shouldn't be all that hard. I'd recomend looking at using a FPGA to implement DMA transfer and addressing for the DACs. They could be put on their own card along with a headphone amplifyer circuit.
I've always wanted a encryptor/decryptor board that is a huge FPGA or similar with a PCI busmaster interface. It would be better if it could hold a large number of differnt keys that the OS then can control the setting and use of. The key space would be write only memory so they can't be read back by any process. Only written to. The board would operate in burst mode when takling on the PCI bus. It would transfer in a block of memory, encrypt it, then burst mode transfer it out. It's operation would be much like a DMA controller, but with the encryption added in. By using a FPGA one can change the programming of the board as new methods come out. One would need a program protect jumper, but that is easy to do.
In Minneapolis it's unfortunatly overcast... This really bummed me out as the proton levels were some of the highest I've seen in quite awhile. On NASA's Space Weather site they posted an alert. Proton levels were up over 20 per cc when they are usually down in the 1-5 range. The solar wind was also up much higher than normal. As I'm posting this the protons/cc is only 1.2. Not very good for Aurora.
I couldn't get through when using www.slashdot.org, but when I tried slashdot.org it worked. There seams to be alot of name server dropouts on the net lately. You guys need to keep your name servers code up to date. The Bind that comes with RedHat 6.1 on the CD has a known root exploit that is actively being exploited. You need to upgrade to BIND 8.2.2 patchlevel 5.
We need a site within the jusridiction of the 6th Circuit Court of Apeals to post the code.
I control my firewall and can't locate the site's IP address. Something's up. Possibly 2600 needs to get better redundancy in it's name servers.
There is a very important difference here. In your analogy you were the one to place the "gun"/"code" there, not someone else. 2600 is not the one who is publishing the DeCSS code, but only the one pointing to where it can be found.
Compiled binaries can be looked at as a translation of the original work. In this light they could be protected as free speach. This would also protect code from a code generator. The input code is free speach, so the output code would also be.
As an alternate: We need to come up with a general purpose program that knows about many different encryption methods and file formats, and how to get around them. As part of this program it should have the ability to deal with new methods and file formats with either loadable modules or easily extensible source. It should have automated tools for dealing with all known encryption methods, and all known wrappers to those encryption methods.
The program would be a general purpose tool. It would contain generic tools for analysing and decrypting encrypted data in files, streams, etc. Once a specific format is determined a template could be built to automate the decryption process for that format. The point is the tool is generic, and not specific to any one system. The templates allow the transfer of knowlege of how to deal with specific systems.
The path forward is clear to me. We need to come up with a general purpose program that knows about many different encryption methods and file formats, and how to get around them. As part of this program it should have the ability to deal with new methods and file formats with either loadable modules or easily extensible source. It should have automated tools for dealing with all known encryption methods, and all know wrappers to those encryption methods.
There was a precursor to Visicalc even. Visicalc is just the one that made it to the top the quickest and was available on a cheep computer.
All it takes is a break in a specific gene to cause a tumor. Look at all the other genes and you can do a reliable match.
I just remembered that at one of the former places I worked they used an intermediary system to allow some really old systems to be served by their main SAN array. Doing something like this may be a possibility to handle systems unsupported by a NAS system. I'd first check to see if it was possible, but it's worth a try. The intermediary system served an old network file system format to the old boxes while actually getting the files via NFS from the SAN system. It did slow things down some, but it worked.
This may end up being the solution to stay under budget, but you do lose alot going this way. One thing is well integrated automated backups. Some of the NAS and SAN products have very good backup abillities built in. Even on an hour by hour basis. Great for those absent minded users who delete critical reports just before they are due.
If you decide to roll your own a few things to remember:
I hope this helps some. In the long run I feel you will likely be happier with a NAS or SAN system.
This is extreemly important. I have so many tools on my resume because I was always been willing to say, shure I'll try, learn, and do it. It sure helps that I bothered to learn the fundamental computer science theories and am willing to learn more.
My own technical book collection is now a couple of shelves larger than the 4 foot wide floor to ceiling book case I originally allocated for it. Sure some of the books are now old and dusty, but then I've been collecting for 2 decades. The dustier ones will soon be cataloged evicted to boxes till I get a larger place. Sure it's expensive to maintain your own persomal technical library, but I've found it indespensible. I use books as refferences nearly daily. I consider it much more important to know how to research and look up something than it is to know it off the top of the head. More often than not the memory isn't quite all there, but you know that printed page ratains what it was told very faithfully. It's also much easier to hand to somebody else so you can off load the task and get on to some more important task.
Some of us fully under stand Microsoft's browser war tatics, but refuse to lower ourselves to their level. The point of your resume is to show how good you are at doing your job. If was looking at your site to determine if you were a suitable canidate for a web coder or admin I would reject you immediatly. For the plain and simple reason you don't show you know how to communicate to the general web user population. Further more it shows that you place confrontation above communication. Both traits are bad in an employee.
Plain text or simple HTML are very good universal formats. If somebody can't deal with either of them, are you sure you want to work for them?
I once maintained my resume in MS Word format, but decided that it was a waste maintaining a Windows system at home just so I could maintain my resume. I now have my resume posted on my web site. I keep the HTML in it and my web site simple as it is there for information transfer, not glitz. If someone needs it in text format I will transfer it over to text buy having Lynx translate it. The resume is currently in flux as I'm updating it for a new job search. I'm looking for a place that will allow me to only work 30 hours a week preferably in web security or programming.
Minor detail, very minor... :}