NTP v RIM (Research In Motion): maker of Blackberry

Note that in 1991 email was still in its infancy.

Research in Motion (RIM) started in 1995 as a pager company, but soon developed the Blackberry device for wireless reception of email. The RIM 900 came out in 1996, using first-generation wireless data networks. The RIM 950 came out in 1998, with the keyboard resembling making it vaguely resemble the blackberry fruit. The first modern Blackberry device was released in 1999 or 2000, with at least some genuine Internet connectivity. Blackberries did not become smartphones until 2002.

The early blackberries could have email sent to them, as if they were pagers that accepted email messages. This was not exactly a new idea; I almost bought a similar system (basically a pager, but capable of receiving email sent to a special address) in 1996.

In 2000 NTP notified RIM that it was infringing on NTP's patents. RIM thought otherwise. A lawsuit was brought shortly thereafter.

See http://www.spectrum.ieee.org/mar06/3087 (written before the final settlement).

On the one hand, NTP had indeed patented the idea of sending email to portable wireless devices. On the other hand, all the components for doing that (general email forwarding, IP over wireless links) were already well-known. There are a few design issues in how to forward email to devices that are essentially pagers, not full-fledged TCP/IP hosts, but those seem pretty trivial. And Blackberries had native Internet connectivity relatively early.

Campana died in October 2004.

RIM settled for $612 million in Mar 2006 , after an original verdict of $33 million. Part of the final settlement is that there are no ongoing royalty payments, so this isn't as egregious as it might seem. Still, the settlement amount just seemed to balloon.

The patent had been challenged with the USPTO. The PTO retracted one or two of the patents in Feb 2006, but the presiding judge in the case (Judge Spencer) refused to stay the case pending further USPTO findings. But the USPTO findings were moving against NTP. Here is a RIM update on the patent-office reexamination; it is not final and it is also a press release.

Just a day after a judge in the NTP-RIM patent fight said that he would not wait for the US Patent Office to complete a review of NTP's patents, that same Patent Office announced a "non-final" ruling on one of the NTP patents, suggesting that the original patent might not be valid. -- techdirt.com

Summary

There are two ways of viewing Campana's invention:

1. As any way of sending email wirelessly. The problem with this theory is that sending email over wireless links wasn't exactly novel. Sending email that way to "consumer" devices might be, but the real issue there was who was going to pony up the cash to build a wireless network for emails. That's more of a business innovation.
2. A specific way of sending wireless emails to a special device so they are received by a laptop attached to that device. The problem with this theory is that's not exactly what Blackberry built. Their devices received emails themselves, though early versions did also allow reading email on your laptop.

NTP argued that their system was a "push" system while existing systems were "pull" systems. As a description of an invention, this is malarkey. Once a device is connected, it receives the email. The only difference between wireless and modem communications is that, with the latter, a connection takes a good deal more work to set up.

RIM may have made some strategic errors in their legal case. Their law firm was slow, in a court that prided itself on speed. And one exhibit was falsely described. Inadvertent, probably, but still.

For a while, Judge Spencer threatened to shut Blackberry's network down with an injunction. The US government filed a petition that they needed Blackberries to work. Didn't matter.

In the end of the day, the patent office decided Campana's patent failed for 2 above, and also because it was not novel. They categorically rejected the idea that the business success of the Blackberry by itself pointed to its novelty. Blackberry, in the PTO view, simply chose to invest.

The case was before Judge James Spencer of the US District Court for Eastern Virginia, known as the home of the "rocket docket": part of the court's culture was strict adherence to timetables. RIM engaged the mega-lawfirm of Jones,Day, which irritated Judge Spencer regularly (probably with delay tactics). This was a classic Bad Idea. However, it also appears that Judge Spencer had little if any awareness of the central debates regarding software patents.

Business-method patents had just been approved in 1998, but Campana's patents did not mention these. Instead, Campana's patents were a confusing mix of system claims (for hardware) and method claims (for processes).

Case: NTP, Inc. v. Research In Motion, Ltd., No. 3:01CV767, 2003 WL 23100881 (E.D. Va. Aug. 5, 2003)

During the trial, RIM put on a demo of some supposedly pre-Campana text-messaging software, BUT a major part of it was post-Campana's-patent. Another bad idea. However, there actually was pre-Campana text-messaging software! It is even cited in the '592 patent.

RIM failed to prove the patent invalid in court. The burden of proof was on them.

The original verdict was for $33 million. RIM could have been liable for triple damages if infringement was found to be "willful". However, Judge Spencer compromised here on 42%, raising the award to $47 million. (Theoretically, a sincere belief in a patent's lack of validity is not a defense against "willful" infringement, though some judges do compromise here.)

RIM appealed Judge Spencer's decision to the Federal Circuit court.

The core problem with RIM: they did implement a system very much like Campana's idea. The only new idea is essentially that effort is made to make the blackberry completely transparent to your laptop's normal email software (ie blackberry email is eventually forwarded, through the blackberry, into your laptop, but is also viewable (and can be responded to) on the blackberry itself.

The core problem with NTP's wireless email patents: the only new element is the use of RF links in an email network. (The store-and-forward idea is acknowledged as prior art). BUT:

Prior art clearly includes the following: 

    1. Use of RF for IP links
    2. Email based on ANY set of IP links
    3. Use of RF to send short text messages to pagers

There is room for innovation in terms of identifying the nearest wireless hub, but that does not appear to be part of the patent. The patent does include lots of detail about switches and gateways, all of which appears to be standard according to RFC 821.

One central idea of the NTP patent: if a wireless unit is out of range, messages are stored at Network Operations Center (NOC). However, this idea is a fundamental part of the original SMTP standard, RFC 821 (August 1982). Specifically, RFC 821 implicitly calls for a forwarding node to hold messages when retrying to contact the next hop, and the final server in the line holds messages indefinitely when waiting for the user to connect and download email.

Whether or not this storage strategy was really the main idea, RIM seized on it when they lost their case; they developed an alleged "workaround" that handled message storage differently:
    http://www.businessweek.com/technology/content/feb2006/tc20060210_490726.htm
When you're out of range, RIM stores your messages at their Network Operations Center, or NOC. The workaround would store them on the senders' server, that is, the RIM NOC apparently wouldn't accept messages from sendersuntil the destination was available and prepared to accept delivery. Or something like that.

Note that if the alleged workaround still violated NTP's very vague patents, it would take another multi-year trial to establish that. One possiblility to consider is that the workaround was intended only as a negotiating tactic when trying to settle on a dollar amount for royalties: it would allow RIM to avoid a shutdown injunction and thus would give them some leverage. And, indeed, the settlement of $600 million wasn't much larger than what NTP had won in court with the added benefit of a RIM shutdown.

On the face of it, NTP's patents were for a SYSTEM, that is, for software. But the system claim is rather weak; all the components were pre-existing. The patent makes much more sense as a BUSINESS METHOD patent, but apparently this issue was never raised through the course of the trial.

One problem with a claim that NTP's patents were for a business method is that such patents were not granted until a 1998 court case. NTP's patents were applied for in 1991.

It seems surprising to me, but this distinction (between system and method patents) played little if any role in the trial. But that is perhaps because I see RIM's system as a business innovation, not a technical one.

Despite the well-established idea of store-and-forward delivery of email over any type of IP links, Judge Spencer allegedly claimed (I don't have a source for this, but remember reading it) that the case "wasn't even close". How could this be?

One possibility is that the judge saw the Campana patent as a non-IP way of delivering email over the last link. That is, neither NTP nor RIM was really bringing the Internet to handheld browsers; they were extracting email messages and then sending them over a proprietary protocol. That might make sense as a genuine invention.

The other view is that the judge saw Campana as patenting email to wireless devices, and RIM was selling exactly that. The problem here is that "email to wireless devices" barely qualifies as an invention at all.

However, we're again up against the "inevitability" issue here: the idea itself is obvious, and what RIM really brought to the table was the capital to build the appropriate wireless network.

Nobody disputes that RIM developed their system independently.

Patents for genuine hardware electronics methods are not an issue here: modulation techniques receiver--cpu interaction NTP's patents covered only the system, not the devices.

Real issue: patent system often ignores the fundamental CS principle of abstraction (ie building networks with any links) (If this was a business-method patent, that might make this issue irrelevant.)

• some of RIM's legal tactics and anti-patent arguments were seen as delaying and/or uncooperative
• The court is only interested in the legal issue at hand; once the verdict is in, the possibility that it is wrong no longer matters. The Supreme Court has tried to some degree to retain this legal principle even in death-penalty cases; nobody is going to chuck it for a patent case.

District Court issues

[quotes from NTP's Memorandum of Points and Authorities in Opposition to RIM's First (or Second) Motion for Summary Judgment]

1. RIM seemed focused on obtaining summary judgment before the "Markman hearing" (or claim-construction hearing), a hearing at which the judge rules on the meaning of various patent claims. This seems awkward. NTP's reply:

RIM's preemptive May 3. 2002 motion for partial summary judgment flies in the face of its prior arguments and representations to the Court. The motion is an attempt to short-circuit the Court's procedural schedule regarding claim construction and to burden NTP with briefing before RIM even files its responsive claim chart and prior art statement. RIM's motion - which RIM admits will likely "miss the mark altogether" - seeks to force the Court to engage in a wasteful, piecemeal, incomplete and ultimately fruitless claim interpretation exercise before the disputed claim terms are briefed or even identified.

2. RIM introduced the Zabarsky prior art (involving text-pager technology; below) rather late in the game. It is not clear why they didn't pick up on this earlier. Perhaps the idea of positioning their devices as pagers didn't occur to them, though they were originally a pager company.

Push me, pull you

3. NTP was an advocate of the "push" idea:

The Campana patents bridged this email-wireless divide by providing universal connectivity for email between wired and wireless systems. For the First time, email sent to a user at his or her normal electronic mail system could be "pushed" to the user's mobile processor in a format suitable for standard email operations such as viewing, replying and forwarding. The user no longer needed to find his or her email; instead, the email would find the user.

This "push" idea, though, is not new: it's what happens when a cellular network calls you (the call is "pushed" to you), and it's what happens in SMTP whenever the next hop is reachable. The last sentence sounds seductive, but again it is difficult to see the innovation here except in the context of actually building a wireless email network.

It is worth noting here (perhaps with some sadness) that the courts seemed to seize on this "push/pull" distinction as important. For example, another NTP description that makes the invention appear very deep is:

Campana opened access between the world of landline-based electronic mail systems and the wireless world. Campana taught the ability to "push" the email stored in the user's mailbox on the email server all the way to a mobile destination processor operated by that user. As Campana recognized, the wireless user would be unable to periodically request email because of all of the drawbacks cited above (e.g., uncertainty. delay and inconvenience/cost). Thus, the wireless user would be best served by a system that delivered email without the need for any request from the user - similar to the way that a server delivers email to the user's desktop computer when Outlook requests it.

But is this meaningful? It seems likely that anyone using RF links for email in 1995 would have found all the ideas here obvious. However, the notion that laptop users had to dial in to retrieve email was also something that many people felt was "inevitable".

NTP also focused on their system as a way of addressing the "inconvenience of dialing up": the difficulty finding a compatible phone jack, and the possibility that there was no email to be received. This seems to be a very tame description when compared to the utility of receiving email while walking around. Campana did not anticipate that part!

4: Note that all of the above amounts to a business justification for believing that there would be a market for small RF-capable email devices. If you accept that all the pieces for RF-based email were already in place, this can be interpreted as suggesting that the real innovation isn't the invention, but RIM's marketing of it.

5. RIM tried to claim their software was "not an email system", to evade Campana's patent claims that his patents covered an email system. RIM's claim might be true in some precise sense, in that it is clearly an add-on to an email system. RIM's precise claim was

The Blackberry relay and the Redirector software are not a part of any email system. They do not alter or modify any existing email system with which they may be used. Rather, they are peripheral components....

To which NTP replied:

The blackberry Redirector software is part of the email system on which it is installed. ... Indeed, RIM admits that the redirector software cannot operate independently and has no utility unless and until it is installed on an email server ....

Until now, RIM has always characterized the blackberry ... as part of an electronic mail system. ... Indeed, this Court is the only entity to which RIM has ever asserted that the Blackberry is not part of an electronic email system.

This is an interesting argument. If Campana's patent is for an entire system, what if RIM just implements the part that corresponds to Campana's addition? Generally, that would still qualify as infringing.

6. RIM tried to interpret some NTP claims very literally; for example, by claiming that the two phrases below (one referring to prior art) were equivalent. NTP argued that the crucial "the" made the difference. RIM's argument was that NTP's patent required both an RF and a wireline transmission to the same processor. Clearly, the Campana patent wording is very vague; in fact, it is vague even by the execrable standards of the patent world. Who is being obfuscatory here, RIM or NTP?

A system for transmitting originated information from one of a plurality of originating processors in an electronic mail system to at least one of a plurality of destination processors in the electronic mail system comprising...

a RF information transmission network for transmitting the originated information to at least one RF receiver which transfers the originated information to the at least one of the plurality of destination processors

7. RIM argued that NTP's patent required the RF receiver and destination processor to be distinct units. That seems specious, in that consolidating components is routine.

8. RIM tried to argue that their wireless units acted as pagers: email was not addressed to the blackberry itself, but only to the user's usual email address. The blackberry system intercepted the email and forwarded it to the blackberry.

First, this kind of forwarding was standard by 1990; most unix email systems provided for .forward files to specify such forwarding. However, RIM has a point: the Campana patents assume that the RF nodes have email addresses. This is not quite true of the Blackberry system: the email address is separate.

Note that the core issue of lack of novelty and obviousness was never raised directly. Indirectly, it appears in the guise that NTP's patents are to be construed very narrowly because the basic ideas were already extant.

Appellate court notes; ruling of Aug 2, 2005

Somehow, in the appellate case RIM is constantly on the defensive with relatively farfetched claims. What happened to the central idea that Campana's patents were about to be overturned? Unfortunately, this is a fact of life in appellate cases: you can't appeal the facts themselves, but only points of law (often in the form of an argument that the District Court misinterpreted the plain language of the patent).

In their appeal, RIM appears to be trying to narrow the scope of the NTP claims. Somehow the assertion that there was prior art for

• RF links for IP data
  
• email over ANY kind of link
  

Perhaps they should have focused on the patent's own claim of prior art in trying to limit the patent claims??

RIM argued before the appellate court that the district court erred in construing the claim terms:
    (a) "electronic mail system" (appearing in the '960, '670, and '172 patents);
    (b) "gateway switch" (appearing in the '960 patent); and
    (c) "originating processor" and "originated information" (appearing in the '960, '670, and '592 patents).

It seems clear that the district court didn't grasp the generality of any of these three terms, but the appeals court did NOT overrule. Instead they found

... the court looks to those sources available to the public that show what a person of ordinary skill in the art would have understood disputed claim language to mean

RIM argues there are two ordinary meanings of "electronic mail system": a broad definition that encompasses "communicating word processors, PCs, telex, facsimile, videotex, voicemail and radio paging systems (beepers)" and a narrow definition that defines the term in the context of "pull" technology. They were apparently trying to argue that the blackberry system was not an "electronic mail system" in the narrow sense, and therefore wasn't covered by the patents.

The appellate court cited Tanenbaum, Computer Networks, a classic text.

The court also wrote:

The message is next sorted by the recipient's ISP mail server into the recipient's particular "mailbox," where it is stored until the recipient initiates a connection with the server and downloads the message off the server onto his or her personal machine. This configuration is commonly referred to as a "pull" system because emails cannot be distributed to the user's machine without a connection being initiated by the user to "pull" the messages from the mail server.

Again the court buys into the idea that this simple bit of marketing terminology is somehow significant and important.

Campana's particular innovation was to integrate existing electronic mail systems with RF wireless communications networks.

A message originating in an electronic mail system may be transmitted not only by wireline but also via RF, in which case it is received by the user and stored on his or her mobile RF receiver.

Is the court suggesting that this is a "push" system? Yes, in fact. But this was a major misunderstanding of prior art.

The BlackBerry system uses "push" email technology to route messages to the user's handheld device without a user-initiated connection.

How is a blackberry different from a laptop with persistent Internet connectivity, limited to port 25 (email)?

An important issue for the court was that blackberries were NOT seen as email endpoints. Rather, they were seen as portable intermediate nodes: they would receive email, they could display the email, but the email's ultimate destination was the user's laptop (via some cable). This is an interesting strategy, in that it makes the blackberry transparent to the laptop that is receiving the email. However, it is not part of the patent debate either!

The appeals court agreed with the district court that the latter's interpretation of "electronic email system" as including the blackberry system was entirely reasonable. This degenerated into more push/pull debate, but part of the issue was that Campana himself tried to characterize an "electronic mail system" as a wire system in order to make his system appear different from Zabarsky's.

The appeals court said, [p 19 of pdf version]

Campana described prior art "electronic-mail services" as "basically a wireline - to - wireline, point-to-point type of system" (emphasis in the court's quote). The use of the term "basically" suggests that an electronic mail system may include other types of connections, including wireless connections. Moreover, Campana provided an example of one prior art electronic mail system in commercial use .... In this prior art electronic mail system, "groups of processors ... may be distributed at locations which are linked by the [PSTN]. The individual processors may be portable computers with a modem which are linked to the [PSTN] through wired or RF communications as indicated by a dotted line" [Campana quotes from 5436960]

Note that the appeals court is essentially granting here that RF links in email were prior art! See that patent, paragraph beginning "FIG. 1 illustrates a block diagram". Note also that, in the images, Fig 1 appears twice, and in one the "RF information transmission network" is deleted. RF links to end-users were never shown. Figures are in http://pld.cs.luc.edu/ethics/campana/960 and http://pld.cs.luc.edu/ethics/campana/451.

As for the contested term "originating processor", the appeals court says "We do not hold that the 'originating processor' is always the processor on which text of the email message was created".[p 23] That is, the blackberry is still an originating processor in the sense of Campana's patents even if the message was created on the associated laptop.

uspto.gov -> patents -> patft (uspto.gov/patft) Search by patent number:
http://patft.uspto.gov/netahtml/PTO/srchnum.htm

Look at patents

6198783 System for wireless serial transmission of encoded information Modulation techniques

6067451 Electronic mail system with RF communications to mobile processors. See the Appendix, under Background Art, for prior art. Note that in Figure 3, some of the underlying telecom infrastructure is shown ("closest LATA switches"). The first non-prior-art diagram is Figure 8 (page 9). (2 of 9 NTP v RIM claims)

Diagrams, and some text pages in .bmp format, are at http://pld.cs.luc.edu/ethics/campana.

6272190 System for wireless transmission and receiving of information and method of operation thereof

4644351: Zabarsky patent, possible prior art. Note that this is cited in the '592 patent. Paging is also cited there as prior art, in the paragraph beginning, "FIG. 2 illustrates a diagram of a prior art network"

A communications system for carrying messages via a radio channel between one central site of a plurality of central sites and a plurality of two-way remote data units is disclosed. Each central site has a radio coverage area and each remote unit has a unique address and association with one of the central sites. When a message addressed to one of the remote units is received in a central site, a file of remote unit addresses is searched to find the location and central site association of the remote unit to which the message is addressed. If an address match is found indicating that the remote transceiver is in the coverage area of the message-receiving central site, the addressed message is stored and transmitted in that site. If an address match is found indicating that the remote transceiver is in another central site, the addressed message is conveyed to that site for transmission.

This would seem to cover delivering text to specific end-users; eg paging, but also email through non-IP channels as well.

Here are two primary claims. The "patentese" is unfortunate and confusing, but the core claim in both cases is using RF links to transmit email.

claim 248 of patent 6067 451

246. In a system comprising a communication system which transmits electronic mail containing information, with the electronic mail being inputted to the communication system from a plurality of processors, a RF system and an interface connecting the communication system to the RF system with the information contained in the electronic mail and an identification of a RF device in the RF system being transmitted from the interface to the RF system and broadcast by the RF system to an identified RF device, the identified RF device comprising:

a RF receiver, which receives the information when the identification of the device is detected in a broadcast by the RF system to the RF receiver; and

a memory, coupled to the RF receiver, which stores the information received by the RF receiver contained in the electronic mail inputted to the communication system.

247. The RF device in accordance with claim 246 further comprising:

a processor, coupled to the memory, which after the information has been outputted from the memory, processes the information.

248. The RF device in accordance with claim 247 further comprising:

at least one application program, executed by the processor, which processes the information.

Fig. 8 is the first non-prior-art figure. It is described under the "BEST MODE FOR CARRYING OUT THE INVENTION" heading.

Certainly Campana appears to be patenting the use of RF links in email.

claim 150 of patent 6317592

150. In a communication system comprising a wireless system which communication system transmits electronic mail inputted to the communication system from an originating device which executes electronic mail programming to originate the electronic mail, mobile processors which execute electronic mail programming to function as a destination of electronic mail, and a destination processor to which the electronic mail is transmitted from the originating device and after reception of the electronic mail by the destination processor, information contained in the electronic mail and an identification of a wireless device in the wireless system are transmitted by the wireless system to the wireless device and from the wireless device to one of the mobile processors, the wireless device and one mobile processor comprising:

a wireless receiver connected to the one mobile processor with the one mobile processor receiving the information contained in the electronic mail after the identification of the wireless device is detected by the wireless receiver in a broadcast by the wireless system.

Patent reexaminations (from http://en.wikipedia.org/wiki/NTP,_Inc.)

• 90/006,491 filed on 12-26-2002
• 90/006,493 filed on 12-26-2002
• 90/006,494 filed on 12-26-2002
• 90/006,495 filed on 12-26-2002: '592 patent, decision 11/10/2009; final rejection of claims 1-764.
  
• 90/006,678 filed on 06-24-2003:  '670 patent
  
• 90/006,680 filed on 06-24-2003
• 90/006,681 filed on 06-24-2003
• 90/007,723 filed on 09-16-2005: 5,625,670 patent, rejection affirmed in part. Of NTP's 328 claims on appeal, however only 24 claims (...) include the requirement that the RF receiver has its own memory for storing received information. None requires a [visual] display in the receiver and none requires that the wirelessly transmitted [email] is received by the RF receiver when it is not attached to a destination processor.
  
• 90/007,726 filed on 09-22-2005: final USPTO rejection of '451 patent (claims 1-341, 393-437) affirmed December 3, 2009 (claims 342-392 were "cancelled"). On page 5, it is stated, NTP argues that we must adopt the claim interpretation applied by the US District Court ... in the infringement litigation between NTP and RIM involving [this] patent.... The argument is without merit. Also note the second paragraph on page 10: "We have not overlooked the fact .... To the extent that the USPTO interprets claim language broader than the E.D. Va. [Judge Spencer's court], NTP now lives with its litigation choice." Basically, the USPTO interpreted the claim broadly, found prior art for the broad claim, and rejected the patent.
  
• 90/007,735 filed on 09-28-2005: BPAI rejection 11/10/2009. This is the '172 patent. 

• 5,625,670
  
• 5,819,172
• 6,067,451
  

To look these up, start at http://portal.uspto.gov/external/portal/pair. Actual documents are under the "image file wrapper" tab. Don't forget "select new case" as appropriate!

In the '451 rejection, the patent examiners found multiple problems. Here is one relating to breadth of construction, from page 16:

As quoted above, NTP's counsel explained during oral hearing that NTP has described in its specification a gateway switch that is said to have additional functionalities than known types of gateway switches. That, however, does not cause prior art types of gateway switches to cease to be gateway switches. If NTP wanted the term "gateway switch" to cover only the specific gateway switch implementation, the one with additional functionalities as described in the specification, NTP was free to make a clear and deliberate limiting definition in the specification. It did not do so.

In other words, the patent's imprecise language cannot be allowed to work in its favor. NTP used the term "gateway switch" in important claims without making it clear that they had in mind a specific meaning.

On page 26, the Telenor 1989 patent is considered. Here the ruling is partly in NTP's favor. The reference is a tech report; does this qualify as a "publication"?

Stig Kaspersen et al, Norwegian Telecommunications Administration, Mobile Data Network Description (1989) (Volumes 1-4, 6-8 (there is no Volume 5))

The Perkins patent 5,159,592, 1992, "describes a system for coupling wireless migrating users to a network operating in accordance with TCP/IP type protocol". That is, Perkins described abstract wireless email delivery via TCP/IP.

Starting at page 36, the USPTO found that the "Perkins" prior art included all features of the NTP system, at least when the NTP system was "broadly construed". Also, NTP tried to argue that Perkins wasn't an "electronic mail system", by construing that term very narrowly.

See also page 114, regarding "obviousness". NTP made the following claims as to why their approach was an improvement:

1. the RF receiver is detachable from the destination processor and operates to wirelessly receive the email messages while it is detached...
2. the RF receiver includes its own memory to store the received email messages intended for the destination processor and does not require power from the destination processor to receive and store those messages.
3. the RF receiver provides reception and reveiew of email messages without need of the destination processor for which the email messages are intended.