I'm quite skeptical of tariffs due to their history but some sort of reciprocal system seems to be needed. On a per nation basis low barriers to trade in both directions, high barriers to trade in both directions, but not low barriers in one direction and high barriers in the other. That is what really needs to be addressed, so its a little different than the historical protect a specific product/industry tariff. Its more of a protect an entire industrial base sort of thing.
And yet Trump keeps telling his rubes that jobs are coming back. They are not coming back. Workers will be replaced by robots.
Germany shows that it is not that bad, they have been heavily into automation for a while. Some different jobs are created, a lot of welders lost but a few electrical technicians added sort of thing. Economically speaking, robots or not the money is still being spent in the country and benefitting the country in various indirect ways. As opposed to outsourcing where all the indirect benefits go overseas.
How he's going to force companies to manufacture in the USA without adding legislation (because he said he would reduce legislation against corporations) is beyond me...
I think he said he would introduce some sort of tariff scheme when a trading partner's markets are not "open". It sort of sound like a reciprocal system, low barriers in both direction or high barriers in both directions, but no more low in one direction and high in the other. I'm not going to debate the wisdom of any tariff scheme but its clearly something quite independent of lower corporate taxes and lower regulations on corporations. If you are going to bash him at least get the criticisms correct.
It might not make sense at the present but you have to look at decades from now were most of the Western nations would have automated and will produce cheap crap themselves instead of importing them from China.
But it still doesn't make sense at present. And my bet is that in the future, those Western nations will develop that newer automation and then build and deploy it in China because that's where the world's industrial base will be.
Recent history has shown us that an industrial base is quite mobile. Provide manufacturers an incentive to move that is also approved/tolerated by consumers and the base will move. What may keep manufacturing in China is the "engineered" exchange rate that make everything available at a 25-30% discount.
It might not make sense at the present but you have to look at decades from now were most of the Western nations would have automated and will produce cheap crap themselves instead of importing them from China.
Automation won't keep manufacturing in China, maintaining parity with the west won't do it. Outsourcing is a royal PITA and there are many problems. There has to be a huge saving to offset the overhead and inefficiencies of outsourcing to make it economical. A wage gap was once part of the "savings" but that is shrinking and robots won't offer much savings either.
The other "savings" is basically that everything sold in China is at a 25-30% discount due to the "engineering" of the exchange rate. This may maintain outsourcing, not robots.
Apologies for communicating with three different individuals.
... and in all three, you seem almost desperate to have someone acknowledge that you are an insider with super-meaningful knowledge.
I am not an insider, nor have any special knowledge. It is quite well known that SOC has a wide latitude in gear selection. I merely saw a single instance of this well known practice. Apologies if your anti-military industrial complex meme or whatever failed. Perhaps there will be an F-35 post for you soon and you can find some joy.
Nope, the real difference is in the ability and willingness to navigate the military procurement process.
Not necessarily in this case, note "United States Army's Special Operations Command". They get a ton of say in gear. A friend's brother had a small company that made some photographic gear for the civilian market, SCUBA divers in particular, and SOC types got interested in it and the normal military procurement process was not involved. SOC can bypass the "military procurement process". They are not just war fighters, they are also involved in finding, developing and evaluating new gear and technology.
GP is right, All the vendors market to the guys wearing stars. If the general likes it then that's what we buy, doesn't matter what the grunts think.
Note "United States Army's Special Operations Command", that works entirely different. A friend's brother made some specialized photographic gear for the civilian market. SOC guys heard about it, visited, asked to evaluate it. They made some suggestions. These were incorporated into the design. They then told the guys wearing stars "we want this" and then "suits" got involved for the paperwork. Selection, evaluation and decision for this gear was made by "operators".
In my experience decisions like this are typically made because somebody high up likes their iPhone and doesn't want to have to learn how to use an Android phone. Sounds overly simplistic, but I've seen it happen too many times.
Bad guess. Note "United States Army's Special Operations Command", they get a lot of say in what equipment they use. A friend's brother made some unique camera equipment. SOC guys thought it interesting. The only people this small company every saw during evaluation were "operators". The "suits" did not get involved until the "operators" said "we want this". What you say may be true for normal military procurement, but its very different for SOC.
> Windows and Presentation Manager having a nearly identical API.
One major porting issue was that the screen co-ordinates were 'upside down'. Windows origin is upper-left corner. PM origin is lower-left.
That is not a major porting issue. That is a very minor thing, well, unless one writes some terrible code.
OS/2 PM had a much richer API but this was _later_ added to Windows.
By the way, I'm referring to Windows 3 not Windows 2.
> OS/2 was the upgrade path from DOS and the Windows was a temporary thing for people with legacy hardware/software
What kept people on Windows was Windows/386 which allowed multiple DOS-boxes to run the DOS software that they wanted to use. They could run WordPerfect 5.1 and Lotus123 at the same time along with the DOS accounting system.
OS/2 at the time only had a single 'penalty box'.
No. Users doing such things were so rare they are statistically insignificant. It was a classic chicken-and-egg thing. People didn't want to change apps, developers didn't want to port apps without a user base. DOS had the network effect advantage over OS/2. The only way to migrate users would have been to phase out DOS and ship OS/2 as the installed OS. But Microsoft changed their minds about partnering with IBM and about OS/2. Well, sort of, OS/2 NT, aka OS/2 3.0, was renamed Windows NT. So in a way DOS/Win3/Win9x was phased out and replaced by (OS/2)Windows NT.
According to Microsoft OS/2 1.x with Presentation Manager was the "upgrade path" from DOS. For users stuck with legacy software they were going to add a comparable GUI to DOS called Windows.
Windows had already been available for nearly three years before Presentation Manager was released with OS/2 1.1, and was introduced only three months after the OS/2 development agreement between MS and IBM was signed.
None of that changes the fact that Microsoft was telling developers, including me, that OS/2 was the upgrade path from DOS and the Windows was a temporary thing for people with legacy hardware/software and that migrating them to OS/2 with Presentation Manager would be easy since porting your apps would be easy, Windows and Presentation Manager having a nearly identical API. As I said there was a window of time where Microsoft was telling people OS/2 was the replacement for DOS, Presentation Manager the replacement for Windows.
And I had seen Windows since v1 too. Prior to v3 I don't think it was used much beyond allowing multiple DOS sessions to run.
The PM API was designed to be similar to the Windows API, not the other way around, and still had some substantial differences.
Its not that simple. Windows v3 had a lot of changes and you can't measure these things by product announcements and release dates. OS/2 Presentation Manager may have been influencing Windows 3.0 behind the scenes.
Having similar APIs was helpful, but they were different enough to make a common code base impractical.
As someone who had common code bases for Windows and MacOS I'd say you are mistaken. I'm not sure what Microsoft was saying but it may have been more of a porting your app from DOS to OS/2 thing. Remember, DOS was the dead end, just maintenance, OS/2 was the future... in that brief window of time.
I'm perfectly happy with your decision. However even if multitasking and multithreading were added they would not have to break backwards compatibility, they could merely extend the API. Legacy apps would not know or care and just run with a single thread. But this is just hypothetical, not a suggestion. If you had an itch to go in that direction I'd say create a FreeOS2 based on 1.x. Hell, there may be some commercial viability to such a project too.
Oh... and damn you;-)... you are the main reason my better DOS and BIOS books, and the Pentium MMX 166, survive garage cleanings and take up valuable space.
Well, there was a "modern MS DOS", it was MS OS/2 1.x.
Except that OS/2 was a multitasking, protected mode operating system...
That was part of what made it "modern".
... from IBM...
And from Microsoft
... and MS-DOS wasn't any of those things.
OS/2 1.x was described by Microsoft as a modern OS designed to replace DOS.
OS/2 1.1, released just 11 months later, came with the promised Presentation Manager GUI, further extending its abilities beyond MS-DOS.
Extending its abilities, also known as "modernizing". According to Microsoft OS/2 1.x with Presentation Manager was the "upgrade path" from DOS. For users stuck with legacy software they were going to add a comparable GUI to DOS called Windows. The Windows and Presentation Manager APIs were nearly identical, a convenience for developers as described by Microsoft. Windows was just temporary. Then the market ignored OS/2 1.x and stayed with DOS, Microsoft then reconsidered Windows and their partnership with IBM. I think we know how the story goes from there. The fact remains, for a little while, OS/2 1.x was the modern OS to replace DOS according to Microsoft.
Yeah, if that level of granularity was used every time there was a security vulnerability related to software that runs on Windows then it might be relevant.
Not to mention the fact that Ubuntu isn't linux: It's a linux distribution that expressly provides an entire software stack, including the software that got hacked here.
Don't be obtuse. "Linux" is most commonly used to refer to the complete server or desktop environment. When Linux fans are championing and encouraging people to switch their server or desktop to Linux they are referring to the entire environment not merely the kernel. Just as when Windows gets hacked and its something in the "software stack" and not the kernel itself, often something from a 3rd party not Microsoft. Matter of fact when the only "Linux" thing in an environment is the Linux kernel we tend not to call it "Linux" at all, for example Android. So don't start with this "Linux" only refers to the kernel nonsense, that is not how the word is used, and that includes within the Linux community.
Where the good assembly language programmer beats the compiler is usually not in the instruction scheduling. It is more typically leveraging knowledge that can not be communicated to the compiler about the desired implementation. High level programmers are constrained by their respective languages. The assembly language programmer is not so constrained.
Furthermore the architectural differences are sometimes a bogus complaint. For example where I last wrote non-trivial amounts of assembly was to get a few more fps out of a couple of video games. This was only necessary for low end machines, not the most recent architecture. If my optimizations were not needed on the more recent architectures that was not a problem, those systems were just fine performance wise. FYI I benchmarked this assembly code over several generations of x86 architecture over the years. While the performance benefits of assembly decreased over the generations it was always still a win, I did not have to update it. However when the original architecture that needed the assembly was no longer a supported target I turned off the assembly and let the C/C++ implementation get compiled so that maintenance would be easier (i.e. I would not be needed).
In the day of viable superoptimizers or superoptimizer-generated peephole optimizers and viable evolutionary/exploratory/search-based compilers, you should need to know assembly even less than ever before. Remember, the machine can try out new things both faster than you and cheaper than you.
Contrary to popular myth C/C++ code can be written in ways that favor one architecture over another. One reason to understand the architecture, assembly language, is to write better C/C++ code for your target.
People suck at macro optimizations in assembly. If your assembly program is more than 25 instructions, you're probably not going to do a better job than a compiler.
Good assembly language programmers are not trying to out-optimize the compiler at instruction scheduling. Good assembly language programmers are leveraging information that can not be communicated to the compiler.
Yes, most 64 bit chips also run 32 bit software, so no, there's no problem finding testing hardware, that's absurd.
Nobody cares if 32-bit ISOs work on 64-bit CPUs because those people are already using a 64-bit ISO. The 32-bit ISO has to be tested on ancient hardware like i686, which there's a shortage of.
No, 32-bit operating systems run just fine on 64-bit CPUs in the x86 world. Some people do so for improved performance. 64-bit sometimes has a performance hit.
In Latin, not so much in English. Yes, in English if the context is discipline in the Roman Legion, otherwise, 1 in 10 is a historical anachronism not a primary definition.
The author is not a very good writer. I believe the word he was looking for is annihilated. This is science. What words mean is important.
As politicians should leave science to the scientists, so should scientists leave the English language to the English major types and not try to misapply their love of metric prefixes like "deci".:-)
Slashdot Mirror
I'm quite skeptical of tariffs due to their history but some sort of reciprocal system seems to be needed. On a per nation basis low barriers to trade in both directions, high barriers to trade in both directions, but not low barriers in one direction and high barriers in the other. That is what really needs to be addressed, so its a little different than the historical protect a specific product/industry tariff. Its more of a protect an entire industrial base sort of thing.
And yet Trump keeps telling his rubes that jobs are coming back. They are not coming back. Workers will be replaced by robots.
Germany shows that it is not that bad, they have been heavily into automation for a while. Some different jobs are created, a lot of welders lost but a few electrical technicians added sort of thing. Economically speaking, robots or not the money is still being spent in the country and benefitting the country in various indirect ways. As opposed to outsourcing where all the indirect benefits go overseas.
How he's going to force companies to manufacture in the USA without adding legislation (because he said he would reduce legislation against corporations) is beyond me...
I think he said he would introduce some sort of tariff scheme when a trading partner's markets are not "open". It sort of sound like a reciprocal system, low barriers in both direction or high barriers in both directions, but no more low in one direction and high in the other. I'm not going to debate the wisdom of any tariff scheme but its clearly something quite independent of lower corporate taxes and lower regulations on corporations. If you are going to bash him at least get the criticisms correct.
It might not make sense at the present but you have to look at decades from now were most of the Western nations would have automated and will produce cheap crap themselves instead of importing them from China.
But it still doesn't make sense at present. And my bet is that in the future, those Western nations will develop that newer automation and then build and deploy it in China because that's where the world's industrial base will be.
Recent history has shown us that an industrial base is quite mobile. Provide manufacturers an incentive to move that is also approved/tolerated by consumers and the base will move. What may keep manufacturing in China is the "engineered" exchange rate that make everything available at a 25-30% discount.
It might not make sense at the present but you have to look at decades from now were most of the Western nations would have automated and will produce cheap crap themselves instead of importing them from China.
Automation won't keep manufacturing in China, maintaining parity with the west won't do it. Outsourcing is a royal PITA and there are many problems. There has to be a huge saving to offset the overhead and inefficiencies of outsourcing to make it economical. A wage gap was once part of the "savings" but that is shrinking and robots won't offer much savings either.
The other "savings" is basically that everything sold in China is at a 25-30% discount due to the "engineering" of the exchange rate. This may maintain outsourcing, not robots.
Really - three nearly identical posts ...
Apologies for communicating with three different individuals.
... and in all three, you seem almost desperate to have someone acknowledge that you are an insider with super-meaningful knowledge.
I am not an insider, nor have any special knowledge. It is quite well known that SOC has a wide latitude in gear selection. I merely saw a single instance of this well known practice. Apologies if your anti-military industrial complex meme or whatever failed. Perhaps there will be an F-35 post for you soon and you can find some joy.
The real difference is in the software.
Nope, the real difference is in the ability and willingness to navigate the military procurement process.
Not necessarily in this case, note "United States Army's Special Operations Command". They get a ton of say in gear. A friend's brother had a small company that made some photographic gear for the civilian market, SCUBA divers in particular, and SOC types got interested in it and the normal military procurement process was not involved. SOC can bypass the "military procurement process". They are not just war fighters, they are also involved in finding, developing and evaluating new gear and technology.
GP is right, All the vendors market to the guys wearing stars. If the general likes it then that's what we buy, doesn't matter what the grunts think.
Note "United States Army's Special Operations Command", that works entirely different. A friend's brother made some specialized photographic gear for the civilian market. SOC guys heard about it, visited, asked to evaluate it. They made some suggestions. These were incorporated into the design. They then told the guys wearing stars "we want this" and then "suits" got involved for the paperwork. Selection, evaluation and decision for this gear was made by "operators".
In my experience decisions like this are typically made because somebody high up likes their iPhone and doesn't want to have to learn how to use an Android phone. Sounds overly simplistic, but I've seen it happen too many times.
Bad guess. Note "United States Army's Special Operations Command", they get a lot of say in what equipment they use. A friend's brother made some unique camera equipment. SOC guys thought it interesting. The only people this small company every saw during evaluation were "operators". The "suits" did not get involved until the "operators" said "we want this". What you say may be true for normal military procurement, but its very different for SOC.
> Windows and Presentation Manager having a nearly identical API.
One major porting issue was that the screen co-ordinates were 'upside down'. Windows origin is upper-left corner. PM origin is lower-left.
That is not a major porting issue. That is a very minor thing, well, unless one writes some terrible code.
OS/2 PM had a much richer API but this was _later_ added to Windows.
By the way, I'm referring to Windows 3 not Windows 2.
> OS/2 was the upgrade path from DOS and the Windows was a temporary thing for people with legacy hardware/software
What kept people on Windows was Windows/386 which allowed multiple DOS-boxes to run the DOS software that they wanted to use. They could run WordPerfect 5.1 and Lotus123 at the same time along with the DOS accounting system.
OS/2 at the time only had a single 'penalty box'.
No. Users doing such things were so rare they are statistically insignificant. It was a classic chicken-and-egg thing. People didn't want to change apps, developers didn't want to port apps without a user base. DOS had the network effect advantage over OS/2. The only way to migrate users would have been to phase out DOS and ship OS/2 as the installed OS. But Microsoft changed their minds about partnering with IBM and about OS/2. Well, sort of, OS/2 NT, aka OS/2 3.0, was renamed Windows NT. So in a way DOS/Win3/Win9x was phased out and replaced by (OS/2)Windows NT.
According to Microsoft OS/2 1.x with Presentation Manager was the "upgrade path" from DOS. For users stuck with legacy software they were going to add a comparable GUI to DOS called Windows. Windows had already been available for nearly three years before Presentation Manager was released with OS/2 1.1, and was introduced only three months after the OS/2 development agreement between MS and IBM was signed.
None of that changes the fact that Microsoft was telling developers, including me, that OS/2 was the upgrade path from DOS and the Windows was a temporary thing for people with legacy hardware/software and that migrating them to OS/2 with Presentation Manager would be easy since porting your apps would be easy, Windows and Presentation Manager having a nearly identical API. As I said there was a window of time where Microsoft was telling people OS/2 was the replacement for DOS, Presentation Manager the replacement for Windows.
And I had seen Windows since v1 too. Prior to v3 I don't think it was used much beyond allowing multiple DOS sessions to run.
The PM API was designed to be similar to the Windows API, not the other way around, and still had some substantial differences.
Its not that simple. Windows v3 had a lot of changes and you can't measure these things by product announcements and release dates. OS/2 Presentation Manager may have been influencing Windows 3.0 behind the scenes.
Having similar APIs was helpful, but they were different enough to make a common code base impractical.
As someone who had common code bases for Windows and MacOS I'd say you are mistaken. I'm not sure what Microsoft was saying but it may have been more of a porting your app from DOS to OS/2 thing. Remember, DOS was the dead end, just maintenance, OS/2 was the future ... in that brief window of time.
I'm perfectly happy with your decision. However even if multitasking and multithreading were added they would not have to break backwards compatibility, they could merely extend the API. Legacy apps would not know or care and just run with a single thread. But this is just hypothetical, not a suggestion. If you had an itch to go in that direction I'd say create a FreeOS2 based on 1.x. Hell, there may be some commercial viability to such a project too.
... and damn you ;-) ... you are the main reason my better DOS and BIOS books, and the Pentium MMX 166, survive garage cleanings and take up valuable space.
Oh
Well, there was a "modern MS DOS", it was MS OS/2 1.x.
Except that OS/2 was a multitasking, protected mode operating system ...
That was part of what made it "modern".
... from IBM ...
And from Microsoft
... and MS-DOS wasn't any of those things.
OS/2 1.x was described by Microsoft as a modern OS designed to replace DOS.
OS/2 1.1, released just 11 months later, came with the promised Presentation Manager GUI, further extending its abilities beyond MS-DOS.
Extending its abilities, also known as "modernizing". According to Microsoft OS/2 1.x with Presentation Manager was the "upgrade path" from DOS. For users stuck with legacy software they were going to add a comparable GUI to DOS called Windows. The Windows and Presentation Manager APIs were nearly identical, a convenience for developers as described by Microsoft. Windows was just temporary. Then the market ignored OS/2 1.x and stayed with DOS, Microsoft then reconsidered Windows and their partnership with IBM. I think we know how the story goes from there. The fact remains, for a little while, OS/2 1.x was the modern OS to replace DOS according to Microsoft.
... before he decided there's just no such thing as "modern DOS" ...
Well, there was a "modern MS DOS", it was MS OS/2 1.x.
Yeah, if that level of granularity was used every time there was a security vulnerability related to software that runs on Windows then it might be relevant.
Not to mention the fact that Ubuntu isn't linux: It's a linux distribution that expressly provides an entire software stack, including the software that got hacked here.
Don't be obtuse. "Linux" is most commonly used to refer to the complete server or desktop environment. When Linux fans are championing and encouraging people to switch their server or desktop to Linux they are referring to the entire environment not merely the kernel. Just as when Windows gets hacked and its something in the "software stack" and not the kernel itself, often something from a 3rd party not Microsoft. Matter of fact when the only "Linux" thing in an environment is the Linux kernel we tend not to call it "Linux" at all, for example Android. So don't start with this "Linux" only refers to the kernel nonsense, that is not how the word is used, and that includes within the Linux community.
...they first cause comment syntax to twist their panties into a bunch.
Huh, I thought it was tabs vs spaces first, comment formatting second.
80 character lines.
I thought we all agreed to set our terminal windows to 120 characters?
...they first cause comment syntax to twist their panties into a bunch.
Huh, I thought it was tabs vs spaces first, comment formatting second.
Where the good assembly language programmer beats the compiler is usually not in the instruction scheduling. It is more typically leveraging knowledge that can not be communicated to the compiler about the desired implementation. High level programmers are constrained by their respective languages. The assembly language programmer is not so constrained.
Furthermore the architectural differences are sometimes a bogus complaint. For example where I last wrote non-trivial amounts of assembly was to get a few more fps out of a couple of video games. This was only necessary for low end machines, not the most recent architecture. If my optimizations were not needed on the more recent architectures that was not a problem, those systems were just fine performance wise. FYI I benchmarked this assembly code over several generations of x86 architecture over the years. While the performance benefits of assembly decreased over the generations it was always still a win, I did not have to update it. However when the original architecture that needed the assembly was no longer a supported target I turned off the assembly and let the C/C++ implementation get compiled so that maintenance would be easier (i.e. I would not be needed).
In the day of viable superoptimizers or superoptimizer-generated peephole optimizers and viable evolutionary/exploratory/search-based compilers, you should need to know assembly even less than ever before. Remember, the machine can try out new things both faster than you and cheaper than you.
Contrary to popular myth C/C++ code can be written in ways that favor one architecture over another. One reason to understand the architecture, assembly language, is to write better C/C++ code for your target.
And then there is debugging.
People suck at macro optimizations in assembly. If your assembly program is more than 25 instructions, you're probably not going to do a better job than a compiler.
Good assembly language programmers are not trying to out-optimize the compiler at instruction scheduling. Good assembly language programmers are leveraging information that can not be communicated to the compiler.
Able to run in 32-bit mode? Every single Intel and AMD CPU out there supports it.
Can't the same be said for 16-bit mode?
Why would anyone consider this to be "hard to find"?
The problem is they are testing a complete PC environment. So its not just the CPU, its that AGP 3dfx voodoo card and PCI sound blaster card too.
Yes, most 64 bit chips also run 32 bit software, so no, there's no problem finding testing hardware, that's absurd.
Nobody cares if 32-bit ISOs work on 64-bit CPUs because those people are already using a 64-bit ISO. The 32-bit ISO has to be tested on ancient hardware like i686, which there's a shortage of.
No, 32-bit operating systems run just fine on 64-bit CPUs in the x86 world. Some people do so for improved performance. 64-bit sometimes has a performance hit.
The majourity of ARMs sold are still 32 bit, and there is no reason for an embedded system to go 64 bit, usually.
Would not having a maintained OS be reason enough for them?
Maintained by who, the desktop and server distro maintainers or the embedded distro maintainers? They aren't necessarily the same groups.
So we can start calling magazines clips now?
Yes if you are using Latin. :-)
Decimated means to kill 1 in 10.
In Latin, not so much in English. Yes, in English if the context is discipline in the Roman Legion, otherwise, 1 in 10 is a historical anachronism not a primary definition.
The author is not a very good writer. I believe the word he was looking for is annihilated. This is science. What words mean is important.
As politicians should leave science to the scientists, so should scientists leave the English language to the English major types and not try to misapply their love of metric prefixes like "deci". :-)
Hah! No-one is gullible enough to still believe such nonsense are they?
The B-Ark seems unlikely to you?