I am an instrument rated pilot with 550 hours and *almost* commercial rated (I've got more than twice the experience required, just need to sign up for the test) and I fly a Cessna 210 or a Cirrus several times a week.
I encourage people NOT to use flight simulators to prepare for initial private pilot training. It is next to useless and teaches bad habits. The private pilot training is all about teaching you how to eyeball it. You fly in VFR (Visual Flight Rules). You barely need the instruments. It is all about looking out the window and learning about weather, judgement, physics, etc. PC based flight sims don't teach you any of this. The view (the most important part) is extremely limited even with todays modern graphics, tuning VOR's with the mouse is just not realistic, and you don't have any control forces or wind noise. Rudder is completely ignored in PC flightsims. You end up relying totally on the instruments without learning anything but how to read instruments which is a trivial skill taught new pilots in just a few minutes.
Flying a PC with a cheap plastic joystick in your hands is nothing like flying a noisy, vibrating, machine where you have a huge view all around you and can feel the aerodynamic forces on the yoke in front of you and the G's on your butt and while trying to keep track of exactly where you are in relation to that class bravo airspace with controllers barking instructions at you on the radio. And this is just stuff the first time student pilot has to learn to deal with to say nothing of actual instrument flying in bad weather.
Before I started flying I used to play with PC based flight sims and I liked to think it was something like the real thing but now that I can look back on it I realize I was fooling myself. I learned a *LOT* more about how to fly an airplane from radio control airplanes. Build the plane yourself, learn something about airframes, control surfaces, flutter, weight and balance, etc. Then actually fly it and learn something about preflighting, takeoffs and landings, airspeed/energy control, the need to be smooth on the controls, making timely corrections, spins and stalls, etc. Much more educational to the potential pilot.
Of course as long as you use consider the PC flight sim as nothing more than a sophisticated game then it's pretty good. They have certainly put a lot of work into things like MS FS. I first used it on my Apple ][c way back when SubLogic produced it.
All that being said, I live in San Diego and fly out of Montgomery Field and I am always looking for people to join me in the cockpit so if anyone wants to go up for a short little spin around town and have a go at flying the airplane (Yes, I will give you complete control of the plane if you are willing) just drop me a line at treed@copilotconsulting.com and check out my photo gallery of flying pics at:
I remember when they said the same thing about 80M drives... "Wow, 80M? That's way too big! What are you going to do with it all? How will you back it up? What if it fails?" It was silly then and it is silly now.
This makes it look like they are announcing a new product or something. Actually, they are announcing an R&D program which they HOPE will reduce the cost of solar cells. Hardly headline news.
I'm not sure what happened with your LVM install but I have been able to do many of the things you could not. You use MD to do RAID-1. LVM and MD work together. You can have root on LVM but it is a bit of a pain so I don't usually bother. I am running swap on LVM right now. The Linux LVM implementation is quite nice and mature.
Others have mentioned the necessity of RAID 5 in a setup like this but let me point out that you don't want to fsck 1T of disk. I have had to watch the fsck of 500G of disk back before we had journalled fs and it was terrible. When we started attaching many terabytes of disk to Linux boxes we needed a better solution. So you will want to use a journalled fs. Reiserfs is my favorite. Then you will not want to have to backup/restore when you decide your current partitioning layout was a bad idea or just generally want to shuffle things around so be sure to use LVM also. I use LVM on all of my machines, even desktops, and it has really made life easier. Often you will need more room on/home but notice that/var has a couple gig unused and with LVM you just shrink/var and expand/home all without reboot and you are good to go.
You cannot blindly compare keysizes to determine strength of encryption. 128 bits on a symmetric cipher like that implemented in SSL (probably what the bank was referring to when requiring 128 bit encryption for online banking) is pretty darn good. The 2048 and 4096 etc bit keys you hear about are usually for RSA which needs a lot more bits to get the same security. I suspect you are comparing apples and oranges.
SE Linux is integrated into 2.6 and a patch to 2.4. It GREATLY improves the security of a Linux box. If someone gets root (or some other uid shell) through a buffer overflow they can no longer take over the whole system. Odds are they cannot do anything. How is this possible? By running every process in a security context carefully restricted to least priviledge through a system of mandatory access controls. If you want to see how effective this is for yourself please telnet to:
The Freenet Project is the solution to this problem. It is a decentralized anonymous publishing system. You can publish web pages, music, movies, and all sorts of things your local government may not like. Freenet is actively being used in China to bring information to the people. It has had performance problems this past year but seems to be on an upswing. I was able to retrieve several 90M files from it with a decent download rate although I had to retry a couple of times.
Also note that the Freenet project needs money to continue to pay its full time programmer. It is GPL software but it is very challenging code to work on and really needs dedicated people. So go to The Freenet Project Homepage and download a copy of the latest version and hopefully make a donation!
It's either freedom in the form of Freenet or lawsuits from the RIAA/MPAA/Your local government. Take your pic.
Not really a comment on "storage" but just a comment on something that has constantly bugged me when someone says "let's put it in a database!"
A filesystem is a special case of a database. So it is perfectly acceptable to store your data into a filesystem. Some people seem to think everything has to be put into a relational database or that is it somehow cool to do so. I have seen people store loads of graphics as BLOBS in databases. Someone once suggested storing a ton of MP3's in a database. Most recently someone said (and this isn't the first time) that we should store all of the emails in a database. It's just another unnecessary layer of complication, especially when you are going to be referencing the email/graphic/mp3 by name all the time anyway (fs's like reiserfs index on name so it's blazingly fast) and not by a bunch of other pieces of meta-data. And if you are going to need to do lookups by various bits of meta-data then store the meta-data in a db and also store a record pointing to the actual file on disk. I have done that lots of times and it works great.
...before someone really is killed due to M$'s negligence. Sure, one could argue that they should have applied patches and that it isn't M$'s fault but tell that to the jury. When surviving relatives see the potential for a profitable liability suit they are going to go after the biggest pockets and that is M$.
The Unix History Graphing Project could use some updating if Linux really does have (legally obtained) code from Sequent, AIX, Irix, etc. It would be really great of Linux were responsible for the de-fragmenting of Unix! It looks like we are heading that way...
Freenet does not have this centralization problem. And a very good new version just came out. I have been using both but because torrents are such a pain to find I have found freenet to be more useful. The freenet guys said bittorrent would run into this problem. I am surprised it has happened so soon.
...to really understand all of the problems with this idea. For one thing paying passenger aircraft have to be certified and one of the requirements is that the pilot be able to overpower any installed autopilot system. They would have to change the rules. But this sort of thing would introduce so much expense and complication and additional safety risks and new modes of failure that it will not be implemented. It's hard enough just to get something like TCAS or even GPS installed. It saddens me to think that my two passions in life (computers and flying) are also two of the most misunderstood fields around.
Broadcast *will not work*. You won't find squat and the network will get overloaded. Just like Gnutella. I'm also not sure why broadcast is needed for anonymity. Freenet does not broadcast and is anonymous. It proxies through a bunch of machines (number defined by the HTL of the request) according to the routing scheme.
Broadcast search does not scale. This has been proven both in theory and in practice. This is why freenet has a routing algorithm based on the hash key.
They are right, of course. Which is precisely why you should all use Freenet which IS anonymous, totally distributed, and has a much better distribution model in that popular content is automatically cached all over the net and all downloads are automatically "swarmed" so you don't get stuck downloading Madonna's music off of someones 56k modem. Using 60 download threads I can consistantly get 90k/s on recently inserted files, no matter what the connection speed of the person who made them available.
Very interesting. I'll have to check it out. It does indeed sound a lot like User Mode Linux. I may disagree with the idea that UML emulates anything. As far as I know it does not. It runs a kernel in a normal user process space so it should be just as far as any userland application. I'm reading the FAQ on this and a few things concern me. It seems to be an enhanced chroot jail. But there have been ways published to break out of chroots. Has something been done about that? I have a great deal of confidence in UML because it traps the whole affair in its own user mode protected memory space like a process and I have a great deal of confidence in SE Linux because of the tremendous amount of study and theoretical work that has gone into it plus the tight integration of the security module into the kernel. I will have to do some more reading to see how this thing compares.
Slashdot Mirror
Slashdot as a source of fair and balanced coverage?!?!
I bet you watch Fox News too!
...he had used Freenet.
Filename: diebold_memo.html
Key: CHK@gvarXrMPVzH8eJVALa-VJEWliqgQAwI,~fu6drzFntkje
I have successfully retrieved it so you probably can also. There also seems to be a pdf version in there. They can both be found in frost.
I am an instrument rated pilot with 550 hours and *almost* commercial rated (I've got more than twice the experience required, just need to sign up for the test) and I fly a Cessna 210 or a Cirrus several times a week.
I encourage people NOT to use flight simulators to prepare for initial private pilot training. It is next to useless and teaches bad habits. The private pilot training is all about teaching you how to eyeball it. You fly in VFR (Visual Flight Rules). You barely need the instruments. It is all about looking out the window and learning about weather, judgement, physics, etc. PC based flight sims don't teach you any of this. The view (the most important part) is extremely limited even with todays modern graphics, tuning VOR's with the mouse is just not realistic, and you don't have any control forces or wind noise. Rudder is completely ignored in PC flightsims. You end up relying totally on the instruments without learning anything but how to read instruments which is a trivial skill taught new pilots in just a few minutes.
Flying a PC with a cheap plastic joystick in your hands is nothing like flying a noisy, vibrating, machine where you have a huge view all around you and can feel the aerodynamic forces on the yoke in front of you and the G's on your butt and while trying to keep track of exactly where you are in relation to that class bravo airspace with controllers barking instructions at you on the radio. And this is just stuff the first time student pilot has to learn to deal with to say nothing of actual instrument flying in bad weather.
Before I started flying I used to play with PC based flight sims and I liked to think it was something like the real thing but now that I can look back on it I realize I was fooling myself. I learned a *LOT* more about how to fly an airplane from radio control airplanes. Build the plane yourself, learn something about airframes, control surfaces, flutter, weight and balance, etc. Then actually fly it and learn something about preflighting, takeoffs and landings, airspeed/energy control, the need to be smooth on the controls, making timely corrections, spins and stalls, etc. Much more educational to the potential pilot.
Of course as long as you use consider the PC flight sim as nothing more than a sophisticated game then it's pretty good. They have certainly put a lot of work into things like MS FS. I first used it on my Apple ][c way back when SubLogic produced it.
All that being said, I live in San Diego and fly out of Montgomery Field and I am always looking for people to join me in the cockpit so if anyone wants to go up for a short little spin around town and have a go at flying the airplane (Yes, I will give you complete control of the plane if you are willing) just drop me a line at treed@copilotconsulting.com and check out my photo gallery of flying pics at:
my flying photo gallery.
I remember when they said the same thing about 80M drives... "Wow, 80M? That's way too big! What are you going to do with it all? How will you back it up? What if it fails?" It was silly then and it is silly now.
If you smoke after sex you're doing it too fast.
This makes it look like they are announcing a new product or something. Actually, they are announcing an R&D program which they HOPE will reduce the cost of solar cells. Hardly headline news.
"Can you hear me YET?"
Hmm...I've just read the LVM howto, followed the linux-lvm mailing list, and used it for a year or so. No great secrets to reveal.
I'm not sure what happened with your LVM install but I have been able to do many of the things you could not. You use MD to do RAID-1. LVM and MD work together. You can have root on LVM but it is a bit of a pain so I don't usually bother. I am running swap on LVM right now. The Linux LVM implementation is quite nice and mature.
Others have mentioned the necessity of RAID 5 in a setup like this but let me point out that you don't want to fsck 1T of disk. I have had to watch the fsck of 500G of disk back before we had journalled fs and it was terrible. When we started attaching many terabytes of disk to Linux boxes we needed a better solution. So you will want to use a journalled fs. Reiserfs is my favorite. Then you will not want to have to backup/restore when you decide your current partitioning layout was a bad idea or just generally want to shuffle things around so be sure to use LVM also. I use LVM on all of my machines, even desktops, and it has really made life easier. Often you will need more room on /home but notice that /var has a couple gig unused and with LVM you just shrink /var and expand /home all without reboot and you are good to go.
You cannot blindly compare keysizes to determine strength of encryption. 128 bits on a symmetric cipher like that implemented in SSL (probably what the bank was referring to when requiring 128 bit encryption for online banking) is pretty darn good. The 2048 and 4096 etc bit keys you hear about are usually for RSA which needs a lot more bits to get the same security. I suspect you are comparing apples and oranges.
SE Linux is integrated into 2.6 and a patch to 2.4. It GREATLY improves the security of a Linux box. If someone gets root (or some other uid shell) through a buffer overflow they can no longer take over the whole system. Odds are they cannot do anything. How is this possible? By running every process in a security context carefully restricted to least priviledge through a system of mandatory access controls. If you want to see how effective this is for yourself please telnet to:
selinux.copilotconsulting.com
user: root
pass: root
Also note that the Freenet project needs money to continue to pay its full time programmer. It is GPL software but it is very challenging code to work on and really needs dedicated people. So go to The Freenet Project Homepage and download a copy of the latest version and hopefully make a donation!
It's either freedom in the form of Freenet or lawsuits from the RIAA/MPAA/Your local government. Take your pic.
Not really a comment on "storage" but just a comment on something that has constantly bugged me when someone says "let's put it in a database!"
A filesystem is a special case of a database. So it is perfectly acceptable to store your data into a filesystem. Some people seem to think everything has to be put into a relational database or that is it somehow cool to do so. I have seen people store loads of graphics as BLOBS in databases. Someone once suggested storing a ton of MP3's in a database. Most recently someone said (and this isn't the first time) that we should store all of the emails in a database. It's just another unnecessary layer of complication, especially when you are going to be referencing the email/graphic/mp3 by name all the time anyway (fs's like reiserfs index on name so it's blazingly fast) and not by a bunch of other pieces of meta-data. And if you are going to need to do lookups by various bits of meta-data then store the meta-data in a db and also store a record pointing to the actual file on disk. I have done that lots of times and it works great.
...before someone really is killed due to M$'s negligence. Sure, one could argue that they should have applied patches and that it isn't M$'s fault but tell that to the jury. When surviving relatives see the potential for a profitable liability suit they are going to go after the biggest pockets and that is M$.
The Unix History Graphing Project could use some updating if Linux really does have (legally obtained) code from Sequent, AIX, Irix, etc. It would be really great of Linux were responsible for the de-fragmenting of Unix! It looks like we are heading that way...
Freenet Project
:)
And then give yourself an A.
Freenet does not have this centralization problem. And a very good new version just came out. I have been using both but because torrents are such a pain to find I have found freenet to be more useful. The freenet guys said bittorrent would run into this problem. I am surprised it has happened so soon.
...to really understand all of the problems with this idea. For one thing paying passenger aircraft have to be certified and one of the requirements is that the pilot be able to overpower any installed autopilot system. They would have to change the rules. But this sort of thing would introduce so much expense and complication and additional safety risks and new modes of failure that it will not be implemented. It's hard enough just to get something like TCAS or even GPS installed. It saddens me to think that my two passions in life (computers and flying) are also two of the most misunderstood fields around.
Tracy R Reed
PP-ASEL-IA and soon to be CP and CFI
Broadcast *will not work*. You won't find squat and the network will get overloaded. Just like Gnutella. I'm also not sure why broadcast is needed for anonymity. Freenet does not broadcast and is anonymous. It proxies through a bunch of machines (number defined by the HTL of the request) according to the routing scheme.
Broadcast search does not scale. This has been proven both in theory and in practice. This is why freenet has a routing algorithm based on the hash key.
Of course it involves a fat pipe and DivX.
They are right, of course. Which is precisely why you should all use Freenet which IS anonymous, totally distributed, and has a much better distribution model in that popular content is automatically cached all over the net and all downloads are automatically "swarmed" so you don't get stuck downloading Madonna's music off of someones 56k modem. Using 60 download threads I can consistantly get 90k/s on recently inserted files, no matter what the connection speed of the person who made them available.
Very interesting. I'll have to check it out. It does indeed sound a lot like User Mode Linux. I may disagree with the idea that UML emulates anything. As far as I know it does not. It runs a kernel in a normal user process space so it should be just as far as any userland application. I'm reading the FAQ on this and a few things concern me. It seems to be an enhanced chroot jail. But there have been ways published to break out of chroots. Has something been done about that? I have a great deal of confidence in UML because it traps the whole affair in its own user mode protected memory space like a process and I have a great deal of confidence in SE Linux because of the tremendous amount of study and theoretical work that has gone into it plus the tight integration of the security module into the kernel. I will have to do some more reading to see how this thing compares.