Slashdot Mirror
IBM Flushes Restroom Patent
jdkane writes "As reported in this CNet news article IBM has quietly eliminated a patent it received on a method for determining who gets to use the bathroom next.
I say Kudos to IBM because it is a relief (no pun intended) to see some common sense prevail in the patent news.
A funny quote from the article is "But just because the patent office granted this and other questionable patents doesn't mean the system is broken".;)"
You might laugh now about this patent, but I work in Operations Research and such data is vitally important for many businesses. The required number and size of the stalls must determinded a priori and well-timed refill of toilet paper is extremely important from a logistics/cost point of view. Makashura and Miller have proven this optimizing problem NP-hard and the IBM patent contains an appromation algorithm with a garantueed maximum error of 1/5 of the original cost which is very good. If the PR deparment of IBM wouldn't be so stupid the would make BIG bucks with this patent.
It seems that they make the same error here which they made with the personal computer: letting things go without enough patent coverage, giving the rise to power competitors like Microsoft or Compaq. Not very clever.
Owner of a Mensa membership card.
The best suggestion I saw on /. was to pay examiners for each application they could reject.
Another one: patents should be backed up by significant and provable documentation of the actual invention process.
Finally, there should be a bounty (payable by the patenter) for anyone able to break a patent by proving prior art.
The alternative is going to be a general constipation of innovation in those countries that allow software patents.
My blog
What does the lady from the patent office mean?!
"We dont know who issue the patent".....is it so tough to track who is the issuing officer. I am surprised it isn't done.
Anyways the reporter should have asked why such an absurd patent was issued in the first place and probably put her in the spot by giving many examples of such crappy patents being issued. Sheesh, its almost like the patent office wants us to think its a clerical error.....and the best is they get away with it.
vv
Not only that but the technology as such is crap. Pardon the pun.
Here in Sweden automated next-in-line ticket machines are everywhere, and having to suffer them one major flaw comes immediately to light; lower efficiency.
You see, what was before a line, with the next person steping up to the counter (or toilet) in all of two seconds, now takes thirty or more. You first have to find the correct teller, then mosy on over there.
Add to that the fact that the people servicing the tellers now can take a brake or fiddle with God knows what (since there is no line of customers eagerly awaiting service). Before the 'improvement' the person behind the counter had to wait for a natural lull, before being able to file here (or his) fingernails, now it can be done with a hundred people waiting in line. Oh progress.
As an example of the lowered efficiency I can take the case of returning books at the local library used to be handled by one librarian, now that they've automated it, it takes three! Same amount of work.
About the only institution that has gotten it right is the (state) liqour stores, where the staff presses the button too soon (so to speak) and actually keep a line of two persons at each teller: the customer being serviced, and the next one. That increases throughput considerably. Wouldn't work here though. You'd have to press the button before you were all wiped off and squeaky clean for that to work in an airplane.
Stefan Axelsson
OK - let's think about this: why would IBM care about people using the bathrooms? Well, this process is a direct analogy to CPU resource management. The article mentions estimated wait time, ability to make and cancel resource reservations, and dangers of numerous stalled processes.
I would bet that some engineer was trying to explain the new resource management algorithm, and used the bathroom analogy. They then registered the patent out of humor, or to ensure nobody later claimed their algorithm was prior art as bathroom usage.
10b||~10b -- aah, what a question!
Ok, clearly someone is way underage to be submitting stories to slashdot. The patent office handles literally hundreds of thousands of applications a year. I challenge anyone to find that large of a system with an error rate of zero. From internet routing to CNN to squirrels in headlights, with a large enough sample size, you are going to be able to find problems. With this rationalle, everything is broken.
Now, I've got some real objections to the current patent system. I think lots of things are patentable that shouldn't be. However, even if the laws were changed and patents really tightened up, I guarantee you that some weird ones will still slip through.
So, can we please see an end to the /. sport of finding the most ridiculous patents and waving them around as evidence that patents are "broken"? Even if patents were overhauled and the most extreme /. views prevailed, these same examples would still exist. Sometimes, the issue is the bureaucracy and human error, not a "broken" system.
That said, I do think the system is broken; this is just a case of getting the right conclusion from the wrong evidence ("It would be dangerous to jump off a tall building because it's hard to see exactly where you'd land").
Cheers
-b
If I wanted a sig I would have filled in that stupid box.
The best suggestion I saw on /. was to pay examiners for each application they could reject.
/. comments by them. And I'd rather have a system that awards too many patents, requiring the lowly peon to challenge the big goliath in a court of law to get a patent busted, than have a system that's abused the other way, preventing the lowly peon from getting deserved patents, while their well-financed competition steals their idea and puts it in practice much faster with their larger capital backing. (disclaimer: I am the lowly peon)
/. crowd. We all know that these big businesses and their patents suck. But think of a million-dollar idea, (like a Jump to Conclusions mat..) and think of what you'd have to do to start a company to go in to business to exploit the idea, assuming you don't already own one. Put yourself in the position of the lowly peon with a ticket to glory, instead of a role of angry-slashdotter-reacting-to-abuse-by-large-corpo rations, and think of how grateful you are for each protection afforded to the small patenter, and how loathe you would be to "reform" those protections away... It's eye-opening.
I think someone pointed out last time, too, that almost all patent applications are rejected the first time around. They send it back to you with a list of reasons they rejected it, you fix it, and send it back. Maybe a couple times, until it's eventually accepted. So it wouldn't really make sense to pay them for each time they sent it back...
IANAPTOPE, but I read a lot of
Finally, there should be a bounty (payable by the patenter) for anyone able to break a patent by proving prior art.
Now that's interesting... but again, that would even further skew the patent system from one that tries to put the lowly peon on equal footing with the Established Interests. Example: OmniCorp has a Patent subdivision of their Legal org. The PTO requires an additional $2k (?) bounty deposit to be staked when first patenting something. That's chump change for OmniCorp, but it may be very tricky for John Q. DocBrown to pony up the deposit, in addition to his now-steeper filing fees (thx PTO reorg). So DocBrown is afraid to file his patent, for fear of losing 3-large if his idea turns out to be stupid, instead of just one.
And when do you get your bounty-deposit back? 20 years later, when the patent expires? Now, the PTO could pay the bounty instead, since they kinda screwed up, but I don't think that would help anyone. You might put in place a system where the PE no longer has any reason to approve a valid app, that may be close to prior art, for fear of getting it busted in court and racking up a demerit point.
Another one: patents should be backed up by significant and provable documentation of the actual invention process.
Umm, it's not that way now? You mean including the inspirational moment, or what? Because you have to have everything else well backed up, if you want to assert that you're the actual inventor, if anyone tries ripping you off.
I've got a suggestion for the