Quote of the Day

[elfs]

Quote of the day: "It should be noted that no ethically-trained software engineer would ever consent to write a DestroyBaghdad procedure. Basic professional ethics would instead require him to write a DestroyCity procedure, to which Baghdad could be given as a parameter."

This is exactly why geeks don't "get" patent law. They understand that it exists and they can play it as well as anyone, but the underlying rules of patenting something don't make sense.

For example, the idea of patenting VOIP (Voice Over IP) confuses geeks because it's just packetizing something over TCP/IP: the hard part was finding the right abstraction for TCP/IP, VOIP is just a specific application of that great abstraction, and is technically "easy" compared to the thought processes that went into creating TCP/IP in the first place. Why anyone should get a patent for applying a parameter to a working procedure puzzles the hell out of us. The only "interesting" aspect to VOIP is that there's finally enough end-to-end bandwidth to make it viable.

Comments

[lucky-otter.livejournal.com]

There are actually other interesting aspects to VoIP. For instance, how do get the packets there as fast as possible over the real internet? It's not using TCP, that's for sure. Of course, using UDP or something, the packets may not arrive in order, so you need to deal with that. And packet drops. Do you want to dynamically scale the bit rate of the encoding as the bandwidth varies? Are you going to do that by using an encoding that gracefully degrades as you send/receive fewer chunks of it so it can all be done below the audio layer, or will you actually change how it's encoded?

Those are the simple parts, too. It gets harder when you consider multicasting. They're all questions which any competent network programmer could answer today, but I wonder if that was true when VoIP was initially developed. The non-obvious constraint for patents is the hardest one to apply.

[fallenpegasus]

All that stuff is, indeed, interesting. And it's all worked out in the RTP RFCs, the drafts of which all predate the Verizon patents.

The patent that Verizon really used to shut down Vonage with wasn't even that interesting. It was basically a patent on the concept of the ENUM mapping protocol, the RFC of which predates their patent by a number of years.

[gromm.livejournal.com]

Basic professional ethics would instead require him to write a DestroyCity procedure, to which Baghdad could be given as a parameter.

How is this related to ethics? This has much more to do with the complete unwillingness to do mounds and mounts of work over and over again.

the idea of patenting VOIP (Voice Over IP) confuses geeks because it's just packetizing something over TCP/IP:

Heh. I suspect that the real issue is that geeks think they know everything. Personally, as I make VOIP happen at work, I know that it's a non-trivial application. In fact, dozens have tried and failed. You could say the same thing about power distribution. It's simple to generate power for your house. Technically all that's needed is to hook a DC motor to something that spins and attach it to your house (assuming everything in it is DC). But the logistics of providing an entire town or state with electrical power is a non-trivial exercise. The thing with human beings though is that we're all terribly inclined to say "Oh, well that's *easy*!"

[brainchaos.net (from livejournal.com)]

This has much more to do with the complete unwillingness to do mounds and mounts of work over and over again.

Is that not the ethos of a well-trained software engineer?

And I don't believe Elf said VOIP is trivial. Only that, given an existing data-transfer protocol, it is obvious.

[lovingboth]

"A sort of furry animal" - a cat defined by a patent lawyer...

[rfreebern.livejournal.com]

I'd never thought about this, but you're so right. Neat. Thanks.

[srmalloy.livejournal.com]

With the recent unanimous SCOTUS decision on a patent infringement suit over a patent on an adjustable gas pedal, where the decision stated that incremental advances of an already patented entity are not automatically patentable, I'm waiting to see how long it takes for someone to realize that, with that decision, SCOTUS has gutted a chunk of the pharmaceutical industry. As an example, I take Hyzaar, which is a single pill containing both Cozaar (losartan potassium), a blood pressure medication, and HCTZ (hydrochlorothiazide), a diuretic. No generic exists yet for Hyzaar, because the patent on it has yet to expire, unlike Cozaar, for which the patent expired some time ago. Under the decision from SCOTUS, since losartan and HCTZ are often prescribed together, particularly for diabetics (the combination helps prevent loss of kidney function), Hyzaar as a single pill combining both medications would not meet the 'non-obvious' requirement to be patentable -- allowing other drug companies to immediately begin producing generic versions. There are a large number of 'combination' medications like this that could be thrown open for generic substitutes as soon as someone sees the implications and starts throwing patent challenges.