Net2phone Sues Skype

robyannetta writes "Net2phone is suing Skype for patent infringement, arguing Skype violated patent 6,108,704 for 'the exchange of IP addresses between processing units in order to establish a direct communications link between the devices via the Internet.'"

DCC? Direct IM?

[ZxCv]

Any more examples of "the exchange of IP addresses between processing units in order to establish a direct communications link between the devices via the Internet.'"?

Maybe Net2phone will go after AOL and the guy who wrote mIRC, too.

Let me get this straight...

[bytesmythe]

They got a patent on dynamic routing tables? Applications have been saving databases of IP addresses for ages. Technically, any two devices that create an IP connection do this, since they save each other's IP addresses in a data structure that describes the connection parameters.

Un-freakin'-believable.

More like STUN

[Eivind]

This sounds more like STUN than like DNs really.

Normally, two computers that are both nated have no way of directly contacting eachother.

STUN is a method by which they still can, by getting information on ip and portnumber for the other side from a third source. (in this case the Skype-network)

If we both know, that we're listening for UDP packets on :33115 and you on :22056 then we can communicate, trough double NAT if need be by both sending a single packet to the other. This will cause the NAT-routers to set up mappings, so that any response packets coming from the other side will get trough.

Kinda sorta. For a more precise explanation see: http://en.wikipedia.org/wiki/STUN [wikipedia.org]

I wonder how the patent filing-date compares to the earlier draft-versions of the RFC...

Re:DCC? Direct IM?

[interiot]

Try any IP protocol out there... IP headers include the to/from address, so that the receiver of a message can turn around and reply. So RFC 791 is prior art for this.

If I got this patent right, it means...

[layer3switch]

comm link between two devices initiating PTP link by email from sender to middle-tier switch/server to query the route to the final destination which will respond with on/off status to middle-tier switch/server and back to original sender.

I don't think, the summary does the justice, however broad none the less. I'm not even sure if Skype works this way. But I think, the key word here is "E-Mail". I also think, it's safe to say, since it's using "E-Mail", there is no original invention here. It's just using whatever is already invented. So what's so patentable about this?

blah, it's late... I might be totally wrong on this.

Re:It's brilliant!

[larytet]

Skype is not a public company. eBay is, but the stock performs so-so for some time already

Re:It's brilliant!

[Vo0k]

Extra profit, no investment. Borrow 100 shares for $100 each, sell them for $100, get $10k. Sue, buy 100 shares for $50, spending $5k, return them, keep the other $5k - or invest them in another 100 shares which you'll sell for $200/piece after the court battle, resulting in $25k profit with zero up-front investment instead of $15k profit from $5k investment needed if you skip step one. Plus additional panic on the market resulting from mass sale of the shares will result in the shares reaching even lower price in step 3.

Re:I'm Confused

[Halo1]

You're quote the summary paragraph, which is (largely) irrelevant as far as what is covered by the patent. You have to look at the claims. The independent claims (claims which do not refer to another claim) are the broadest, and often claim 1 is the broadest of them all. In this case:

1. A computer program product for use with a computer system, the computer system executing a first process and operatively connectable to a second process and a server over a computer network, the computer program product comprising:

a computer usable medium having program code embodied in the medium, the program code comprising:

program code for transmitting to the server a network protocol address received by the first process following connection to the computer network;

program code for transmitting, to the server, a query as to whether the second process is connected to the computer network;

program code for receiving a network protocol address of the second process from the server, when the second process is connected to the computer network; and

program code, responsive to the network protocol address of the second process, for establishing a point-to-point communication link between the first process and the second process over the computer network.

It's indeed quite possible that IRC is prior art for this. And although it corresponds roughly to the summary paragraph in this case, this is often not the case.

That said, what's most objectionable about this is that it's a software patent, not so much whether or not it's new or non-obvious. After all, there's no macro-economic rationale for having software patents, just some legalistic arguments and some based on "natural rights" (although the patent system is an economic policy tool, and not something to "reward" or "justly treat" people; "sweat of the brow" is neither a sufficient nor a necessary condition for obtaining a patent).

Re:It's DNS!

[hypnagogue]

RFC dated April 1997, based on draft documents dating back to February 1995, based on public work on Dynamic DNS dating from 1994. This one is a no-brainer, overwhelming prior art exists -- DynDNS doesn't begin to touch it.

Take a look at BOOTP from 1985 (client transmits MAC address to BOOTP server announcing it's presence, BOOTP server returns an IP address, routed and ARP complete the query portion transforming public presence identity, in this case a publicly routable IP address, to the MAC address).

Prediction: Net2Phone loses case, Skype loses money, lawyers make money.