If there were 193 vetoes, do you think we would ever get changed over to IPv6?
Since gridlock is (IMHO) the natural state of the UN. I expect they would have sat on v4 until things fell apart.
As an Internet systems administrator, I am personally aware of the thousands of attacks per day on my systems from various places in China. If Huawei is so great, how come they tolerate and allow chinese hackers to attack our country on such a grand scale. There is of course the question of whether these attacks are sanctioned by the Communist Party. And I guess as well we should ask if we want to buy critical infrastructure components from a communist country. As far as the UnitedStates is concerned, I think we should by equipment made in the USA. And our neighbor to the north might want to consider that as well.. Chinese telecom equipment is in no way superior to our own, and perhaps only cheaper. But do we want to skimp on such important infrastructure?
I am with you, Mac Pro 8-cores, 16GB-ram 32TB of HD, 46" 1920x1200 HD flat panel. The workstation of our dreams.
Too bad the MacPro2,1 will never run Mountain Lion due to 32-bit boot rom that cannot be upgraded.
"4G of memory just isn't enough anymore"
I remember when Microsoft wanted us to move to Windows 3.1 (with enhanced mode), and 4M just wasn't enough any more. And at that time, 16MB of ram costed $1200.
Now they have a thousand times more memory, and that isn't enough. Considering that Windows 3.1 had TCP/IP and OLE, I can't think of what Microsoft has added to the Windows functionality that was complicated enough to have eaten all that memory and still be hungry.
Now I admit that loading hundreds of patches on top of XP does bloat it's software inventory, and maybe along the way, Microsoft could have recompiled those patches into the product and simplified the rube goldberg creation that is the highly patched XP of today.
Considering the huge amount of XP that s still installed and running, I think Microsoft should have spawned a division to continue sustained engineering on it for another five years. In our current economic condition, we can hardly thank Microsoft for forcing another round of hardware and software upgrades.
Normally I would agree that they would do what serves them best.
But then, why would they want to do business with a company that is smiling out of one side of it's mouth,
and snarling out of the other. Maybe Apple should have considered the consequences before suing their panel
supplier, and blocking their distribution, continents at a time.
I doubt that Samsung will suffer appreciably the loss of Apple as a client, because there are most likely plenty of manufacturers standing in line for display panel allocations. Microsoft for instance.
Farnell is not so great. They lost the first shipment to me and it took them 6 months to refund my money. Their business practices $#&%. I will never do business with them again.
Nope, it was the stealth cruiser. One of the Navy's most closely guarded secrets, an outgrowth of the Philadelphia Experiment.
The cruiser actually becomes invisible, but it has such an intense electro-magnetomic(tm) field, it grabbed right on to the sub, and gave it a big hug.
After many years of MSDN Universal, I just got tired of amorphous collections of fading API's mixed with massive amounts of deprecated tools from five of more versions back. After buying thousands of dollars of version specific Microsoft Press books that were almost obsolete by the time they were printed. A friend at Microsoft told me about shopping carts full of hard disk drives so when developers blew up their development systems, they could pop a new drive in and be back up and ready to continue coding. There is a cost to eating your own dog food. The constantly changing directions that Microsoft recommended for development caused continual need to re-architect applications. VS may be the best IDE for coding Windows apps, but that isn't saying much. What pisses me off the most is the fact that VS will practically write the app for you if you hit the right combinations of keystrokes, but the inability to back up caused me a lot of heartache. And Petzold gives me a stomach ache, from resting to book on my stomach while reading it.
My problem with VS (which has existed for many versions) is that once you run the wizard, declare your wishes, out pops 38,000 lines of code that you become instantly responsible for. If that code wasn't buggy things might be better, but if I coded 38,000 lines of code without testing it, I would call that a "Major Hack".
The entire point of CP/M was the separation of the hardware interface layer from the kernel layer. Before that applications had to contain floppy drivers. As processors progressed, DRI had operating systems which more fully utilized the advanced features of the x86 CPU's (protected mode boundary protection and memory protection for one). They had common back-end code generator language technology and front ends for the half dozen common programming languages, and their graphics products were coming along nicely. In terms of innovation, Digital Research outclassed Microsoft substantially. IBM's gift of monopoly status for DOS on the PC platform allowed Microsoft to run roughshod over DRI despite significant technical merit on the DRI side. Also DRI had CP/M versions that ran on Z8000, 680x0 in addition to the x86. DRI's ability to parse their technologies into hardware specific and generic modules had great promise for the operating systems and the languages, and significant developer support all over the world. Sure with all the DOS monopoly money, Microsoft eventually bought Lattice C and called it Microsoft C. They perverted Java and called it J++. They declared we might as well code in BASIC and.NET made that possible giving us the stellar Vista core experience. And Microsoft has left behind a trail littered with six month learning curves that didn't result in viability for developers after all. So why should we believe that Microsoft has the magic bullet for Phones, Pads, or servers?
I don't care for how Bill Gates became rich, at the expense of an entire industry, but he had some vision and it worked out well for him. Some young guy in the grocery store told me recently that Bill single handedly gave us the personal computer. There was no arguing with the guy, because we were in the kool-aid department of the store, and his cart was full. Steve Balmer is in my opinion, the Anti-Jobs. He is running -80 on the charisma scale. No one I know believes anything he has to say. I was told once, if you want to make lots of money, you have to help others do so as well. It sees to me that Microsoft has a long history of turning on it's strategic partners. These days it's turning on the developers and bailing on VS in favor of HTML5/script. Also turning on it's OEM partners by going into the hardware business. Their devices will always have better versions of software than they release to their OEM's. As an example, in the early days, they licensed MFC to compiler companies in order to let them target Windows. But there was always a newer more capable version of MFC bundled with Microsoft's compiler. Balmer just doesn't inspire the customers. They think they need Windows and Office, but in general I think WIndows/Office owners don't feel pride in their choice of computing equipment.
Turn up the variac if you want the computer to go faster, as the microcode is on the drum, and if the drum rotates faster, instructions execute faster. And keep plenty of spare tubes around because the MTBF on the tubes was about 45 minutes.
I recommend you have two or three computers...
1) at least one Windows box.
2) at least one Linux box.
3) at least one Mac.
At least one of these should have a 46" flat panel display with 1920x1200 resolution
You will often want to display a large schematic, and several component data sheets (mandatory while debugging).
Two large displays are better than one.
New age test equipment often comes in the form of USB pods, and the host software varies in requirements
from Windows XP onward. Logic analyzers like the Saleae that have Mac compatible software are very cool.
Windows version requirements for inexpensive test equipment often include downlevel or unsupported Windows versions.
Thank you tekrat, I have noticed the same thing. Recently I restored an IBM-PC/AT and was amazed that it performed tasks for me quickly, and never made me wait more than a few seconds. Here we are with machines a thousand times faster, a thousand times more ram, thousands of times more storage, and still they make us wait. I have known for decades that Windows was a cycle hog. Windows NT was briefly fast during certain beta cycles.
When I edit a program, and compile it, I know those sounds from the hard disk are the machine actually doing what I told it, and not scanning, indexing, caching, and doing things that modern systems do, not on a time available basis, but in direct competition with user directed operations.
I do love the vintage systems, from Apple ][e through PC clones with ISA16 motherboards. I would say the Compaq Deskpro 386/16 was the end of the lovable old machines.
I was watching the ISO stack for a long time. Marshall Rose was involved with a sample implementation which was big.
I seem to remember that the minimum functioning ISO stack was something like 80K back when you were lucky if you had 640K.
The possibility of creating embedded systems that included the stack was prohibitively expensive at the time.
I remember a time when only one or several vendors had an ISO stack you could license into a product.
My attempts to understand it were hampered by the critical information spread evenly through a huge number of expensive documents.
If you bought one, it referenced another,and there was no end in sight to the number of documents you would have to buy and absorb before you would have critical mass on an implementation, or enough of one to do anything.
Marshall predicted that we would need a period of transition to move from the toy internet to the ISO version.
After a while we didn't hear so much about it.
Some of us are questioning whether we can depend on Apple for leadership. At the moment we can depend on them to charge more, make more profit, and show less responsibility for the working conditions where Apple hardware is made. I have a 3.0GHz 8-Core Xeon Mac Pro that cannot run Mountain Lion because the machine has a down-level boot rom that apparently cannot be updated, and I cannot afford to buy another Mac Pro right now. I am heart-sore about what has happened to my feelings about Apple in the last several years. What is your view on Apple's future relationship to it's customers and development partners? You're still my hero Woz.
This year I happened to be a paid up member of the Apple Developer program for Mac OS X. After I paid, I went to their web site and downloaded my signing keys, for the installer and for the application. It seems to me that sending the keys over the internet at all is a gross security violation. Off the top of my head, I don 't see a practical was of transporting these keys from Cupertino to a worldwide collection of developers. I agree that the signing keys should never be on a machine connected to the Internet. What is wrong with this picture, and how do we make this better?
You shouldn't trust the USB on this card. It manifests in various ways. On my pi, it causes random keystrokes to be auto-repeated hundreds of times. Just try deleting one line of a source file in vi and you will see your source code vanish like magic. The USB trouble also touches the network interface, since it is the connection between the ethernet and the CPU. in my opinion, the pi is unusable, and a gigantic rip-off. They shipped many thousands of these things, knowing full well about the USB problem, which is a hardware problem primarily. Although they would like to fix this in software, so far no joy there.
Weight is such a factor in notebook applications, and helium filled drives could enable a new class of devices weighing significantly less than existing hard drivers...
How do you build a supercomputer out of processor modules that cannot reliably communicate with each other.
The ethernet connectivity of the pi is based on a small module that attaches to the USB.
I don't get it...
Slashdot Mirror
I thought the universe was patented by root@univese.org.
If there were 193 vetoes, do you think we would ever get changed over to IPv6? Since gridlock is (IMHO) the natural state of the UN. I expect they would have sat on v4 until things fell apart.
As an Internet systems administrator, I am personally aware of the thousands of attacks per day on my systems from various places in China. If Huawei is so great, how come they tolerate and allow chinese hackers to attack our country on such a grand scale. There is of course the question of whether these attacks are sanctioned by the Communist Party. And I guess as well we should ask if we want to buy critical infrastructure components from a communist country. As far as the UnitedStates is concerned, I think we should by equipment made in the USA. And our neighbor to the north might want to consider that as well.. Chinese telecom equipment is in no way superior to our own, and perhaps only cheaper. But do we want to skimp on such important infrastructure?
I am with you, Mac Pro 8-cores, 16GB-ram 32TB of HD, 46" 1920x1200 HD flat panel. The workstation of our dreams. Too bad the MacPro2,1 will never run Mountain Lion due to 32-bit boot rom that cannot be upgraded.
"4G of memory just isn't enough anymore" I remember when Microsoft wanted us to move to Windows 3.1 (with enhanced mode), and 4M just wasn't enough any more. And at that time, 16MB of ram costed $1200. Now they have a thousand times more memory, and that isn't enough. Considering that Windows 3.1 had TCP/IP and OLE, I can't think of what Microsoft has added to the Windows functionality that was complicated enough to have eaten all that memory and still be hungry. Now I admit that loading hundreds of patches on top of XP does bloat it's software inventory, and maybe along the way, Microsoft could have recompiled those patches into the product and simplified the rube goldberg creation that is the highly patched XP of today. Considering the huge amount of XP that s still installed and running, I think Microsoft should have spawned a division to continue sustained engineering on it for another five years. In our current economic condition, we can hardly thank Microsoft for forcing another round of hardware and software upgrades.
Normally I would agree that they would do what serves them best. But then, why would they want to do business with a company that is smiling out of one side of it's mouth, and snarling out of the other. Maybe Apple should have considered the consequences before suing their panel supplier, and blocking their distribution, continents at a time. I doubt that Samsung will suffer appreciably the loss of Apple as a client, because there are most likely plenty of manufacturers standing in line for display panel allocations. Microsoft for instance.
Call me when it can take out the trash.
Farnell is not so great. They lost the first shipment to me and it took them 6 months to refund my money. Their business practices $#&%. I will never do business with them again.
> an outgrowth of the Philadelphia Experiment
That was a silly movie.
-- BMO
The sequel was worse.
Nope, it was the stealth cruiser. One of the Navy's most closely guarded secrets, an outgrowth of the Philadelphia Experiment. The cruiser actually becomes invisible, but it has such an intense electro-magnetomic(tm) field, it grabbed right on to the sub, and gave it a big hug.
After many years of MSDN Universal, I just got tired of amorphous collections of fading API's mixed with massive amounts of deprecated tools from five of more versions back. After buying thousands of dollars of version specific Microsoft Press books that were almost obsolete by the time they were printed. A friend at Microsoft told me about shopping carts full of hard disk drives so when developers blew up their development systems, they could pop a new drive in and be back up and ready to continue coding. There is a cost to eating your own dog food. The constantly changing directions that Microsoft recommended for development caused continual need to re-architect applications. VS may be the best IDE for coding Windows apps, but that isn't saying much. What pisses me off the most is the fact that VS will practically write the app for you if you hit the right combinations of keystrokes, but the inability to back up caused me a lot of heartache. And Petzold gives me a stomach ache, from resting to book on my stomach while reading it.
Sweet :-)
My problem with VS (which has existed for many versions) is that once you run the wizard, declare your wishes, out pops 38,000 lines of code that you become instantly responsible for. If that code wasn't buggy things might be better, but if I coded 38,000 lines of code without testing it, I would call that a "Major Hack".
The entire point of CP/M was the separation of the hardware interface layer from the kernel layer. Before that applications had to contain floppy drivers. As processors progressed, DRI had operating systems which more fully utilized the advanced features of the x86 CPU's (protected mode boundary protection and memory protection for one). They had common back-end code generator language technology and front ends for the half dozen common programming languages, and their graphics products were coming along nicely. In terms of innovation, Digital Research outclassed Microsoft substantially. IBM's gift of monopoly status for DOS on the PC platform allowed Microsoft to run roughshod over DRI despite significant technical merit on the DRI side. Also DRI had CP/M versions that ran on Z8000, 680x0 in addition to the x86. DRI's ability to parse their technologies into hardware specific and generic modules had great promise for the operating systems and the languages, and significant developer support all over the world. Sure with all the DOS monopoly money, Microsoft eventually bought Lattice C and called it Microsoft C. They perverted Java and called it J++. They declared we might as well code in BASIC and .NET made that possible giving us the stellar Vista core experience. And Microsoft has left behind a trail littered with six month learning curves that didn't result in viability for developers after all. So why should we believe that Microsoft has the magic bullet for Phones, Pads, or servers?
My word processor is Wordstar and it runs on CP/M-80 v2.2c (N8VEM). :-)
I don't care for how Bill Gates became rich, at the expense of an entire industry, but he had some vision and it worked out well for him. Some young guy in the grocery store told me recently that Bill single handedly gave us the personal computer. There was no arguing with the guy, because we were in the kool-aid department of the store, and his cart was full. Steve Balmer is in my opinion, the Anti-Jobs. He is running -80 on the charisma scale. No one I know believes anything he has to say. I was told once, if you want to make lots of money, you have to help others do so as well. It sees to me that Microsoft has a long history of turning on it's strategic partners. These days it's turning on the developers and bailing on VS in favor of HTML5/script. Also turning on it's OEM partners by going into the hardware business. Their devices will always have better versions of software than they release to their OEM's. As an example, in the early days, they licensed MFC to compiler companies in order to let them target Windows. But there was always a newer more capable version of MFC bundled with Microsoft's compiler. Balmer just doesn't inspire the customers. They think they need Windows and Office, but in general I think WIndows/Office owners don't feel pride in their choice of computing equipment.
Turn up the variac if you want the computer to go faster, as the microcode is on the drum, and if the drum rotates faster, instructions execute faster. And keep plenty of spare tubes around because the MTBF on the tubes was about 45 minutes.
I recommend you have two or three computers... 1) at least one Windows box. 2) at least one Linux box. 3) at least one Mac. At least one of these should have a 46" flat panel display with 1920x1200 resolution You will often want to display a large schematic, and several component data sheets (mandatory while debugging). Two large displays are better than one. New age test equipment often comes in the form of USB pods, and the host software varies in requirements from Windows XP onward. Logic analyzers like the Saleae that have Mac compatible software are very cool. Windows version requirements for inexpensive test equipment often include downlevel or unsupported Windows versions.
Thank you tekrat, I have noticed the same thing. Recently I restored an IBM-PC/AT and was amazed that it performed tasks for me quickly, and never made me wait more than a few seconds. Here we are with machines a thousand times faster, a thousand times more ram, thousands of times more storage, and still they make us wait. I have known for decades that Windows was a cycle hog. Windows NT was briefly fast during certain beta cycles. When I edit a program, and compile it, I know those sounds from the hard disk are the machine actually doing what I told it, and not scanning, indexing, caching, and doing things that modern systems do, not on a time available basis, but in direct competition with user directed operations. I do love the vintage systems, from Apple ][e through PC clones with ISA16 motherboards. I would say the Compaq Deskpro 386/16 was the end of the lovable old machines.
I was watching the ISO stack for a long time. Marshall Rose was involved with a sample implementation which was big. I seem to remember that the minimum functioning ISO stack was something like 80K back when you were lucky if you had 640K. The possibility of creating embedded systems that included the stack was prohibitively expensive at the time. I remember a time when only one or several vendors had an ISO stack you could license into a product. My attempts to understand it were hampered by the critical information spread evenly through a huge number of expensive documents. If you bought one, it referenced another,and there was no end in sight to the number of documents you would have to buy and absorb before you would have critical mass on an implementation, or enough of one to do anything. Marshall predicted that we would need a period of transition to move from the toy internet to the ISO version. After a while we didn't hear so much about it.
Some of us are questioning whether we can depend on Apple for leadership. At the moment we can depend on them to charge more, make more profit, and show less responsibility for the working conditions where Apple hardware is made. I have a 3.0GHz 8-Core Xeon Mac Pro that cannot run Mountain Lion because the machine has a down-level boot rom that apparently cannot be updated, and I cannot afford to buy another Mac Pro right now. I am heart-sore about what has happened to my feelings about Apple in the last several years. What is your view on Apple's future relationship to it's customers and development partners? You're still my hero Woz.
This year I happened to be a paid up member of the Apple Developer program for Mac OS X. After I paid, I went to their web site and downloaded my signing keys, for the installer and for the application. It seems to me that sending the keys over the internet at all is a gross security violation. Off the top of my head, I don 't see a practical was of transporting these keys from Cupertino to a worldwide collection of developers. I agree that the signing keys should never be on a machine connected to the Internet. What is wrong with this picture, and how do we make this better?
You shouldn't trust the USB on this card. It manifests in various ways. On my pi, it causes random keystrokes to be auto-repeated hundreds of times. Just try deleting one line of a source file in vi and you will see your source code vanish like magic. The USB trouble also touches the network interface, since it is the connection between the ethernet and the CPU. in my opinion, the pi is unusable, and a gigantic rip-off. They shipped many thousands of these things, knowing full well about the USB problem, which is a hardware problem primarily. Although they would like to fix this in software, so far no joy there.
Weight is such a factor in notebook applications, and helium filled drives could enable a new class of devices weighing significantly less than existing hard drivers...
How do you build a supercomputer out of processor modules that cannot reliably communicate with each other. The ethernet connectivity of the pi is based on a small module that attaches to the USB. I don't get it...