Opinion
Civil Case No. 06-cv-00102-LTB-CBS.
January 22, 2008
ORDER
In this patent infringement case, Plaintiff, Tandberg Data Corporation, asserts that Defendant, Hewlett-Packard Company, infringed upon its U.S. Patent No. 5,349,481 (the `481 patent). Hewlett-Packard counterclaims that Tandberg infringed upon its U.S. Patent No. 5,761,032 (the `032 patent), and U.S. Patent No. 6,469,850 (the `850 patent). These patents concern technology related to computer data storage devices.
The matter is now before me on the parties' cross-memorandums in support of their respective proposed constructions related to the disputed claim terms in the `481 patent and the `032 patent. I note that after filing, the parties have agreed on the definition of a previously disputed claim term in the `850 patent. In addition to the parties' briefs and related attachments, I have the benefit of the demonstrative evidence and oral arguments presented at a hearing on January 4, 2008. As such, I make the following determinations of claim definition.
I. LAW
Patent infringement cases requires a two-step analysis. See CCS Fitness, Inc. v. Brunswick Corp., 288 F.3d 1359, 1365 (Fed. Cir. 2002); Johnson Worldwide Assocs. v. Zebco Corp., 175 F.3d 985, 988 (Fed. Cir. 1999). At issue here is the first step which requires my determination, as a matter of law, of the correct scope and meaning of disputed claim terms which determine the extent of the patent grant. Markman v. Westview Instruments, Inc., 52 F.3d 967, 979-81 (Fed. Cir. 1995), aff'd, 517 U.S. 370, 116 S.Ct. 1384, 134 L.Ed.2d 577 (1996); see also Lizardtech, Inc. v. Earth Resource Mapping, Inc., 35 Fed. Appx. 918, 923 (Fed. Cir. 2002).As a "baseline" rule, the words of a claim are generally given their ordinary and customary meaning. Phillips v. AWH Corp., 415 F.3d 1303, 1312 (Fed. Cir. 2005) ( en banc). "[O]rdinary and customary meaning of a claim term is the meaning that the term would have to a person of ordinary skill in the art in question at the time of the invention." Id. "In some cases, the ordinary meaning of claim language as understood by a person of skill in the art may be readily apparent even to lay judges, and claim construction in such cases involves little more than the application of the widely accepted meaning of commonly understood words." Id. at 1314. However, because the meaning of a claim term as understood by persons of skill in the art is often not immediately apparent, and because patentees frequently use terms idiosyncratically, the court looks to "those sources available to the public that show what a person of skill in the art would have understood disputed claim language to mean." Innova/Pure Water, Inc. v. Safari Water Filtration Systems, Inc, 381 F.3d 1111, 1116 (Fed. Cir. 2004).
In assessing the correct scope and meaning of disputed claim terms, I am first to look at the intrinsic evidence of record; the patent itself, including the claims, the specification and, if relevant and in evidence, the prosecution history. Interactive Gift Exp., Inc. v. Compuserve Inc., 256 F.3d 1323, 1331 (Fed. Cir. 2001) ; Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1582 (Fed. Cir. 1996) ("intrinsic evidence is the most significant source of the legally operative meaning of disputed claim language"). The hierarchy among the intrinsic evidence means that I start with the claim language. Id.; see also Phillips v. AWH, supra, 415 F.3d at 1312 (the court first "looks to the words of the claims themselves . . . to define the scope of the patented invention"). I then look to the rest of the intrinsic evidence, beginning with the specification and concluding with the prosecution history. Interactive Gift Exp. v. Compuserve, supra, 256 F.3d at 1331.
If the claim language is clear on its face, consideration of the remaining intrinsic evidence is restricted to determining if a deviation from such clear language is specified. A deviation may be necessary if: 1) "a patentee [has chosen] to be his own lexicographer and use terms in a manner other than their ordinary meaning" or 2) if a patentee has "relinquished [a] potential claim construction in an amendment to the claim or in an argument to overcome or distinguish a reference." Interactive Gift Exp. v. Compuserve, supra, 256 F.3d at 1331 ( quoting Vitronics v. Conceptronic, supra, 90 F.3d at 1582; Elkay Mfg. Co. v. Ebco Mfg. Co., 192 F.3d 973, 979 (Fed. Cir. 1999)); see also Phillips v. AWH, supra, 415 F.3d at 1316.
If the meaning of the claim limitations is apparent from the totality of the intrinsic evidence, then the claim has been construed. Id. If, however, a claim limitation is still not clear, I may look to extrinsic evidence to help resolve the lack of clarity. Id. at 1317. Extrinsic evidence "consists of all evidence external to the patent and prosecution history, including expert and inventor testimony, dictionaries, and learned treatises." Id. ( quoting Markman v. Westview Instruments, supra, 52 F.3d at 980). For example, technical dictionaries may provide a court better understanding of the underlying technology and the way in which one of skill in the art might use the claim terms. Phillips v. AWH, supra, 415 F.3d at 1318. Likewise, extrinsic evidence related to expert testimony can be useful to a court for a variety of purposes, "such as to provide background on the technology at issue, to explain how an invention works, to ensure that the court's understanding of the technical aspects of the patent is consistent with that of a person of skill in the art, or to establish that a particular term in the patent or the prior art has a particular meaning in the pertinent field." Id.
Finally, I note that "[t]he inquiry into the meaning that claim terms would have to a person of skill in the art at the time of the invention is an objective one." Innova/Pure Water v. Safari Water Filtration Systems, supra, 381 F.3d at 1116.
II. TANDBERG'S `481 PATENT
Tandberg's `481 patent is entitled "Apparatus and Method for Distorted Track Data Recovery by Rewinding and Re-Reading the Tape at a Slower Than Normal Speed." The patent relates to both an apparatus and a method for error recovery when reading information or data recorded and stored on magnetic tapes.
Magnetic tape (or the "storage medium") is a thin plastic ribbon coated with a thin layer of magnetic material used to store computer data. The magnetic tape is rolled onto a hub and stored in a cartridge that also houses another hub on which the tape is unwound.
Recording or writing onto a magnetic tape is done by "write heads" that selectively energize the tape, in response to an incoming data stream, leaving a data track. As is relevant here, the data tracks are written onto the magnetic tape diagonally at an angle to the tape's edge, resulting in "helical tracks." Helical tracks are parallel rows of magnetic signals written diagonally across the magnetic tape, with each track written by a single pass of a write head. "Reading heads" are devices that detect the magnetic signals. Read heads are mounted to a rotating drum, which traverses the moving tape along "azimuthal paths."
Tape drives, which are used to read the data stored on the magnetic tapes, contain error-correcting systems to attempt to overcome errors during the reading process. Patent `481 relates to the reading of a magnetic tape by a tape drive and a method by which the drive can recover a read error. Specifically, when an error is encountered, the tape drive slows the magnetic tape and passes the read head over the tape at a slower speed in order to attempt to recover from the error.
A. Disputed Claim Terms
The relevant claim language at issue is set forth in claim 1 of the `481 patent, which provides the following:
Apparatus for reading information recorded on a storage medium in helical tracks, with each track comprising a plurality of blocks of data with each block having unique block-identifying information, said apparatus comprising:
a transport for selectively transporting the storage medium in forward and reverse directions of medium travel;
a rotatable drum having a portion of a peripheral surface thereof contiguous with the storage medium as the storage medium is transported there past by the transport;
a reading head mounted on the drum for traversing original azimuthal paths across the storage medium and for reading blocks including the block-identifying information recorded along at least a portion of each of the original azimuthal paths;
a controller which uses the block-identifying information to determine whether any read errors occurred during the traversal of the original azimuthal paths, and which requires that the read head re-read portions of the storage medium where the error occurred by traversing modified azimuthal paths, the modified azimuthal paths being separated from one another by a distance less than the track pitch. [11:4-29]
The bolded terms are those terms at issue here that require construction or interpretation.
B. Block of Data Unique Block-Identifying Information
Tandberg asserts that a "block of data," as set forth above, is a "complete and independent group of magnetic signals" written to a magnetic tape. Tandberg further maintains that the "unique block-identifying information" — as assigned to each block of data — is "information associated with a block of data in a track that distinguishes it from other data blocks within that track." Conversely, Hewlett-Packard asserts that a "block of data" consists of "a complete and independent unit of data which can be read independently of any other data." Hewlett-Packard further asserts that "unique block-identifying information" is information located in the header of a block of data "which distinguishes that block of data from all other blocks of data on the tape."
In essence, Tandberg is asserting that a block of data is solely a unit of information, while Hewlett-Packard maintains that a block of data is a unit of information that, in addition, is distinct or unique from all other data. As such, it is Tandberg's position that a block of data is different from other blocks of data contained on the same tract only, while Hewlett-Packard's construction is that a block of data is different from other blocks of data on the entire magnetic tape.
I start my analysis with the language of the claim which provides, in relevant part, patent protection for an "[a]pparatus for reading information recorded on a storage medium in helical tracks, with each track comprising a plurality of blocks of data with each block having unique block-identifying information . . ." [11:4-7] In this case, the claim language that specifies that "each track comprising a plurality of blocks with each block having unique block-identifying information" indicates that, at the least, the block-identifying information must be unique to each block in its track.
In support of its proposed construction that the block-identifying information must be unique to every block of data on the entire magnetic tape, Hewlett-Packard refers me to the following language in the specification: "[e]ach block is a complete and independent entity which can be read independently of any other data blocks." [5:32-34] This language comes from "one embodiment of the invention" as illustrated in Figure 3. [5:8-9] Figure 3 shows four separate helical tracks, each track containing eight blocks of data, and each of the blocks of data illustrated has a unique identifier to all of the other blocks depicted.
In addition, Hewlett-Packard also maintains that each block of data on a tape must have unique identifying information because such requirement is necessary to the read error detection method and the re-reading operation of patent `481. Hewlett-Packard asserts that from the specification, each of the blocks must be independently readable because the identifying data from each block must be stored separately in an allocation table "as each block is read" and because the purpose of the disclosed re-reading operation is to "read previously-missed blocks on the track." [3:21 10:67] Hewlett-Packard urges that the specification serves to "define the term by implication." See Irdeto Access, Inc. v. Echostar Satellite Corp., 383 F.3d 1295, 1300 (Fed. Cir. 2004) ("[e]ven when guidance is not provided in explicit definitional format, the specification may define claim terms by implication such that the meaning may be found in or ascertained by a reading of the patent documents"); see also Bell Atlantic Network Services, Inc. v. Covad Communications Group, Inc., 262 F.3d 1258, 1273 (Fed. Cir. 2001) (finding the patentees defined a term by implication via its consistent use throughout the patent specification).
In response, Tandberg argues that there is nothing inherent about the method of patent `481 that dictates a conclusion that all the blocks on the tape must be unique — only that they must be unique within a track in order to detect read errors within that track. In addition, Tandberg notes that a possible embodiment of the invention, as is relied on here by Hewlett-Packard, cannot be used to limit the claim definitions at issue. See e.g. Agfa Corp. v. Creo Products Inc., 451 F.3d 1366, 1376 (Fed. Cir. 2006) ("this court has repeatedly rejected the contention that depiction of a single embodiment in a patent necessarily limits the claims to that depicted scope"); SanDisk Corp. v. Memorex Products, Inc., 415 F.3d 1278, 1286 (Fed. Cir. 2005) ("without more the court will not limit claim terms to a preferred embodiment described in the specification").
I agree with Tandberg that the language of the claim, which is the baseline for my analysis, in and of itself supports only a conclusion that the unique requirement of the blocks of data is limited to each track. I cannot read the unambigious claim language any other way. To the extent that Hewlett-Packard asserts that all blocks of data contained on a entire tape must be unique in order for the read process disclosed in the specifications to work, I conclude that while the preferred embodiment of the invention disclosed in the specification describes a process that might optimally operate with each block on a tape containing unique information, I cannot read the specification as contemplating or mandating such requirement. Although Figure 3 illustrates four adjacent tracks with unique information to every block, it does not necessarily generalize to a conclusion that the patent requires such unique information to all the blocks of data on a tape.
I am also not persuaded by Hewlett-Packard's argument that the consistent use of the word "unique" throughout the patent without the modifier "to a tract" defines the term by implication under Bell Atlantic v. Covad Communications, supra. However, to the extent that Hewlett-Packard asserts that the unique identifying information must be contained in the "header" portion of the block of data, I note that in the summary portion the patent describes that "[e]ach tract comprises a plurality of blocks of recorded data, with each block having unique blockidentifying information provided in a block header." [3:10-13] The specification is consistent in its depiction of such information in the header portion of the block of data. [5:40]
Therefore, I conclude that in patent `481 a "block of data" is a complete and independent group of magnetic signals written to a magnetic tape, and that "unique block-identifying information" is information associated with a block of data in a track that distinguishes it from other data blocks within that track, and contained in the header portion of the block of data.
C. Reading
The parties next dispute the construction of the word "reading" as related to the `481 patent. Tandberg asserts that it means merely "acquiring magnetic signals from the tape," while Hewlett-Packard maintains that it should be construed to mean "detecting and processing data stored in blocks on the tape, including stripping header information and storing it in an allocation table, and saving user data into memory for presentation to a user or utilization device." Hence, Tandberg maintains that "read" should mean only the acquisition of magnetic signals from the tape by the read head, while Hewlett-Packard contends that reading includes both the acquisition and the processing of the signals acquired.
The relevant language of claim 1 of patent `481 discloses an "[a]pparatus for reading information recorded on a storage medium in helical tracks . . . said apparatus comprising" of many things, including: "a reading head mounted on the drum for traversing original azimuthal paths across the storage medium and for reading blocks including the block-identifying information recorded along at least a portion of each of the original azimuthal paths." [11:4-20]
Tandberg asserts that the claim language supports a determination that reading is merely the initial step of the process, comprised of only the acquisition of signals, and that the remainder of the claim then describes the processing steps after the initial reading or acquiring of signals. In support of its proposed construction — that the term reading includes acquiring and processing data — Hewlett-Packard argues that the term "reading information" set forth in the claim language means the steps of taking the data off the magnetic tape and then processing it by stripping header information and storing it in an allocation table, and saving user data into memory for presentation to a user.
Hewlett-Packard maintains that "read" or "reading" is used throughout the specifications to describe the act of processing information — as opposed to just acquiring data — in that the specification consistently refers to "reading" as including both the detection and processing of data. First, the summary portion of the patent notes that "[a]s each block is read, the block-identifying information is stripped from the block header" and then goes on to describe the steps necessary to process the information or data obtained. [3:21-30] See C.R. Bard, Inc. v. U.S. Surgical Corp., 388 F.3d 858, 864 (Fed. Cir. 2004) ("statements that describe the invention as a whole, rather than statements that describe only preferred embodiments, are more likely to support a limiting definition of a claim term").
Hewlett-Packard further maintains that this proposition is echoed in the body of the specification, which describes the preferred embodiment of the invention as illustrated in the drawings. For example, in describing Figure 5 the specification states that: "[t]he read head is connected to apply signals read from the tape to read circuits clock (See FIG. 5). Data to be read is received at the read circuits and, together with a clock signal, is coupled to a read demodulator and deserializer and to a read deformatter. In the read deformatter, certain header and referencing signals are removed from the data stream so that the recovered user data can be applied in block form to a data buffer." [6:19-30] (emphasis added). Hewlett-Packard argues that the "data to be read" language indicates that data is first obtained, and then processed by being "read." Additionally, Hewlett-Packard refers to Figure 8, which depicts a flowchart that "illustrates steps executed by the controller in connection with the reading operation of the drive system." [7:41-43]
In response, Tandberg argues that the inventors did not limit the term "read" in the claim to include processing, and that Hewlett-Packard's improper attempt to do so comes from a passage in the specification that describes only a single embodiment of the invention. Tandberg also asserts that the specification indicates that "upon each rotation of the drum, the read head acquires signals from eight blocks . . .". [7:20-21] Furthermore, it maintains that Figure 8 merely depicts "Read Tape" as a single, initial step before other actions are accomplished.
The language of claim 1 does not contemplate the process of analyzing the data read or acquired by the read head. Although processing the data is a necessary step to the invention, the exact or particular way that data is processed as described in the patent summary and specification is not claimed as part of the patent. See Smith Nephew, Inc. v. Ethicon, Inc., 276 F.3d 1304, 1311 (Fed. Cir. 2001) ("[a] claim is not defective when it states fewer than all of the steps that may be performed in practice of an invention"). Including the steps of the data processing into the "read" term would constitute importing a described, but unclaimed, preferred embodiment into the patent by improperly expanding the definition of read. See Acumed LLC v. Stryker Corp., 483 F.3d 800, 805 (Fed. Cir. 2007) (describing an improper "attempt to import a feature from a preferred embodiment into the claims"); see also Phillips v. AWH, supra, 415 F.3d at 1323 ("although the specification often describes very specific embodiments of the invention, we have repeatedly warned against confining the claims to those embodiments"). I conclude that in the context of patent `481 the term "read" is different, and has different meaning, than the term "reading operation" as the process depicted in the illustrations and flow charts of the preferred embodiment in the specification.
I am also not persuaded by HP's contention that the extrinsic evidence, in the form of technical dictionary definitions, supports a construction that the word "reading" in claim 1 of patent `481 includes processing the data acquired. See generally Itel Corp. v. VIA Techs., Inc., 319 F.3d 1357, 1367 (Fed. Cir. 2003) ("[w]hen an analysis of intrinsic evidence resolves any ambiguity in a disputed claim term, it is improper to rely on extrinsic evidence to contradict the meaning so ascertained"). In the context of this patent, I conclude that the intrinsic evidence reveals that the claim language contemplates "read" and "reading" to mean acquiring magnetic signals from the tape.
D. Read Error
The meaning of a "read error," as set forth in the `481 patent, is also in dispute. Tandberg asserts that a read error is the "failure to confirm whether the data in a track was accurately read," while Hewlett-Packard argues that a read error is the "failure to accurately read a block of data."
The claim language in claim 1 of patent `481 describes an apparatus that includes, among other things, "a controller which uses the block-identifying information to determine whether any read errors occurred during the traversal of the original azimuthal paths." [11:21]
Hewlett-Packard's argument in support of its definition that a read error constitutes a failure to accurately read a block of data is, in essence, that a read error occurs when a block identification is not found in the allocation table. Hewlett-Packard again refers to the operation or process outlined in the summary portion of the specification and the description of the preferred embodiment which outlines the steps for processing that data. Specifically, Hewlett-Packard points to the language in the specification where it describes a case in which a block identification is not located in the allocation table and, as a result, an attempt is made to re-read the tape. [3:44-48 and 8:9-19]
In response, Tandberg argues that the claim language does not limit a "read error" to the process described in the specification. Tandberg also asserts that Hewlett-Packard's attempt to limit a read error to a single block of data is too narrow; that while a failure to read a block of data could result in a read error, it is not necessarily true. It maintains that a mis-read of a single missing block is just one embodiment described in the specifications. [3:44-46] Tandberg asserts that the entire invention is directed toward the recovery of an error in the helical track of a magnetic tape. Tandberg notes that the `481 patent is titled "Apparatus and method for distorted track data recovery . . ." [1:1], and points to several places in the specifications where distorted track data is referenced, as opposed to any error or distortion in an individual block of data. [ See 3:7-10, 5:66, 3:1, 10:51]
I agree with Tandberg that HP's proposed definition of read error — as the "failure to accurately read a block of data" — is reliant on an erroneous analysis or processing of the data read. Whether or not a block of data is found in the allocation table is a non-claimed processing step in patent `481. As such, I determine that a "read error" means the failure to confirm whether the data in a track was accurately read.
E. Re-read
Finally, the parties dispute the meaning or construction of the term "re-read" in patent `481. Tandberg argues that it means "to read again," while Hewlett-Packard asserts that is means "to repeat the read operation performed during the first read attempt."
Claim 1 of patent `481 describes an apparatus that includes, among other things, "a controller which uses the block-identifying information to determine whether any read errors occurred during the traversal of the original azimuthal paths, and which requires that the read head re-read portions of the storage medium where the error occurred . . ." by doing so at a slower speed, in that the read is done on modified azimuthal paths. [11:21-26]
Hewlett-Packard's proposed construction requires a repeat of the initial reading operation, and thus mandates a second attempt to read the user data recorded on the magnetic tape. Tandberg agrees that the patent teaches a second read of the tape, but argues that it is a "different" read in that it goes slower and the azimuthal paths taken by the read head are closer together. Hewlett-Packard agrees that the second read (or re-read) occurs at a different speed. Rather, it maintains that the re-read must accomplish what was done on the first or initial read — that is, an attempt to read the user data and make it available to a user or utilization device.
Again, because I have determined that read means acquiring the magnetic signals only, I further conclude that re-read means a subsequent attempt to acquire the missed signals. Any processing of the signals or data is an unclaimed step of patent `481. However, I agree with Hewlett-Packard that the second or subsequent attempt to acquire the missed data requires an attempt to acquire signals from the same data that was initially read incorrectly or where the data was not accurately read. The final paragraph of the specification — just prior to the various claims — notes that various alterations in form and detail may be made without departing from the spirit and scope of the invention, and then gives an example of changing the speed of the slow tape read "in order for the read head to read previously-missed block on a track." [10:67-68]
Therefore, the term "re-read" in patent `481 means "to read again" which, in turn, means to attempt to acquire magnetical signals from the data that was initially read incorrectly.
III. HEWLETT-PACKARD'S `032 PATENT
Hewlett-Packard's `032 patent is entitled "Computer Structure With Modular Housings." The patent protects a frame structure containing a library system holding multiple drives used to read magnetic tapes. The drives are mounted within housings that are detachably mounted for easy removal from the frame or the "primary structure." Each housing has its own handle and cooling fan. The structure also contains a robotic media inserter which is a device capable of moving storage media, such as magnetic tapes, from a drive to a storage location and then back again.
A. Disputed Claim Terms
The relevant claim language is set forth in claim 1 of the `032 patent, which provides as follows:
A multiple drive library computer structure, comprising:
(a) a primary structure to which is mounted a number of electronic and mechanical subsystems, said subsystems including a robotic media inserter;
(b) two or more modular drive housings, detachably mounted to said primary structure, on each of which is mounted a cooling fan, and in each of which are mounted one or more SCSI storage drives, each SCSI storage drive having an electrical connection face opposite a media-inserting face, wherein each of said modular drive housings holds the media-inserting faces of its SCSI storage drives in abutment to, and in cooperating engagement with, the robotic media inserter. [8:41-55]
The bolded terms are those terms that require construction or interpretation.
B. Modular Electronic and Mechanical Subsystems
Hewlett-Packard proposes that the term "modular" should be construed to mean "of standardized dimensions to facilitate removal and replacement." Tandberg proposes that the terms be construed to mean "removable without interrupting system operations." Likewise, Tandberg further asserts that "electronic and mechanical subsystems" consists of a group of components such that SCSI drives (the type of tape drive used here) may be removed from the primary structure without interrupting system operation. Hewlett-Packard argues, however, that "electronic and mechanical subsystems" merely means an electronic and mechanical "component or collection of components organized to perform a function or set of functions."
The relevant language from claim 1 of patent `032 provides that the invention consists of a multiple drive library computer structure comprising of "a primary structure to which is mounted a number of electronic and mechanical subsystems . . ." and "two or more modular drive housings."
Hewlett-Packard asserts that this language supports only the application of the ordinary meaning of modular which it argues is "designed with standardized units or dimensions, as for easy assembly and repair or flexible arrangement or use." Likewise, Hewlett-Packard argues that "electronic and mechanical" requires no construction and the parties stipulate that a subsystem is a component or collection of components organized to perform a function or set of functions.
Tandberg acknowledges that the language of the claim does not, in and of itself, support the limitation it proposes; specifically that modular must mean "removable without interrupting system operations" or that the term "electronic and mechanical subsystems" means a group of components such that drives may be removed from the primary structure without interrupting system operation. Tandberg's proposed construction requires that the claim must be read to incorporate the primary feature of the invention; specifically that the structure allows for individual drives to be removed (for repair, replacement, etc.) without interrupting the system operations. In other words, the removal of one drive would not interrupt the use of the remaining drives in the system. This ability is commonly called "hot swapping" or "hot or online repair."
Hewlett-Packard maintains that the term "modular" — which it notes is only associated with the housing and not the hard drive — solely means "ease of removeability." As such, while the ability to remove and replace a drive while other drive remain running (hot swapability) is only one way the library system disclosed in the patent is modular. Hewlett-Packard asserts that the patent also covers or contemplates another way to achieve modularity; specifically, to turn off the system, take out a particular housing, and then turn the system back on. Hewlett-Packard argues that these two aspects of modularity in the system are illustrated by the language in the background and summary portions of the specification. The "Background of the Invention" portion of the specification sets forth the objectives of the invention. Hewlett-Packard notes that the first relevant paragraph relates to the ability to remove a modular housing by turning off the system; it provides that "a primary object of this invention [is] to significantly reduce the down time of a multiple drive library computer system upon the failure of one or more of the library's storage drives. In the case of a removable media type library, it is believed that library down time can be eliminated." [2:1-5] Hewlett-Packard then asserts that the following paragraph contemplates hot swapability as an additional objective of the invention; it provides that "[i]t is a further object of this invention to allow a host to have continual access to one or more good storage drives of a multiple drive library computer system while one or more failed drives of the library are repaired." [2:6-9] As such, Hewlett-Packard maintains that hot swapability is only one benefit — not a patent requirement — of the modular library system disclosed. Hewlett-Packard says that the patent claims only the structure, which is modular in nature, and that hot swapability is simply an advantage disclosed in one way to use the system.
In contrast, Tandberg argues that the specification language clearly requires the hot swapability limitation, via the term "modular" or the term "electronic and mechanical subsystems," by a clear and intentional disclaimer or disavowal of claim scope. Regarding the term "modular," Tandberg first refers to the abstract of patent `032 which provides that: "the frame structure with modular housings allows good drives in a multiple library drive system to remain online, and accessible to the host, during a repair of one or more failed drives." In the background section of patent `032, the specification describes the invention as a new computer frame structure with modular housings where "repair of a failed drive may occur while one or more good drives in the library system remain online. In this manner, a host may have continual and uninterrupted access to information stored within the library." [1:7-14] The specification also describes "a purpose" of this invention as "an online repairable system . . ." [4:63-64] In "one method of repairing one or more drives . . .", "[t]he failed drive may be serviced while the library system remains online." [6:39, 6:65-7:9] Finally, the "method of repairing" a failed drive outlined in the specification is described as an "unconventional manner" for removal of a "modular drive" when the "robotic media inserter is not interrupted as it continues to feed media to the library's good drives." [6:39 7:37-44].
Tandberg also relies on the prosecution history of patent `032 which it maintains estops Hewlett-Packard from arguing that the claims are not limited to a "hot" or "online" repairable system. The patent examiner first rejected patent `032 on the basis that the Corfits Patent (U.S. Patent No. 4,853,830) disclosed the same type of library frame and "[i]t would have been obvious to one of ordinary skill in that art . . . to modify a state of the art multiple drive library computer system to include modular drive housing to improve servicing as taught by Corfits." As a result, the applicant responded as follows:
Applicant does not believe that Corfits obviates his claims for the following reasons. First, application of the Corfits teachings to the instant application would destroy the function of, and prevent the intended use of, applicant's online repairable drive system and associated structure.
Corfits teaches how to align electrical connectors to facilitate completion of a "blind" electrical connection (column 1, lines 6-10). In an online repairable drive system, storage drives are mounted so that their media-inserting faces are blindly abutted to a robotic media inserter. In this manner, the robotic media inserter may insert and/or remove media from the storage drives.
If the teachings of Corfits were adapted to the online repairable drive system disclosed, a blind abutment of storage drives and robotic media inserter would not be possible while maintaining the entire system in an online state. It is impossible to make both a blind electrical connection and a blind robotic abutment in an online repairable drive system.
Second, the instant application relates to online drive repair, and more specifically, to the online repair of a single SCSI storage drive contained in a multiple drive library computer system. In providing a solution to the problem of online drive repair, problems related to the removal and cooling of SCSI storage drives needed to be addressed. Prior to the disclosure at issue, a means of removing a single SCSI storage drive from a library of storage drives, while keeping the other SCSI storage drives online, did not exist.
As a result, Tandberg asserts that the prosecution history results in a "clear and unmistakable disavowal" of claim scope and acts to limit the claims of patent `032 to a hot repairable system. Tandberg argues that while the most logical place and best supported location for reading the hot or online repair system limitation into the claims is through the term "modular," the limitation should also be read into the phrase "electrical and mechanical subsystems" based on the same reasoning. See generally MBO Laboratories, Inc. v. Becton, Dickinson Co., 474 F.3d 1323, 1330 (Fed. Cir. 2007) (interpreting the term "immediately" to include an essential feature of the invention as set forth in the specification and the prosecution history).
Hewlett-Packard asserts that its response to the rejection based on the Corfits patent is mischaracterized by Tandberg. Hewlett-Packard argues that the response served to amend the claims to add: (1) a robotic media inserter; (2) a cooling fan to the drive housings; and (3) positioning the media inserting face of the SCSI drive "in abutment to, and in cooperating engagement with, the robotic media inserter." As such, Hewlett-Packard argues that there is no clear disavowal of the claim scope that serves to narrow the claims of patent `032 to a hot swapable system only. See Middleton, Inc. v. Minnesota Mining Mfg. Co., 311 F.3d 1384, 1388 (Fed. Cir. 2002) (requiring a clear and unambiguous disclaimer in the prosecution history). I disagree.
My reading of the patent history and the examiner's rejection based on the Corfits patent supports Tandberg's contention that the hot swapability aspect of patent `032 is not just a benefit, but rather it constitutes the very essence of the invention after being rejected based on Corfits. The prosecution history shows that hot swapability is the very reason that patent `032 is different and desirable over the Corfits patent. Hewlett-Packard's contention that it is the structural aspects of the library storage system that define the invention is not persuasive; rather, such structural requirements (the individual housing fans and handles, and the "abutment relationship") are structural adaptations necessary to achieve hot swapability. Hewlett-Packard's contention that the discussion of hot swapability in the prosecution history served only to illustrate the "mental state" of the inventors to eliminate the "daisy chain" construction (which, in essence, mandates an inability to remove a broken drive for repair without shutting down the entire system) of the prior art actually supports a conclusion that the hot swapability of the modular system constitutes its novelty. Finally, I disagree with Hewlett-Packard's assertion that the specification language reveals that the patent contemplates two ways — hot swapability and also turning off the system, removing the particular housing, and then turning the system back on — to achieve modularity. My reading of the descriptions of the objectives of the invention leads to the conclusion that all the objectives of the patent refer to the attempt to eliminate downtime by allowing the system to continue to work while a drive is removed and does not disclose "turning off" the system as an alternative way to achieve the modularity.
I conclude that the specifications as a whole and, more importantly, the prosecution history, reveal a clear or unambiguous and deliberate limitation in scope of patent `032 to a hot swapable system. MBO Laboratories v. Becton, Dickinson Co., supra, 474 F.3d at 1330. As such, I find that the term "modular" in patent `032 necessarily encompasses that concept and, thus, is defined as removable without interrupting system operations. Furthermore, because I have construed the hot swapability aspect of patent `032 to be encompassed in the term "modular," it is not necessary for me to incorporated it in the alternative term "electronic and mechanical subsystems." Thus, an "electrical and mechanical subsystem" merely means an electronic and mechanical component or collection of components organized to perform a function or set of functions, as proposed by Hewlett-Packard.
B. In abutment to
Finally, Hewlett-Packard asserts that "in abutment to" means, in the context of patent `032, "positioned next to." Tandberg argues that it means close enough to have physical contact; specifically "such that it touches."
The relevant claim language provides that the multiple drive library computer structure includes two or more modular drive housings "wherein each of said modular drive housings holds the media-inserting faces of its SCSI storage drives in abutment to, and in cooperating engagement with, the robotic media inserter." [8:53] Both parties agree that the phrase is not used or further defined in the specification or prosecution history. Therefore, they both rely on arguments extrinsic to the patent in support of their proposed definitions.
Tandberg asserts that the ordinary meaning of the word "abut" is to touch. In response, Hewlett-Packard refers me to general purpose dictionaries for its contention that abut can mean both "to touch" and "to be next to." Consequently, Hewlett-Packard relies upon a design argument in support of its contention that a person in the ordinary skill in the art would understand that "in abutment to," as used here, could not mean to touch because if the robotic media inserter touched the faces of the tape drives, "problems would arise in moving the robotic media inserter between various drives and storage locations." Hewlett-Packard relies on the deposition testimony of Tandberg's Fed. Rule Civ. P. 30(b)(6) witness to confirm this assertion.
Tandberg maintains that the testimony of its Fed.R.Civ.P. 30(b)(6) witness merely provides that a system that touches "would not be ideal" but that the testimony does not provide that it cannot be the case. In support of its interpretation, Tandberg argues that because the claim language requires that the robot media inserter be both (1) "in abutment to" and (2) "in cooperating engagement with" the media inserting face, the phrase "in abutment to" must mean something different than "in cooperating agreement with." The parties stipulate that "in cooperating agreement with" means "positioning so as to facilitate the transfer of storage media (e.g. a tape cartridge) between the robotic media inserter and the tape drive." Because the robotic media inserter must be positioned and located, pursuant to the definition of the phrase "in cooperating agreement with," the phrase "in abutment to" must mean something different; it must mean physically touching. Based on the same reasoning, Tandberg also argues that Hewlett-Packard's proposed definition of "positioned next to" is identical to the definition of "in cooperating agreement," improperly rendering the phrase meaningless.
However, as argued by Hewlett-Packard, such definitions are not mutually exclusive; a media inserting face could be positioned next to the robotic media insert, but not positioned in a manner that would facilitate transfer of the storage media. Rather, Hewlett-Packard contends that the term "cooperating engagement" relates mainly to function, while "in abutment to" relates to position. I agree.
In sum, I conclude that the term "in abutment to," as set forth in patent `032, means positioned next to.
III. HEWLETT-PACKARD'S `850 PATENT
Hewlett-Packard's `850 patent is entitled "Automatic Splaying Picker Finger." The patent protects a method or way to disengage a cartridge picker, that is contained within a data storage library system, from a data cartridge (holding at least one storage media) after it has deposited the cartridge in a drive or storage slot. Specifically, the patent describes how the "thumb assembly" and "finger" of the cartridge picker move together to disengage from the cartridge.
A. Disputed Claim Terms
The relevant claim language is set forth in claim 1 of the `850 patent, which provides a "method for disengaging a cartridge picker from a data cartridge, comprising:"
moving a thumb assembly in said cartridge picker in a first direction to insert said data cartridge in a cartridge receiving devise; and
moving said cartridge picker in a second direction, said second direction being substantially perpendicular to said first direction, until a finger pivotally mounted to said thumb assembly is withdrawn from a notch in said data cartridge. [14:4-15]
After initially contesting the meaning of the term "substantially perpendicular" in the briefing — essentially disagreeing on the scope of the word "substantial" — counsel indicated at oral argument that they had agreed to a definition and now stipulate as to its meaning. As such, I determine that the term "substantially perpendicular" in patent `850 means "approximately, but not necessarily exactly, at a right angle from a first reference point" as stipulated by the parties.
ACCORDINGLY, I make the following determinations of claim definition:
1) In U.S. Patent No. 5,349,481, the term "block of data" is defined as a complete and independent group of magnetic signals written to a magnetic tape;
2) In U.S. Patent No. 5,349,481, the term "unique block-identifying information" is defined as information associated with a block of data in a track that distinguishes it from other data blocks within that track, and contained in the header portion of the block of data;
3) In U.S. Patent No. 5,349,481, the terms "read" and "reading" are defined as acquiring magnetic signals from the tape;
4) In U.S. Patent No. 5,349,481, the term "read error" is defined as the failure to confirm whether the data in a track was accurately read;
5) In U.S. Patent No. 5,349,481, the term "re-read" is defined as to read again which, in turn, means to attempt to acquire magnetical signals from the data that was initially read incorrectly.
6) In U.S. Patent No. 5,761,032, the term "modular" is defined as removable without interrupting system operations;
7) In U.S. Patent No. 5,761,032, the term "electronic and mechanical subsystems" is defined as an electronic and mechanical component or collection of components organized to perform a function or set of functions;
8) In U.S. Patent No. 5,761,032, the term "in abutment to" is defined as positioned next to; and
9) In U.S. Patent No. 6,469,850, the term "substantially perpendicular" is defined as approximately, but not necessarily exactly, at a right angle from a first reference point.