Heres the definition I got from wikipedia: "The term cyborg, a portmanteau of cybernetic organism, is used to designate an organism which is a mixture of organic and mechanical (synthetic) parts. Generally, the aim is to add to or enhance the abilities of an organism by using technology."
Now the first sentence means its an organism thats both flesh and machine. This means the machine should be an innate part of the organism, to the level its not the organism anymore without the machine. I suppose pacemakers and artificial hearts will make the person a cyborg. However the second part means use of technology makes the organism a cyborg. That means wearing contact lenses makes me a cyborg.
Now 'bionic eyes' which really is a small screen in your helmet, and the kind of exoskeletons that we've seen from military contractors are really add-ons that the user will remove at night before going to bed. Even prosthetics are generally removed (fake teeth) at bedtime, which disqualifies the organism from being a cyborg in the first sense of the word. If you accept the second sense, then I'm the 6 dollar man.
I'd choose to use the first meaning of the word, and I'd include people with artificial hearts there. We already have Borgs around us and its completely normal. No use using funky words to oversell an article... "Exoskeletons a reality" will be a better fit, only if the skeletons are being actively marketed already. We've seen tonnes of exoskeleton articles "vaporware" for years in slashdot, now theyre using words where they dont belong to shock the crowd more. OK OK I get it show me a real soldier in combat in an exoskeleton already!
(and I'll tell you its still not the Borg that has existed for years among us).
Well QoS has existed forever in most posix operating systems out there for network interfaces. Theres your network QoS.
Now I wonder if without using virtualization, on the command line, you could limit memory available to a process. One of the big 3 UNIXen out there probably has that.
Imagine a color flag. Its encrypted by an organization. When that flag arrives in the email, your user agent puts up a color flag or icon or whatever, big enough to be noticed, next to the email.
Now the organization is affiliated with the user agent makers like mozilla and microsoft.. so only encrypted emails from that organization are read and used. Companies etc pay a small fee to the organization, and give them a string (name) and ip (from and reply-to servers, the dns domain name). Their smtp gateway is this special organization which checks the dns claimed name, ip, name string etc to make sure the company is not fooling anyone, and adds the flag before sending it off. Companies can pay more for a 'higher' flag, so that emails from banks etc are more expensive to (attempt to) fake.
Doesnt work? Or how about this?
The company uses the special 'organization' as the smtp gateway. The organization checks the source IP against its member database (maybe the smtp requires auth) and strips the header of EVERYTHING except the subject, quoted name, and the user part of the from email address. It then rebuilds the whole header clean with the provided info and fires the email off. User agent checks if the email is from that special organization (can be just an IP check or smtp auth), and gives the email a different color in the list. Now each member company to the organization pays a minimal sum per email (1 cent?) to discourage mass mailers.
Where does the money go? If its a nonprofit, then a list of charities or telecom standards organizations etc.
Not worth it at all except for slashdot-article posts. You can buy an Athlon64 cpu + motherboard + 512mb ram for $250 and thats equal in power to dozens of PII machines, which will use up $250 worth additional power in under a year. Consider issues of space, the fact that its way more work and unreliability, and its obvious the pace of development is so fast that using 10 year old machines is never worth it. If you need to run many single-threaded apps, its still better to use one Athlon64 machine... but even better to use a dual-core or an Ultrasparc T1 with 8 cores, each core working equal to many PII machines.
The only use (slower) PIII and older machines have, is to be sent to poorer countries with knoppix CDs, since they have no money but enough labour to work and fix the computers. Electricity is generally cheaper there too (Iran, Pakistan, central asian, african) so it becomes worthwhile. So instead of building megaclusters, please donate them to poorer countries.
You can go to the Qandahari Bazaar of Quetta today, and you'll see dentists lined up on the footpath fixing peoples teeth. There are cars and rickshaws blaring away a few feet away from them. Dentists themselves are stocky muscular dudes in the same traditional dress, shoes taken off sitting on the cloth mat and sometimes with a made-in-china loupe holding boiled metal tools that they sharpen using knife sharpeners or simply ceramic bricks.
They obtain their tools from the organic waste of hospitals of Karachi which are sold on trolleys in the bazaars there. You see thousands of scalpels and the likes lined up under the sun sold for Rs 5 (10c) or less even to the grand public. Get up real close and in the crevices of the handle you'll notice dried up blood.
But the dentists DO boil their tools sometimes before your eyes on gas cookers, on the footpath. You'll occasionally hear a moan where a tooth is getting right out... real men dont need anasthetic.
With my full dental insurance here in Toronto, I still am put on long holds, have to fill out way more paperwork, and in the end, its still an italian surgeon who remarkably resembles the Qandahari Bazaar surgeons complete with hairy forearms, who pulls the teeth. Even the tools look the same. So stop pretending we've advanced that much!
Embedded Linux, even as ucLinux with uclibc etc is still fat for an embedded OS. Others like ecos, freertos etc are a much better fit. And so are the nice commercial offerings like qnx palmos and vxworks. You just have to add expensive ram and flash to make it boot linux in any form, and it cant be a 16-bitter. The only reason you see Linux being used is for hardware support and its nice networking, which is almost matched now by ecos' offerings.
I think for most small devices Linux is just too big. However it works better on routers and PVRs where space is not an issue. I will not buy a Linux watch or cellphone anytime soon, unless the battery can support it.
Most other 'fast' VMs out there only pass the kernel system calls to the vm subsystem which translates it. Everything else is more or less direct in a protected memory space, and the binaries are natively run on the hardware (ie vmware).
Xen is better. The virtualization and memory protection all happen within the hardware. All hardware access can be translated and sent to a Xen 'controller' which manages hardware between virtual OSes. Everything else is direct-to-hardware access and binaries are natively run on the CPU.
So I dont know what you mean by if it has been tested. Of course Windows has been tested on the i386 cpu since its inception, and so has all its x86 binaries. x86 software is optimized for x86 CPUs which is why it'll run real fast on a Xen box. Faster than vmware since what little virtualization is done (very unlike bochs) in software is then done in hardware.
Of course Xen has already been 'tested' if you will, between the BSDs and Linux and its definitely fast. I dont know if it has been tested on the POWER architecture, but POWER has hypervisor in hardware for AIX and its LPARs are definitely fast.
Xen itself is not the threat. Hardware Hypervisors are. Xen is like the software driver for the hypervisor technologies, and could be in theory replaced by other such apps.
Hypervisors will completely replace virtual machines in most markets. In others where strict timing is required or is measured, and where the server OS should not be preemptively disturbed, or where more memory is required than is available, or where standardized virtualized hardware is required, or in the emulation of different hardware/architectures, vm software will still have its place. The majority will simply use Xen.
I wonder if in a few years the standard way to install Windows on a server will be to install NetBSD + Xen and then Windows on top of it. Even better, if the BIOS will be replaced by a small NetBSD or Linux distro with Xen. This way a running OS can be moved live to a different machine mainframe-style... only on servers, desktops and maybe laptops.
Maybe I can move my live running game on the desktop to the laptop when I have to leave.
Yeah. My testdrive account has already expired... and I didnt need it since I bought a PARISC workstation from ebay, its memory and disks and got it up and running.... then I read about it on slashdot!
Most of the airspace below 12500 feet in north america is class X (dont remember X), where you can fly around anywhere without a previously declared plan. You need a mode C transponder, but youre free to fly VFR. Thats reflective of the freedom provided to you. Certain regions, cities, airports etc are more restricted, but the default piece of ground is this VFR class.
Looks like this class might be eliminated completely to allow drones to fly around anywhere. Which means a general aviation airplane will have to always file a flightplan and possibly remain on IFR, except on airport approaches, where they can request a VFR type approach. Flying will never be the same.
Its easy to sell this to the general public. "We dont want to let anyone fly just anywhere" and "we could use the extra security" and "War against terrorism" whatever that means. But somewhere in the future Americans will realize what they lost.
Same reason why Linux is not #1
on
Why Windows is Slow
·
· Score: 3, Insightful
Its precisely Windows' legacy support that it holds the market share. Make a new binary format, take away all the previous apps ability to run, and suddenly Windows has lost the real edge, the real reason why everyone doesnt switch to another OS. Linux/BSD are awesome, except too many apps run only on Windows. Many apple and Linux fans are sitting on Win32 machines right now because theres that one app that has no equivalent in Linux/OSX. Games are a significant part of those apps.
Say Windows switches to a new binary format for a new processor and asks all other software and driver vendors to follow suit. Many of them wont rerelease their apps. Others will not care. Many driver makers will not bother to produce the new version (I've tried running the AMD64 Windows XP... so I know all this). The result is Linux has the edge suddenly. You dont need to have vendors rerelease drivers, except for the few proprietary drivers (like nvidia).
Microsoft will never do that. AMD64 is giving em enough headaches as it is... and AMD64 actually supports x86 32-bit in-hardware. Take away DOS support, and all the older API in Windows, and suddenly there are more apps available for Linux than for Windows. Suddenly, MSFT stock seems overvalued.
Why? 4GB of RAM means 4kb of ram per molecule (or was it atom?) Computers with 32GB of ram (AIX, sun) arent the most expensive ones around. I dont think each molecule needs 4kb of data including its properties and interactions. Several million particles should be possible in a midsized server's ram entirely.
Simulating life is awesome. Now the next step is to simulate something like an Amoeba in water... let its DNA drive it to 'eat' a food particle, and see how accurate the digestion (and binary replication) is with the input being only the DNA and initial conditions. I wonder what kind of computers are required to simulate all that, in how much time? I'd more gladly donate cpu cycles to this than to SETI.
Next I wonder if the computer can be used to run regression tests to create the ideal bacteria or virii for a given situation. Virii can be built to repair human DNA in various ways... a particularly disadvantageous gene can be switched off throughout the body once infected with the virus.
Of course this only allows Cybernet to have more destroying power once it 'wakes up'.
I would replace mkdir with ps. I dont nearly use mkdir as frequently, maybe during embedded systems simulation...
I use ps head tail and cat alot. su also comes to mind. Others like ifconfig dmesg are frequently used but are more specialized. Of course I'd use joe instead of vi except in minimal systems.
I setup an OpenBSD system in vmware and tried to trim it down as much as possible. Init would just exec bash and nothing more. There were still some kernel processes. I realized you cant have a monoprocessor OS today beside DOS.
So beside DOS, anything else can utilize the second core. As far as its feasibility goes, if it costs twice as much as a monocore, its not worth it. If it costs 20% more, well worth it.
Let it stay at 386 till x64 becomes mainstream enough for most packages to simply be compiled for x64. Then it'll be stuck with x64 for another 15 years.
Seriously enough, x86 means {?86} or anything-86. Linux doesnt run on a 286. It runs on a 386,486,586.. etc etc, and I dont think the Athlon FX 57 quite has a simple number as that. However the 386 code is the common denominator, just as code with NX bit is the common denominator for the OSX.
I fully agree with you. Developing all those commands, ls, cp, mv etc should have taken some effort. And the kernel, wow, thats some piece of work. Other impressive parts of redhat is the glibc and the gcc compiler that can compile on so many platforms. Add some databases to that, and you're competing with Microsoft. They even have a windows system thats pretty advanced and looks great.
Very true. For that reason I recommend data and network administrators over programmers. Support people must be right there on the spot to solve the major problem. If you have 20 unix servers, cant depend on an IT guy in Mumbai 12 time zones away to debug and fix it all immediately. Each company over 50 fulltime employees in N America needs one techie to control all communications and data. This part of the IT world will not die.
And then there are specialized jobs for which (REALLY) companies look for the H-1B visas... like certain driver programmers for certain OSes, certain ERP system's module developers, people who are really good with graphic development in certain packages etc. There may be thousands of good IT guys out of work in N America, none or few might actually have the exact skill the employee is looking for. Its the bread-and-butter IT jobs which will get an American employee nowhere (Windows 2000, Visual Basic, MS Exchange, MS Access), unless its augumented with the skillset of one specific market (medical, manufacturing, finance).
But dont tell everyone that!! We dont want support jobs saturated now.
Slashdot Mirror
I'd be careful with these terms.
Heres the definition I got from wikipedia:
"The term cyborg, a portmanteau of cybernetic organism, is used to designate an organism which is a mixture of organic and mechanical (synthetic) parts. Generally, the aim is to add to or enhance the abilities of an organism by using technology."
Now the first sentence means its an organism thats both flesh and machine. This means the machine should be an innate part of the organism, to the level its not the organism anymore without the machine. I suppose pacemakers and artificial hearts will make the person a cyborg. However the second part means use of technology makes the organism a cyborg. That means wearing contact lenses makes me a cyborg.
Now 'bionic eyes' which really is a small screen in your helmet, and the kind of exoskeletons that we've seen from military contractors are really add-ons that the user will remove at night before going to bed. Even prosthetics are generally removed (fake teeth) at bedtime, which disqualifies the organism from being a cyborg in the first sense of the word. If you accept the second sense, then I'm the 6 dollar man.
I'd choose to use the first meaning of the word, and I'd include people with artificial hearts there. We already have Borgs around us and its completely normal. No use using funky words to oversell an article... "Exoskeletons a reality" will be a better fit, only if the skeletons are being actively marketed already. We've seen tonnes of exoskeleton articles "vaporware" for years in slashdot, now theyre using words where they dont belong to shock the crowd more. OK OK I get it show me a real soldier in combat in an exoskeleton already!
(and I'll tell you its still not the Borg that has existed for years among us).
Well QoS has existed forever in most posix operating systems out there for network interfaces. Theres your network QoS.
Now I wonder if without using virtualization, on the command line, you could limit memory available to a process. One of the big 3 UNIXen out there probably has that.
Been there. Tried it for 50-odd users. Impossible.
Imagine a color flag. Its encrypted by an organization. When that flag arrives in the email, your user agent puts up a color flag or icon or whatever, big enough to be noticed, next to the email.
Now the organization is affiliated with the user agent makers like mozilla and microsoft.. so only encrypted emails from that organization are read and used. Companies etc pay a small fee to the organization, and give them a string (name) and ip (from and reply-to servers, the dns domain name). Their smtp gateway is this special organization which checks the dns claimed name, ip, name string etc to make sure the company is not fooling anyone, and adds the flag before sending it off. Companies can pay more for a 'higher' flag, so that emails from banks etc are more expensive to (attempt to) fake.
Doesnt work? Or how about this?
The company uses the special 'organization' as the smtp gateway. The organization checks the source IP against its member database (maybe the smtp requires auth) and strips the header of EVERYTHING except the subject, quoted name, and the user part of the from email address. It then rebuilds the whole header clean with the provided info and fires the email off. User agent checks if the email is from that special organization (can be just an IP check or smtp auth), and gives the email a different color in the list. Now each member company to the organization pays a minimal sum per email (1 cent?) to discourage mass mailers.
Where does the money go? If its a nonprofit, then a list of charities or telecom standards organizations etc.
Not worth it at all except for slashdot-article posts. You can buy an Athlon64 cpu + motherboard + 512mb ram for $250 and thats equal in power to dozens of PII machines, which will use up $250 worth additional power in under a year. Consider issues of space, the fact that its way more work and unreliability, and its obvious the pace of development is so fast that using 10 year old machines is never worth it. If you need to run many single-threaded apps, its still better to use one Athlon64 machine... but even better to use a dual-core or an Ultrasparc T1 with 8 cores, each core working equal to many PII machines.
The only use (slower) PIII and older machines have, is to be sent to poorer countries with knoppix CDs, since they have no money but enough labour to work and fix the computers. Electricity is generally cheaper there too (Iran, Pakistan, central asian, african) so it becomes worthwhile. So instead of building megaclusters, please donate them to poorer countries.
You can go to the Qandahari Bazaar of Quetta today, and you'll see dentists lined up on the footpath fixing peoples teeth. There are cars and rickshaws blaring away a few feet away from them. Dentists themselves are stocky muscular dudes in the same traditional dress, shoes taken off sitting on the cloth mat and sometimes with a made-in-china loupe holding boiled metal tools that they sharpen using knife sharpeners or simply ceramic bricks.
They obtain their tools from the organic waste of hospitals of Karachi which are sold on trolleys in the bazaars there. You see thousands of scalpels and the likes lined up under the sun sold for Rs 5 (10c) or less even to the grand public. Get up real close and in the crevices of the handle you'll notice dried up blood.
But the dentists DO boil their tools sometimes before your eyes on gas cookers, on the footpath. You'll occasionally hear a moan where a tooth is getting right out... real men dont need anasthetic.
With my full dental insurance here in Toronto, I still am put on long holds, have to fill out way more paperwork, and in the end, its still an italian surgeon who remarkably resembles the Qandahari Bazaar surgeons complete with hairy forearms, who pulls the teeth. Even the tools look the same. So stop pretending we've advanced that much!
How about this data set across time:
/dev/random
cat
or
dd if=/dev/random of=/tmp/file ; sleep 1 ; dd if=/dev/random of=/tmp/file2
Can it compress it even 2x reliably?
Embedded Linux, even as ucLinux with uclibc etc is still fat for an embedded OS. Others like ecos, freertos etc are a much better fit. And so are the nice commercial offerings like qnx palmos and vxworks. You just have to add expensive ram and flash to make it boot linux in any form, and it cant be a 16-bitter. The only reason you see Linux being used is for hardware support and its nice networking, which is almost matched now by ecos' offerings.
I think for most small devices Linux is just too big. However it works better on routers and PVRs where space is not an issue. I will not buy a Linux watch or cellphone anytime soon, unless the battery can support it.
You dont know much about Xen do you?
Most other 'fast' VMs out there only pass the kernel system calls to the vm subsystem which translates it. Everything else is more or less direct in a protected memory space, and the binaries are natively run on the hardware (ie vmware).
Xen is better. The virtualization and memory protection all happen within the hardware. All hardware access can be translated and sent to a Xen 'controller' which manages hardware between virtual OSes. Everything else is direct-to-hardware access and binaries are natively run on the CPU.
So I dont know what you mean by if it has been tested. Of course Windows has been tested on the i386 cpu since its inception, and so has all its x86 binaries. x86 software is optimized for x86 CPUs which is why it'll run real fast on a Xen box. Faster than vmware since what little virtualization is done (very unlike bochs) in software is then done in hardware.
Of course Xen has already been 'tested' if you will, between the BSDs and Linux and its definitely fast. I dont know if it has been tested on the POWER architecture, but POWER has hypervisor in hardware for AIX and its LPARs are definitely fast.
Xen itself is not the threat. Hardware Hypervisors are. Xen is like the software driver for the hypervisor technologies, and could be in theory replaced by other such apps.
Hypervisors will completely replace virtual machines in most markets. In others where strict timing is required or is measured, and where the server OS should not be preemptively disturbed, or where more memory is required than is available, or where standardized virtualized hardware is required, or in the emulation of different hardware/architectures, vm software will still have its place. The majority will simply use Xen.
I wonder if in a few years the standard way to install Windows on a server will be to install NetBSD + Xen and then Windows on top of it. Even better, if the BIOS will be replaced by a small NetBSD or Linux distro with Xen. This way a running OS can be moved live to a different machine mainframe-style... only on servers, desktops and maybe laptops.
Maybe I can move my live running game on the desktop to the laptop when I have to leave.
I'd like to start a company manufacturing such bras.
Who wants to buy stake?
Are you trying to fool us?
Yeah. My testdrive account has already expired... and I didnt need it since I bought a PARISC workstation from ebay, its memory and disks and got it up and running. ... then I read about it on slashdot!
Document my code.
or
Just build all the required functions and classes, and let him connect the dots.
or
Let him use all kinds of compiler optimizations and build the makefiles.
or finally
Let him attempt to build the whole program first. Then 'tweak' it to make it work.
Most of the airspace below 12500 feet in north america is class X (dont remember X), where you can fly around anywhere without a previously declared plan. You need a mode C transponder, but youre free to fly VFR. Thats reflective of the freedom provided to you. Certain regions, cities, airports etc are more restricted, but the default piece of ground is this VFR class.
Looks like this class might be eliminated completely to allow drones to fly around anywhere. Which means a general aviation airplane will have to always file a flightplan and possibly remain on IFR, except on airport approaches, where they can request a VFR type approach. Flying will never be the same.
Its easy to sell this to the general public. "We dont want to let anyone fly just anywhere" and "we could use the extra security" and "War against terrorism" whatever that means. But somewhere in the future Americans will realize what they lost.
Its precisely Windows' legacy support that it holds the market share. Make a new binary format, take away all the previous apps ability to run, and suddenly Windows has lost the real edge, the real reason why everyone doesnt switch to another OS. Linux/BSD are awesome, except too many apps run only on Windows. Many apple and Linux fans are sitting on Win32 machines right now because theres that one app that has no equivalent in Linux/OSX. Games are a significant part of those apps.
Say Windows switches to a new binary format for a new processor and asks all other software and driver vendors to follow suit. Many of them wont rerelease their apps. Others will not care. Many driver makers will not bother to produce the new version (I've tried running the AMD64 Windows XP... so I know all this). The result is Linux has the edge suddenly. You dont need to have vendors rerelease drivers, except for the few proprietary drivers (like nvidia).
Microsoft will never do that. AMD64 is giving em enough headaches as it is... and AMD64 actually supports x86 32-bit in-hardware. Take away DOS support, and all the older API in Windows, and suddenly there are more apps available for Linux than for Windows. Suddenly, MSFT stock seems overvalued.
"The RAM requirements must be insane"
Why? 4GB of RAM means 4kb of ram per molecule (or was it atom?)
Computers with 32GB of ram (AIX, sun) arent the most expensive ones around. I dont think each molecule needs 4kb of data including its properties and interactions. Several million particles should be possible in a midsized server's ram entirely.
Simulating life is awesome. Now the next step is to simulate something like an Amoeba in water... let its DNA drive it to 'eat' a food particle, and see how accurate the digestion (and binary replication) is with the input being only the DNA and initial conditions. I wonder what kind of computers are required to simulate all that, in how much time? I'd more gladly donate cpu cycles to this than to SETI.
Next I wonder if the computer can be used to run regression tests to create the ideal bacteria or virii for a given situation. Virii can be built to repair human DNA in various ways... a particularly disadvantageous gene can be switched off throughout the body once infected with the virus.
Of course this only allows Cybernet to have more destroying power once it 'wakes up'.
I would replace mkdir with ps. I dont nearly use mkdir as frequently, maybe during embedded systems simulation...
I use ps head tail and cat alot. su also comes to mind. Others like ifconfig dmesg are frequently used but are more specialized. Of course I'd use joe instead of vi except in minimal systems.
And I'd add 'more' or 'less'.
I setup an OpenBSD system in vmware and tried to trim it down as much as possible. Init would just exec bash and nothing more. There were still some kernel processes. I realized you cant have a monoprocessor OS today beside DOS.
So beside DOS, anything else can utilize the second core. As far as its feasibility goes, if it costs twice as much as a monocore, its not worth it. If it costs 20% more, well worth it.
Pentium still sounds advanced compared to the base architecture: i386
Let it stay at 386 till x64 becomes mainstream enough for most packages to simply be compiled for x64. Then it'll be stuck with x64 for another 15 years.
Seriously enough, x86 means {?86} or anything-86. Linux doesnt run on a 286. It runs on a 386,486,586.. etc etc, and I dont think the Athlon FX 57 quite has a simple number as that. However the 386 code is the common denominator, just as code with NX bit is the common denominator for the OSX.
I fully agree with you. Developing all those commands, ls, cp, mv etc should have taken some effort. And the kernel, wow, thats some piece of work. Other impressive parts of redhat is the glibc and the gcc compiler that can compile on so many platforms. Add some databases to that, and you're competing with Microsoft. They even have a windows system thats pretty advanced and looks great.
How could Redhat ever do that?
"There is a limit to what you can outsource"
Very true. For that reason I recommend data and network administrators over programmers. Support people must be right there on the spot to solve the major problem. If you have 20 unix servers, cant depend on an IT guy in Mumbai 12 time zones away to debug and fix it all immediately. Each company over 50 fulltime employees in N America needs one techie to control all communications and data. This part of the IT world will not die.
And then there are specialized jobs for which (REALLY) companies look for the H-1B visas... like certain driver programmers for certain OSes, certain ERP system's module developers, people who are really good with graphic development in certain packages etc. There may be thousands of good IT guys out of work in N America, none or few might actually have the exact skill the employee is looking for. Its the bread-and-butter IT jobs which will get an American employee nowhere (Windows 2000, Visual Basic, MS Exchange, MS Access), unless its augumented with the skillset of one specific market (medical, manufacturing, finance).
But dont tell everyone that!! We dont want support jobs saturated now.
So Microsoft has stagnated so much, some people there are killing innovation now!
I wonder how much more innovation Microsoft plans to kill this way in the future? I hope they kill Windows 2000 innovation to benefit WINE or ReactOS.
To the uninitiated: Microsoft has repeatedly called opensourcing killing innovation.