SCO has also alleged that some of their copyrighted code has been directly incorporated into the Linux source. Direct copying is almost certainly grounds for SCO to collect money, once SCO establishes that the code is indeed theirs. I can easily imagine that there are places where both SCO and Linux have incorporated code under a BSD-like license, which would make it look identical but would not involve any violation of SCO copyright.
Of course, no one else is going to see SCO's code until at least the discovery phase of a lawsuit gets under way, so it'll be a while before we have any public facts about this.
I think it would be fascinating to work out just where the $20 per CD ends up going. Some of the possibilities:
Depending on where you live, anywhere from 0 to 7% for sales tax.
Some fraction for the people manning the store. What's their aggregate salary? Are there benefits beyond that, like medical insurance for the managers? There's certainly FICA and Medicare taxes, some visible, some not. How many CDs do they sell per hour?
Rent on the space. Electric bill. Cleaning the facilities. Is there a toilet for employees -- water and sewer bill? How many phone lines? Payroll and tax accounting? Insurance? How many CDs do they sell per month?
Where were the disks pressed? How many people's hands did they pass through to get there? How many gallons of fuel burned?
Inventory taxes at various places -- the store, some warehouse, etc.
Personally, I suspect that an surprisingly small fraction of the money actually ends up directly in the hands of the record companies. I keep waiting for the record companies to figure out that burnable CDs and the Internet potentially provide them with the means to cut out a great portion of the middlemen expenses.
IIRC, a typical DVD playback bitstream averages just under 5 Mbps and spikes to nearly 10 Mbps during complex sequences. I will assert that we are a long ways from people getting sustained high-bandwidth streams like that delivered to their homes over any of the current broadband technologies -- cable modem, DSL, powerline, or wireless. Hence DVD playback will have a significant advantage in video quality for a considerable time. It certainly does when compared to, say, today's digital cable.
Yes, advanced codecs like JVT will improve on those rates. But even with a 3:1 improvement, we are looking at a stream that averages 1.7 Mbps and spikes as high as 3.3. Affordable broadband using any of the current technologies cannot deliver even that reduced level of traffic to a significant portion of households on a continuous basis.
They are so common in this industry, we have our own special name for them: Killer Apps.
Let's review, shall we?
VisiCalc
WordPerfect
Mosaic
I think this list is interesting from two directions. First, because of the way that I would classify the apps from the users' perspective: smart paper, smart paper, and smart notebook/magazine. What other broad classes of killer app are there? Games and several other things go in the category of "smart TV". Besides paper and TV, what other "smart thing" does software provide?
Second, from the point of view of the elimination of support jobs. 25 years ago at Bell Labs, an engineer took handwritten documents to the typing pool; over the course of less than three years, troff made the typing pool disappear and the engineers were made responsible for their own typing. Software has repeated that process in lots of other places: telephone switching, mail sorting, precision machining, etc. How far can it go?
Keep in mind the scale on which companies of this size are operating. I know that a typical year at any RBOC involves hundreds of thousands of purchases totaling billions of dollars. For Qwest local service, equipment purchases are spread across organizations in 14 states. Each of those items must be correctly classified and entered into the system. Data input alone requires a large number of clerks. Even if the input error rate is low, there are so many entries that there are a large number of errors in total. When I was buying lab equipment and had to review various reports, I would guess that as many as 5% of our purchases were placed in the wrong category. Most commonly, two digits in the multi-digit category code were transposed, or one of the digits was off-by-one. And our stuff was pretty simple compared to multi-use equipment used for real service.
All states require their own special accounting reports from their local phone companies, so custom systems are involved. Those have to be developed and maintained (regulators are really good at coming up with new and different regulations that have to be met). Again, despite the best efforts and test plans, errors creep in.
Now do that for 60 years -- yes, some states require some assets such as cables to be depreciated over periods as long as 60 years. So you have to track some stuff for that long.
First, I am not surprised about the inaccurate accounting of equipment at an RBOC, or any large telephone company. Because of the regulatory history (and local phone companies are probably more heavily regulated in terms of funky accounting practices than any other kind of company), an RBOC has to actually keep multiple sets of books. Consider depreciation on a piece of switching equipment. The IRS may require that it be depreciated over the course of ten years for tax purposes. The state in which it is located may require that it be depreciated over the course of thirty years for rate-setting purposes. Depending on the services for which it is used, there may be additional FCC accounting rules. It may have an actual usable lifetime of twelve years, then is replaced, but gets refurbished and reinstalled in a different state that has yet another set of rules. If it is used in providing multiple services, there may be multiple sets of conflicting accounting rules that apply to it. If it is destroyed, some of the rules may require that it be carried on the books until it is fully depreciated, even though it no longer exists.
For the RBOCS, keep in mind that serious regulation started in 1934, and there were 23 local companies operating under the AT&T banner. Then those companies were consolidated into seven in 1984, and have further combined into just three. Could anyone have kept accurate track of equipment and accounting for it under those conditions?
Several years ago, I was in a similar position working in the research part of a large telecom company. I couldn't talk the lawyers into GPL, but did manage to convince them to let me distribute freely, where the license allowed use and modification but not redistribution (that was as far as I could get them to go at the time).
The arguments that convinced them to allow me to go forward were about the business aspects of trying to market the software. When would the company make a firm decision to move forward on marketing it? Who would pay for the marketing effort? Who would pay to develop documentation, training materials and all the other things that went into supporting a product? Who would pay for ongoing maintenance to fix user-reported problems? Which organization would take on those jobs (I was in research and was going to stay there, it wouldn't be me)?
Once they realized that the company wasn't in the software product business, and wasn't going to be in that business, they were willing to listen to me about the non-monetary benefits of distributing the software.
Many years ago, it came out that 98% of the state legislators in New Jersey were lawyers of one sort or another. IIRC, this made headlines right after they had voted down, by a very large majority, a tort reform act that would have made it marginally more difficult to collect more from an insurance company than the face value of the policy. At that time, it was not uncommon for juries to require an insurance company to pay (for example) $1M on a $100K policy...
The insurance company my father worked for at the time quit doing business in NJ because of such awards, and the state's refusal to enforce some sort of sane contract law...
Indeed. And to make them cheaper while still maintaining the profit for the artist and the label, you have to reduce the distribution costs. And one easy way to do that is to... use the Internet and those millions and millions of CD burners out there. I'd like to think that there's some way to set things up so the end user can pay $5 and download a burn-once file. You could put some sort of little shop in the malls where people w/o computers could have a CD burned while they wait, but that shop would still depend on download-and-burn (with caching?) rather than trucks and warehouses.
Re:Interesting idea, but will it work?
on
Foiling Cinema Pirates
·
· Score: 3, Interesting
In the movies, when you see a scene with a television in it, why are there no such artifacts?
There are no artifacts because the TV is a specialized device and the video it is showing is synchronized with the movie camera. Watch the credits at the end of the movie for "24 fps video" or something similar.
Bought the system, got the $100 rebate within three weeks without any further action on my part. I suppose there's some reason having to do with financial reporting why they would prefer to have $500 in revenue followed by an immediate $100 expense...
Many years ago (pull it off of the "old textbook" shelf, copyright is... 1976, I am getting so old), Kernighan and Plauger wrote "Software Tools." One of my favorite statements made in the introduction is
Every program in this book has been run and carefully tested, directly from the text itself, which is in machine-readable form.
I suppose some of the code fragments can't be checked this way, but at least testing the bigger pieces of code can be automated. Of course, this assumes that at some point the files are turned over to the publisher who does not mess with them any more. Not a safe assumption -- I wrote an article for an IEEE journal and approved the "final" version, which was then mangled by another editor and sent to press. I spent weeks trying to explain to people that I could write better than that...
Electrical power distribution has horrible losses in long-distance transmission. Assuming the material is cheap enough, and can be used in cables, and that "room temperature" covers normal outside temperatures as well, you could get much more efficient power transmission. Reduce the number of new generating plants needed, reduce emissions, etc.
And here is a story I can personally verify, because I know the person it involves. This person (who will remain anonymous but is not a coward) was extremely allergic to Brazil Nuts. Dangerously allergic. Eating Brazil Nuts could kill him. And he ate something that contained no Brazil Nuts -- or so it said -- but contained something that had been genetically modified. It had Brazil Nut genes. He was hospitalized.
At the very least, this stuff should be labeled -- it should be manditory that consumers be able to find out not only what food is GE, but what genes were included.
This is an excellent point. A good example of how old laws don't adequately protect us (or allow us to protect ourselves) when new technology is introduced. I'm sure that the brazil nut genes code one or more proteins that have a good effect -- maybe some sort of disease or pest resistance in the plant. But they also code proteins (or perhaps it's the same proteins) that cause the allergic reaction.
Now come the tough questions. How many millions of proteins are there? How many of them trigger severe allergic reactions in some people? How big would the label have to be to list all of the proteins contained in a product? How big if it listed just the proteins known to cause severe allergic reactions in some people? How many people can/would keep track of all the proteins to which they are allergic? What happens if a GE plant (for example) ends up with a genome that codes for new proteins that did not previously occur in nature? Should the inventor be responsible for testing to determine if anyone is allergic to those new proteins?
I don't have any idea what the answers ought to be, but as a minimum, it seems that this particular product should be labeled so that people allergic to brazil nuts would know that they are allergic to whatever is in there. I know that doctors err on the side of caution; my daughter is allergic to a particular antibiotic, and the doctors will not prescribe it or any of several related drugs for her.
The amount of CO2 released by human activities could still be significant, assuming the other sources have constant production.
Back in the 70's, I was peripherally involved in a study funded by the feds to try and figure out where all the CO2 released by human activities was going. The estimates of the amount released since 1900 were large enough that it should have increased the percentage of CO2 in the atmosphere by a measurable amount -- but it hadn't, or at least not by nearly as much as it should have.
Anyone got historical measurements of the percentage of CO2 in the atmosphere? Has it increased measureably of late?
After reading the marked-up version of the Colorado bill, several things seems clear:
The original intent of the bill was to go after people stealing cellular phone service, and possibly DBS video service. After mark-up, it has become much broader.
Using a NAT firewall to allow multiple computers in your home access to the Internet would not appear to be a violation unless your ISP is somehow deprived of money or forbids such arrangements in the TOS. So long as all of the computers are in your home, YOU ARE NOT CONCEALING THEIR LOCATION. I think you might even make an argument based on distributed processing that YOU ARE NOT CONCEALING THEIR IDENTITY. Consider the case where the browser process runs on the server but the display is rendered on another machine somewhere on the LAN. Would that be illegal?
Stealing service is a misdemeanor, but you have to pay for the service you've stolen. Providing equipment or knowledge that allows someone to steal service is a felony.
For me, the last item is the scariest. So long as one cable company in the state forbids the use of NAT firewalls, once someone buys a cable router at Circuit City and uses it illegally, it would appear that Circuit City has now committed a felony offense. Similarly, anyone who writes a FAQ on how to use a Linux box as a cable router will be guilty as soon as someone uses such a box illegally after following the instructions. Ditto for Microsoft and sharing Internet connections.
The earlier version that applied to cell phones and possibly DBS video appeared more reasonable, at least in the sense that actions would clearly be either legal or illegal anywhere in the state. Particularly with regard to cable modem service and the variations in TOS between cable companies, some actions would be legal in one place and not in another, which strikes me as being bad law.
I mean, I know you can probably detect a NAT box by looking at all the packets, but how many ISPs are going to do this? Honestly?
All of the big cable companies are deploying or looking at deploying equipment that examines each and every packet sent or received by a subscriber (with the possible exception of stuff that stays within a single CMTS). The primary purpose of this is to measure volume, that is, how many gigabytes per month did the subscriber use, enabling volume-sensitive pricing. The same equipment can, however, identify traffic types and even monitor individual TCP or UDP flows. Statistical analysis of those flows should make it possible to identify subscribers operating multiple computers behind a NAT firewall with a high degree of accuracy.
Doing that detailed analysis for all subscribers all the time may not be practical, so it may be restricted to subscribers that raise other, simpler flags like high volumes. When I was employed by a cable company, I always argued that we really shouldn't care what the user was doing, just how much bandwidth they used. The most common argument for outlawing NAT boxes was not lost revenues, but the expense of tech support time wasted looking for problems in the cable network that were actually in the subscriber's LAN.
I think there is a mechanism in physical engineering that does not carry across well into software engineering organizations that follow SEI/CMM or other programming methodologies. In building construction you have the architect who applies both functional as well as creative/artistic qualities to the work. Then you have the (civil or other ) engineers that apply more of their discipline to make sure that the building works and does not fall down and meets other requirements. This is oversimplified as there is a necessity for creativity at all points of constructions, but I don't think this analogy carries well to software organizations.
Actually, there's some amount of evidence that it does carry over fairly well. In fact, in The Mythical Man-Month, Brooks makes the point that creativity is involved in all levels of the job. IBM experimented with chief programmer teams, and teams of teams, 30 years ago. They got phenomenal (for the time) improvements in productivity and quality. This was usually attributed to the fact that all key decisions were being made by one individual (after consulting with his second-in-command privately) who was consistent about those choices. The problem with using the approach widely was that, contrary to popular belief amongst programmers, there weren't enough of them with the talent and skills to be chief programmers.
IANAL, and this comment is confined strictly to H1-B workers (not work done offshore). I believe that the law requires that companies employing H1-B workers must pay them wages conforming to the prevailing rate in the area. That is, it's supposed to be illegal to bring the workers in and undercut the wages being paid to citizens or green-card holders. The purpose of the law was to provide a temporary fix to a worker shortage, and specifically not to hold down wage expenses for the companies.
It may be good for Sun's shareholders and/or customers, but it's illegal.
Much as I hate to admit it, the household server is still running 2.0.36. IP masquerade and firewall for the household LAN's Internet access via cable modem, Samba print service for the other machines, backup storage for kids' schoolwork, and an antique version of Apache whose main function is to provide access to the Perl scripts that allow multiple people to share an old SCSI scanner. Haven't had a monitor connected to it for years, and the BIOS is old enough that you can tell it to ignore the fact that the keyboard check fails at boot time. The old AT power supply comes back up without any manual action after a power failure -- no idea how many of those it's been through, but the ext2 file system doesn't seem to have ever lost anything.
I recall reading an article by a company doing a complete systems analysis of a modern car prior to designing an all-electric vehicle. One of the interesting sidelights of the analysis was that further reductions in air drag, at least for sedans, were pretty hard to come by in practice. The limiting factors were the air flows required for heat dissipation from the radiator, the muffler/catalytic converter, and the brake surfaces. IIRC, enclosing the wheel wells almost completely would have given the biggest improvement, but without the air flow the brake rotors melted under heavy use...
Where do you put a complex Excel spreadsheet that handles a range of inputs? These get used for multi-million dollar decisions all the time.
I got kicked out of a meeting one time when I asked to see the test plan for the spreadsheet on which the company was basing a $100M decision. I believe the comment that got me kicked was something like "If I wrote code for the implementation like that, you'd fire my ass!" implying that I expected the business analyst to have made some reasonable effort to be sure the complex financial calculations were correct. I have seen a spreadsheet where an intermediate result was multiplied by the constant 10 rather than 100, and the error was uncovered rather late in the process (like just before presenting to the CFO).
My own opinion is that if it's going to run more than once and the results are supposed to be right across a range of inputs (certainly true for a business case with sensitivity analysis on the key variables), it's programming. Processing a Web form such that there are complex and persistent side effects (eg, a credit card charge) is programming, no matter what it's written in.
Exactly. Intel needs to sell lots of leading edge chips with good profit margins. In practice, that means selling them into households or to business desktops, not into the back-office servers. Which uses more processors -- the render farm at Pixar, or the millions of DVD players used to watch Toy Story 2? Not necessarily a good example, as the nature of consumer electronics tends to favor putting the decoder into low-cost low-power custom silicon rather than software on a high-end processor, but you get the idea.
I tell my friends at Intel that research into the next mass-market killer app is at least as important to the company's future as the next generation of chip design or fab. What apps will require lots of people to upgrade to a better, faster processor? Real-time speech recognition might be one, if it could be embedded effectively in the UI for other widely-used apps. Apple has tried to convince everyone that they should edit digital video (those Jeff Goldblum commercials) but I suspect the reality is that not many people are really ready to put the creative effort into that. Any other candidates?
If someone can really upgrade the entire power grid infrastructure to support high-speed data service for the whole country for $100M, then they're going to be very rich. The cable industry has spent billions of dollars upgrading their plant to two-way hybrid-fiber-coax arrangements in order to support HSD with cable modems. If someone can provide a similar service for a fraction of the cost, they should be able to undercut the cable companies' prices and capture the entire market. If you can show a real working demonstration, lots of people will lend you $100M against your future profit stream.
The fact that there are no cities where you can buy such service has always made me suspicious that either (a) the stuff doesn't work in the field the way it does in the lab or (b) the actual cost figures are completely different.
Slashdot Mirror
Of course, no one else is going to see SCO's code until at least the discovery phase of a lawsuit gets under way, so it'll be a while before we have any public facts about this.
-
Depending on where you live, anywhere from 0 to 7% for sales tax.
-
Some fraction for the people manning the store. What's their aggregate salary? Are there benefits beyond that, like medical insurance for the managers? There's certainly FICA and Medicare taxes, some visible, some not. How many CDs do they sell per hour?
-
Rent on the space. Electric bill. Cleaning the facilities. Is there a toilet for employees -- water and sewer bill? How many phone lines? Payroll and tax accounting? Insurance? How many CDs do they sell per month?
-
Where were the disks pressed? How many people's hands did they pass through to get there? How many gallons of fuel burned?
-
Inventory taxes at various places -- the store, some warehouse, etc.
Personally, I suspect that an surprisingly small fraction of the money actually ends up directly in the hands of the record companies. I keep waiting for the record companies to figure out that burnable CDs and the Internet potentially provide them with the means to cut out a great portion of the middlemen expenses.Yes, advanced codecs like JVT will improve on those rates. But even with a 3:1 improvement, we are looking at a stream that averages 1.7 Mbps and spikes as high as 3.3. Affordable broadband using any of the current technologies cannot deliver even that reduced level of traffic to a significant portion of households on a continuous basis.
Second, from the point of view of the elimination of support jobs. 25 years ago at Bell Labs, an engineer took handwritten documents to the typing pool; over the course of less than three years, troff made the typing pool disappear and the engineers were made responsible for their own typing. Software has repeated that process in lots of other places: telephone switching, mail sorting, precision machining, etc. How far can it go?
All states require their own special accounting reports from their local phone companies, so custom systems are involved. Those have to be developed and maintained (regulators are really good at coming up with new and different regulations that have to be met). Again, despite the best efforts and test plans, errors creep in.
Now do that for 60 years -- yes, some states require some assets such as cables to be depreciated over periods as long as 60 years. So you have to track some stuff for that long.
Some problems are hard just because of scale.
For the RBOCS, keep in mind that serious regulation started in 1934, and there were 23 local companies operating under the AT&T banner. Then those companies were consolidated into seven in 1984, and have further combined into just three. Could anyone have kept accurate track of equipment and accounting for it under those conditions?
The arguments that convinced them to allow me to go forward were about the business aspects of trying to market the software. When would the company make a firm decision to move forward on marketing it? Who would pay for the marketing effort? Who would pay to develop documentation, training materials and all the other things that went into supporting a product? Who would pay for ongoing maintenance to fix user-reported problems? Which organization would take on those jobs (I was in research and was going to stay there, it wouldn't be me)?
Once they realized that the company wasn't in the software product business, and wasn't going to be in that business, they were willing to listen to me about the non-monetary benefits of distributing the software.
The insurance company my father worked for at the time quit doing business in NJ because of such awards, and the state's refusal to enforce some sort of sane contract law...
Indeed. And to make them cheaper while still maintaining the profit for the artist and the label, you have to reduce the distribution costs. And one easy way to do that is to... use the Internet and those millions and millions of CD burners out there. I'd like to think that there's some way to set things up so the end user can pay $5 and download a burn-once file. You could put some sort of little shop in the malls where people w/o computers could have a CD burned while they wait, but that shop would still depend on download-and-burn (with caching?) rather than trucks and warehouses.
Bought the system, got the $100 rebate within three weeks without any further action on my part. I suppose there's some reason having to do with financial reporting why they would prefer to have $500 in revenue followed by an immediate $100 expense...
Electrical power distribution has horrible losses in long-distance transmission. Assuming the material is cheap enough, and can be used in cables, and that "room temperature" covers normal outside temperatures as well, you could get much more efficient power transmission. Reduce the number of new generating plants needed, reduce emissions, etc.
Now come the tough questions. How many millions of proteins are there? How many of them trigger severe allergic reactions in some people? How big would the label have to be to list all of the proteins contained in a product? How big if it listed just the proteins known to cause severe allergic reactions in some people? How many people can/would keep track of all the proteins to which they are allergic? What happens if a GE plant (for example) ends up with a genome that codes for new proteins that did not previously occur in nature? Should the inventor be responsible for testing to determine if anyone is allergic to those new proteins?
I don't have any idea what the answers ought to be, but as a minimum, it seems that this particular product should be labeled so that people allergic to brazil nuts would know that they are allergic to whatever is in there. I know that doctors err on the side of caution; my daughter is allergic to a particular antibiotic, and the doctors will not prescribe it or any of several related drugs for her.
Back in the 70's, I was peripherally involved in a study funded by the feds to try and figure out where all the CO2 released by human activities was going. The estimates of the amount released since 1900 were large enough that it should have increased the percentage of CO2 in the atmosphere by a measurable amount -- but it hadn't, or at least not by nearly as much as it should have.
Anyone got historical measurements of the percentage of CO2 in the atmosphere? Has it increased measureably of late?
Hours of sleep I require per night:
a) Eight or more (0 points)
b) Six (2 points)
c) Four (4 points)
d) Sleeping makes me irritable (10 points)
-
The original intent of the bill was to go after people stealing cellular phone service, and possibly DBS video service. After mark-up, it has become much broader.
-
Using a NAT firewall to allow multiple computers in your home access to the Internet would not appear to be a violation unless your ISP is somehow deprived of money or forbids such arrangements in the TOS. So long as all of the computers are in your home, YOU ARE NOT CONCEALING THEIR LOCATION. I think you might even make an argument based on distributed processing that YOU ARE NOT CONCEALING THEIR IDENTITY. Consider the case where the browser process runs on the server but the display is rendered on another machine somewhere on the LAN. Would that be illegal?
-
Stealing service is a misdemeanor, but you have to pay for the service you've stolen. Providing equipment or knowledge that allows someone to steal service is a felony.
For me, the last item is the scariest. So long as one cable company in the state forbids the use of NAT firewalls, once someone buys a cable router at Circuit City and uses it illegally, it would appear that Circuit City has now committed a felony offense. Similarly, anyone who writes a FAQ on how to use a Linux box as a cable router will be guilty as soon as someone uses such a box illegally after following the instructions. Ditto for Microsoft and sharing Internet connections.The earlier version that applied to cell phones and possibly DBS video appeared more reasonable, at least in the sense that actions would clearly be either legal or illegal anywhere in the state. Particularly with regard to cable modem service and the variations in TOS between cable companies, some actions would be legal in one place and not in another, which strikes me as being bad law.
Doing that detailed analysis for all subscribers all the time may not be practical, so it may be restricted to subscribers that raise other, simpler flags like high volumes. When I was employed by a cable company, I always argued that we really shouldn't care what the user was doing, just how much bandwidth they used. The most common argument for outlawing NAT boxes was not lost revenues, but the expense of tech support time wasted looking for problems in the cable network that were actually in the subscriber's LAN.
It may be good for Sun's shareholders and/or customers, but it's illegal.
Much as I hate to admit it, the household server is still running 2.0.36. IP masquerade and firewall for the household LAN's Internet access via cable modem, Samba print service for the other machines, backup storage for kids' schoolwork, and an antique version of Apache whose main function is to provide access to the Perl scripts that allow multiple people to share an old SCSI scanner. Haven't had a monitor connected to it for years, and the BIOS is old enough that you can tell it to ignore the fact that the keyboard check fails at boot time. The old AT power supply comes back up without any manual action after a power failure -- no idea how many of those it's been through, but the ext2 file system doesn't seem to have ever lost anything.
I recall reading an article by a company doing a complete systems analysis of a modern car prior to designing an all-electric vehicle. One of the interesting sidelights of the analysis was that further reductions in air drag, at least for sedans, were pretty hard to come by in practice. The limiting factors were the air flows required for heat dissipation from the radiator, the muffler/catalytic converter, and the brake surfaces. IIRC, enclosing the wheel wells almost completely would have given the biggest improvement, but without the air flow the brake rotors melted under heavy use...
Where do you put a complex Excel spreadsheet that handles a range of inputs? These get used for multi-million dollar decisions all the time.
I got kicked out of a meeting one time when I asked to see the test plan for the spreadsheet on which the company was basing a $100M decision. I believe the comment that got me kicked was something like "If I wrote code for the implementation like that, you'd fire my ass!" implying that I expected the business analyst to have made some reasonable effort to be sure the complex financial calculations were correct. I have seen a spreadsheet where an intermediate result was multiplied by the constant 10 rather than 100, and the error was uncovered rather late in the process (like just before presenting to the CFO).
My own opinion is that if it's going to run more than once and the results are supposed to be right across a range of inputs (certainly true for a business case with sensitivity analysis on the key variables), it's programming. Processing a Web form such that there are complex and persistent side effects (eg, a credit card charge) is programming, no matter what it's written in.
I tell my friends at Intel that research into the next mass-market killer app is at least as important to the company's future as the next generation of chip design or fab. What apps will require lots of people to upgrade to a better, faster processor? Real-time speech recognition might be one, if it could be embedded effectively in the UI for other widely-used apps. Apple has tried to convince everyone that they should edit digital video (those Jeff Goldblum commercials) but I suspect the reality is that not many people are really ready to put the creative effort into that. Any other candidates?
The fact that there are no cities where you can buy such service has always made me suspicious that either (a) the stuff doesn't work in the field the way it does in the lab or (b) the actual cost figures are completely different.
Or were you being sarcastic?