Cost to build a solar thermal plant is currently $2 to $3 per watt, resulting in a cost of 9 to 12 cents per kWh delivered. Reference. That's in contrast to 2.5 to 5 cents per kWh for gas or coal plants. Expected advances (within the next few decades) should bring the solar cost down to 4 to 5 cents.
You said that system inefficiencies (batteries, inverters, etc.) are not applicable to solar thermal plants. How so? Thermal storage requires energy transfer, and converting heat to electricity isn't 100% efficient. You'll also need inverters and step-up transformers to get any generated electricity to the grid. The literature isn't terribly clear on whether "transmission line losses" include the losses you incur in converting your source's power stream into alternating current at whatever frequency and voltage the grid is expecting.
I didn't want to imply that solar plants (photovoltaic or solar thermal) should be dismissed out of hand. I was more laughing at the idea of covering the entire state of Arizona with solar cells. I'll will admit that I did get a little carried away.
I think solar can become an important part of our energy needs, but I doubt that it will become the major source or even a primary source in the near future.
It's unfortunate that I fucked up the math by a factor of 1,000 (the hazards of using a calculator rather than writing the stuff down). Fortunately I screwed up uniformly, so the cost numbers match the generation numbers.
The system we're talking about, using my wildly optimistic numbers and covering 5% of the state (340 million kWh) is roughly equivalent to a 15,000 Megawatt power plant. That's about four times as much power as is generated by the Palo Verde nuclear plant west of Phoenix.
In any event, even covering 5% of Arizona (that'd be 5,700 square miles, probably about the size of the military training ground there between Gila Bend and Yuma) would have some serious environmental effects.
I can see it now...
Protestor: "No Nukes!"
Advocate: "Would you rather build four nuclear plants, or permanently destroy 5,700 square miles of pristine desert by covering it with mirrors?"
Egads, where'd my math go? Sometimes I want to delete and start over....
At 17% efficiency, 2 trillion kWh becomes 340 billion kWh. 25% system loss brings that to 255 billion kWh. At 100 kWh per day per household, that'd be 2.5 billion households. Okay, so you could generate enough to power every house in the country about 2,500 times.
I don't know the power requirements of industry. Is it 2,500 times that of personal requirements?
Still, that'd be a hell of an expensive system: $750 trillion? ouch!
The average daily sunlight hitting the surface of the Earth in most of Arizona is about 6.1 kWh/m^2 (kilowatt hours per meter squared). In the extreme southwest, it's about 6.7. See this article [it's a PDF] for full details.
If you use an average of 6.5 kWh/m^2 for the whole state (it's high, but it'll do for these purposes), you end up with a potential of almost 2 trillion kWh of energy produced every day. Supporting structure will cut that down a bit, of course, but let's use the 2 trillion number.
There are several problems that we'll have to overcome before solar energy can be used for electricity on a large scale:
Today's high efficiency solar cells can convert about 17% of the energy that hits them to electricity. That 2 trillion kWh becomes 3.4 million kWh in a hurry.
Power is generated only during daylight hours. When the sun goes down, the lights go out. Some means of energy storage (batteries, etc.) must be implemented in order to keep the lights on.
System inefficiencies can decrease the amount of available power by as much as 50%. Battery storage and power inverters aren't terribly efficient. I'll be optimistic and figure that we can cap the system losses at 25%, leaving us with 255 million kWh delivered to the transmission lines.
A good BOE number for household energy consumption is 100 kWh per day. So using my best case estimates above, and assuming no transmission line losses (which usually are around 30 to 40 percent and would be more if the transmission lines reached across the entire country), covering the entire state of Arizona with solar cells would provide electricity to 2.55 million households. You couldn't even power California.
At retail, the cost of photovoltaic modules is about $5 per watt. The literature doesn't really say if that's per watt delivered, or per watt generated (i.e. before system losses or after). If we assume that it's after system losses and that government could work a miracle and actually pay less than retail (say $1 per watt) then for our fictional 255 million kilowatt system, that'd be a paltry $255 billion for the solar cells.
The photovoltaic cells make up only 25 to 50 percent of the entire cost of a system. Taking that into consideration, cost of the entire system would be between $750 billion and $1 trillion.
Manufacturing photovoltaic cells involves the use of many hazardous chemicals (mostly the same as used by the semiconductor industry).
Energy storage systems have many toxic materials, are prone to leakage, have limited duty life, and are expensive to dispose of safely.
Manufacturing photovoltaics requires a lot of energy. Payback time (i.e. the cell generating as much energy as it cost to produce) is from six months to ten years, depending on the cell's efficiency and where it's deployed.
Photovoltaics have a limited lifecycle, and become less efficient as they get older. The entire array would have to be replaced in 20 years or less. Batteries would have to be replaced on a regular basis, too. I'd figure an annual reserve for replacement of $50 to $75 billion.
You'd need an army of people or one damned impressive machine to clean the faces of the modules in order to prevent accumulated dirt from further degrading system efficiency.
You can't discount the environmental effects of permanently depriving 114,000 square miles of sunlight.
I'm not saying that it's impossible, but it doesn't look too terribly practical today, or in the near future.
So what you're saying is that Microsoft's IE team is exhibiting technical leadership by implementing procedures that improve the quality of their product? And they're doing this before any of the open source development teams?
Wordperfect - barely breathing Quattro Pro - dead Paradox - dead DR-DOS - dead?
Novell, a company whose mission for the past 15 years seems to have been "Buy Microsoft's competition and run it into the ground" has purchased one of the few Linux desktops that could potentially give Microsoft a run for its money.
Yeah, and that's especially annoying when software developers aren't aware of SHGetFolderLocation [microsoft.com] and hard code directory names like "Program Files".:-P
Of course it's Microsoft's fault that some clueless software developer failed to read the documentation and hard coded something like "C:\Program files". A programmer doing something similarly stupid on any other system, of course, is an idiot. But on Windows it's the operating system's fault.
Bull. If this were being done by some research laboratories funded by IBM or AT&T people would be very interested to see what ideas they were trying out and what results they were getting.
The primary point being that they would be getting results.
Has anybody here, other than the reviewer, actually read the book? The reviewer says:
Blunden [the author] makes sure to cover every topic related to virtual machines in extreme depth.
Fine, that I could believe...maybe...in a 650-page book. But then, he goes on to say:
Initially he starts slowly, and introduces the reader to how CPUs work, how memory works, how paging works, and how almost any other system process you can imagine works. Nothing is missed out. Multitasking, threads, processes, porting.. he covers it all.
All in just 2 chapters? Furthermore:
...a number of advanced topics (garbage collection, memory management, assembler construction, paging, token parsing) are dealt with in a very easy to understand way.
Followed by a complete VM implementation, with assembler and debugger, including clear and pertinent code examples in C/C++? All in a 650-page book? Maybe the writing is really, really small?
This book must be a tour de force of clear, concise writing. Either that, or the reviewer's ability to discriminate between hand waving and thorough discussion of a topic is limited, at best.
If this book really is as good as the reviewer says, then I absolutely must get it. But, somehow, it sounds too good to be true.
Don't be too quick to judge MIT, as they probably outsource the fulfillment.
Also, don't be too sure it's the software that's the problem. Many data entry systems allow the operator to input the ZIP code. The city name is displayed either in an edit box or a drop-down list where the operator is expected to verify it, and change it if not correct. This could be (and likely is) a simple case of operator error.
(Imagine that coming from me, who thinks most software sucks rocks.)
2) Write a book that agent sells to publisher for 10% royalty. Publisher sells book to stores for $35.00, of which you receive 10%, or $3.50. Your agent takes 10% off of that, leaving you with $3.15 per book sold.
You need to sell lots of copies to get rich writing a book.
I was on the team that created Jack Nicklaus Golf 4 for Accolade. When the game was released in the fall of 1996, it had the most realistic physics of any golf simulation. That physics engine lasted through Jack 5 and Jack 6--perhaps longer...I haven't kept up with the game.
Our physics guy wasn't a physicist, but rather an engineer with a very strong physics and math background. He spent months studying the physics of golf and building a simulation system to test it long before we had a game that he could plug it into. Creating the physics for that game was quite an education for him, and for me because we'd talk about the issues that come up. As somebody else in this thread mentioned, the equations get you most of the way there, but the computer simulation is working in discrete time steps, and the equations break down at some point. Then you have to fudge it.
If all you're doing is coding, then since that's all logic, physics should fit right in there with it.
In my experience, it takes a very good programmer to understand and simulate a physical system. Most programmers try to fudge it from the start rather than apply the physical equations. The result is typically very convoluted code with lots of exceptions and an unrealistic system. There's a huge difference between implementation logic (programming) and a consistent and logical overall systems design.
The Kaypro 10, if I remember correctly, was a 10 MHz 80286. An IBM AT clone. This was my first foray into PC-compatibles. Before the AT came out, my 8 MHz Z-80 running TurboDOS blew away any of the PClones.
I think you meant the Kaypro II--a blue steel box with a handle and fold-up keyboard. This box came out about 6 months after the Osborne I.
I still have my Osborne I, which I purchased for $1795 back in December of 1981. I pulled it out of the closet the other day just to marvel at how far we've come in 20 years. My little Kyocera phone/PDA (20 MHz, 8 MB) is five times as fast and has 128 times the memory. In fact, the Kyocera has enough RAM to hold the contents of 80 Osborne I diskettes.
Those headers are only reliable if nobody futzs with them. And futzing with them (or deleting them) is trivial. It's also trivial to spoof your IP address. I could create an email message here on my machine, add a couple of headers that say it was routed through Microsoft's and Sun's SMTP servers, and deliver it directly to your ISP's SMTP server with a spoofed IP that looks like I'm sending it from slashdot's IP address. Your IPS's server might attempt a reverse DNS lookup (probably wouldn't), and even if it failed, it'd probably just mark the thing as "unverified." There would be no way to trace that message back to me.
You assume that the engine is all of the code in a game. This may be true (although I doubt it) of a FPS game, but in many other types of games, the "engine" either doesn't exist or is a relatively minor part of the game. In a golf simulation, for example, the rendering engine makes up less than 20% of all the code. The rest is made up of user interface, physics, and rules. For a puzzle game, it's even more skewed away from the rendering engine.
Re:Cheap, efficient power vs. the A-bomb
on
Fission in a Box
·
· Score: 1
Southern California Edison is already testing microturbines: small, clean-burning natural-gas power plants that cost a relatively inexpensive $25,000 to $200,000 to build, depending on size.
Sorry, but that $25,000 figure just feels wrong. You couldn't buy the land to put the thing on for $25,000. I might believe $250,000 to $2,000,000.
In a free and stable society filled with reasonable people who are a) able to take criticism, constructive or not and b) willing to accept the consequences of their actions, then anonymous speech isn't required.
So much for the little guy being able to have a web-streaming radio station.
Huh? Nothing's preventing the "little guy" from having a web-streaming radio station. Nothing in the corporations' or the union's actions prevents somebody else from broadcasting.
Unless you meant listening to the station. Oh well. Radio's free anyway--paid for through advertising by the same evil corporations everybody claims is fleecing the public.
The moderator was right. Funny. Hardly factual in most cases. For example:
Basic everywhere. At time of smalltalk and lisp, they pushed basic. Thanks for that. They ruined my life.
It was a lot easier to get a Basic interpreter into a 4K ROM than a Lisp interpreter or Smalltalk environment. Basic is what people wanted--it was a lot closer to FORTRAN than either of the others.
The paperclip. This harmed millions of users.
Mentioning that paperclip makes you look as stupid as it is. It's just stupid and annoying...not harmful. Everybody knows it's stupid and knows how to turn it off. It would have been forgotten by now except MS bashers like to parade it out just to make themselves look stupid.
And FAT, the Fragmented Allocation Table ? Who should pay for the countless hours morons spend looking DEFRAG.EXE painfully moving blocks around ?
The FAT existed in CP/M, and probably before that. It was very effective and efficient for the diskettes and miniscule hard drives at the time. MS's only fault here is perhaps not providing an alternative in 1987 or so when 40 MB drives started showing FAT's limitations.
And the windows API?
It's hardly the worst. Certainly no worse in general than the glibc API, or the C standard library API for that matter. There are bone-headed design decisions in every one of them.
Has anyone else tried to actually buy a copy of Windows without buying a computer system? I don't mean an upgrade package, which requires that you already own Windows. MS will not sell the average consumer an original license for Windows. You can only purchase such a license from an OEM with a computer system.
Wrong! I was in Best Buy yesterday, and saw the following:
Windows 2000 Professional: $299. No pre-existing software requirements.
Windows 2000 Professional Upgrade: $199. Requires Windows 95/98/ME/NT 3.51 or later.
Actually, King has stated that the first three installments will be $1 each, and the fourth through eighth will be $2 each. Any remaining installments will be free. So the total you'll pay is $13.00. That's about half of what you'd pay for the book if it was printed and distributed through a publisher.
And of course nobody would actually pay money to support free software, either, because there is a huge nebulous "them" out there writing it for free, and somebody else is paying for the servers and bandwidth charges.
The first release of Kylix is the Delphi 5 IDE with a new cross-platform component library called CLX ("clicks"). It is very cool. I've seen non-trivial Windows/VCL apps ported from Delphi 5 to Kylix in very short periods of time (a couple of hours to a couple of days, depending on the app).
There's no real difference between a programmer with no real Linux experience doing Kylix and a programmer with no real Windows experience doing Delphi. In both cases they're junior programmers who learn from the more experienced. The senior programmers, on the other hand, will have to rely on their C knowledge in order to glean useful information from the vast amount of C/C++ code that's available for Linux. This is no different than the early days of Borland Pascal for Windows and (later) Delphi.
Re-read the article again. I don't think there was a 'series of "guilties"'.
Microsoft hailed the jury's July 16, 1999, verdict, which found in the company's favor on nearly every claim except the unfair-business-practices charge.
Slashdot Mirror
Cost to build a solar thermal plant is currently $2 to $3 per watt, resulting in a cost of 9 to 12 cents per kWh delivered. Reference. That's in contrast to 2.5 to 5 cents per kWh for gas or coal plants. Expected advances (within the next few decades) should bring the solar cost down to 4 to 5 cents.
You said that system inefficiencies (batteries, inverters, etc.) are not applicable to solar thermal plants. How so? Thermal storage requires energy transfer, and converting heat to electricity isn't 100% efficient. You'll also need inverters and step-up transformers to get any generated electricity to the grid. The literature isn't terribly clear on whether "transmission line losses" include the losses you incur in converting your source's power stream into alternating current at whatever frequency and voltage the grid is expecting.
I didn't want to imply that solar plants (photovoltaic or solar thermal) should be dismissed out of hand. I was more laughing at the idea of covering the entire state of Arizona with solar cells. I'll will admit that I did get a little carried away.
I think solar can become an important part of our energy needs, but I doubt that it will become the major source or even a primary source in the near future.
It's unfortunate that I fucked up the math by a factor of 1,000 (the hazards of using a calculator rather than writing the stuff down). Fortunately I screwed up uniformly, so the cost numbers match the generation numbers.
The system we're talking about, using my wildly optimistic numbers and covering 5% of the state (340 million kWh) is roughly equivalent to a 15,000 Megawatt power plant. That's about four times as much power as is generated by the Palo Verde nuclear plant west of Phoenix.
In any event, even covering 5% of Arizona (that'd be 5,700 square miles, probably about the size of the military training ground there between Gila Bend and Yuma) would have some serious environmental effects.
I can see it now...
Protestor: "No Nukes!"
Advocate: "Would you rather build four nuclear plants, or permanently destroy 5,700 square miles of pristine desert by covering it with mirrors?"
Egads, where'd my math go? Sometimes I want to delete and start over....
At 17% efficiency, 2 trillion kWh becomes 340 billion kWh. 25% system loss brings that to 255 billion kWh. At 100 kWh per day per household, that'd be 2.5 billion households. Okay, so you could generate enough to power every house in the country about 2,500 times.
I don't know the power requirements of industry. Is it 2,500 times that of personal requirements?
Still, that'd be a hell of an expensive system: $750 trillion? ouch!
If you use an average of 6.5 kWh/m^2 for the whole state (it's high, but it'll do for these purposes), you end up with a potential of almost 2 trillion kWh of energy produced every day. Supporting structure will cut that down a bit, of course, but let's use the 2 trillion number.
There are several problems that we'll have to overcome before solar energy can be used for electricity on a large scale:
I'm not saying that it's impossible, but it doesn't look too terribly practical today, or in the near future.
So what you're saying is that Microsoft's IE team is exhibiting technical leadership by implementing procedures that improve the quality of their product? And they're doing this before any of the open source development teams?
Say it isn't so!
Let's recap some of Novell's previous purchases:
Wordperfect - barely breathing
Quattro Pro - dead
Paradox - dead
DR-DOS - dead?
Novell, a company whose mission for the past 15 years seems to have been "Buy Microsoft's competition and run it into the ground" has purchased one of the few Linux desktops that could potentially give Microsoft a run for its money.
Might as well cede the desktop to Microsoft.
According to my trusty American Heritage Dictionary:
re*dun*dant: 1. Exceeding what is necessary or natural; superfluous. 2. Needlessly repetitive; verbose.
So both "a=a" and dead code are properly characterized as redundant.
Informative? Save us from clueless moderators.
:-P
Yeah, and that's especially annoying when software developers aren't aware of SHGetFolderLocation [microsoft.com] and hard code directory names like "Program Files".
Of course it's Microsoft's fault that some clueless software developer failed to read the documentation and hard coded something like "C:\Program files". A programmer doing something similarly stupid on any other system, of course, is an idiot. But on Windows it's the operating system's fault.
Bull. If this were being done by some research laboratories funded by IBM or AT&T people would be very interested to see what ideas they were trying out and what results they were getting.
The primary point being that they would be getting results.
Has anybody here, other than the reviewer, actually read the book? The reviewer says:
...a number of advanced topics (garbage collection, memory management, assembler construction, paging, token parsing) are dealt with in a very easy to understand way.
Blunden [the author] makes sure to cover every topic related to virtual machines in extreme depth.
Fine, that I could believe...maybe...in a 650-page book. But then, he goes on to say:
Initially he starts slowly, and introduces the reader to how CPUs work, how memory works, how paging works, and how almost any other system process you can imagine works. Nothing is missed out. Multitasking, threads, processes, porting.. he covers it all.
All in just 2 chapters? Furthermore:
Followed by a complete VM implementation, with assembler and debugger, including clear and pertinent code examples in C/C++? All in a 650-page book? Maybe the writing is really, really small?
This book must be a tour de force of clear, concise writing. Either that, or the reviewer's ability to discriminate between hand waving and thorough discussion of a topic is limited, at best.
If this book really is as good as the reviewer says, then I absolutely must get it. But, somehow, it sounds too good to be true.
Don't be too quick to judge MIT, as they probably outsource the fulfillment.
Also, don't be too sure it's the software that's the problem. Many data entry systems allow the operator to input the ZIP code. The city name is displayed either in an edit box or a drop-down list where the operator is expected to verify it, and change it if not correct. This could be (and likely is) a simple case of operator error.
(Imagine that coming from me, who thinks most software sucks rocks.)
Actually, it's:
2) Write a book that agent sells to publisher for 10% royalty. Publisher sells book to stores for $35.00, of which you receive 10%, or $3.50. Your agent takes 10% off of that, leaving you with $3.15 per book sold.
You need to sell lots of copies to get rich writing a book.
I was on the team that created Jack Nicklaus Golf 4 for Accolade. When the game was released in the fall of 1996, it had the most realistic physics of any golf simulation. That physics engine lasted through Jack 5 and Jack 6--perhaps longer...I haven't kept up with the game.
Our physics guy wasn't a physicist, but rather an engineer with a very strong physics and math background. He spent months studying the physics of golf and building a simulation system to test it long before we had a game that he could plug it into. Creating the physics for that game was quite an education for him, and for me because we'd talk about the issues that come up. As somebody else in this thread mentioned, the equations get you most of the way there, but the computer simulation is working in discrete time steps, and the equations break down at some point. Then you have to fudge it.
If all you're doing is coding, then since that's all logic, physics should fit right in there with it.
In my experience, it takes a very good programmer to understand and simulate a physical system. Most programmers try to fudge it from the start rather than apply the physical equations. The result is typically very convoluted code with lots of exceptions and an unrealistic system. There's a huge difference between implementation logic (programming) and a consistent and logical overall systems design.
The Kaypro 10, if I remember correctly, was a 10 MHz 80286. An IBM AT clone. This was my first foray into PC-compatibles. Before the AT came out, my 8 MHz Z-80 running TurboDOS blew away any of the PClones.
I think you meant the Kaypro II--a blue steel box with a handle and fold-up keyboard. This box came out about 6 months after the Osborne I.
I still have my Osborne I, which I purchased for $1795 back in December of 1981. I pulled it out of the closet the other day just to marvel at how far we've come in 20 years. My little Kyocera phone/PDA (20 MHz, 8 MB) is five times as fast and has 128 times the memory. In fact, the Kyocera has enough RAM to hold the contents of 80 Osborne I diskettes.
Those headers are only reliable if nobody futzs with them. And futzing with them (or deleting them) is trivial. It's also trivial to spoof your IP address. I could create an email message here on my machine, add a couple of headers that say it was routed through Microsoft's and Sun's SMTP servers, and deliver it directly to your ISP's SMTP server with a spoofed IP that looks like I'm sending it from slashdot's IP address. Your IPS's server might attempt a reverse DNS lookup (probably wouldn't), and even if it failed, it'd probably just mark the thing as "unverified." There would be no way to trace that message back to me.
You assume that the engine is all of the code in a game. This may be true (although I doubt it) of a FPS game, but in many other types of games, the "engine" either doesn't exist or is a relatively minor part of the game. In a golf simulation, for example, the rendering engine makes up less than 20% of all the code. The rest is made up of user interface, physics, and rules. For a puzzle game, it's even more skewed away from the rendering engine.
Southern California Edison is already testing microturbines: small, clean-burning natural-gas power plants that cost a relatively inexpensive $25,000 to $200,000 to build, depending on size.
Sorry, but that $25,000 figure just feels wrong. You couldn't buy the land to put the thing on for $25,000. I might believe $250,000 to $2,000,000.
What's the source of your information?
The Tao of Programming: http://www.softpanorama.org/Bulletin/Humor/tao_of_ programming.shtml
In a free and stable society filled with reasonable people who are a) able to take criticism, constructive or not and b) willing to accept the consequences of their actions, then anonymous speech isn't required.
So much for the little guy being able to have a web-streaming radio station.
Huh? Nothing's preventing the "little guy" from having a web-streaming radio station. Nothing in the corporations' or the union's actions prevents somebody else from broadcasting.
Unless you meant listening to the station. Oh well. Radio's free anyway--paid for through advertising by the same evil corporations everybody claims is fleecing the public.
The moderator was right. Funny. Hardly factual in most cases. For example:
Basic everywhere. At time of smalltalk and lisp, they pushed basic. Thanks for that. They ruined my life.
It was a lot easier to get a Basic interpreter into a 4K ROM than a Lisp interpreter or Smalltalk environment. Basic is what people wanted--it was a lot closer to FORTRAN than either of the others.
The paperclip. This harmed millions of users.
Mentioning that paperclip makes you look as stupid as it is. It's just stupid and annoying...not harmful. Everybody knows it's stupid and knows how to turn it off. It would have been forgotten by now except MS bashers like to parade it out just to make themselves look stupid.
And FAT, the Fragmented Allocation Table ? Who should pay for the countless hours morons spend looking DEFRAG.EXE painfully moving blocks around ?
The FAT existed in CP/M, and probably before that. It was very effective and efficient for the diskettes and miniscule hard drives at the time. MS's only fault here is perhaps not providing an alternative in 1987 or so when 40 MB drives started showing FAT's limitations.
And the windows API?
It's hardly the worst. Certainly no worse in general than the glibc API, or the C standard library API for that matter. There are bone-headed design decisions in every one of them.
Has anyone else tried to actually buy a copy of Windows without buying a computer system? I don't mean an upgrade package, which requires that you already own Windows. MS will not sell the average consumer an original license for Windows. You can only purchase such a license from an OEM with a computer system.
Wrong! I was in Best Buy yesterday, and saw the following:
Windows 2000 Professional: $299. No pre-existing software requirements.
Windows 2000 Professional Upgrade: $199. Requires Windows 95/98/ME/NT 3.51 or later.
Actually, King has stated that the first three installments will be $1 each, and the fourth through eighth will be $2 each. Any remaining installments will be free. So the total you'll pay is $13.00. That's about half of what you'd pay for the book if it was printed and distributed through a publisher.
And of course nobody would actually pay money to support free software, either, because there is a huge nebulous "them" out there writing it for free, and somebody else is paying for the servers and bandwidth charges.
The first release of Kylix is the Delphi 5 IDE with a new cross-platform component library called CLX ("clicks"). It is very cool. I've seen non-trivial Windows/VCL apps ported from Delphi 5 to Kylix in very short periods of time (a couple of hours to a couple of days, depending on the app).
There's no real difference between a programmer with no real Linux experience doing Kylix and a programmer with no real Windows experience doing Delphi. In both cases they're junior programmers who learn from the more experienced. The senior programmers, on the other hand, will have to rely on their C knowledge in order to glean useful information from the vast amount of C/C++ code that's available for Linux. This is no different than the early days of Borland Pascal for Windows and (later) Delphi.
Re-read the article again. I don't think there was a 'series of "guilties"'.
Microsoft hailed the jury's July 16, 1999, verdict, which found in the company's favor on nearly every claim except the unfair-business-practices charge.