Who cares? Its not like Corel was a REAL linux distro.
Re:So let me get this straight, Apple...
on
Quicktime In Linux
·
· Score: 1
Okay, so you have a point about the MegaHertz Myth, but how can PowerPC compete with 2 Ghz systems shipping today and dual/quad+ 2Ghz systems on their way before christmas? I want some real world benchmarks to show me that these PowerPC chips are more than hype, that they really are supercomputers! Then I'll think about spending the extra $1000 over the cost of my cheap/modular PC hardware to get into a Mac. Now if all the Macs came with titanium cases it might be different. (MacOS is quite nice!)
I download free and legal music off the net by bands that don't think their CDs should cost $17 a piece. In fact most of them reallize that I can burn a CD for about $.30. Now you, some stupid record exec, trying to sell me crappy over-hyped music, have the balls to tell me I'm doing something illegal because your profit margins are down? Fuck you! Tell me you think I'm a thief to my face, motherfucker. I'd beat you within an inch of your life. Your high tower is falling and its my fault, so don't drag me down with you and your poor management. Real economics relies on supply and demand, and guess what? With this internet you have infinite supply, and your demand is dropping because you refuse to admit it. Sorry your monopoly on audio content had to be smashed by a free alternative, the internet and people who care.
I'm just glad the moderation system on/. doesn't decide which patches get added to the linux kernel. Just imagine how buggy linux would become if we all had a say in what gets added and what doesn't. But perhaps we'd get linux to beat windows in 3d frame rates and downtime.
True and I agree with you for the most part, but I disagree with this statement:
"If I were to write an OS kernel and use all the GNU tools to provide a userland, RMS would like take away my freedom to call the resulting product whatever I like."
I just think RMS is a bit insisting and perhaps acts childish from time to time, but deep down I bet he knows the right thing to do and isn't going to make it mandetory to include "GNU/" in front of any GPLed title or anything like that. He just wants more attention since Linux has gotten a nice spotlight from the media, and I think he deserves that attention, because it was GNU and the free software foundation and the GPL license that made linux what it is today: not just the kernel, but an OS.
Just think about linux for a sec. Its amazing. Its an alternative to a commercial operating system for just about any type of computer out there. Its runs mostly on GNU software and has several alternative choices for almost any application. Its mostly stable. And almost all the work put into it came from volunteers. Those volunteers, whether they like it or not, contributed to a GNU project. And it has bloomed wonderfully, don't you think?
Why not? They are using GPLed code. That by definition is GNU. Remember GNU is NOT UNIX! Its the software, the code, the utilities and tools that run on unix-like systems, occationally including windows. Its the ideology behind the code. Its SO MUCH MORE than a simple name. By naming your product GNU/whatever it states that you use and hopefully contribute to free software and GPLed code, which in turn means you tend to play fair and work for the product and the people, not for your investors, customers, and money. It means you are willing to give something away for free, to help people, expecting nothing in return, but more than likely they'll modify your code and give something back to you: a better product. The GPL is all about the product. Its a different way of thinking and doing business. Its the choice of the GNU generation.:)
Actually, I had another thought. Remember when you were a kid and someone did something to you that you didn't like. Often you would go tell your mommy and she'd get the other kid in trouble, specially if it was your brother or sister. Well, adults don't act like adults when it comes to requesting or giving recognition or playing fair, morally and ethically. If something doesn't go their way they write articles about it, post their opinions on message boards and in corporations they go to the parent, or manager, and attempt to get that other kid in trouble. Why can't we learn to simply confront the kid, stand up for our own rights and respect theirs and talk instead of preach, play instead of work, and just get along.
I have to disagree. I can understand how some small time developer who starts an odd application might want to maintain control of the application when it becomes popular. But something as crutial as libc should definitely be maintained by the top developers of the project. An SC that would look after the interrests of the user base as well as legal issues as they do for glibc. If the main developer always wanted to maintain control of a project, knowing it would be a fundamental piece of the GNU/GPL OS, whatever kernel it runs on, they should understand the consequences and decide not to release their code under the GPL. Personally, if I could program like that, I would not want to maintain control of a project that big.
The alternative is to branch off the GPLed code into a new tree and remove Mr. Drepper from its management, start directing new developement towards the new branch, etc. But this has huge political and moral problems. People should just understand, specially at Drepper's level, how to get along, play fair, and speak their mind face to face instead of behind eachothers backs. Save the back stabbing and childish bickerring for corporate america.
Well, strictly speaking, the kernel is the OS. It manages all the processes and hardware on your system. You merely need a few daemons to give the added functionality of networking, logging, and services that provide for the general health of your system. But in this case, calling the kernel the OS, you still have a lot of contributed (GPLed) code that went into the linux kernel. Would it be better to say all the names of the developers in alphabetic order as the name of the OS or simply attach GNU, which defines the legal, ethical and moral idealism of the license, code and OS. Afterall its the choice of the GNU generation. But at least its a choice.
I think the point is NO ONE OWNS LINUX. So you can call it what you want, but calling it Linux gives credit to Linus. Similarly since a lot of GNU based or GPL based work has went into the linux kernel by contributers other than Linus as well as the daemons, libraries and apps that make up the running / usable system, an easy way to attribute that work to more than just one person is to say GNU/linux. But go ahead and be rude to all those developers, just remember they're not writing this OS for you!
You should not call windows "Windows98" because windows in no way qualifies as a "part of the Operating System". Instead you should call windows "dos".
The difference is Stallman is asking, the BSD license would require the "GNU inside" logo, like M$ and Intel do. If you don't like Stallman's request, make your own Linux distro and never mention GNU. He can't and won't do anything about it, but complain. I don't care. I like hearing his complaints. At least he cares enough to complain.
* Smit and the ODM database are not pluses for AIX. They create frustration and configuration problems when smit doesn't work right. A simply configuration script that modified flat files and worked 100% of the time (preferrably universal to all linux systems) would go a long way to make administration easier.
* AIX often takes >30 minutes to boot on older hardware.
* I have to hand it to you, though, once you get it configured and running its quite stable. I've never had a problem with any RS/6000s besides the regular reconfigs when changing networks.
In my opinion Linux is ready for enterprise-class workloads, but you have to understand how to create scalable distributed and parallel networks either using load balancing or clustering depending on your needs.
Linux is excellent for putting a bunch of cheap yet powerful PCs as the horsepower behind any mission critical network service or application. The problems most people face are poor planning. Every system will crash given a hardware outage or rolling blackout or extremely poorly written software (at least possible on linux (stupid kernel modules), perhaps not on AIX). The important thing is to design your network to be scalable and fault tolerant, so it does not rely on any one PC as a server, but instead a load balanced virtual address that will always be answerred by some system. Chances are you will notice a system or two fail and replace them before you lose your whole web presence or access to your DB, etc.
They should push Deus Ex out the door whether its ready or not. That's the game I've been paying them for this whole time. I hope they can keep afloat until they release it.
Better yet why don't you put a book on your table and learn something so you can afford these games. I can't see how any geek who can afford a computer can't afford a small piece of software to go with it. I agree with Taco, I bought a few games from Loki that I haven't had time to really play yet. But one day I know I will, and the added bonus of buying loki software is you don't need to search for your CD to play them. They're installed on your harddrive and will sit there until you're good and ready.
What's the problem? Would you rather save your money for your next upgrade to windows?
Constants such as the gravitational constant are based on our perceptions and lots of mathematical evidence. Perhaps in distance corners of our universe the fabric of space/time is warped. We know that gravity actually bends the fabric of space, causing light to bend and applying forces to massive objects, etc. If the fabric is thinner or somehow different than our solar system perhaps the constants that define our equations should actually be functions based on higher mathematical models similar to superstrings. But I'm not even an amateur physicist... just a thought.
Well, physics, like math, attempts to be exact and accurate. But whenever we discover otherwise more complex functions need to be drafted. I doubt the average human brain is fully capable of understanding the intricate details of how the universe really works on a functional / mathematical level. Most people, americans, don't even understand the basics of newtonian physics, let alone superstrings. Perhaps something like a genetic program could solve these problems a bit easier and faster?
Yes you heard me right. You doubt moore's law? Now my math is skewed a bit, that should have read 60 FPS at the resolution this quadro system was renderring, but I suspect we'll far surpass moore's law these next few years. Exponential growth baby!
How long do you think it'll take before home systems will be able to render movie quality video in real time 60FPS? We're already pushing over 300FPS on quake 3. I don't know how many polygons get drawn there, and ray tracing is a lot more CPU intensive. But imagine this. A P2P network where a few nodes render the image for you. These nodes each render a 64x64 pixel area of the frame and your system puts together the pixels 60 times a second to create your smooth video. You can prerender a few frames in advance as long as you have the network bandwidth, processor power and software. I think we could possibly make something like that in 4 to 5 years. Hell, I've only been working in the field 4 to 5 years and we've made considerable advances I would have never imagined. Heh, Gigahertz processors. Who needs one of those? There already are 1.8 Ghz P4s. And I bet we'll see some dual 2 Ghz Athlon systems before the end of the years. That's 4 GB/s memory to processor bandwidth with > 1 GB of system ram and 1 Gbps network capacity over copper (and the ability to use it) in a HOME COMPUTER this christmas.
Well, think of it this way, you want to render 60 frames at lets say 1280x1024 with 32 bit color depth. That's 60x1280x1024x32 bits of data a second, not including the math behind the ray tracing, which I know nothing about... But anyhow that's 2.4 GB of data per second. Almost managable with modern technology.
We're not too far off from renderring movie quality video real time, in my opinion. The question is what kind of content do we want to render.
I largely agree with both ideas. I have no information about how much loki pays to companies to port their games or how much it costs them to port them. But one simple concept they should take into account is they need to price and sell their product above the cost of manufacturing and distribution in order to make a profit. If they are making a profit then they will pull through these tough times and eventually clear out their debts. If they are not making a profit maybe they should develope new games or find alternatives to the costs OR raise their prices... or possibly lower prices and increase marketting depending on their customers.
All I know is most loki games are not sold in stores, so go to www.lokigames.com and buy some if you are a good linux geek.
Well, we've already surpassed moore's law and are doubling performance about every 8 months currently. And the rate at which we double performance, or release a new genereation of tech, is increasing. So I would predict at our current rate we'd be seeing home systems capable of renderring movie quality images at 60 FPS in 4-5 years, tops, not taking into account that the rate is increasing. And there could still be a leap in performance and technology if something like a quantum computer or public P2P processing software becomes mainstream by then.
Well, people are turning out not to be much more creative than a computer, and computers are a lot faster than we are. So I could see how a computer could create realistic models, textures, lighting, shading, animation and sound using clustering or P2P technology and genetic programming in future games. At the rate graphics and processor performance increases we'll be seeing less and less hardcoded graphics and more dynamic / generated data to max out the fill rate.
I bought almost every game loki made. They all install easily through a simply gui option. Work on a hacked version of slackware, upgraded kernel and X of course. And never once did they ask for the CD to be in the freakin drive to run the game. That, for me, is worth more than the cost of a full priced game. But loki also sells their games often below retail prices and ships within a week, usually. I will continue to support them until the end. But if you lazy stupid RedHat users would get off your ass and grab a few games NOW the end won't come and linux will not only continue to be a viable gaming platform, but also a stable one.
*flame bait* Heh heh, we all know RedHat users couldn't find / -name vmlinu\* -exec rm {}\; if their life depended on it. And most would probably execute that command to find out.;)
Because unlike some other linuxs RedHat has very few similarities to a real unix OS. It hides configuration files in obscure directories not at all like the Sun and BSD standards. Redhat has changed their version of automount at least 3 times in the last 4 releases, which wouldn't be a problem if they all used the same file format. (that kinda makes it tough for enterprise wide installs) You can't simply upgrade, since gnome breaks between releases and requires you to delete your.gnome directory, losing your look and feel, etc. And the list goes on. So my opinion as a Sr. Sys Admin is to skip RedHat altogether and go with a more respectable distribution such as Debian or Suse, or Slackware if you are a masochist like myself. Suse even comes with a book that will teach you all about linux and configuration (including the version installed on your system) without the need to buy this ol' one.
If they hadn't tried to make pdf secure by using a ROT13 type of encryption scheme maybe it wouldn't have been cracked so easily and they wouldn't be in the mess they are now. Security through obscurity does not work because someone, somewhere, eventually will crack it. I'm not a security expert, but a little common sense will tell you obscurity is a lot easier to crack than brute force attacks on 1024-bit encryption, at least until I can buy a quantum computer. Then you will simply need to increase the security by using larger, more complex encryption. As long as you stay ahead of the technology currently available, when you implement a security feature, it will be mostly secure, but keep in mind P2P software and clusters. Then you will find that your weakest point isn't the encryption but your users passwords.
Slashdot Mirror
If ignorance is bliss you'd think American Christians would be the happiest people on earth.
Who cares? Its not like Corel was a REAL linux distro.
Okay, so you have a point about the MegaHertz Myth, but how can PowerPC compete with 2 Ghz systems shipping today and dual/quad+ 2Ghz systems on their way before christmas? I want some real world benchmarks to show me that these PowerPC chips are more than hype, that they really are supercomputers! Then I'll think about spending the extra $1000 over the cost of my cheap/modular PC hardware to get into a Mac. Now if all the Macs came with titanium cases it might be different. (MacOS is quite nice!)
I download free and legal music off the net by bands that don't think their CDs should cost $17 a piece. In fact most of them reallize that I can burn a CD for about $.30. Now you, some stupid record exec, trying to sell me crappy over-hyped music, have the balls to tell me I'm doing something illegal because your profit margins are down? Fuck you! Tell me you think I'm a thief to my face, motherfucker. I'd beat you within an inch of your life. Your high tower is falling and its my fault, so don't drag me down with you and your poor management. Real economics relies on supply and demand, and guess what? With this internet you have infinite supply, and your demand is dropping because you refuse to admit it. Sorry your monopoly on audio content had to be smashed by a free alternative, the internet and people who care.
I'm just glad the moderation system on /. doesn't decide which patches get added to the linux kernel. Just imagine how buggy linux would become if we all had a say in what gets added and what doesn't. But perhaps we'd get linux to beat windows in 3d frame rates and downtime.
True and I agree with you for the most part, but I disagree with this statement:
"If I were to write an OS kernel and use all the GNU tools to provide a userland, RMS would like take away my freedom to call the resulting product whatever I like."
I just think RMS is a bit insisting and perhaps acts childish from time to time, but deep down I bet he knows the right thing to do and isn't going to make it mandetory to include "GNU/" in front of any GPLed title or anything like that. He just wants more attention since Linux has gotten a nice spotlight from the media, and I think he deserves that attention, because it was GNU and the free software foundation and the GPL license that made linux what it is today: not just the kernel, but an OS.
Just think about linux for a sec. Its amazing. Its an alternative to a commercial operating system for just about any type of computer out there. Its runs mostly on GNU software and has several alternative choices for almost any application. Its mostly stable. And almost all the work put into it came from volunteers. Those volunteers, whether they like it or not, contributed to a GNU project. And it has bloomed wonderfully, don't you think?
Why not? They are using GPLed code. That by definition is GNU. Remember GNU is NOT UNIX! Its the software, the code, the utilities and tools that run on unix-like systems, occationally including windows. Its the ideology behind the code. Its SO MUCH MORE than a simple name. By naming your product GNU/whatever it states that you use and hopefully contribute to free software and GPLed code, which in turn means you tend to play fair and work for the product and the people, not for your investors, customers, and money. It means you are willing to give something away for free, to help people, expecting nothing in return, but more than likely they'll modify your code and give something back to you: a better product. The GPL is all about the product. Its a different way of thinking and doing business. Its the choice of the GNU generation.
*end plug*
Actually, I had another thought. Remember when you were a kid and someone did something to you that you didn't like. Often you would go tell your mommy and she'd get the other kid in trouble, specially if it was your brother or sister. Well, adults don't act like adults when it comes to requesting or giving recognition or playing fair, morally and ethically. If something doesn't go their way they write articles about it, post their opinions on message boards and in corporations they go to the parent, or manager, and attempt to get that other kid in trouble. Why can't we learn to simply confront the kid, stand up for our own rights and respect theirs and talk instead of preach, play instead of work, and just get along.
just a thought
I have to disagree. I can understand how some small time developer who starts an odd application might want to maintain control of the application when it becomes popular. But something as crutial as libc should definitely be maintained by the top developers of the project. An SC that would look after the interrests of the user base as well as legal issues as they do for glibc. If the main developer always wanted to maintain control of a project, knowing it would be a fundamental piece of the GNU/GPL OS, whatever kernel it runs on, they should understand the consequences and decide not to release their code under the GPL. Personally, if I could program like that, I would not want to maintain control of a project that big.
The alternative is to branch off the GPLed code into a new tree and remove Mr. Drepper from its management, start directing new developement towards the new branch, etc. But this has huge political and moral problems. People should just understand, specially at Drepper's level, how to get along, play fair, and speak their mind face to face instead of behind eachothers backs. Save the back stabbing and childish bickerring for corporate america.
Well, strictly speaking, the kernel is the OS. It manages all the processes and hardware on your system. You merely need a few daemons to give the added functionality of networking, logging, and services that provide for the general health of your system. But in this case, calling the kernel the OS, you still have a lot of contributed (GPLed) code that went into the linux kernel. Would it be better to say all the names of the developers in alphabetic order as the name of the OS or simply attach GNU, which defines the legal, ethical and moral idealism of the license, code and OS. Afterall its the choice of the GNU generation. But at least its a choice.
I think the point is NO ONE OWNS LINUX. So you can call it what you want, but calling it Linux gives credit to Linus. Similarly since a lot of GNU based or GPL based work has went into the linux kernel by contributers other than Linus as well as the daemons, libraries and apps that make up the running / usable system, an easy way to attribute that work to more than just one person is to say GNU/linux. But go ahead and be rude to all those developers, just remember they're not writing this OS for you!
You should not call windows "Windows98" because windows in no way qualifies as a "part of the Operating System". Instead you should call windows "dos".
The difference is Stallman is asking, the BSD license would require the "GNU inside" logo, like M$ and Intel do. If you don't like Stallman's request, make your own Linux distro and never mention GNU. He can't and won't do anything about it, but complain. I don't care. I like hearing his complaints. At least he cares enough to complain.
As a sys admin I can offer some more info:
* Smit and the ODM database are not pluses for AIX. They create frustration and configuration problems when smit doesn't work right. A simply configuration script that modified flat files and worked 100% of the time (preferrably universal to all linux systems) would go a long way to make administration easier.
* AIX often takes >30 minutes to boot on older hardware.
* I have to hand it to you, though, once you get it configured and running its quite stable. I've never had a problem with any RS/6000s besides the regular reconfigs when changing networks.
In my opinion Linux is ready for enterprise-class workloads, but you have to understand how to create scalable distributed and parallel networks either using load balancing or clustering depending on your needs.
Linux is excellent for putting a bunch of cheap yet powerful PCs as the horsepower behind any mission critical network service or application. The problems most people face are poor planning. Every system will crash given a hardware outage or rolling blackout or extremely poorly written software (at least possible on linux (stupid kernel modules), perhaps not on AIX). The important thing is to design your network to be scalable and fault tolerant, so it does not rely on any one PC as a server, but instead a load balanced virtual address that will always be answerred by some system. Chances are you will notice a system or two fail and replace them before you lose your whole web presence or access to your DB, etc.
Hell yeah!
They should push Deus Ex out the door whether its ready or not. That's the game I've been paying them for this whole time. I hope they can keep afloat until they release it.
Better yet why don't you put a book on your table and learn something so you can afford these games. I can't see how any geek who can afford a computer can't afford a small piece of software to go with it. I agree with Taco, I bought a few games from Loki that I haven't had time to really play yet. But one day I know I will, and the added bonus of buying loki software is you don't need to search for your CD to play them. They're installed on your harddrive and will sit there until you're good and ready.
What's the problem? Would you rather save your money for your next upgrade to windows?
Constants such as the gravitational constant are based on our perceptions and lots of mathematical evidence. Perhaps in distance corners of our universe the fabric of space/time is warped. We know that gravity actually bends the fabric of space, causing light to bend and applying forces to massive objects, etc. If the fabric is thinner or somehow different than our solar system perhaps the constants that define our equations should actually be functions based on higher mathematical models similar to superstrings. But I'm not even an amateur physicist... just a thought.
Well, physics, like math, attempts to be exact and accurate. But whenever we discover otherwise more complex functions need to be drafted. I doubt the average human brain is fully capable of understanding the intricate details of how the universe really works on a functional / mathematical level. Most people, americans, don't even understand the basics of newtonian physics, let alone superstrings. Perhaps something like a genetic program could solve these problems a bit easier and faster?
Yes you heard me right. You doubt moore's law? Now my math is skewed a bit, that should have read 60 FPS at the resolution this quadro system was renderring, but I suspect we'll far surpass moore's law these next few years. Exponential growth baby!
How long do you think it'll take before home systems will be able to render movie quality video in real time 60FPS? We're already pushing over 300FPS on quake 3. I don't know how many polygons get drawn there, and ray tracing is a lot more CPU intensive. But imagine this. A P2P network where a few nodes render the image for you. These nodes each render a 64x64 pixel area of the frame and your system puts together the pixels 60 times a second to create your smooth video. You can prerender a few frames in advance as long as you have the network bandwidth, processor power and software. I think we could possibly make something like that in 4 to 5 years. Hell, I've only been working in the field 4 to 5 years and we've made considerable advances I would have never imagined. Heh, Gigahertz processors. Who needs one of those? There already are 1.8 Ghz P4s. And I bet we'll see some dual 2 Ghz Athlon systems before the end of the years. That's 4 GB/s memory to processor bandwidth with > 1 GB of system ram and 1 Gbps network capacity over copper (and the ability to use it) in a HOME COMPUTER this christmas.
Well, think of it this way, you want to render 60 frames at lets say 1280x1024 with 32 bit color depth. That's 60x1280x1024x32 bits of data a second, not including the math behind the ray tracing, which I know nothing about... But anyhow that's 2.4 GB of data per second. Almost managable with modern technology.
We're not too far off from renderring movie quality video real time, in my opinion. The question is what kind of content do we want to render.
I largely agree with both ideas. I have no information about how much loki pays to companies to port their games or how much it costs them to port them. But one simple concept they should take into account is they need to price and sell their product above the cost of manufacturing and distribution in order to make a profit. If they are making a profit then they will pull through these tough times and eventually clear out their debts. If they are not making a profit maybe they should develope new games or find alternatives to the costs OR raise their prices... or possibly lower prices and increase marketting depending on their customers.
All I know is most loki games are not sold in stores, so go to www.lokigames.com and buy some if you are a good linux geek.
Well, we've already surpassed moore's law and are doubling performance about every 8 months currently. And the rate at which we double performance, or release a new genereation of tech, is increasing. So I would predict at our current rate we'd be seeing home systems capable of renderring movie quality images at 60 FPS in 4-5 years, tops, not taking into account that the rate is increasing. And there could still be a leap in performance and technology if something like a quantum computer or public P2P processing software becomes mainstream by then.
Well, people are turning out not to be much more creative than a computer, and computers are a lot faster than we are. So I could see how a computer could create realistic models, textures, lighting, shading, animation and sound using clustering or P2P technology and genetic programming in future games. At the rate graphics and processor performance increases we'll be seeing less and less hardcoded graphics and more dynamic / generated data to max out the fill rate.
I bought almost every game loki made. They all install easily through a simply gui option. Work on a hacked version of slackware, upgraded kernel and X of course. And never once did they ask for the CD to be in the freakin drive to run the game. That, for me, is worth more than the cost of a full priced game. But loki also sells their games often below retail prices and ships within a week, usually. I will continue to support them until the end. But if you lazy stupid RedHat users would get off your ass and grab a few games NOW the end won't come and linux will not only continue to be a viable gaming platform, but also a stable one.
*flame bait* Heh heh, we all know RedHat users couldn't find / -name vmlinu\* -exec rm {}\; if their life depended on it. And most would probably execute that command to find out.
Because unlike some other linuxs RedHat has very few similarities to a real unix OS. It hides configuration files in obscure directories not at all like the Sun and BSD standards. Redhat has changed their version of automount at least 3 times in the last 4 releases, which wouldn't be a problem if they all used the same file format. (that kinda makes it tough for enterprise wide installs) You can't simply upgrade, since gnome breaks between releases and requires you to delete your
If they hadn't tried to make pdf secure by using a ROT13 type of encryption scheme maybe it wouldn't have been cracked so easily and they wouldn't be in the mess they are now. Security through obscurity does not work because someone, somewhere, eventually will crack it. I'm not a security expert, but a little common sense will tell you obscurity is a lot easier to crack than brute force attacks on 1024-bit encryption, at least until I can buy a quantum computer. Then you will simply need to increase the security by using larger, more complex encryption. As long as you stay ahead of the technology currently available, when you implement a security feature, it will be mostly secure, but keep in mind P2P software and clusters. Then you will find that your weakest point isn't the encryption but your users passwords.