USPTO Grants Bezos Patent On '60s-Era Chargebacks 144
theodp writes "Chargebacks on computing resources are certainly nothing new, dating to the '60s. But five decades later, the USPTO has deemed Amazon CEO Jeff Bezos' invention — Dynamic Pricing of Web Services Utilization — worthy of a new patent. From the patent: 'Utilization of a storage resource may be measured in terms of a quantity of data stored (e.g., bytes, megabytes (MB), gigabytes (GB), etc.) per unit of time (e.g., second, day, month, etc.). Similarly, communication bandwidth utilization may be measured in terms of a quantity of data transmitted per unit of time (e.g., megabits per second). Processing resource utilization may be measured as an aggregate number of units of processing effort (e.g., central processing unit (CPU) cycles, transactions, etc.) utilized, or as a rate of processing effort utilization per unit of time (e.g., CPU cycles or transactions per second).' Sound familiar, Greyglers? Another example of why it's not wise to grant software patents when people don't know much about computer history."
Misleading summary (Score:5, Informative)
The patent is actually for utilizing a predictive process to change pricing based upon expected future load. Still not necessarily new, but very different than the summary implies.
Re:Brainless (Score:5, Informative)
So who is more brainless?
You're brainless for not reading the patent claims, and theodp is brainless for his nonstop anti-Amazon patent tirade [google.com] (372 results)
Here are the relevant parts from all of the links in the summary:
Wikipedia time-sharing article: "Users were charged rent for the terminal, a charge for hours of connect time, a charge for seconds of CPU time, and a charge for kilobyte-months of disk storage." Yeah, ok... that's metered service, just like your power bill or long distance charges.
Plato history article: Uh, this mentions getting a 50th anniversary Plato-style Google logo on Google. Nothing relevant on this link.
The actual patent abstract: "A method and system for dynamic pricing of web services utilization. According to one embodiment, a method may include dynamically predicting utilization of a web services computing resource that is expected to occur during a given interval of time, and dependent upon the dynamically predicted utilization, setting a price associated with utilization of the web services computing resource occurring during the given interval of time. The method may further include providing the price to a customer. "
Does charging for CPU time and resources involve dynamic prediction? No? I didn't think so. Lousy try, theodp. Better luck next time.
Re:Misleading summary (Score:3, Informative)
Maybe misleading summary but what you just said is old, old, old - part of my job when started -71, and old already then! The problem today seems to be - add a couple of words, like WEB or Internet, to some old idea and , voilà, you can have a patent.
Seriously - is the USPO the only government office which, when not skilled on the issue, is supposed and/or allowed to make bad decisions anyway? Doesn't sound good or even very useful?
"Predictive process" has been used in any kind of business forever "to change pricing based upon expected future .. whatever" - especially what I did, combining insurance profits with having, charging and selling computer resources. It's very interesting and much what the patent describes is really out of the textbooks - maybe just too old textbooks, not any more in libraries, even in universities? Will be big again - you have to charge in cloud? Maybe that's why they really did go after a patent. Of course it can't hold - even abstracts as WEB are either defined as "interlinked documents" or not really defined at all, just needs a couple of million $$ to fight the patent and then you are free to do what you / most large sw/hw/cloud/etc service corporations have been doing 40+ years! If you still have any money left to do anything - LOL!
Besides, the mistake often is the load - you have to look much more than bits, bytes, seconds, etc - someone else is charging your resoures, be they hw, electircity, taxes, and so on - so, "It's complicated", sometimes changing daily - think currencies and almost any business today!
Re:Shit! (Score:5, Informative)
FFS, someone should take a hatchet to the US PTO. Don't they need to reduce the budget or something?
The USPTO is entirely fee-funded. Your tax dollars don't pay the examiners' salaries.
On a side note, if you do know something about history and technology, and you'd like to put your money where your mouth is and improve the quality of patent examination, the USPTO is currently hiring qualified individuals with expertise in electrical, computer, and biomedical engineering. US citizenship required. In addition to standard federal benefits and a salary that can reach $100k in about three years, the USPTO has the federal government's flagship telework program, which allows you to work from home, anywhere in the country, once you meet certain qualifications.
See http://usptocareers.gov/ [usptocareers.gov] for more info.
Re:O'RLY (Score:3, Informative)
I don't know if the actual patent is a good idea either, but to be fair, it is somewhat narrower than just charging for computer time/resources. It appears to specifically cover only charging for computer time/resources in a dynamic way based on a prediction of their utilization. So e.g. the cost for one CPU-minute would go up or down depending on the system's estimate of current contention for CPU-minutes. It seems that has probably been used in some other markets (e.g. in some electric markets), but it's possible that it's novel when applied to computer resources. Possibly obvious to a practitioner skilled in the art, but invalidation for obviousness seems uncommon these days, at least at the USPTO level.
A Futures Market in Computer Time (Harvard, 1968) (Score:4, Informative)
A Futures Market in Computer Time [ucsb.edu], Communications of the ACM, June 1968: "An auction method is described for allocating computer time that allows the price of computer time to fluctuate with the demand...if the computer ever is idle, its price automatically becomes attractively low."
Bilski (Score:3, Informative)
I think this coming Monday is the last day of the current SCOTUS session, and is expected to be the day that they give a ruling on the Bilski case. If we wish with all our might, we may hear on Monday that software and algorithms are not considered patentable material.