They measure resistance conducted through the human body. Any standard electronic voltmeter can be programmed to measure resistance in the low ohm range.
Does this work on anyone, or are there potential problems with variations from one person to another?
I worked a summer job once in a lab where, to kill some time, we used the equipment to measure basic electrical properties of ourselves.
As I recall, my skin resistance (measured from one hand to the other) was a small fraction of everyone else's.
I also seemed to have a much greater capacitance.
Would I be a good bus, or a bad one?
Aeron chairs??? Those things dig into your legs! OW!
Unless your body proportions are extremely unusual, this means you have the wrong size chair.
They come in small, medium and large.
While I am not particularly tall overall (5'10"), my upper legs are out-of-proportion -- too long.
The small and medium Aeron chairs did indeed dig into the backs of my legs, but the large size, normally for people over six feet tall, fit very nicely.
Lots of futzing around with the back support and other adjustments also necessary to make the chair comfortable.
If these folks built this thing for peanuts (compared to NASA budgets), NASA will seem ridiculously ineffective... It's not like they deserve this kind of treatment, but the question will be raised for sure.
Raised by idiots, perhaps.
While there may be many things to criticize about NASA, comparing SpaceShipOne to their efforts is definitely apples and oranges.
Suborbital.
Minimal payload capacity.
Has NASA designed anything to that kind of spec since the early 60s?
I applaud Scaled Composites' achievements, but... wake me up when they've got a cheap way to lift heavy loads to LEO.
Given that capability, space begins to get interesting.
Sounds like the guy has a case. The only iffy point would be: does inserting a CD count as authorization?
Is the jewel box clearly marked?
How about the CD itself?
Do retail people who sell the disk explain it to you?
If everything is clearly marked that inserting this disk in your computer may result in the installation of additional software, the insertion may be authorization.
Otherwise, forget it -- if you don't know, you can't possibly be authorizing.
A quick check with Google didn't provide an answer -- does anyone know what was the earliest implementation of Forth for the 8080?
It might not have been as early as Gates' BASIC, but given the ease of implementation, I would expect it to have appeared fairly quickly.
Also, a similar and more complete language chart is available
here.
What belongs in the OS is the concept of a sandbox.
The ability to establish (and the default should be quite restrictive) a set of rules for what an executable can and cannot do.
So when the mail client launches it, the executable can write to the screen, but not write to the disk (or at least nowhere but the sandbox version of the disk).
So when that executable attempts to access the network, nothing happens.
There's nothing really wrong with allowing an executable to run, so long as it is restricted as to what it can do.
Windows (and DOS before it) have never really shaken off the basic concept of the single-user machine not attached to a network.
The first chapter is pretty much a detailed covering of the entire Florida election scandal - illegally purging voters from the rolls, allowing a private company to do the purging, illegally requiring felons from other states to petition for voting rights they never lost, etc.
Indeed, the worst offenses in Florida appear to be with processes outside of the actual casting and counting of ballots.
I would like to read fewer stories about counting ballots and more about fixing the problems of people who were denied their right to vote, or granted that right inappropriately.
Such problems were not confined to Florida; recall that Missouri had various officials and judges overruling one another about when polling places could be closed, or not.
TTBOMK, all of the unofficial recounts done by the newspapers who gained access to the Florida ballots indicated that Bush won by various small margins.
Those recounts included different scenarios: recounting all counties, recounting only select Democratic-leaning counties, using different "hanging chad" standards, etc.
Add to your list of "correct" ways to mark a ballot the requirement to choose N or fewer from a list of M(>N) candidates.
At least here in Colorado, this type of situation shows up on the ballot regularly for things like school board elections.
Actually, the ballots always say "Choose three...";
I've assumed, but never verified by checking the laws, that ballots with only one or two candidates selected are still valid.
Surely the same institution that came up with a distributed computing software project such as Folding@Home can handle a menial financial and record-keeping software project.
Assuming that this is a typical situation where a business (and Stanford is a business, at least over on the administration side) is spec'ing a financial system, consider some of the requirements that the system has to meet.
Thousands of employees must be paid promptly and the checks must be correct.
The Congress and the IRS change the withholding rules regularly -- who will be responsible for making sure that all the necessary changes are installed so that the first January checks are correct?
Once per year or so all the employees get to make choices about medical plans, etc.
All such elections must be handled properly. Relevant information must be forwarded on some medium in the appropriate formats for the insurance companies, the medical providers, etc.
Such information may come under HIPPA regulations -- are any of the students or faculty experts on those requirements?
Lots of employees spend money on behalf of the university and then file forms for payment (eg, if you're on a trip and have to pay a taxi).
The IRS mandates a specific audit trail that must be maintained by the system for such payments.
The people who know the details of the audit trail that must be maintained this year tend to work for companies like PeopleSoft and the big accounting houses, not at research/teaching universities.
Without saying anything bad about the students or faculty of the Stanford CS department, who have done many impressive pieces of software over the years, they are NOT qualified to put together a system that meets all of the legal and accounting requirements.
They are NOT organized to meet the ongoing support needs for such a system (as mentioned, who is responsible when the IRS changes 437 rules for next year?).
They are NOT staffed adequately to produce and administer training for all the clerks that have to use the system.
The coding for the system may in fact be straightforward -- but the legal and accounting requirements are definitely non-trivial.
Re:The 802.11 basic standard does not support voic
on
Is VOIP Over WLAN DOA?
·
· Score: 1
Most phone calls aren't meant to be productive. They're used to stay in touch with loved ones, chat with friends, that sort of thing.
Quite possibly true (although until very recently, I think you would be surprised at the fraction of all calls that were business calls).
What I meant was that as a researcher looking at questions of what was technically possible, and how different protocols might work, and what services might be offered for which users would be willing to pay a premium, investigations of multiple media were much more interesting than plain old phone service.
Re:The 802.11 basic standard does not support voic
on
Is VOIP Over WLAN DOA?
·
· Score: 2, Informative
I started doing VOIP applied research while employed at one of the big telephone companies twelve years ago (at least).
The same basic rules that applied then apply now.
VOIP can consistently provide telephony-like service -- high-quality audio, low round-trip latency -- under one of two conditions:
(1) some form of QOS, or
(2) swamp the problems with bandwidth.
For most LAN situations (not all), switched Ethernet is sufficient to provide the "swamping" solution.
Unfortunately, wireless CSMA is shared and seldom gets into the "swamping" category.
QOS can take a variety of forms, from packet-by-packet prioritization, to traffic segregation and call-by-call admission control.
I'll admit to being surprised at the use of IP for plain old telephone service.
I always thought the interesting applications were those involving multiple media -- shared slides (with pointers and scribbling), shared apps, some low bitrate video, etc.
I remember a whole series of very effective four- and five-way meetings between the developers of an authoring system in Denver and the people using it to write training materials in Minnepolis.
Secondly, the regulations on phone companies...I wonder how much of the problem is due to cities only allowing one set of copper, and all the regulations preventing outsiders from joining the system? Just look at what's going on with VOIP.
As I have posted before, I do not believe there is any industry in the US regulated more, or by more different entities, than phone service.
And of course, many of today's regulatory disputes are grounded in yesterday's well-meaning regulatory requirements.
Since it was originally difficult to interconnect networks and numbering plans, you got franchise areas with only one provider per area.
Since it was desirable that phone service be available to everybody, you get subsidies:
long distance subsidized local service,
urban service subsidized rural,
business service subsidized residential, etc.
Once you have monopolies and subsidies, there are incentives to game the system.
When you separate long distance and local service, a company that can provide long-distance service but avoid the local subsidies has a cost advantage.
When you allow other companies to provide business-only local service, they have a cost advantage if they can avoid the residential subsidy.
When new companies introduce technology such as cellular or VOIP, which provide voice transport but for whom it is difficult to provide expensive capabilities such as 911 that are required of the older companies, you get regulatory fights.
If you roll the FCC back to a function like the IANA, where *all* they get to do is keep track of who owns which spectrum in which areas, that would be a dandy start.
Scaling back the FCC's responsibilities may be reasonable, but not quite that far. Consider a couple of instances.
First, because we're talking about broadcast media here, there is some need to restrict actions before they happen, rather than strictly after. For example, the FCC sets standards on radiated power levels for consumer devices.
Suppose a vendor introduces a wonderful new device, but the design is such that it degrades over-the-air television for anyone within 400 yards of the operating device.
Who sues whom?
Does the TV station sue the manufacturer?
Do they sue the individual owners?
Do private individuals have to identify which neighbor is the culprit and individually sue them?
Does the court issue an initial restraining order?
Does that stop the company from selling any more hardware, or does it stop the individuals from using the devices already sold?
Does the sheriff knock down your door and turn off your device?
What if it takes 60 days to come up on the court's schedule?
Second, the FCC has been saddled with lots of responsibilities other than spectrum allocation over the years.
Lots and lots of regulations on telephone companies, for example -- local number portability to name one.
This is a case where, based on actual experience, both the wireline and cell phone companies had zero interest in making it possible for the end user to keep a phone number that they had used for years when they changed carriers.
It works now (well, mostly works) ONLY because the rule was imposed on the companies from the outside.
it's a drop in the bucket compared to their $60 Billion in cash. It is just a simple "cost of doing business" for them.
It's not the fine that's the big deal in this case, it's the requirement that they build and sell a version of Windows without Media Player bundled and/or integrated.
Assuming that it stands up on appeal, it sets a precedent that MS cannot arbitrarily bundle and/or integrate what were applications "into" the OS.
And that they have to reveal the APIs so that other firms can develop components that can be used in place of the MS ones.
Indeed.
To paraphrase George Bernard Shaw, "We know what I am, now we're just negotiating the price.":^)
When I leased myself to the Fortune 100 company, I knew that I would have to perform certain acts that I might find somewhat objectionable.
As an individual, I am opposed to the idea that software can be patented, but I don't have sufficient personal integrity to refuse to work for a company that applies for such patents.
At least the company's strategy on patents was one of defense -- if the company held a patent on a particular process or architecture, no one else could obtain such a patent and make the company stop using their own invention.
TTBOMK, and I feel reasonably sure that I would know if it had happened, the company never took any offensive actions WRT "my" patents.
We readily shared the code with other research groups, under conditions that they could not redistribute it (never could convince the lawyers of the value in "just give it away").
Seriously, the company I work for, has a function in our COMMERCIAL software package that requires a triple-click in order to do something. It's been there for about 8 years now - so we already got prior art =P
Semiseriously:^), did your company publish the triple-click in anything available to the general public (ie, technical journal)?
If the description appeared only in material with limited distribution (eg, user manual available only to those who licensed the product), you may not have actually "disclosed" your art, in which case it may not count.
While IANAL, this might well leave you in the position of having to now apply for a patent on your own and show invention prior to the date that MS claims they invented it.
If necessary, can your company document when the triple-click idea was conceived and document that you made diligent (ie, continuous) efforts to implement said idea after it was conceived?
Surprisingly few companies these days require engineers to maintain the necessary records to build a good patent case: stitched rather than loose-leaf lab notebooks, written in ink, entries dated, initial entries on inventions properly witnessed.
25+ years ago, when I went to work at Bell Labs, all new hires were required to attend a class given by the legal department on how to keep such records.
No no, I'm implying it's probably easy to buy a patent office.
Did you mean to say it's easy to buy "the" USPTO?
Personally I doubt that, but having been through the process multiple times, I would agree that large corporations (with large budgets) have advantages that may give them a higher success rate on their applications.
They can afford to hire experienced law firms employing lawyers who have engineering backgrounds and understand intimately the language that the patent office requires.
The USPTO has a large number of requirements as to how documents should be prepared, and are not required to assist applicants.
They can afford to do some of the background checking that all applicants are supposed to do.
The USPTO requires that you perform such a search and reveal any possibly relevant findings to them.
They can afford the additional fees that follow when a rejected application is modified and resubmitted.
Yeah but you're not a major corporation either. I'm quite sure their patents almost never get rejected.
Corporations cannot, legally, invent things.
It's one of the few rights that are denied to them.
All patent applications must have an inventor who is a natural person (or multiple inventors who are).
The corporate name on the filings as assignee was a Fortune 100 company.
If/when we reach the state where artificial intelligence programs are capable of creating new things, it will be interesting to see who gets credit for their creations.
just how many other devices have similar "hidden" features, just waiting to be hacked. I suspect it's a lot.
Nor is this a recent phenomenon.
I remember a Usenet discussion from many years ago about one line of VCRs.
There were three models, call them basic, standard, and deluxe.
As you progressed up the line, the available features (vertical tracking adjustment, frame-by-frame advance) increased, as did the price.
This was in the days when those features all used analog controls.
Turned out the only difference in the three models was the outside case;
the high-end case had additional holes drilled in it and labeled buttons or knobs attached.
Someone provided a set of drawings that showed precisely where to drill holes and which Radio Shack knobs could be used to convert the basic or standard model into the deluxe.
Is there anywhare a statistic pf granted vs. denied patants by the PTO. Otherwise it seems save to assume the PTO is just rubberstamping...
Not a meaningful statistic (sample of one), but...
My name is on four different software or software architecture patents.
Of the four, two were denied on initial filing.
Of the two that were denied,
one was granted after several claims covered by prior art were removed, and
the other was granted after writing a several-page submission that showed how the prior art cited by the patent office did not apply to our situation.
Casual conversation with other people listed as inventors on patent applications made by the giant corporation where we worked suggested that my experience was not unusual -- a substantial portion of applications seem to be initially rejected.
Let's say a police officer were to appear at my door without a warrant, wanting to search my house. If I allow him to enter, can he use anything incriminating that he finds as evidence, even though he conducted the search without a warrant?
In the days of my youth (sharing a house with two other guys who might have who knows what kind of illegal substances about), the conventional wisdom was that once you let the officer in the front door there was little you could do to constrain their subsequent actions.
When the police knocked at my door last year (looking for a reported runaway who was friendly with my daughter), I still found myself automatically saying, "I'll be happy to step outside and discuss matters with you officer."
But without a warrent, they don't come through the door.
At some point I expect (or at least I hope) that the portions of the Patriot Act that allow officers to enter and search without showing a judge sufficiently probable cause to get a warrent are ruled unconstitutional.
So once the dam is complete, it will be over 1/8th the power
the country uses and will help reduce dependence on fossil fuels.
We agree on the dam's output.
We differ in what we were comparing that output to.
You compare it to their current production of electricity.
Today, China is a poor country (on average) but is growing rapidly.
They already have electric power shortages.
If they continue to develop, their per capita energy usage will increase rapidly, and a large fraction of the new usage will be electricity.
I compared the dam's output to a guess about China's total energy demands if they succeed in becoming a "rich" country.
I guessed (conservatively, I think) that they would need per capita energy consumption at half the level of Japan today.
Compared to that value, the dam's output is pretty small.
As I mentioned, it is not practical to meet ALL energy needs with electricity -- but possibly excluding transportation, much of the growth in energy use will be electricity.
None of this is intended to deny that the Three Gorges Dam is a hell of an impressive civil engineering project -- it is.
Nor to deny your environmental point -- not only you, but I and millions in Japan (where acid rain due to China burning coal is becoming a serious problem) are happy to see a major hydro source of electricity come on line.
But in the longish view (say, 30 years?), and assuming that China continues to grow at anything like its current rate, the dam's output is a very small fraction of the energy that China will need.
We know that over the last 100 years that the world-wide temp has gone up by roughly 1 degree. But before that time period, there is no climate data at all.
There are no direct measurements, but there is a considerable indirect record.
During parts of the 1700s, the Thames River froze over on a regular basis, with ice thick enough to support a horse and sleigh.
Ask any of the British folks who read Slashdot how long it has been since that happened.
Around 900, the climate in the Northern Hemisphere was warm enough that vikings could colonize Greenland.
IIRC, some of the pressure for the Vikings to spread was due to a population boom caused by increased crop yields due to warmer weather.
Using both computer models and knowledge of the relationship between temperature and grain yields, it is possible to make shrewd guesses about the mean temperatures at those times.
Microsoft's value proposition is product based, i.e. sell lots of units of software product to make profit.
OSS value proposition is service based. Give the software away, and provide services, consulting etc. to help companies deploy and use the software.
So, given this distinction, how does OSS move into the consumer market?
At least in the US, consumers appear to be much more motivated to buy a thing rather than a service.
This is one of the problems that Tivo faces: people are willing to buy a $300 PVR, but are much less enthusiastic about buying the $20/month service that goes with it.
US consumers buying a computer assume that the device will just work -- they don't need to buy a service contract, or hire a consultant.
Very few of them are interested in buying a complete OS package separately from the computer.
18.2 Million Kilowatts , ie. 18.2 Billion watts of power.
Interesting.
That's 14 watts per person.
Assume the generators run continuously and you get 14 watt-years per person or
12.27 KWH per person.
Using 3,412 BTU per KWH,
20,681,000 BTU per short ton of coal, and
2.2 kg per pound,
that's the equivalent of about 1.84 kg of coal per person per year.
Japan's total energy consumption in 2003 was equivalent to 5,224 kg of coal per person per year.
Let's assume the goal is half of that -- that allows for China's target not being "as developed" as Japan is today, plus allowing for increased energy efficiency in the future -- and the dam's per person output is about 0.07% of the goal.
Put another way, if China were to use hydro power exclusively to meet the energy goal -- impractical, of course -- they would need on the order of 1400 Three Gorges dams.
I myself have no feel for numbers this big and always have to work it all out in order to see if something is really "big" or not:^)
Slashdot Mirror
Does this work on anyone, or are there potential problems with variations from one person to another? I worked a summer job once in a lab where, to kill some time, we used the equipment to measure basic electrical properties of ourselves. As I recall, my skin resistance (measured from one hand to the other) was a small fraction of everyone else's. I also seemed to have a much greater capacitance. Would I be a good bus, or a bad one?
Unless your body proportions are extremely unusual, this means you have the wrong size chair. They come in small, medium and large. While I am not particularly tall overall (5'10"), my upper legs are out-of-proportion -- too long. The small and medium Aeron chairs did indeed dig into the backs of my legs, but the large size, normally for people over six feet tall, fit very nicely. Lots of futzing around with the back support and other adjustments also necessary to make the chair comfortable.
Raised by idiots, perhaps. While there may be many things to criticize about NASA, comparing SpaceShipOne to their efforts is definitely apples and oranges. Suborbital. Minimal payload capacity. Has NASA designed anything to that kind of spec since the early 60s? I applaud Scaled Composites' achievements, but... wake me up when they've got a cheap way to lift heavy loads to LEO. Given that capability, space begins to get interesting.
Is the jewel box clearly marked? How about the CD itself? Do retail people who sell the disk explain it to you? If everything is clearly marked that inserting this disk in your computer may result in the installation of additional software, the insertion may be authorization. Otherwise, forget it -- if you don't know, you can't possibly be authorizing.
Also, a similar and more complete language chart is available here.
What belongs in the OS is the concept of a sandbox. The ability to establish (and the default should be quite restrictive) a set of rules for what an executable can and cannot do. So when the mail client launches it, the executable can write to the screen, but not write to the disk (or at least nowhere but the sandbox version of the disk). So when that executable attempts to access the network, nothing happens. There's nothing really wrong with allowing an executable to run, so long as it is restricted as to what it can do. Windows (and DOS before it) have never really shaken off the basic concept of the single-user machine not attached to a network.
Indeed, the worst offenses in Florida appear to be with processes outside of the actual casting and counting of ballots. I would like to read fewer stories about counting ballots and more about fixing the problems of people who were denied their right to vote, or granted that right inappropriately. Such problems were not confined to Florida; recall that Missouri had various officials and judges overruling one another about when polling places could be closed, or not.
TTBOMK, all of the unofficial recounts done by the newspapers who gained access to the Florida ballots indicated that Bush won by various small margins. Those recounts included different scenarios: recounting all counties, recounting only select Democratic-leaning counties, using different "hanging chad" standards, etc.
Add to your list of "correct" ways to mark a ballot the requirement to choose N or fewer from a list of M(>N) candidates. At least here in Colorado, this type of situation shows up on the ballot regularly for things like school board elections. Actually, the ballots always say "Choose three..."; I've assumed, but never verified by checking the laws, that ballots with only one or two candidates selected are still valid.
Assuming that this is a typical situation where a business (and Stanford is a business, at least over on the administration side) is spec'ing a financial system, consider some of the requirements that the system has to meet.
Without saying anything bad about the students or faculty of the Stanford CS department, who have done many impressive pieces of software over the years, they are NOT qualified to put together a system that meets all of the legal and accounting requirements. They are NOT organized to meet the ongoing support needs for such a system (as mentioned, who is responsible when the IRS changes 437 rules for next year?). They are NOT staffed adequately to produce and administer training for all the clerks that have to use the system. The coding for the system may in fact be straightforward -- but the legal and accounting requirements are definitely non-trivial.
Quite possibly true (although until very recently, I think you would be surprised at the fraction of all calls that were business calls). What I meant was that as a researcher looking at questions of what was technically possible, and how different protocols might work, and what services might be offered for which users would be willing to pay a premium, investigations of multiple media were much more interesting than plain old phone service.
I'll admit to being surprised at the use of IP for plain old telephone service. I always thought the interesting applications were those involving multiple media -- shared slides (with pointers and scribbling), shared apps, some low bitrate video, etc. I remember a whole series of very effective four- and five-way meetings between the developers of an authoring system in Denver and the people using it to write training materials in Minnepolis.
As I have posted before, I do not believe there is any industry in the US regulated more, or by more different entities, than phone service. And of course, many of today's regulatory disputes are grounded in yesterday's well-meaning regulatory requirements. Since it was originally difficult to interconnect networks and numbering plans, you got franchise areas with only one provider per area. Since it was desirable that phone service be available to everybody, you get subsidies: long distance subsidized local service, urban service subsidized rural, business service subsidized residential, etc.
Once you have monopolies and subsidies, there are incentives to game the system. When you separate long distance and local service, a company that can provide long-distance service but avoid the local subsidies has a cost advantage. When you allow other companies to provide business-only local service, they have a cost advantage if they can avoid the residential subsidy. When new companies introduce technology such as cellular or VOIP, which provide voice transport but for whom it is difficult to provide expensive capabilities such as 911 that are required of the older companies, you get regulatory fights.
Scaling back the FCC's responsibilities may be reasonable, but not quite that far. Consider a couple of instances.
First, because we're talking about broadcast media here, there is some need to restrict actions before they happen, rather than strictly after. For example, the FCC sets standards on radiated power levels for consumer devices. Suppose a vendor introduces a wonderful new device, but the design is such that it degrades over-the-air television for anyone within 400 yards of the operating device. Who sues whom? Does the TV station sue the manufacturer? Do they sue the individual owners? Do private individuals have to identify which neighbor is the culprit and individually sue them? Does the court issue an initial restraining order? Does that stop the company from selling any more hardware, or does it stop the individuals from using the devices already sold? Does the sheriff knock down your door and turn off your device? What if it takes 60 days to come up on the court's schedule?
Second, the FCC has been saddled with lots of responsibilities other than spectrum allocation over the years. Lots and lots of regulations on telephone companies, for example -- local number portability to name one. This is a case where, based on actual experience, both the wireline and cell phone companies had zero interest in making it possible for the end user to keep a phone number that they had used for years when they changed carriers. It works now (well, mostly works) ONLY because the rule was imposed on the companies from the outside.
It's not the fine that's the big deal in this case, it's the requirement that they build and sell a version of Windows without Media Player bundled and/or integrated. Assuming that it stands up on appeal, it sets a precedent that MS cannot arbitrarily bundle and/or integrate what were applications "into" the OS. And that they have to reveal the APIs so that other firms can develop components that can be used in place of the MS ones.
Indeed. To paraphrase George Bernard Shaw, "We know what I am, now we're just negotiating the price." :^)
When I leased myself to the Fortune 100 company, I knew that I would have to perform certain acts that I might find somewhat objectionable.
As an individual, I am opposed to the idea that software can be patented, but I don't have sufficient personal integrity to refuse to work for a company that applies for such patents.
At least the company's strategy on patents was one of defense -- if the company held a patent on a particular process or architecture, no one else could obtain such a patent and make the company stop using their own invention.
TTBOMK, and I feel reasonably sure that I would know if it had happened, the company never took any offensive actions WRT "my" patents.
We readily shared the code with other research groups, under conditions that they could not redistribute it (never could convince the lawyers of the value in "just give it away").
Semiseriously :^), did your company publish the triple-click in anything available to the general public (ie, technical journal)?
If the description appeared only in material with limited distribution (eg, user manual available only to those who licensed the product), you may not have actually "disclosed" your art, in which case it may not count.
While IANAL, this might well leave you in the position of having to now apply for a patent on your own and show invention prior to the date that MS claims they invented it. If necessary, can your company document when the triple-click idea was conceived and document that you made diligent (ie, continuous) efforts to implement said idea after it was conceived? Surprisingly few companies these days require engineers to maintain the necessary records to build a good patent case: stitched rather than loose-leaf lab notebooks, written in ink, entries dated, initial entries on inventions properly witnessed.
25+ years ago, when I went to work at Bell Labs, all new hires were required to attend a class given by the legal department on how to keep such records.
Did you mean to say it's easy to buy "the" USPTO? Personally I doubt that, but having been through the process multiple times, I would agree that large corporations (with large budgets) have advantages that may give them a higher success rate on their applications. They can afford to hire experienced law firms employing lawyers who have engineering backgrounds and understand intimately the language that the patent office requires. The USPTO has a large number of requirements as to how documents should be prepared, and are not required to assist applicants. They can afford to do some of the background checking that all applicants are supposed to do. The USPTO requires that you perform such a search and reveal any possibly relevant findings to them. They can afford the additional fees that follow when a rejected application is modified and resubmitted.
Corporations cannot, legally, invent things. It's one of the few rights that are denied to them. All patent applications must have an inventor who is a natural person (or multiple inventors who are). The corporate name on the filings as assignee was a Fortune 100 company.
If/when we reach the state where artificial intelligence programs are capable of creating new things, it will be interesting to see who gets credit for their creations.
Nor is this a recent phenomenon. I remember a Usenet discussion from many years ago about one line of VCRs. There were three models, call them basic, standard, and deluxe. As you progressed up the line, the available features (vertical tracking adjustment, frame-by-frame advance) increased, as did the price. This was in the days when those features all used analog controls. Turned out the only difference in the three models was the outside case; the high-end case had additional holes drilled in it and labeled buttons or knobs attached. Someone provided a set of drawings that showed precisely where to drill holes and which Radio Shack knobs could be used to convert the basic or standard model into the deluxe.
Not a meaningful statistic (sample of one), but... My name is on four different software or software architecture patents. Of the four, two were denied on initial filing. Of the two that were denied, one was granted after several claims covered by prior art were removed, and the other was granted after writing a several-page submission that showed how the prior art cited by the patent office did not apply to our situation. Casual conversation with other people listed as inventors on patent applications made by the giant corporation where we worked suggested that my experience was not unusual -- a substantial portion of applications seem to be initially rejected.
In the days of my youth (sharing a house with two other guys who might have who knows what kind of illegal substances about), the conventional wisdom was that once you let the officer in the front door there was little you could do to constrain their subsequent actions. When the police knocked at my door last year (looking for a reported runaway who was friendly with my daughter), I still found myself automatically saying, "I'll be happy to step outside and discuss matters with you officer." But without a warrent, they don't come through the door. At some point I expect (or at least I hope) that the portions of the Patriot Act that allow officers to enter and search without showing a judge sufficiently probable cause to get a warrent are ruled unconstitutional.
We agree on the dam's output. We differ in what we were comparing that output to. You compare it to their current production of electricity. Today, China is a poor country (on average) but is growing rapidly. They already have electric power shortages. If they continue to develop, their per capita energy usage will increase rapidly, and a large fraction of the new usage will be electricity.
I compared the dam's output to a guess about China's total energy demands if they succeed in becoming a "rich" country. I guessed (conservatively, I think) that they would need per capita energy consumption at half the level of Japan today. Compared to that value, the dam's output is pretty small. As I mentioned, it is not practical to meet ALL energy needs with electricity -- but possibly excluding transportation, much of the growth in energy use will be electricity.
None of this is intended to deny that the Three Gorges Dam is a hell of an impressive civil engineering project -- it is. Nor to deny your environmental point -- not only you, but I and millions in Japan (where acid rain due to China burning coal is becoming a serious problem) are happy to see a major hydro source of electricity come on line. But in the longish view (say, 30 years?), and assuming that China continues to grow at anything like its current rate, the dam's output is a very small fraction of the energy that China will need.
There are no direct measurements, but there is a considerable indirect record. During parts of the 1700s, the Thames River froze over on a regular basis, with ice thick enough to support a horse and sleigh. Ask any of the British folks who read Slashdot how long it has been since that happened. Around 900, the climate in the Northern Hemisphere was warm enough that vikings could colonize Greenland. IIRC, some of the pressure for the Vikings to spread was due to a population boom caused by increased crop yields due to warmer weather. Using both computer models and knowledge of the relationship between temperature and grain yields, it is possible to make shrewd guesses about the mean temperatures at those times.
So, given this distinction, how does OSS move into the consumer market? At least in the US, consumers appear to be much more motivated to buy a thing rather than a service. This is one of the problems that Tivo faces: people are willing to buy a $300 PVR, but are much less enthusiastic about buying the $20/month service that goes with it. US consumers buying a computer assume that the device will just work -- they don't need to buy a service contract, or hire a consultant. Very few of them are interested in buying a complete OS package separately from the computer.
Interesting. That's 14 watts per person. Assume the generators run continuously and you get 14 watt-years per person or 12.27 KWH per person. Using 3,412 BTU per KWH, 20,681,000 BTU per short ton of coal, and 2.2 kg per pound, that's the equivalent of about 1.84 kg of coal per person per year. Japan's total energy consumption in 2003 was equivalent to 5,224 kg of coal per person per year. Let's assume the goal is half of that -- that allows for China's target not being "as developed" as Japan is today, plus allowing for increased energy efficiency in the future -- and the dam's per person output is about 0.07% of the goal. Put another way, if China were to use hydro power exclusively to meet the energy goal -- impractical, of course -- they would need on the order of 1400 Three Gorges dams.
I myself have no feel for numbers this big and always have to work it all out in order to see if something is really "big" or not :^)