Slashdot Mirror
Patent Attempt on some forms of Dynamic Web Posting
Scott Ainsworth writes "ZDNet has a new story on a company patenting distributed processing. This time it is about distributing the generation of Dynamic Web pages accross multiple servers. The story is here. " The company apparently says that they hold the patent on many of the load-balancing technqiues that companies use on the Web. *sigh*
I suspect (although I have not done the research) that there is quite a bit of prior art for this patent. Unfortunatly digging up configurations for the web servers from 3 years ago may be quite difficult.
Of course what we really need is a documented channel for challenging patents that are too obvious/trivial.
I guess this is as good a time as any for the obligatory "Why are they only looking into infringement opportunities now? Certainly they've seen these kind of server setups popping up all over the place for a couple of years now, why did they wait so long? It must be the money."
I read the internet for the articles.
We have to wait a little while for the Supreme Court to rule, but...
:-)
If source code is Speech, protected by the First Amendment, then software patents cannot legally restrict it, and we can post patent-breaking SRPMs with impunity, right?
This would be fun to litigate.
"...the firmament sheweth his handiwork" (Ps. 19:1)
Firmament Science and Engineering
Standing on the Solid St
Uhhh, read this quote from InfoSpinner:
.. but I think we're in agreement that no matter how you read it the technologies have been around long before Apr 96.
"When the browser makes the request to a Web server,
the patent covers the notion of redirecting that request to
other machines [application servers] within a cluster and
having those other machines dynamically produce a
response,"
No mention of load balancing here. Redirection fits into the above if you set cluster size=1.
This is me playing devil's advocate in any case... and they probably couldn't stand up in court -- but that won't stop them from trying to intimidate others.... the granted patent may be more specific so InfoSpinner might get shot down in going the whole nine yards as indicated above.
Doesn't this patent simply describe apache (which existed before 1996) ? Apache uses a single parent process to receive all requests. This parent passes off requests to its children, which often generate dynamic content.
This patent described the process IBM implemented
in it's CICS transaction processing system in the
early 1970's. CICS implemented a system called
Multi-Region Operation, where a Terminal Owning
region would route transactions to various
Application Owning regions, which would further
route data requests to various Data Owning regions.
Change the name of the Terminal Owning region to
a Web Server and IBM's CICS is an exact duplicate
of the patent.
Thanx Doug...
It seems to me that this kind of thing can be easily prevented for future patents. People in the open source movement code because they -enjoy- coding, they like experimenting with ideas. How many of us have written throwaway programs just to 'see if I can do it'? How many of those ideas are now patented by somebody?
Don't throw away those 'see if I can' programs. Publish them on the 'net - preferably digitally signed to a newsgroup archived by DejaNews et. al. Just be sure to clearly label your program 'proof of concept' so that nobody thinks it's supposed to be more than that. And if everyone puts the phrase 'proof of concept' in the keywords of their postings, a search of deja-news will quickly spit out prior-arts to challenge future patents with - assuming, of course, that some enterprising soul doesn't start a prior-art archive somewhere with thousands of these little programs.
Remember that most of these patents are founded on sloppy little programs just like that, that prove the concept but are utterly useless in the real world. And remember that it isn't prior art unless it's published.
--Parity
--Parity
'Card carrying' member of the EFF.
When it comes to enforcing such patents, technical content makes little difference. What matters is mainly a tradeoff between management time spend on dealing with litigation vs. licensing cost. As long as the licensing cost is kept below the cost of management time, the patent will get licensed, valid or not.
This is an economic problem: patents are enormously valuable and impose risk on others; their cost compared to their value is tiny (although the cost of entry for making any money off them is fairly high, since they need to be prepared carefully and be backed up by a believable legal team in order to actually get licensing revenue).
And the solution, I believe, has to be economic, too: there should be real costs and risks to the patent holder if a patent is later invalidated because of obviousness or prior art. For example, people who paid licensing fees should be able to have them returned, legal fees should be reimbursed, companies who lost revenue because their management was kept busy through legal tactics should be able to recover expenses and damages, etc.
Changing the patent office itself won't help, and neither will increasing patent fees (that will only keep out the small inventor). The beauty of making patent holders liable if their patents are invalidated is that it is self regulating and puts the burden of careful evaluation on the patent submitter, where it should be.
There are two things that can be usefully done:
Come on...
;)
The US Patent office has issued so many unenforceable patents it is not even funny.
If this patent went to court, it would be smashed apart. As far as I am aware, you can only patent a PROCESS. In software terms, this would mean patenting code.. not the IDEA for a code base. Ideas are un-patentable.
But then.. we already have copyright for code, since it can be considered a written work. So what the hell is the patent office doing issuing patents for software based solutions?
Let me put it another way.. you can patent the subsystem of a car, but you have to patent the apparatus that performs that work.. you can't patent the -idea- of a windshield wiper. You can only patent a particular windshield wiper assembly. You can even patent it all the way down to the rubber used in the wiper. But if Joe Cars sees your wiper, and comes up with his own method of making windshield wipers, you have no claim.
Am I totally off base here? Hopelessly off my legal rocker? Please correct me, legions of legal liasons, and I shall recant.
----
"Eternal vigilance is the price of Freedom."
It seems they're claiming that they have a patent on the process of pushing web requests off to other servers which reside in a pool of available servers in a manner such that no server is overloaded. Hence load balancing.
Even though this is more specific than redirecting or spawning new threads, it is still a very obvious process. I can't understand why the patent office is letting things like this slide. Why is such a simple process considered new when you put the word "web" in there somewhere?
This exact thing has been done for years, but now that it's a web server (ooh, new technology!) it's patentable. I wish that the patent office would realize what all this technology they're reviewing means and what the implications of these broad patents are.
-- atomly
My co-workers and I have gone around and around on this topic over many a cold soyburger down in the corporate cafeteria (I hate turbo fish!). We went back and forth on the value of patents.
;-)
Personally, I think patents are very important. They encourage two things:
1) Innovation, by protecting the exclusive right to exploit an idea for a period of time, allowing a person to profit from his/her cleverness without getting trounced by someone with more money than ideas.
2) Improvement in the state of the art. Part of the patent is an exhaustive description of the invention or process. (AFAIK, specific machines and specific processes are patentable). When the patent expires, there's the full description in the patent -- go to it folks, have a field day!
This system works reasonably well for machines and industrial processes. The length of the patent (I'm not a lawyer, I know it is several years, and the the exact number varies from country to country -- how many years in the US? It's like 7 or 9 or something like that) makes sense for something you to "tool up" to make.
My thought was that software, which has very low capital requirements to produce, and which, therefore, changes much more rapidly than machines and industrial processes, could use its own patent. A software patent would expire in 12-18 months. This is much more in line with the rate at which change occurs in our industry.
This isn't a fully-fleshed out idea, just one of those ideas tossed out like turbo fish over lunch. Still, I think such a system would be better than applying current patents, and might even be better than copyright. Copyright has the virtue of protecting only an exact code version, but the defect of lasting much much longer than needed (60 years, isn't it?)
Well, I'll let the amateur Matlocks point out how ignorant of the law I am...
The patent is available at:
http://www.patents.i bm.com/details?pn=US05894554__&language=en
Review:
On a side note, they reference a patent from Oracle which seems to claim that they invented the idea of having a web server retrieve data from a database through a stored procedure. It seems anyone can get a patent for anything these days.
Scott Severtson
Software Developer
Auragen Communications
scotty@auragen.com
Scott Severtson
Senior Architect, Digital Measures
So, if I remember correctly, IBM used a system called "Womplex" to host the 1996 Olympic website. After receiving an HTTP request, the system would send out a message on a private WAN to several servers placed around the world and have them each ping the recipient, and then determine what the "best" webserver for the user to be on the resulting ping times. It would then feed the user a dynamically-built web page with all of the links pointing to the server that had the lowest latency to them.
There was a C|Net article about this dated June, '96 (two months after the Infospinner filing), but the system obviously was built and running at that point. Sounds like prior art to me. Did anyone dig through InfoSpinner's filing enough to know?
Until recently, I was a patent attorney. I have written and secured software patents. IMHO, the patent system is hopelessly flawed, and I am not sure any form of patent protection is justified. I keep reading the Slashdot discussions regarding each new software patent, and I am waiting for people to wake up to what the patent system is about.
People have this notion that a patent is granted on "real" inventions (you know, light-bulbs, transistors, etc). The idea is that we reward those who solve the big problems. Unfortunately, that is not how the system works. I have read thousands of patents, and I can tell you that very few of them solve a problem in a new way. The vast majority of the time, what happens is this: someone, upon stumbing on a new (to them) problem, quickly figures out the most obvious, straight-forward solution. The solution they find is the same one that most people in the same field would come to when they stumble onto the same problem. But because this particular individual was first to find the problem (in a fast moving area of technology, they may have found the problem a month or less before their competitors), they were first with the solution. They are, under our system, entitled to a patent, and entitled to exclude others from their solution. It simply doesn't matter (practically) that it was a simple solution and took no real spark of creativity. What matters is that this person was first.
Follow the logic here: Everything was new at some point. Under our patent laws everything (at the time it was new) could have been patented and held up for a period of twenty years. As long as the person who came up with it first applies for a patent, forget about using it for twenty years. Does this promote science and the useful arts? In a sideways kind of way, it might. But generally (and overwhelmingly in an area like software), it cripples innovation.
I have absolutely no idea whether this particular patent is valid. What I do know is that there is no reason in principle (under our patent system) that the person who first applied the notion of load-balancing to web-servers could not get a perfectly valid patent. What I see in this discussion is people trying desperately to explain why this particular patent may not be valid. The key is this: why do you want so much for it to be invalid? Because it would be wrong to stop programmers from applying a common-sense scheme like load-balancing to web-servers just because someone else was the first to run up against this particular problem. In other words, those who are trying to innovate can only hope that the patent system does not work as designed. In still other words, our patent system is wrong.
I hope you will all forgive me for venting, but I have been close to this problem for a long time, and I hope that the day will come when ordinary people come to realize the nature of our patent system and dismantle it (or at least gut it so that it is no more harmful than copyright).
-Steve, a recovering patent attorney
Democracy is a poor substitute for liberty.
The world would not be were it is today without despotism and monarchy, without slavery and religious oppression, or without war and arms races. Just because these things have been important, I would even say necessary, for taking us to the level we are at today, does not mean that we should hold on to them.
Their is am undeniable fundamental opposition between Intellectual Property and freedom of thought, and if we don't decide where we want to be on this issue, we are heading for big trouble. As long as the flows of information in our society were limited and controlled, the system of IP worked, and only then by bare necessity. When information flows freely, not only does the infringement on freedom become clearer: enforcement becomes impossible. The only thing hanging on to Intellectual Property will do to our society is to our data networks what the "War on drugs" has done to our cities.
In matters of freedom, fuck the economy.