By: Daily Tech
Source: http://www.dailytech.com
Category: Patent News
Apparently the U.S. Patent Office has an answer for everything, even breaking the speed of light
The U.S. Patent and Trademark Office is an interesting system. Unlike some other global patent offices, it isn't sustained by any sort of tax allotment. Rather, it sustains itself with the fees it charges. The system essentially creates a fun game for the USPTO in that the more patents it grants, the more patents are filed. And the more patents that are filed the more fees are collected. Hence, the more patents that are granted, the more fees that are collected.
Of course the patents office can only review so many patents. So if it wants to grant more patents than it has employees to handle, it will need to simply approve patents that aren't thoroughly reviewed. Granting insufficiently reviewed patents should help to increase the number of bad patents.
I. Modeling the Number of Bad Patents as a Constraint Satisfaction Problem
Constraint satisfaction problems or CSPs are used to model everything from financial decisions to sports season scheduling. The quality of the model is dictated by your constraints. It is easy to over-constrain or under-constrain. Model quality is also determined by how well you estimate linear constants based on various experiments.
So let's look at the above description of how the USPTO can profit off bad patents, but put it in reasonable math terms.
Model in Symbolic Logic:
GOAL:
Max(R = F - B)
CONSTRAINTS:
1. F = a * G
2. B = r* L + b * S
3. P >= ( G - c * A )
4. P >= 0
5. S >= d * e * (G-P) + e * P
6. S >= 0
7. L >= k * l * (G-P) + l * P
8. L >= 0
Goal in Words:
You seek to maximize revenue (R), which is increased as you take in more fees (F) and decreased by backlash (B) against the patent office in terms of lawsuits, patent review requests, etc.
Constraining equation 1 in words:
Fees (F) increase as the patents granted (G) increase. (a) is a constant coefficient representing a modeled relationship between the number of granted patents and how many extra fees are collected due to increased filing if companies and individuals know they're more likely to receive a grant.
Constraining equation 2 in words:
Backlash (B) comes from the number of granted silly (or your other favorite "S" word, e.g. stupid, etc.) patents (S) plus the number of granted patents with incorrect literature reviews (L). (r) and (b) are constants representing the scaling of public backlash for the two types of bad patents. The more r*L + b*S the USPTO puts out, the more backlash (B) will be generated.
Constraining equation 3 in words:
Poorly reviewed patents (P) are found by taking the number of patents granted (G) and subtracting the number of Attorneys (A) times the number of patents each attorney can review (A)
Constraining equation 4 in words:
The number of poorly reviewed patents is at least (0) as a negative number of poor reviews is nonsensical.
Constraining equation 5 in words:
The number of silly patents (S) (typically an non-novel patent) is equal to the frequency of silly filings (e) times the number of poorly reviewed patents (P). As some well reviewed silly patents may be granted on accident, a constant (e) is utilized to indicate how likely a reviewed silly submission is accidentally granted. So an additional number of silly patents is added for the frequency of silly filings (e) times the frequency of grants of reviewed silly filings (d) times the number of well reviewed patents (G-P).
Constraining equation 6 in words:
The number of granted silly patents (S) is at least (0) as a negative number of grants is nonsensical.
Constraining equation 7 in words:
The number of patents with incorrect literature reviews (L) (e.g. prior art goes unnoticed) is equal to the frequency of filings that ignore existing literature (l) times the number of poorly reviewed patents (P). As some earnest reviews may miss prior art on accident, a constant (k) is utilized to indicate how likely a review is to miss prior art in the case of a bad patent. So an additional number of incorrect literature reviews is added for the frequency of filings that ignore existing literature (l) times the frequency actual reviewed patents are approved accidentally ignoring literature (d) times the number of well reviewed patents (G-P).
Constraining equation 8 in words:
The number of granted patents with incorrect literature reviews (L) is at least (0) as a negative number of grants is nonsensical.
Overall Justification of Model:
This is a pretty simplistic model in that everything has to scale linearly and it operates under the assumption that the USTPO's goal is to maximize earnings (fees-backlash). But it seems reasonably accurate, given the patent office's inherent business-like operation and the assumption that the goal of every business is to produce maximum profits.
Shortcoming:
There is one important flaw in the model. The frequency of filed silly patents (e) and filed patents that ignore prior art (l) should increase as the number of grants increase. However, if additional constraints were added to relate (e) and (l) to (G) by a scaling relationship, it would make the CSP non-linear.
Ultimately this could result in more patents being filed (if backlash is exceeded by the gains in fees) or less patents being filed (if the fees are exceeded by the backlash).
Conclusions:
This model offers two key observations:
The model has a tendency to accept a lot of bad of patents, depending on the state of the office and public.
The ony thing holding it back from accepting all bad patents (hence obtaining the best case maximum on profits) is the amount of public backlash.
II. "Silly" Patents
Alright enough math, let's take a look at some interesting patents...
Today we present to you some jewels from the USPTO. Remember, these folks are the law of the land and decide whether it's legal or not for someone to build something:
Patenting the slingshot...
"Handheld Water Balloon Catapault" (U.S. Patent 4,922,884)
Filed: Mar. 1989 Granted: May 1990
Now at this point you're probably thinking "How is this thing a catapault? It has no cranked tension! Isn't this essentially just a plastic slingshot, a crude precursor to the bow and arrow?"
Well silence your skepticism at its slightly misleading terminology and prepare to be amazed at the novel contribution:
"...implementing a yolk of greater clearance, thus allowing an enlarged projectile puch to be attached. These features allow projectiles, of greater mass such as water balloons... Combined with... a guard which protects the fingers from the occasional contact which may occur during projectile launch."
Now at this point you're probably thinking they've somehow made a catapault out of hardened egg yolks. No, they have not. It appears that the author meant to put "yoke", as in "handle", but instead put "yolk" as in "yummy egg". This occured a good five times in the patent. And he was patenting a slingshot. It smells like a lawsuit against Dennis the Menace and Bart Simpson may be incoming.
Patenting nature...
"A Jumping Snail" (U.S. Design Patent 435,610)
Filed: June 1999 Granted: Dec 2000
Quite literally what it sounds like, this patent is a stuffed animal resembling a snail, which is poorly drawn in a series of imaginative images by someone whom we're guessing never had any formal college background in art. We're not sure if snails are in violation of this design, but should human laws be applied to gastropods they're going to be crushed in court, particularly after they don't show up as a result of their general snailish qualities. But seriously if there's one thing the Japanese horror film Uzumaki taught us, it's that snails are not suitable children's toys as they quite possibly might be trying to kill you.
Patenting use of a playground apparatus...
"Method of Swinging on a Swing" (U.S. Patent U.S. 6,368,227 B1)
Filed: Nov. 2000 Granted: April 2002
If you're wondering, yes this patent describes playing on a swing. While it's hard to fault the author St. Paul, Minn.-resident Steven Olson (for the record, he's probaby one of these guys) for engaging in such a nostalgic activity, one might worry that someone might challenge this patent given that Mr. Olson didn't really provide much in the way of prior art for his novel play "technology". A prior art claim seems especially likely if 18th century French Rococo artist Jean-Honore Fragonarde (see above) returns from the grave. But fortunately the USPTO cleared that up for us. See that "B1"? That means it's the first of its kind, or as the USPTO, "No previously published pre-grant publication."
Well if you were worried this patent was going to turn out to make the USPTO look willfully incompetent, thankfully they cleared it up for us by examining the issue and concluding that no one had ever "invented" swinging on a swing before. Clearly Mr. Fragonard needs to get right back in his time machines, head straightaway to his own era, and stop stealing the 20th century's ideas. That's some USPTO justice there!
Warp speed 10 on my Mark...
"Hyper-Light Speed Antenna" (U.S. Patent 6,025,810)
Filed: Oct 1997 Granted: Feb. 2000
While all these patents concerned modern technology somewhat indirectly as:
a) They covered "technology" inventions.
b) They show the likely quality of the approvals process for more serious invention claims.
They don't cover any sort of hard tech. That's why we've saved the best to last. You see, the USPTO has verified that Einstein was wrong and that one of the most fundamental laws of physics: E=MC^2 was just a bunch of baloney.
Meet the patent on the hyper-light speed antenna. That's right, the author claims to have found a way to send electromagnetic waves at a speed faster than light. He does not mention whether this is in a medium or not, so clearly he's shattered the speed of light in any setting, including the absolute speed of light in the vacuum. Guess what skeptics, the U.S. Patent and Trademark Office says he's right. It took them three years to verify his claims, but now they've gave resounding approval to the filing. Goodbye fundamental law of physics.
Some mistakenly believed that the patent was assigned to a U.S. Doctor in Aurora, Colo. That's understandable as the patent assignee is David L. Strom, while there's an orthopedic surgeon named David E. Strom. But it appears that the filer is instead possibly a real estate agent in the town whose middle initial matches that of the filer. That's right, it may be a real estate agent who trumped all those stuffy physics professors who told you about all that laws of physics hooey. Looks like we can finally go Star Trek on the universe, assuming we find enough dilithium crystals to power this sophisticated gadget.
And yes, for the record, these are all real patents, follow the links. Hopefully this inspires you to file some good patents of your own.
Source: http://www.dailytech.com/Fun+Patents+of+the+Week+HyperLight+Speed+Antenna+and+More/article23691.htm
Source: http://www.dailytech.com
Category: Patent News
 |
| Patent News |
The U.S. Patent and Trademark Office is an interesting system. Unlike some other global patent offices, it isn't sustained by any sort of tax allotment. Rather, it sustains itself with the fees it charges. The system essentially creates a fun game for the USPTO in that the more patents it grants, the more patents are filed. And the more patents that are filed the more fees are collected. Hence, the more patents that are granted, the more fees that are collected.
Of course the patents office can only review so many patents. So if it wants to grant more patents than it has employees to handle, it will need to simply approve patents that aren't thoroughly reviewed. Granting insufficiently reviewed patents should help to increase the number of bad patents.
I. Modeling the Number of Bad Patents as a Constraint Satisfaction Problem
Constraint satisfaction problems or CSPs are used to model everything from financial decisions to sports season scheduling. The quality of the model is dictated by your constraints. It is easy to over-constrain or under-constrain. Model quality is also determined by how well you estimate linear constants based on various experiments.
So let's look at the above description of how the USPTO can profit off bad patents, but put it in reasonable math terms.
Model in Symbolic Logic:
GOAL:
Max(R = F - B)
CONSTRAINTS:
1. F = a * G
2. B = r* L + b * S
3. P >= ( G - c * A )
4. P >= 0
5. S >= d * e * (G-P) + e * P
6. S >= 0
7. L >= k * l * (G-P) + l * P
8. L >= 0
Goal in Words:
You seek to maximize revenue (R), which is increased as you take in more fees (F) and decreased by backlash (B) against the patent office in terms of lawsuits, patent review requests, etc.
Constraining equation 1 in words:
Fees (F) increase as the patents granted (G) increase. (a) is a constant coefficient representing a modeled relationship between the number of granted patents and how many extra fees are collected due to increased filing if companies and individuals know they're more likely to receive a grant.
Constraining equation 2 in words:
Backlash (B) comes from the number of granted silly (or your other favorite "S" word, e.g. stupid, etc.) patents (S) plus the number of granted patents with incorrect literature reviews (L). (r) and (b) are constants representing the scaling of public backlash for the two types of bad patents. The more r*L + b*S the USPTO puts out, the more backlash (B) will be generated.
Constraining equation 3 in words:
Poorly reviewed patents (P) are found by taking the number of patents granted (G) and subtracting the number of Attorneys (A) times the number of patents each attorney can review (A)
Constraining equation 4 in words:
The number of poorly reviewed patents is at least (0) as a negative number of poor reviews is nonsensical.
Constraining equation 5 in words:
The number of silly patents (S) (typically an non-novel patent) is equal to the frequency of silly filings (e) times the number of poorly reviewed patents (P). As some well reviewed silly patents may be granted on accident, a constant (e) is utilized to indicate how likely a reviewed silly submission is accidentally granted. So an additional number of silly patents is added for the frequency of silly filings (e) times the frequency of grants of reviewed silly filings (d) times the number of well reviewed patents (G-P).
Constraining equation 6 in words:
The number of granted silly patents (S) is at least (0) as a negative number of grants is nonsensical.
Constraining equation 7 in words:
The number of patents with incorrect literature reviews (L) (e.g. prior art goes unnoticed) is equal to the frequency of filings that ignore existing literature (l) times the number of poorly reviewed patents (P). As some earnest reviews may miss prior art on accident, a constant (k) is utilized to indicate how likely a review is to miss prior art in the case of a bad patent. So an additional number of incorrect literature reviews is added for the frequency of filings that ignore existing literature (l) times the frequency actual reviewed patents are approved accidentally ignoring literature (d) times the number of well reviewed patents (G-P).
Constraining equation 8 in words:
The number of granted patents with incorrect literature reviews (L) is at least (0) as a negative number of grants is nonsensical.
Overall Justification of Model:
This is a pretty simplistic model in that everything has to scale linearly and it operates under the assumption that the USTPO's goal is to maximize earnings (fees-backlash). But it seems reasonably accurate, given the patent office's inherent business-like operation and the assumption that the goal of every business is to produce maximum profits.
Shortcoming:
There is one important flaw in the model. The frequency of filed silly patents (e) and filed patents that ignore prior art (l) should increase as the number of grants increase. However, if additional constraints were added to relate (e) and (l) to (G) by a scaling relationship, it would make the CSP non-linear.
Ultimately this could result in more patents being filed (if backlash is exceeded by the gains in fees) or less patents being filed (if the fees are exceeded by the backlash).
Conclusions:
This model offers two key observations:
The model has a tendency to accept a lot of bad of patents, depending on the state of the office and public.
The ony thing holding it back from accepting all bad patents (hence obtaining the best case maximum on profits) is the amount of public backlash.
II. "Silly" Patents
Alright enough math, let's take a look at some interesting patents...
Today we present to you some jewels from the USPTO. Remember, these folks are the law of the land and decide whether it's legal or not for someone to build something:
Patenting the slingshot...
"Handheld Water Balloon Catapault" (U.S. Patent 4,922,884)
Filed: Mar. 1989 Granted: May 1990
Now at this point you're probably thinking "How is this thing a catapault? It has no cranked tension! Isn't this essentially just a plastic slingshot, a crude precursor to the bow and arrow?"
Well silence your skepticism at its slightly misleading terminology and prepare to be amazed at the novel contribution:
"...implementing a yolk of greater clearance, thus allowing an enlarged projectile puch to be attached. These features allow projectiles, of greater mass such as water balloons... Combined with... a guard which protects the fingers from the occasional contact which may occur during projectile launch."
Now at this point you're probably thinking they've somehow made a catapault out of hardened egg yolks. No, they have not. It appears that the author meant to put "yoke", as in "handle", but instead put "yolk" as in "yummy egg". This occured a good five times in the patent. And he was patenting a slingshot. It smells like a lawsuit against Dennis the Menace and Bart Simpson may be incoming.
Patenting nature...
"A Jumping Snail" (U.S. Design Patent 435,610)
Filed: June 1999 Granted: Dec 2000
Quite literally what it sounds like, this patent is a stuffed animal resembling a snail, which is poorly drawn in a series of imaginative images by someone whom we're guessing never had any formal college background in art. We're not sure if snails are in violation of this design, but should human laws be applied to gastropods they're going to be crushed in court, particularly after they don't show up as a result of their general snailish qualities. But seriously if there's one thing the Japanese horror film Uzumaki taught us, it's that snails are not suitable children's toys as they quite possibly might be trying to kill you.
Patenting use of a playground apparatus...
"Method of Swinging on a Swing" (U.S. Patent U.S. 6,368,227 B1)
Filed: Nov. 2000 Granted: April 2002
If you're wondering, yes this patent describes playing on a swing. While it's hard to fault the author St. Paul, Minn.-resident Steven Olson (for the record, he's probaby one of these guys) for engaging in such a nostalgic activity, one might worry that someone might challenge this patent given that Mr. Olson didn't really provide much in the way of prior art for his novel play "technology". A prior art claim seems especially likely if 18th century French Rococo artist Jean-Honore Fragonarde (see above) returns from the grave. But fortunately the USPTO cleared that up for us. See that "B1"? That means it's the first of its kind, or as the USPTO, "No previously published pre-grant publication."
Well if you were worried this patent was going to turn out to make the USPTO look willfully incompetent, thankfully they cleared it up for us by examining the issue and concluding that no one had ever "invented" swinging on a swing before. Clearly Mr. Fragonard needs to get right back in his time machines, head straightaway to his own era, and stop stealing the 20th century's ideas. That's some USPTO justice there!
Warp speed 10 on my Mark...
"Hyper-Light Speed Antenna" (U.S. Patent 6,025,810)
Filed: Oct 1997 Granted: Feb. 2000
While all these patents concerned modern technology somewhat indirectly as:
a) They covered "technology" inventions.
b) They show the likely quality of the approvals process for more serious invention claims.
They don't cover any sort of hard tech. That's why we've saved the best to last. You see, the USPTO has verified that Einstein was wrong and that one of the most fundamental laws of physics: E=MC^2 was just a bunch of baloney.
Meet the patent on the hyper-light speed antenna. That's right, the author claims to have found a way to send electromagnetic waves at a speed faster than light. He does not mention whether this is in a medium or not, so clearly he's shattered the speed of light in any setting, including the absolute speed of light in the vacuum. Guess what skeptics, the U.S. Patent and Trademark Office says he's right. It took them three years to verify his claims, but now they've gave resounding approval to the filing. Goodbye fundamental law of physics.
Some mistakenly believed that the patent was assigned to a U.S. Doctor in Aurora, Colo. That's understandable as the patent assignee is David L. Strom, while there's an orthopedic surgeon named David E. Strom. But it appears that the filer is instead possibly a real estate agent in the town whose middle initial matches that of the filer. That's right, it may be a real estate agent who trumped all those stuffy physics professors who told you about all that laws of physics hooey. Looks like we can finally go Star Trek on the universe, assuming we find enough dilithium crystals to power this sophisticated gadget.
And yes, for the record, these are all real patents, follow the links. Hopefully this inspires you to file some good patents of your own.
Source: http://www.dailytech.com/Fun+Patents+of+the+Week+HyperLight+Speed+Antenna+and+More/article23691.htm
