You seem to think that poor progrmming is the problem. Poor management is the real problem. I can, in the right environment, write code with a defect density of 1 per 10 KSLOC. I've actually done that. But I can't do it on the project I'm on now, because the schedule is way to short to fit all the features that we committed to deliver. So quality goes down.
Before we hold programmers responsible for defects, let's hold program managers responsible. I wonder how many jobs that are currently bid at half what the developers say it will take to do the job would still get bid if the PM knew he could go to jail if things went south?
Ultimately, it is the fault of customers who think they can get good software cheap. I learned a long time ago that if you bid what the job will really cost, you are out of business. But if you bid 3/4 of the real cost, you can convince the customer to take "upgrades" as you go along. And you can tell him that he didn't really specify the feature that he really wants, but we can put it in for so much extra.
At any rate, the best success I've seen is to turn over detailed designs for offshore coders to implement, but even that can be of questionable quality, unless strict supervision is applied.
The problem you describe has to do with trying to ship a detailed design off to the "coders" to implement. Across town or across the world, coding needs to be done by the team that did the design. Unless you built a design that left the coders with no decisions to make, the coders will make decisions. Since the coders had nothing to do with the requirements or design process, they don't know your assumption. So their decisions will suck.
This is where the folks in the original article departed from reality. They think that the "low level" work can be done independently of the high-level work. In fact, the outsourcing that will succeed is the outsourcing of the entire project.
The pattern for moving entire professions overseas is to save some bucks by outsourcing the entry-level work, and live with the pain that comes from garbled communications. But it will become clear that you really need mid-level folks to guide that entry level. But guess what, the entry level workers who have been doing your grunt work are now experienced enough to be mid-level. Recurse twice and the industry is gone overseas completely.
SCO is in over their heads, they know it, but they are going to sling FUD for as long as they can. This is only an attempt to grab some headlines, and a pretty pathetic grab at that.
All of this will be moot soon after the Dec 5 oral arguments. SCO will have to present some evidence that IBM damaged them, or they get tossed out of court. Since the first is unlikely, the second is likely. Bye bye SCO.
Re:Cue Sun Java Desktop (madhatter)
on
Longhorn in 2006
·
· Score: 1
How can it be any cheaper or more readily available if the "5 year old technology" is already paid for and is working? If something is paid for and working, ANY upgrade costs money and time.
True for existing machines. Most shops repalce machine more often than that, however. Do you know anybody still running a 266MHz PIII in an office environment? That's what I had on my desk 5 years ago.
And every time an MS shop buys an office computer, they pay MS several hundred dollars. For large enterprises, that adds up.
The interesting thing is that MS is able to get monopoly prices on very mature (in terms of the market cycle, not defect density) software.
> But "real design" doesn't work, at least, not very well.
>Says who? Works pretty well for me. My team and I are generally *more* productive when we design *before* we code. Ever hear the saying, "Measure seven times, cut once"?
What we are missing here is the context. If you are working in a domain where you have already delivered 17 systems just like the one you are working on, then you can decide on the design up front. (As an aside, in that situation you should really just take a previous system and make it generic. That will pay off the third time you use the generic system.) But if you are in a domain that still has unknowns for your team, it is hard, (how hard depends on the density of the unknowns) to do a decent design (or spec!) up front. You will fill the unkowns with assumptions (sometimes without realizing it). And the assumptions are often wrong.
So in a (small!) document-heavy development, you can expect to spend at least 40% of your effort in planning, requirements, architecture, and detailed design. And still wind up needing 35% rework. One of the nasty traps of document-centric development is that you don't know which representation of the system is canonical. If the code doesn't match the spec, which one gets changed? In a perfect world the spec is right and the code should change. But I've seen too many cases where the coder realized that the designers really didn't understand this business process and asked for something impossible/stupid/silly. So you jump through some hoops to decide to change the spec, and then jump throuch more hoops to actually change the spec & architecture & design. Which costs about 10 times what the code cost. And it holds up progress on the code while the changes are approved.
> I believe that hacking is the best most
ideal way to code.
Then you don't know much about industrial-strength software development. Or XP. XP is a high discipline methodology. Test Driven Development (TDD) is a discipline. Merciless refactoring is a discipline. If you don't do them, you fail. Not requiring documents that (in the XP view) don't contribute as much as they cost does not equate to hacking.
When the f**k are managers going to stop hiring people who actually know more than themselves about programming? Well, it seems to me that most managers I know only do that by accident.
Sun money to swat the SCO mosquito:_____$299,996.50 Sad to say that Sun kicked in 2.5 million $ in the latest quarter, with another 2.5 mill due by November. I had really been hoping that their only involvement was that 300K$ back in January. NOPE.
If you add up all the license fees in the 10Q, it adds up to only Sun and MS have paid SCO anything.
I know a Doctor who is worried about getting laid off for the second time in 3 years.
The historical wisdom is that trade makes everyone better off. But this may very well be different. Consider what happens to the US and Eurozone economies when if the software, chip design, architecture, engineering design, architecture, financial back-office -- hell, all back-office jobs get off-shored in the next 10 to 15 years. We are looking at the destruction of the upper middle class. We will see severe downward pressure on wages, and not just the wages of those in trashed industries. The refugees from those areas will drive down wages across the board. Even Universities will take a hit. What students are going to pay current college costs for educations that prepare them to compete with folks in the developing world. Enrollments decline, profs get sacked, departments go dark.
Folks whose income gets cut in half often go bankrupt. The financial world is in for a huge storm.
Welcome to the Great Depression redux. Wages will stop falling when we all make Indian wages, even CEOs. India and China will be the technology powerhouses of the 21st Century, which will make them the military powerhouses.
gain a new skillset OK, what skillset do you suggest. It must meet the requirement that it cannot be done in India, China, Ireland, or Israel and have the results shipped back to the States by internet. Let's see Finance, going offshore... Engineering, going offshore... Architecture, going offshore...
Every dollar pays off 10-20 times on economic growth.
Sadly, this is not true. NASA is very good at claiming credit for all sorts of things. Do you think NASA developed Teflon? Nope, it was developed in the '30s or '40s. NASA only used it to treat the material in space suits. I once, in the course of a long thread on the benefits of NASA on usenet, took a close look at some of NASA's claims for technology development. They have an unfortunate tendency to give small - 20 or 30K$ grants to small businesses to investigate technologies that are already at the patent stage of development. Then they claim that they were influential in the development of those technologies. The technique is effective. I have even run across folks who think NASA was behind the development of the microprocessor. But I've never seen real numbers on the non-space payoff of the NASA budget. I will give them credit for the satelite industry, but I doubt that is a 10 or 20 times payoff. Besides, that technology is mature.
I agree with your opinion on outsourcing results, but not your conclusions. You get back crap from overseas, but if the Indians outsourced a project to the US, they would get crap back.
The problem is that you can't build a good system without access to the customer. I've been in software development going on 30 years, and I've never seen a spec that didn't have holes. And I've never seen a design that didn't have holes. If the coders are 12 time zones away from the designers/analysts/customers, then they are going to make things up to fill in the holes. Which means that 99 times in 100 the result is crap.
Alistair Cockburn has a very good book Agile Software Development that is about methodology, mostly. He says that he has never seen a methodology that works for outsourcing part of a project, like coding. He says that what will work is outsourcing whole projects, including architecture on down. This sheds some light on IBM's recent announcement that they will be moving high level jobs offshore in the coming years.
It seems to me that the best strategy to defend against frivolous suits is to be ready to countersue. If there were, say, 1000 registered copyrights within the Linux kernal, then SCO would be liable for $150,000,000 statutory damages (plus legal fees) every time someone downloaded Linux from their site after they broke the GPL. So if you contribute to OSS, once in a while pop for the copyright fee and register your work.
(the statutory damages are what the RIAA is claiming when they sue college students for billions)
the core of R&D will undoubtedly stay here in the US And why will R&D stay here? If India and China combined are graduating 300,000 engineers a year, and the US is graduating 60,000, don't you think that R&D is likely to wander offshore as well? I would be willing to bet that the most of the work that large corporations have moved to Asia is R&D. Like Intel setting up a compiler shop in Russia. It is harder to build bespoken apps offshore than packaged products. The apps need customer interface; products are usually driven by the organization's marketing folks assessment of what will sell. And packaged products are the D in R&D. But the ability to do D generates the resources and talent to do the R. And if there are enough Universities to graduate 300K engineers, there are lots of profs looking for research topics.
Of course the offshoring of the next domain will happen quicker that the offshoring of IT.
a straight EE Masters is awfully tempting as it is much harder to outsource those jobs say what!? Most EE's work in Asia. Even before outsourcing started (say in 1978) the average time that a BSEE worked as an EE was less than 10 years. EE and CS are both fields that shed folks as they get older. It is stupid to waste experience, but is is true that once you hit 40, it gets harder and harder to find a job.
BTW, I'm a CSer with 26 years experience. When I was new in the field, I could get a new job in 3 weeks. Now it takes a year or more
300K is the cost. Cost includes overhead -- office space (20$/sq ft/mo is a bargain in the DC area) lights, phone, computer, bandwidth, Social Security tax (employers pay ~7.5% of salary to the SSA)) and General and Administrative (AKA G&A, the money to pay the PHB and his PHB, and all the secretaries, guards, etc).
I've never seen a programmer get paid as much as half of what they cost. I once worked in a shop that charged 3.35 times salary, and only about 10% of that was profit.
Consider that Sun bought a 300K$ license in Feb, a month or more before SCO publicly went after IBM. Sun may be entirely innocent of anti-Linux actions. Or even intent.
300K$ is chump change in corporate circles. That's less than the cost of 2 programmers for a year.
If they got options that are now worth 2 million in the process, I'd say they are good businessfolk.
Only true for those who decide not to use Linux. I'd be willing to bet that at least half of the 9% are moving away from SCO. Who wants to business with a rattlesnake?
It's not my period, but I have read that the US contribution to WWI was small - that the main force of troops arrived about a year after the US declared war, and that their munitions were predominantly provided by the UK. The Germans made a major push in 1918 that was within 50 miles of Paris (poster were up advising evacuation of the city). This advance was stopped by 2 US divisions (1 army, 1 marines) at Chateau-Thierry. US troops were in the thick of Belleau Woods and the 2nd battle of the Marne. The French just didn't have anything left by the summer of 1918, so even 2 US divisions made a world of difference. On a grander scale, the entry of the US into the war forced Germany to take desperate measures, including stripping the best officers and ratings from the surface navy to send to the submarine force (where lots of them died) to try to intercept arriving US troops. The Germans had to win before Americans arrived in force. When the surface fleet was finally ordered to sea (they had been holed up in Kiel since the battle of Jutland), they knew they had little chance to do anything but die, so they revolted (never trust the Navy -- they did in the Czar, too) which in turn triggered civilian unrest. So the Kaiser had to sue for peace, even though the German army still held parts of France.
As to the Soviet Union, this seems unlikely. It was the successes of Germany on the Eastern Front which provoked the revolution, so removing the US from the war would not seem to change this. It sould seem so except that neither the Karinsky government nor the Bolsheviks managed to stop the Germans. The Germans were advancing into Russia even as they sought terms on the Western front. Had they won the war, they would not have left the Bolsheviks in power; they knew Lenin was dangerous.
If it wasn't for US foreign policy, Europe would be one big Germany. Almost true. The one Germany would have a Kaiser. And the Austro-Hungarian Empire would be intact. And the Ottoman Empire. Without US intervention, the Central Powers would have won WWI. So no WWII. No Soviet Union. And no Iraq.
If it wasn't for US foreign policy, there would be no "South" Korea. True. Had not the US given the northern half of Korea to the Soviets to entice them into war with Japan, Korea would be unified. Also true that if Dean Atchison hadn't made a speech that excluded Korea from the areas that the US pledged to defend, there would have been no Korean War.
Alternate histories are fun, but not very relevant. Would the European empires have survived if the Central Powers had won WWI? Not likely, the world was changing no matter what.
Sadly, the world is changing again. The US seems determined to turn itself into an empire. We have had a professional military for 29 years now. We assert, and exercise, the right of preemptive war. Worse than preemptive, because Iraq was never a threat to the US (until the Gulf War, it was our protoge). We went to war because the administration did not like the government of Iraq (and the oil, of course). We went to war because it suited us. Very Imperial.
I don't think there are any historical examples of Republics that became empires without the death of the Republic. I fear that that is what we are seeing in the US, the beginning of the death of the Republic.
I think it is C) waiting to see what cards SCO is holding. IBM has a lot to lose, and they are not going to do anything stupid. They have lots of experience with bet-the-company lawsuits.
I'd bet that IBM is spending its energy digging into who owns what in the Unix world (wasn't the BSD suit sealed?). Getting that nailed down is much more important to the future of the company than exchanging PR accusations with SCO.
There will be a period in this (and every) lawsuit called "discovery." During discovery, SCO gets to grill IBM about relevent matters, and IBM gets to grill SCO. At that point SCO has to show the infringing code. And then IBM will figure out where every line came from, who owns it, and then probably counter sue for tortuous interference with IBM's business (those letters to customers).
In this case it is SCO that has bet the company. I bet they lose.
First, IANAL. But I do know that there are trial lawyers who will take on, say, Big Tobacco, on a contingent fee, hoping for a big jackpot. So if the students of RPI were to get together and claim that each of them had been denied a service by the barratry of the RIAA, and it was worth maybe 5K$ to them. 5K$ per student times the student population is what you sue for. If there are a few more instances this blatent, then invoke the RICO statute and go for triple damages. Anybody know a trial lawyer with time on his hands?
Nonsense. Software is not math. Math is "for each program there exists (or could exist) a specification which makes the program correct." Not very useful. Software is human beings communicating with each other in ambigous natural languages and then trying to convert what they think they understand into a hyper specific computer language that a program (ie compiler) will translate into machine code. The hard part is trying to eliminate all the killer misunderstandings. One of the early Geminis came down several hundred miles from the planned spot because some programmer assumed that there were 24 hours in a day. Not in celestial navigation! Software is hard to do right.
Slashdot Mirror
You seem to think that poor progrmming is the problem. Poor management is the real problem. I can, in the right environment, write code with a defect density of 1 per 10 KSLOC. I've actually done that. But I can't do it on the project I'm on now, because the schedule is way to short to fit all the features that we committed to deliver. So quality goes down.
Before we hold programmers responsible for defects, let's hold program managers responsible. I wonder how many jobs that are currently bid at half what the developers say it will take to do the job would still get bid if the PM knew he could go to jail if things went south?
Ultimately, it is the fault of customers who think they can get good software cheap. I learned a long time ago that if you bid what the job will really cost, you are out of business. But if you bid 3/4 of the real cost, you can convince the customer to take "upgrades" as you go along. And you can tell him that he didn't really specify the feature that he really wants, but we can put it in for so much extra.
Great, now SCO is going to claim you copied that Perl out of System V, and take down /. on a DCMA claim.
At any rate, the best success I've seen is to turn over detailed designs for offshore coders to implement, but even that can be of questionable quality, unless strict supervision is applied.
The problem you describe has to do with trying to ship a detailed design off to the "coders" to implement. Across town or across the world, coding needs to be done by the team that did the design. Unless you built a design that left the coders with no decisions to make, the coders will make decisions. Since the coders had nothing to do with the requirements or design process, they don't know your assumption. So their decisions will suck.
This is where the folks in the original article departed from reality. They think that the "low level" work can be done independently of the high-level work. In fact, the outsourcing that will succeed is the outsourcing of the entire project.
The pattern for moving entire professions overseas is to save some bucks by outsourcing the entry-level work, and live with the pain that comes from garbled communications. But it will become clear that you really need mid-level folks to guide that entry level. But guess what, the entry level workers who have been doing your grunt work are now experienced enough to be mid-level. Recurse twice and the industry is gone overseas completely.
SCO is in over their heads, they know it, but they are going to sling FUD for as long as they can. This is only an attempt to grab some headlines, and a pretty pathetic grab at that.
All of this will be moot soon after the Dec 5 oral arguments. SCO will have to present some evidence that IBM damaged them, or they get tossed out of court. Since the first is unlikely, the second is likely. Bye bye SCO.
How can it be any cheaper or more readily available if the "5 year old technology" is already paid for and is working? If something is paid for and working, ANY upgrade costs money and time.
True for existing machines. Most shops repalce machine more often than that, however. Do you know anybody still running a 266MHz PIII in an office environment? That's what I had on my desk 5 years ago.
And every time an MS shop buys an office computer, they pay MS several hundred dollars. For large enterprises, that adds up.
The interesting thing is that MS is able to get monopoly prices on very mature (in terms of the market cycle, not defect density) software.
If your partner is really hot, then yes
I'm really hot, would you pair with me?
Really, really hot; sweating like a pig, and I need a shave. Come on, let's pair.
> But "real design" doesn't work, at least, not very well. >Says who? Works pretty well for me. My team and I are generally *more* productive when we design *before* we code. Ever hear the saying, "Measure seven times, cut once"?
What we are missing here is the context. If you are working in a domain where you have already delivered 17 systems just like the one you are working on, then you can decide on the design up front. (As an aside, in that situation you should really just take a previous system and make it generic. That will pay off the third time you use the generic system.)
But if you are in a domain that still has unknowns for your team, it is hard, (how hard depends on the density of the unknowns) to do a decent design (or spec!) up front. You will fill the unkowns with assumptions (sometimes without realizing it). And the assumptions are often wrong.
So in a (small!) document-heavy development, you can expect to spend at least 40% of your effort in planning, requirements, architecture, and detailed design. And still wind up needing 35% rework. One of the nasty traps of document-centric development is that you don't know which representation of the system is canonical. If the code doesn't match the spec, which one gets changed? In a perfect world the spec is right and the code should change. But I've seen too many cases where the coder realized that the designers really didn't understand this business process and asked for something impossible/stupid/silly. So you jump through some hoops to decide to change the spec, and then jump throuch more hoops to actually change the spec & architecture & design. Which costs about 10 times what the code cost. And it holds up progress on the code while the changes are approved.
> I believe that hacking is the best most ideal way to code.
Then you don't know much about industrial-strength software development. Or XP. XP is a high discipline methodology. Test Driven Development (TDD) is a discipline. Merciless refactoring is a discipline. If you don't do them, you fail. Not requiring documents that (in the XP view) don't contribute as much as they cost does not equate to hacking.
When the f**k are managers going to stop hiring people who actually know more than themselves about programming?
Well, it seems to me that most managers I know only do that by accident.
Sun money to swat the SCO mosquito:_____$299,996.50
Sad to say that Sun kicked in 2.5 million $ in the latest quarter, with another 2.5 mill due by November. I had really been hoping that their only involvement was that 300K$ back in January. NOPE.
If you add up all the license fees in the 10Q, it adds up to only Sun and MS have paid SCO anything.
I know a Doctor who is worried about getting laid off for the second time in 3 years.
The historical wisdom is that trade makes everyone better off. But this may very well be different. Consider what happens to the US and Eurozone economies when if the software, chip design, architecture, engineering design, architecture, financial back-office -- hell, all back-office jobs get off-shored in the next 10 to 15 years. We are looking at the destruction of the upper middle class. We will see severe downward pressure on wages, and not just the wages of those in trashed industries. The refugees from those areas will drive down wages across the board. Even Universities will take a hit. What students are going to pay current college costs for educations that prepare them to compete with folks in the developing world. Enrollments decline, profs get sacked, departments go dark.
Folks whose income gets cut in half often go bankrupt. The financial world is in for a huge storm.
Welcome to the Great Depression redux. Wages will stop falling when we all make Indian wages, even CEOs.
India and China will be the technology powerhouses of the 21st Century, which will make them the military powerhouses.
gain a new skillset OK, what skillset do you suggest. It must meet the requirement that it cannot be done in India, China, Ireland, or Israel and have the results shipped back to the States by internet.
Let's see Finance, going offshore... Engineering, going offshore... Architecture, going offshore...
See a pattern?
Every dollar pays off 10-20 times on economic growth. Sadly, this is not true. NASA is very good at claiming credit for all sorts of things. Do you think NASA developed Teflon? Nope, it was developed in the '30s or '40s. NASA only used it to treat the material in space suits. I once, in the course of a long thread on the benefits of NASA on usenet, took a close look at some of NASA's claims for technology development. They have an unfortunate tendency to give small - 20 or 30K$ grants to small businesses to investigate technologies that are already at the patent stage of development. Then they claim that they were influential in the development of those technologies.
The technique is effective. I have even run across folks who think NASA was behind the development of the microprocessor. But I've never seen real numbers on the non-space payoff of the NASA budget.
I will give them credit for the satelite industry, but I doubt that is a 10 or 20 times payoff. Besides, that technology is mature.
I agree with your opinion on outsourcing results, but not your conclusions. You get back crap from overseas, but if the Indians outsourced a project to the US, they would get crap back.
The problem is that you can't build a good system without access to the customer. I've been in software development going on 30 years, and I've never seen a spec that didn't have holes. And I've never seen a design that didn't have holes. If the coders are 12 time zones away from the designers/analysts/customers, then they are going to make things up to fill in the holes. Which means that 99 times in 100 the result is crap.
Alistair Cockburn has a very good book Agile Software Development that is about methodology, mostly. He says that he has never seen a methodology that works for outsourcing part of a project, like coding. He says that what will work is outsourcing whole projects, including architecture on down. This sheds some light on IBM's recent announcement that they will be moving high level jobs offshore in the coming years.
It seems to me that the best strategy to defend against frivolous suits is to be ready to countersue. If there were, say, 1000 registered copyrights within the Linux kernal, then SCO would be liable for $150,000,000 statutory damages (plus legal fees) every time someone downloaded Linux from their site after they broke the GPL.
So if you contribute to OSS, once in a while pop for the copyright fee and register your work.
(the statutory damages are what the RIAA is claiming when they sue college students for billions)
the core of R&D will undoubtedly stay here in the US
And why will R&D stay here? If India and China combined are graduating 300,000 engineers a year, and the US is graduating 60,000, don't you think that R&D is likely to wander offshore as well? I would be willing to bet that the most of the work that large corporations have moved to Asia is R&D. Like Intel setting up a compiler shop in Russia. It is harder to build bespoken apps offshore than packaged products. The apps need customer interface; products are usually driven by the organization's marketing folks assessment of what will sell. And packaged products are the D in R&D. But the ability to do D generates the resources and talent to do the R. And if there are enough Universities to graduate 300K engineers, there are lots of profs looking for research topics.
Of course the offshoring of the next domain will happen quicker that the offshoring of IT.
a straight EE Masters is awfully tempting as it is much harder to outsource those jobs
say what!? Most EE's work in Asia. Even before outsourcing started (say in 1978) the average time that a BSEE worked as an EE was less than 10 years.
EE and CS are both fields that shed folks as they get older. It is stupid to waste experience, but is is true that once you hit 40, it gets harder and harder to find a job.
BTW, I'm a CSer with 26 years experience. When I was new in the field, I could get a new job in 3 weeks. Now it takes a year or more
300K is the cost. Cost includes overhead -- office space (20$/sq ft/mo is a bargain in the DC area) lights, phone, computer, bandwidth, Social Security tax (employers pay ~7.5% of salary to the SSA)) and General and Administrative (AKA G&A, the money to pay the PHB and his PHB, and all the secretaries, guards, etc).
I've never seen a programmer get paid as much as half of what they cost. I once worked in a shop that charged 3.35 times salary, and only about 10% of that was profit.
Consider that Sun bought a 300K$ license in Feb, a month or more before SCO publicly went after IBM. Sun may be entirely innocent of anti-Linux actions. Or even intent.
300K$ is chump change in corporate circles. That's less than the cost of 2 programmers for a year.
If they got options that are now worth 2 million in the process, I'd say they are good businessfolk.
Only true for those who decide not to use Linux. I'd be willing to bet that at least half of the 9% are moving away from SCO. Who wants to business with a rattlesnake?
It's not my period, but I have read that the US contribution to WWI was small - that the main force of troops arrived about a year after the US declared war, and that their munitions were predominantly provided by the UK.
The Germans made a major push in 1918 that was within 50 miles of Paris (poster were up advising evacuation of the city). This advance was stopped by 2 US divisions (1 army, 1 marines) at Chateau-Thierry. US troops were in the thick of Belleau Woods and the 2nd battle of the Marne. The French just didn't have anything left by the summer of 1918, so even 2 US divisions made a world of difference.
On a grander scale, the entry of the US into the war forced Germany to take desperate measures, including stripping the best officers and ratings from the surface navy to send to the submarine force (where lots of them died) to try to intercept arriving US troops. The Germans had to win before Americans arrived in force. When the surface fleet was finally ordered to sea (they had been holed up in Kiel since the battle of Jutland), they knew they had little chance to do anything but die, so they revolted (never trust the Navy -- they did in the Czar, too) which in turn triggered civilian unrest. So the Kaiser had to sue for peace, even though the German army still held parts of France.
As to the Soviet Union, this seems unlikely. It was the successes of Germany on the Eastern Front which provoked the revolution, so removing the US from the war would not seem to change this. It sould seem so except that neither the Karinsky government nor the Bolsheviks managed to stop the Germans. The Germans were advancing into Russia even as they sought terms on the Western front. Had they won the war, they would not have left the Bolsheviks in power; they knew Lenin was dangerous.
If it wasn't for US foreign policy, Europe would be one big Germany.
Almost true. The one Germany would have a Kaiser. And the Austro-Hungarian Empire would be intact. And the Ottoman Empire. Without US intervention, the Central Powers would have won WWI. So no WWII. No Soviet Union. And no Iraq.
If it wasn't for US foreign policy, there would be no "South" Korea.
True. Had not the US given the northern half of Korea to the Soviets to entice them into war with Japan, Korea would be unified. Also true that if Dean Atchison hadn't made a speech that excluded Korea from the areas that the US pledged to defend, there would have been no Korean War.
Alternate histories are fun, but not very relevant. Would the European empires have survived if the Central Powers had won WWI? Not likely, the world was changing no matter what.
Sadly, the world is changing again. The US seems determined to turn itself into an empire. We have had a professional military for 29 years now. We assert, and exercise, the right of preemptive war. Worse than preemptive, because Iraq was never a threat to the US (until the Gulf War, it was our protoge). We went to war because the administration did not like the government of Iraq (and the oil, of course). We went to war because it suited us. Very Imperial.
I don't think there are any historical examples of Republics that became empires without the death of the Republic. I fear that that is what we are seeing in the US, the beginning of the death of the Republic.
Sure, the game is "LINUX SYSADMIN!"
It's a game! It's training! It's a Trojan!
I think it is C) waiting to see what cards SCO is holding. IBM has a lot to lose, and they are not going to do anything stupid. They have lots of experience with bet-the-company lawsuits.
I'd bet that IBM is spending its energy digging into who owns what in the Unix world (wasn't the BSD suit sealed?). Getting that nailed down is much more important to the future of the company than exchanging PR accusations with SCO.
There will be a period in this (and every) lawsuit called "discovery." During discovery, SCO gets to grill IBM about relevent matters, and IBM gets to grill SCO. At that point SCO has to show the infringing code. And then IBM will figure out where every line came from, who owns it, and then probably counter sue for tortuous interference with IBM's business (those letters to customers).
In this case it is SCO that has bet the company. I bet they lose.
IANAL
First, IANAL. But I do know that there are trial lawyers who will take on, say, Big Tobacco, on a contingent fee, hoping for a big jackpot.
So if the students of RPI were to get together and claim that each of them had been denied a service by the barratry of the RIAA, and it was worth maybe 5K$ to them. 5K$ per student times the student population is what you sue for.
If there are a few more instances this blatent, then invoke the RICO statute and go for triple damages.
Anybody know a trial lawyer with time on his hands?
Nonsense. Software is not math. Math is "for each program there exists (or could exist) a specification which makes the program correct." Not very useful.
Software is human beings communicating with each other in ambigous natural languages and then trying to convert what they think they understand into a hyper specific computer language that a program (ie compiler) will translate into machine code.
The hard part is trying to eliminate all the killer misunderstandings. One of the early Geminis came down several hundred miles from the planned spot because some programmer assumed that there were 24 hours in a day. Not in celestial navigation!
Software is hard to do right.
Did you also notice that the PDF is dated 2000, and it says she is no longer an employee?