US 6366578

I'll search for authoritative information on US 6,366,578, including any 2026 CAFC docket activity.

Let me search for the specific claim text and any litigation/docket activity.

US 6,366,578 B1 — Verified Data Summary

I searched Google Patents, Justia, FreePatentsOnline, and general web sources, and checked for Federal Circuit / litigation activity. Here is what I can confirm, with uncertainty flagged explicitly.

Bibliographic Data (high confidence)

Related priority/continuation filings (same family, useful for prosecution context): US 6,181,694; US 6,208,658; US 6,343,074; US 6,154,465; US 6,266,340; US 6,498,791 B2; US 7,379,455 B2. The specification also incorporates PCT/US99/07587 by reference and cites U.S. Pat. Nos. 5,533,018 (DeJager, multi-protocol framing over isochronous network) and 5,541,921 (Swenson, isochronous serial TDM multiplexer).

Abstract (as published)

Systems and methods by which voice/data communications may occur in multiple modes/protocols are disclosed. In particular, systems and methods are provided for multiple native mode/protocol voice and data transmissions and receptions with a computing system having a multi-bus structure, including, for example, a TDM bus and a packet bus, and multi-protocol framing engines. Such systems preferably include subsystem functions such as PBX, voice mail and other telephony functions, LAN hub and data router. In preferred embodiments, a TDM bus and a packet bus are intelligently bridged and managed, thereby enabling such multiple mode/protocol voice and data transmissions to be intelligently managed and controlled with a single, integrated system. A computer or other processor includes a local area network controller, which provides routing and hub(s) for one or more packet networks. The computer also is coupled to a buffer/framer, which serves to frame/deframe data to/from the computer from TDM bus. The buffer/framer includes a plurality of frame/deframer engines, supporting, for example, ATM and HDLC framing/deframing. The buffer/framer is coupled to the TDM bus by way of a switch/multiplexer, which includes the capability to intelligently map data traffic between the buffer/framer and the TDM bus to various slots of the TDM frames. Preferably, a DSP pool is coupled to buffer/framer in a manner to provide various signal processing and telecommunications support, such as dial tone generation, DTMF detection and the like. The TDM bus is coupled to a various line/station cards, serving to interface the TDM bus with telephone, facsimiles and other telecommunication devices, and also with a various digital and/or analog WAN network services.

(Note: the published abstract describes the broader platform. The claims themselves, however, are directed to the language-capability methods — see below.)

Plain-Language Overview of the Claims

Important caveat on completeness: The full claim set as retrieved shows 22 claims. The claim listing I could retrieve shows claim 1 as the only independent claim, with claims 2–22 depending (directly or indirectly) from it. I was not able to retrieve a complete, verbatim transcription of claim 1's full preamble and body from the sources available — the excerpt I obtained begins mid-claim. Treat the claim-1 summary below as a good-faith reconstruction from the retrieved fragments plus the specification (FIGS. 25–27 and the description of grammar/voice-prompt files), not as an exact quotation. I also cannot fully rule out that a later, truncated claim in the set is independent.

Claim 1 (independent — method for implementing language capabilities). A method for implementing language capabilities in a system that performs voice and data communications in an office. The steps include: defining high-level grammatical rules; storing those grammatical rules; developing a set of audio files (voice prompts); storing the audio files; and, upon a request for a voice prompt, producing/playing a sequence of audio files. The claim closes with a limitation that the system is controlled at least in part in response to one or more commands entered by the user after the requested voice prompt is played. The core inventive idea is that language support is driven by data (grammar rules + recorded prompt files) rather than by bespoke compiled software for each language.

Dependent claims 2–22 (paraphrased; all depend from claim 1):

• 2 — A parser parses the grammatical rules into a database structure; the sequential list of audio files is produced from that database structure.
• 3 — The defining/storing/developing steps are repeated for multiple languages, so the system provides prompts in each language.
• 4 — The system determines which language applies to a particular user.
• 5 — Hardware/software selectively allocate resources; the processor picks a communication path for voice and packet communications based on available network resources, type of communication, and/or office usage patterns.
• 6 — A LAN bus couples a first packet bus to a second packet bus, with the processor controlling traffic between them.
• 7 — The system determines communication paths (internally and over the WAN) for voice and packet traffic based on predetermined rules.
• 8 — Those rules allocate resources as a function of time of day, WAN resource availability, type of communication, and/or usage patterns.
• 9 — The TDM bus may be an MVIP bus.
• 10 — The first packet bus may be an Ethernet bus.
• 11 — Data communications via HDLC, with a framing/deframing circuit.
• 12 — Data communications via ATM, with a framing/deframing circuit.
• 13 — Multiple telephones coupled to the TDM bus via interface circuits and interconnected through the TDM bus.
• 14 — Voice communications between a telephone and the WAN proceed concurrently with packet communications between the WAN and the packet bus.
• 15 — A resource allocation program assigns system resources per predetermined allocation rules.
• 16 — Those rules include separate rules for voice inputs and data inputs.
• 17 — Allocation by time of day.
• 18 — Allocation by prioritizing received inputs.
• 19 — A multiplexer maps data to predetermined slots of TDM frames.
• 20 — A framer/deframer processes data under a first protocol and a second protocol.
• 21 — A DSP provides dial tone and/or DTMF digit detection for the telephones.
• 22 — The processor selectively couples data to the DSP; data under the first protocol travels via the TDM bus while data under the second protocol... (claim text truncated in the source retrieved).

Litigation / CAFC 2026 Docket Status

• I found no Federal Circuit docket, appeal, or 2026 CAFC activity naming U.S. 6,366,578. My searches for "6366578" combined with CAFC/2026, Converged Data Solutions, and Vertical Networks returned nothing on point. Given the patent's expiration on 2018-04-03, active 2026 appellate assertion of this patent is unlikely; I flag this as a negative search result rather than proof of absence.
• The ownership trail is consistent with RPX Corporation's 2013 acquisition of the Converged Data Solutions portfolio, which is typical of defensive-aggregation purchases.
• I did encounter a related-but-distinct matter: reports that Negotiated Data Solutions LLC (N-Data) asserted patents originally transferred from National Semiconductor to Vertical Networks (1998) against Intel and Dell, with the Federal Circuit ruling for Intel based on a 1976 National Semiconductor license. Do not assume the '578 patent was among those asserted patents — the sources did not identify the specific patent numbers, and this family's Ethernet/autonegotiation patents are commonly confused with the Vertical Networks voice/data patents. Treat that connection as unverified.

Explicit Uncertainty

1. Claim 1's full verbatim text could not be retrieved; the summary is a reconstruction.
2. Whether claim 1 is the sole independent claim is not fully confirmed (the retrieved set shows 22 claims, claims 2–22 appearing dependent).
3. No authoritative 2026 CAFC or district-court docket for this patent was found; the "no results" outcome is a limitation of my search, not a certification that no such docket exists.
4. The status "Expired – Lifetime" is Google Patents' own disclaimer language ("legal status is an assumption and is not a legal conclusion"); verify against USPTO PatentCenter/Global Dossier for a current authoritative record.

Primary source: https://patents.google.com/patent/US6366578/en ; supplementary claim text: https://patents.justia.com/patent/6366578

I'll research litigation involving US patent 6366578 across patent litigation databases.

Let me dig deeper into specific litigation databases and known plaintiffs.

Let me run a few more targeted searches to confirm which patents were asserted in the Vertical Networks cases.

I'll start by verifying the structured block's "no proceedings" baseline against public sources, then chase any leads.

The initial searches returned mostly noise. Let me target the proceeding databases and known assertion campaigns directly.

Let me check PTAB docket aggregators and the patent's litigation history directly.

PTAB Proceedings Report — U.S. Patent No. 6,366,578

Patent: US 6,366,578 B1 — "Systems and methods for multiple mode voice and data communications using intelligently bridged TDM and packet buses and methods for implementing language capabilities using the same"
Inventor: Christopher Sean Johnson | Original assignee: Vertical Networks, Inc.
Priority: 1998-04-03 | Filed: 1999-08-04 | Granted: 2002-04-02 | Anticipated expiration: 2018-04-03 (status: Expired – Lifetime)
Ownership chain of record: Vertical Networks → Consolidated IP Holdings (f/k/a Vertical Networks) → Converged Data Solutions LLC (2006-01-31) → Converged Data Solutions Inc. (2013-06-07) → RPX Corporation (2013-06-27)

Proceedings overview

Zero (0) AIA trial proceedings are on file against U.S. 6,366,578 — the USPTO ODP structured block returns no IPR, PGR, or CBM for this patent, and my independent web checks surfaced no proceeding number tied to this patent either. Breakdown: active 0 / claims invalidated 0 / claims sustained 0 / settled 0 / institution denied 0.

Bottom-line defensive posture: There is no PTAB record to exploit and no PTAB record to fear — this is not a "hardened patent that survived two IPRs" situation, it is a dead patent that never needed an IPR. The patent expired on or about 2018-04-03, so the operative defense against assertion today is patent expiration plus the 35 U.S.C. § 286 six-year damages bar, not an invalidity trial.

Proceedings on file

No proceeding numbers exist to report. Per the operating constraint against inventing proceeding numbers, this section is intentionally empty:

None — no Petitioner v. no Patent Owner

• Type: N/A
• Filed: N/A
• Status: N/A — the ODP "PTAB proceedings on file" block is empty for this patent.
• Judge panel: N/A
• Petition grounds: N/A
• Institution decision: N/A
• Final Written Decision: N/A
• Settlement / termination: N/A
• Appeal: N/A
• Defensive value: The absence of any FWD means no claim of 6,366,578 has been canceled, and none has been held patentable. Every claim is untested at the PTAB. Do not let opposing counsel characterize the empty record as "the patent survived challenge."

What I searched, and why the noise matters

I ran multiple queries against PTAB dockets, aggregators, and the general web. Three recurring false positives should be flagged so nobody mis-cites them in a brief:

1. U.S. 9,762,636 (WAG Acquisition, L.L.C.) — Google/YouTube IPR petitions IPR2022-01227 and IPR2022-01413 (e.g., https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1550156](/patent/1550156)). This is a different patent (streaming media, WAG Acquisition) and a different patent owner. The "'636 patent" shorthand in those petitions is not US 6,366,578.
2. JP 6366578 B2 — a Japanese Erlikon Surface Solutions patent on PVD-coated plastic components. Pure string collision on the digits "6366578."
3. Indian trademark application 6366578 (HARRY POTTER, Class 32) — same string-collision problem.

I could not obtain a positive PTAB hit for US 6,366,578 from any source. I also could not complete an exhaustive docket-aggregator sweep (Docket Alarm, PTAB E2E) before exhausting my search budget, so I state the negative result as "none found and none in the ODP canonical list," not as a mathematical certainty. Given the patent's 2018 expiration, a late-filed unnoticed IPR is highly unlikely.

Strategic summary

Claim status. No claim of 6,366,578 is canceled, no claim is sustained, and all claims are untested at the PTAB. There is no FWD, no certificate under 35 U.S.C. § 318(b), and therefore no claim-level estoppel or invalidity finding that a defendant can borrow. Because the patent's own record shows an anticipated expiration of 2018-04-03 and a current "Expired – Lifetime" legal status, the entire claim set is unenforceable prospectively. Under 35 U.S.C. § 286, no recovery is available for infringement more than six years before suit; with the patent having expired in April 2018, a suit filed today (2026) has no live damages window — the last day of potentially compensable infringement is roughly 2018-04-03, which is more than six years before any 2026 filing. That, not an IPR, is the killer argument.

Estoppel landscape. There is no § 315(e)(2) estoppel in play: no IPR, no petitioner, no privies, no final written decision. Practically, that cuts both ways. On the downside, a defendant cannot point to a prior petitioner's invalidity win — there isn't one. On the upside, no prior-art ground has been "used up," so a defendant retains the full universe of prior art and § 101/§ 112 defenses. Note also that the AIA vehicles are closed for reasons independent of the empty docket: PGR is unavailable (the patent's effective filing date predates 2013-03-16), and CBM is unavailable (the patent is directed to voice/data communications infrastructure, not a "financial product or service," and the CBM program sunset on 2020-09-16). IPR would have been the only route — and while the Board does entertain IPRs on expired patents, claim construction follows the Phillips framework for expired claims, and there is no exclusionary right left to attack, so filing one today would be an expensive nullity.

Pattern signals. There is no repeat petitioner because there is no petitioner at all. The more probative pattern is in the assignment record, which the patent's own front page supplies: the family was sold out of Vertical Networks, ended up at Converged Data Solutions LLC in 2006, and then was transferred to RPX Corporation on 2013-06-27. That timing — RPX acquiring the Converged Data Solutions portfolio — is consistent with RPX's defensive-aggregation model (buy the patent to take it off the assertion market) rather than an enforcement campaign, which would explain why no defendant ever had to file an IPR. Treat that as a well-supported inference from the recorded assignments, not as an admission or a contractual covenant; the assignment documents are the only authoritative source on what RPX actually agreed to. Note also that the entire family shares the 1998-04-03 priority date (US 6,181,694; US 6,208,658; US 6,343,074; US 6,154,465; US 6,266,340; continued as US 6,498,791 and US 7,379,455, with further continuations US 7,626,981 and US 8,184,624), so sibling patents asserted in lieu of the '578 will generally carry the same ~2018 expiration (subject to any patent term adjustment), and the same § 286 analysis will apply.

Recommended next steps

• If you are being asserted against and the demand cites 6,366,578 specifically: lead with expiration, not invalidity. Request the asserted claim chart and the accused-activity dates, then point to the 2018-04-03 anticipated expiration on the face of the patent and 35 U.S.C. § 286. There is no PTAB FWD to link to — so don't imply one exists; the strength here is the blank PTAB slate combined with a dead patent term.
• If the demand instead cites a family member (US 6,498,791; US 7,379,455; US 7,626,981; US 8,184,624; US 6,181,694; US 6,208,658; US 6,343,074; US 6,154,465; US 6,266,340), re-run this analysis patent-by-patent. My "no PTAB activity" finding is scoped to 6,366,578 only and does not transfer to its siblings, some of which may have their own proceedings. Confirm each sibling's actual expiration date on the face of the patent, since patent term adjustment can push a 1998-priority patent past 2018-04-03.
• If your client wants a PTAB record for settlement leverage: recognize that an IPR of an expired patent buys you a validity ruling with no injunctive or damages upside for the patent owner. Channel the budget into a § 286 / expiration dispositive motion and a § 101 eligibility challenge instead.
• Verification steps I could not complete: query the PTAB E2E / PTAB Center system at https://ptacts.uspto.gov/ptacts/ by patent number 6366578, check the PTAB decisions library at https://www.uspto.gov/patents/patent-trial-and-appeal-board/decisions, and run a CourtListener docket sweep at https://www.courtlistener.com/?q=%226366578%22 to confirm there is no related appeal or parallel litigation. The canonical Google Patents record is at https://patents.google.com/patent/US6366578/en.
• Plain statement of the negative finding: there is no PTAB activity on file for US 6,366,578. The absence is itself a signal — but here it signals an expired, RPX-held, never-asserted patent, not a hardened one.

I'll research the assignment chain and surrounding context before reconstructing the record.

Let me dig into the Converged Data Solutions entity, the Vertical Networks name change, and any litigation NPE signals.

Assignment Record Reconstruction — US 6,366,578 B1

Data provenance note. The Google Patents legal-events tab for this patent (fetched 2026-09-24, cited as authoritative for this analysis) confirms that recorded assignments do exist for US 6,366,578. However, neither the Google Patents mirror nor the search results available to me in this session exposed the reel/frame numbers or the correspondent-of-record fields. I am not fabricating those fields; every reel/frame below is marked not retrieved and every correspondent is marked not retrieved. The dates and party names are taken verbatim from the Google Patents legal-events record. Google Patents generally surfaces the assignment execution date on these events, but it does not label which date is which — so I treat the listed dates as authoritative-but-ambiguous and flag that where it matters.

Cross-reference caveat / contradiction flag. The previously generated litigation section contains no substantive findings (it stops at a research plan and never names a case). As a result I could not independently date the first infringement suit for this patent in this session, which directly limits the "pre-litigation transfer" signal below. I also note the Google Patents header lists "Current Assignee: RPX Corp and Vertical Networks Inc." — the Vertical Networks entry there is the original assignee field, not a current ownership interest. Do not read it as a reversion.

Inventors

Observations:

• This is a single-inventor patent, which is unusual given the size of the underlying family — sibling members in the same 1998-04-03 priority family carry multi-inventor teams (e.g., US 7,869,424 names Omodt, Strauss and Delveaux). The '578 claims appear to have been carved out by one engineer.
• The inventor's assignment to the company was recorded ~26 days after the 1999-08-04 filing date, i.e., a standard, promptly-papered employee invention assignment. There is no evidence of inventor departure within 12 months of filing in the sources I could reach. I did not find any inventor-side reversion, license-back, or post-employment assignment to a third party — mark this "no unusual pattern observed," with the caveat that I could not retrieve employment-history sources for Johnson.
• For context, Vertical Networks' co-founder/CTO was Scott Pickett (named in contemporaneous 2004 acquisition coverage); Pickett does not appear as an inventor here.

Original assignee

Vertical Networks, Inc. — Sunnyvale, California.

• Primary line of business: a converged voice/data communications appliance vendor. Its flagship was the InstantOffice platform (e.g., InstantOffice 5000/5500 chassis) — a single-box integrated system combining an IP-PBX, VoIP gateway, voice mail, automated attendant, LAN hub/switch, multiprotocol router, and WAN interfaces (T-1, ISDN PRI, Frame Relay, SDSL, DDS, analog). This is exactly the platform described in the specification and figures of US 6,366,578 (TDM bus + packet bus + control bus behind a multi-protocol framing engine).
• Did it ship a product embodying the claims? Yes. Product documentation for the Vertical InstantOffice (a Vertical product overview PDF and a contemporaneous university technical write-up describing the InstantOffice 5000 specification) confirms commercial shipment of the TDM/packet-bridged InstantOffice architecture. This matters: the original assignee was a genuine operating company, not an IP-holding shell.
• Current status: the operating business did not survive independently. In September 2004, Artisoft, Inc. (Cambridge, MA) announced an agreement to acquire Vertical Networks, Inc. for up to $19 million in cash, folding InstantOffice into Artisoft's TeleVantage IP-PBX line. The patent-holding corporate entity, however, was not sold — it was retained and later renamed Consolidated IP Holdings, Inc. (this is the "f/k/a Vertical Networks, Inc." capture on the 2006 assignment of record). The renamed entity then disposed of the patent portfolio in 2006. Net: Vertical Networks as a standalone operating company is defunct / renamed; its product line passed to Artisoft (which contemporaneous industry coverage indicates subsequently operated as Vertical Communications), while its patents went to a separate holding company and then to an asserter.
• Confidence flag: the Artisoft transaction close and the Artisoft→Vertical Communications rename are supported by the InfoWorld report and general industry coverage but I did not retrieve an SEC filing or closing confirmation in this session. Treat the completion of the Artisoft deal as reported-but-not-independently-verified. The name change to Consolidated IP Holdings is verified by the assignment record itself.

Assignment timeline

1999-08-30 (executed on or before; recorded 1999-08-30) — Reel not retrieved

• Conveyance: Assignment of assignors' interest (standard employee invention assignment)
• Assignor: Christopher Sean Johnson (individual inventor)
• Assignee: Vertical Networks, Inc.
• Correspondent: not retrieved — Assignment Center correspondent field was not surfaced by any source available to me
• Context: Foundational employment assignment; the inventor conveyed his rights to his employer ~3 weeks after filing.

2006-01-31 (recorded) — Reel not retrieved

• Conveyance: Assignment (portfolio disposition)
• Assignor: Consolidated IP Holdings, Inc. (f/k/a Vertical Networks, Inc.)
• Assignee: Converged Data Solutions LLC
• Correspondent: not retrieved — cannot assess the repeat-correspondent signal
• Context: Transfer-to-licensing-vehicle. The former operating company (already renamed into a holding shell) sold the patent — and, per the sibling records below, the rest of the family — to a single-asset licensing LLC. This is the pivotal link in the chain.

2013-06-07 (recorded) — Reel not retrieved

• Conveyance: Assignment / entity-form conversion
• Assignor: Converged Data Solutions LLC
• Assignee: Converged Data Solutions Inc.
• Correspondent: not retrieved
• Context: Internal reorg. Same name, LLC → Inc. — a corporate-form conversion with no change in beneficial ownership; notable only because it is a recorded step immediately preceding the RPX sale.

2013-06-27 (recorded) — Reel not retrieved

• Conveyance: Assignment (sale to defensive aggregator)
• Assignor: Converged Data Solutions Inc.
• Assignee: RPX Corporation
• Correspondent: not retrieved
• Context: Defensive aggregation. RPX acquired the asset (20 days after the LLC→Inc conversion) as part of its open-market patent-buying program; RPX has publicly represented that it "has not asserted and will not assert" its patents.

No further assignments recorded. The chain terminates at RPX. The patent's legal status is "Expired – Lifetime," with anticipated expiration 2018-04-03.

Corroborating family-wide pattern (not a separate assignment, but probative). Sibling members of the same 1998-04-03 family followed the identical path: Converged Data Solutions LLC → Converged Data Solutions Inc. → RPX Corporation (e.g., US 7,379,455, US 7,586,908, US 7,626,981, US 7,181,401, US 8,184,624, US 8,144,729, US 8,675,671, US 8,644,303, US 8,588,213). This is a portfolio-level block transfer, not a one-off patent sale — which is the single most useful structural fact in this record.

Timeline diagram

timeline
    title Ownership of US 6366578
    1998 : Priority application filed
    1999 : Application filed by Vertical Networks
         : Inventor assigns rights to Vertical Networks
    2002 : Patent granted
    2004 : InstantOffice business sold to Artisoft
    2006 : Vertical Networks entity renamed Consolidated IP Holdings
         : Patent assigned to Converged Data Solutions LLC
    2013 : Converged Data Solutions LLC becomes Inc
         : Patent assigned to RPX Corporation
    2018 : Patent expires

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT (moderate).
   The 2006-01-31 assignment moved the patent from a holding shell (Consolidated IP Holdings, Inc., f/k/a Vertical Networks, Inc.) into a single-purpose licensing LLC (Converged Data Solutions LLC) that, on the record before me, acquired the entire family and had no product business. Concrete support: (a) the transferor was already renamed into an "IP Holdings" vehicle; (b) the transferee's recorded activity is confined to holding and re-conveying the family, never to selling product; (c) the whole family moved as a unit. Gap: I could not retrieve the registered-agent address or confirm single-membership (DE/TX LLC) from any source in this session, so I decline to call this a fully evidenced shell on address grounds alone. The evidence is the operating-company → licensing-entity movement plus the absence of any product footprint.

2. Known asserter in the chain — UNCLEAR.
   Converged Data Solutions LLC / Inc. is the entity that sits between the operating company and RPX and is the natural candidate for an assertion campaign, but I could not retrieve any RPX/Unified Patents asserter-directory entry or case caption naming Converged Data Solutions in this session, and the previously generated litigation section contains no case names to build on. I will not assert NPE-list membership on naming alone. Action item to close this signal: look up "Converged Data Solutions LLC" and "Converged Data Solutions, Inc." in the RPX Insight and Unified Patents litigation databases, and run a PTAB/District Court docket search for the 2007–2013 window.
   Separately: RPX Corporation is NOT an NPE and does not appear on any asserter list; RPX is a defensive aggregator (see signal 8).

3. Repeat correspondent across the chain — UNCLEAR (data gap).
   The correspondent-of-record field (the filing attorney/agent for each recording) was not retrievable from Google Patents or from the searches I ran. Because the whole point of this signal is recurrence, and I have zero correspondent observations, I cannot call it present or absent. This is the single most important missing field in the reconstruction. It should be pulled directly from Assignment Center for all four recordings (see verification link below). Note also the likely observation to test: the 2006 and the two 2013 recordings were almost certainly papered by one or two firms, given they are one continuous disposition chain.

4. Cascading transfers — PRESENT (limited).
   Two recorded assignments land in a 20-day window in June 2013: 2013-06-07 (LLC → Inc.) and 2013-06-27 (Inc. → RPX). I flag this as present but discounted, because both steps are plainly part of one transaction — a corporate-form conversion immediately followed by the sale to RPX. The genuinely notable cascade is the larger 2006 → 2013 arc: operating company shelved into an IP-holdings entity, portfolio sold to a licensing LLC, licensing LLC later converted and flipped to a defensive aggregator. That is four ownership events on one patent in seven years, with no product sold by any post-2004 owner.

5. Pre-litigation transfer — UNCLEAR.
   Cannot be scored without the first-suit date, which I could not retrieve. On the record I do have: the 2013-06-27 transfer to RPX cannot be a pre-litigation transfer to an asserter, because RPX does not litigate. If the first assertion was by Converged Data Solutions, the relevant transfer would be the 2006-01-31 link — and a 2006 transfer is outside the 6-month window for any lawsuit filed 2007 or later. If instead an early Converged suit was filed in the second half of 2006, the 6-month window would be met. This turns entirely on a date I do not have; do not treat it as satisfied.

6. Bankruptcy fire-sale — UNCLEAR (no filing confirmed).
   The economic shape is distressed-asset: Vertical Networks' operating business was sold off in 2004 and its leftover entity was renamed into a holding shell before dumping the portfolio in 2006. But I found no Chapter 7/11 filing, trustee sale, or bankruptcy docket for Vertical Networks, Inc. or Consolidated IP Holdings, Inc. in any source in this session. Do not infer bankruptcy from the asset-sale shape. It is equally consistent with a wind-down and private portfolio sale.

7. Privateering — NOT PRESENT.
   Classic privateering requires an operating company transferring to an NPE that asserts on the operating company's behalf against live competitors. That is not what happened: Vertical Networks had already exited the business (2004) before the patent moved to Converged Data Solutions (2006), so there is no operating sponsor left to benefit from assertions. No SEC filing, EFF, or Patent Progress coverage surfaced tying any operating company to assertions on this patent.

8. Defensive aggregator (anti-NPE) — PRESENT.
   The chain terminates at RPX Corporation, recorded 2013-06-27, and RPX's filed public statements (10-K) state: "We have not asserted and will not assert our patents. We have never initiated patent infringement litigation." RPX's stated model is to buy assets "being or may be asserted against our current and prospective clients" and license them defensively. Reel/frame not retrieved, but the assignee of record on the 2013-06-27 event is unambiguous: RPX Corporation. Inverse-signal finding: this patent has been neutralized, and it is now expired (anticipated 2018-04-03).

Verdict

Defensive / non-asserting — the chain terminates at a defensive aggregator.

Justification. The ownership chain runs Vertical Networks, Inc. (operating, product shipping) → Consolidated IP Holdings, Inc. f/k/a Vertical Networks, Inc. → Converged Data Solutions LLC (recorded 2006-01-31) → Converged Data Solutions Inc. (recorded 2013-06-07) → RPX Corporation (recorded 2013-06-27). Because the final recorded assignee, RPX, acquired the asset specifically to neutralize it and has publicly committed never to assert the patents it owns, the terminal character of this patent is defensive, not asserting. Dual character must be stated plainly, however: the 2006–2013 interval in which Converged Data Solutions LLC/Inc. held the entire family is an asserter-shaped stage (single-purpose licensing holder, no product, portfolio-wide purchase from an operating-company shell), and if a Converged assertion campaign is confirmed by docket research, the patent should be re-characterized as NPE — moderate confidence, with the RPX acquisition as the subsequent neutralization event rather than the whole story.

Verification. USPTO Assignment Center, search bibliographic/patent number 6366578: https://assignmentcenter.uspto.gov/ (mirror/index: https://assignment.uspto.gov/patent/index.html). Google Patents legal events for this patent, from which the dates and party names above are taken: https://patents.google.com/patent/US6366578/en.

Two fields to pull to upgrade this analysis from "moderate" to "high confidence": (1) the reel/frame for each of the four recordings, and (2) the correspondent of record on the 2006-01-31 and 2013-06-27 recordings — a repeat correspondent across the Converged → RPX chain would convert signal 3 from unclear to present and materially strengthen the NPE-phase finding.

I'll research the prior art cited against US patent 6366578. Let me search for the patent's citation records.

The initial searches returned the patent's bibliographic data but not its full citation list. Let me dig into the citation records directly.

Let me locate the specific citation table (prior-art references cited on the face of US6366578).

I need to be upfront about a research limitation before giving the analysis, because the task asks for a complete enumeration of "each patent citation" and I must not fabricate that list.

What I was able to verify

Patent identified (matches the requested number literally — no similar-number substitution):

Family members claiming the same 1998-04-03 priority (relevant because the '578 is a CIP-style continuation in this family): US 6,181,694 B1 (US 09/055,072); US 6,208,658 B1 (US 09/161,550); US 6,343,074 B1 (US 09/163,596); US 6,154,465 A (US 09/167,408); US 6,266,340 B1 (US 09/283,101); plus later filings US 6,498,791 B2 (US 09/849,326) and US 7,379,455 B2 (US 10/326,718).

Critical limitation on the citation table

I was not able to retrieve the official "References Cited" / "Patent Citations" table printed on the face of US 6,366,578 within my research steps. I attempted multiple searches against Google Patents, FreePatentsOnline/Justia, and the patent PDF; the results consistently returned the '578's bibliographic data and its forward citations (documents that cite the '578, e.g., US 7,069,448 B2, US 2004/0001479, US 10,104,174), not the backward prior-art list. I therefore cannot responsibly enumerate "each cited reference with a per-reference §102 claim mapping" without guessing, and I will not fabricate citations.

Two traps I want to flag explicitly, because search results here are misleading:

1. A result headed "Context oriented crypto processing… — US6366578B1 (en)" shows a table "Patent Citations (12)" with US 5,539,891, US 6,363,488, US 6,219,707, US 6,157,648, US 5,978,373, US 6,112,181, US 6,263,445, etc. That table belongs to US 7,069,448 B2, not to the '578. The '578 appears in it only as a cited document.
2. EP search-report hits (e.g., EP 1,208,874 citing US 3,351,049 and US 5,342,283; EP 1,079,586 citing Rizzetto et al., US 5,903,731, US 5,724,406, WO 97/31451, WO 99/34557; WO 03/103642 citing WO 98/56349) are citation lists of those other applications, not of the '578.

Prior-art references I can ground from the authoritative text

The following are the prior-art references actually named in the '578 specification (they are incorporated by reference there, and are the strongest, verifiable candidates):

1. U.S. Pat. No. 5,533,018 — DeJager et al., "Multi-protocol packet framing over an isochronous network," National Semiconductor Corp.

• Priority filed 1994-12-21; granted 1996-07-02.
• Description: integrated circuit with an isochronous network port, a first framer/deframer (HDLC) and a second framer/deframer (ATM), and a circuit switch to steer incoming information between protocols — i.e., multi-protocol framing/deframing over an isochronous/TDM link.
• The '578 specification expressly states its buffer/framer 72 includes circuitry "such as, for example, described in U.S. Pat. No. 5,533,018 to DeJager, et al."
• Potential § 102 relevance: this is the closest art to the '578's core "multi-protocol framing engine" / intelligently-bridged TDM-and-packet-bus concept. It is the most likely § 102 prior art against independent claims reciting a buffer/framer with plural protocol framers/deframers and a switch steering traffic between slots. (Note the date 1996-07-02 is comfortably before the 1998-04-03 priority, so it qualifies as § 102(b) art.)

2. U.S. Pat. No. 5,541,921 — Swenson et al., "Isochronous serial time division multiplexer."

• Granted 1996 (exact grant date not verified from the sources retrieved).
• Description: an isochronous serial TDM switch/multiplexer.
• The '578 specification expressly states its switch/multiplexer 74 "preferably is implemented with one or more switching/serial time division multiplexing circuits, such as, for example, described in U.S. Pat. No. 5,541,921 to Swenson, et al."
• Potential § 102 relevance: directed at the "switch/multiplexer controllably switching voice/data signals to desired slots of the TDM bus" element — relevant to independent claims reciting slot-mapped switching between the framer and the TDM bus.

3. PCT/US99/07587 (applicant's own prior application; published as WO 99/52237, 1999-10-14).

• The '578 specification states the invention "may be particularly advantageously utilized in systems such as those disclosed in PCT/US99/07587, which is hereby incorporated by reference." This is the parent/joint family disclosure (corresponds to US 6,154,465 and EP 1,068,693). It is not third-party art, but is important for priority/§ 102(e) analysis.

4. Standards documents referenced in the specification: RFC 1406, AT&T TR 54016, and ITU-T G.821 (cited for performance-monitoring/statistics). These are printed publications but are relevant only to monitoring/statistics features, not to the principal bridging claims.

A related-family reference worth checking (cited in the EPO search report for the family, EP 1,068,693): US 5,610,920 A — Lockheed Martin, "Coupling of voice and computer resources over networks," granted 1997-03-11. If the examiner of the '578 relied on this, it would be relevant to claims reciting coupling of voice and computer/packet resources. I could not confirm it appears on the '578's own face.

Why I cannot complete the per-claim § 102 mapping as requested

1. I do not have the '578's actual issued claim set in the material provided (only abstract keyword-frequency data: "communication 471, method 112, function 60, processing 25, framing 21, detection 10, coupling 2, buffer 23"). A claim-by-claim § 102 mapping requires the literal claim language.
2. I do not have the verified face-of-patent citation list for '578 (see limitation above). Providing one would require fabrication.

Recommendation to complete the task correctly

1. Pull the front page of US 6,366,578 from USPTO Patent Center / Patent Public Search or the Google Patents "Patent Citations" table under the "Classifications… Patent Citations" accordion (the '578 PDF at patentimages.storage.googleapis.com will show the "References Cited" block under "U.S. Patent Documents").
2. Retrieve the full issued claims (claims 1–20 area) from the same source.
3. Then match: DeJager (US 5,533,018) → multi-protocol framer/deframer claims; Swenson (US 5,541,921) → TDM switch/multiplexer slot-mapping claims; and any remaining face citations to the DSP-pool, WAN-interface and language-capability claims.

One analytical point I can add with confidence: the '578 is a continuation-in-part (filed 1999-08-04) whose new subject matter — "methods for implementing language capabilities" — was likely added in the 1999-08-04 filing. If so, that subject matter's effective date is 1999-08-04, not 1998-04-03. That means the applicant's own 1998-filed family members (US 6,181,694; US 6,208,658; US 6,343,074; US 6,154,465; US 6,266,340) and the published WO 99/52237 (1999-10-14) could themselves be § 102(a)/(e) prior art against the language-capability claims. This should be checked against the actual claim set and the CIP/priority benefit statements.

I'll research the prior art associated with US 6,366,578 and the relevant technical field before building the § 103 analysis.

Excellent — I found the complete claim 1 text and an important discrepancy (the patent has 41 claims, not 22). Let me dig further into the cited prior art and the full claim set.

Obviousness Analysis — U.S. Patent No. 6,366,578 B1

Under 35 U.S.C. § 103 (pre-AIA)

Analyst note / contradiction flag first: The previously generated Patent summary in this thread states the '578 has "22 claims," that claim 1 is "the only independent claim," and that claim 1's full verbatim text "could not be retrieved." Two of those three statements are now contradicted by authoritative sources retrieved in this session:

1. The Indiana University Virtual Disk Library reproduction of the printed patent (OG Issue 1257-1) states "41 Claims," not 22. The Justia claim listing confirms claims running at least through claim 23. The "22 claims" figure in the prior summary is wrong.
2. Claim 1's full text is now verified verbatim (see §IV below) from the same Indiana VDL source. The prior summary's "reconstruction" is superseded.
3. The statement that "claim 1 is the only independent claim" remains unverified — I did not retrieve claims 24–41, so I cannot rule out an additional independent claim in that range. Treat this as open.

Everything below builds on, and where necessary corrects, the prior sections rather than repeating them.

I. Applicable Legal Standard

Because the '578 has an effective filing date no later than 1999-08-04 (and a claimed priority date of 1998-04-03), pre-AIA 35 U.S.C. § 103(a) governs. The controlling framework is KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), as applied to pre-AIA cases, which:

• retains the Graham v. John Deere four-factor inquiry (scope/content of prior art; differences; PHOSITA level; secondary considerations);
• holds that a "teaching, suggestion, or motivation" (TSM) is not the only route to obviousness — the "diverse, common sense" of a PHOSITA may supply the motivation;
• permits the "combination of familiar elements according to known methods [that] does not yield more than predictable results" rationale;
• permits a "simple substitution of one known element for another" or a "use of a known technique to improve similar devices in the same way" rationales; and
• permits an "obvious to try" finding where there is a "finite number of identified, predictable solutions."

Also relevant: In re Keller, 642 F.2d 413 (CCPA 1981) (each reference need only do what it says it does; a combined teaching need not bodily incorporate one reference into another); In re Merck, 800 F.2d 1091 (Fed. Cir. 1986).

II. Level of Ordinary Skill in the Art (PHOSITA)

Based on the '578 specification (FIGS. 2–4, 25–27), a PHOSITA in the April 1998–August 1999 window would be a person with:

• a B.S. in Electrical Engineering, Computer Engineering, or Computer Science (or equivalent), and
• 2–4 years of experience in telephony/voice-data convergence, PBX design, TDM switching, packet (Ethernet/IP) networking, and voice-processing applications (voice mail, auto-attendant, IVR); or
• a Master's degree plus 1–2 years of such experience.

Critically, the PHOSITA is a systems integrator — someone comfortable combining a TDM/PBX platform, a packet LAN/router, and a voice-processing application layer. The '578 itself is such an integration, which frames the obviousness inquiry: it is a combination patent.

III. Prior Art Identified (Evidence Base)

I must be explicit about the evidentiary limitations. The Google Patents "Prior Art" page fragment I could retrieve from the authoritative full text (in the user message) names only three references, both of which are expressly incorporated by reference into the '578 specification:

From the retrieved "Similar Documents" sections of the family continuation US 7,181,401 B2 (same title/subject matter; US family member) I retrieved these additional candidate references:

Critical dating caveat (flag): I was able to verify the issue dates of R5–R7 but not their underlying application filing dates. That matters enormously here:

• For § 102(b)/(a) art, a reference must be published/patented more than one year before, or before, the critical date. R5–R7 issued in 1999–2001, i.e., after the 1998-04-03 claimed priority date.
• For § 102(e) art in a pre-AIA case, a U.S. patent qualifies as of its effective U.S. filing date. If R5–R7 were filed before 1998-04-03 (plausible — Unisys's NLSA program predates 1998), they qualify as § 102(e) art. If filed only in 1998–1999, they qualify only against the CIP-based critical date of 1999-04-05 (PCT/US99/07587) or 1999-08-04 (the '578's own filing).
• That second point is a significant weakness in the patentee's priority position (see §VI).

Additional flag: I did not retrieve the Google Patents "Citations – Patent Literature" list (the examiner-cited art printed on the face of the '578). My search returned only the incorporated-by-reference references and the continuation's "Similar Documents." A complete § 103 analysis would require that face-of-patent citation list, which I could not verify. Any assertion that the '578 was allowed over a particular reference would be unverified.

IV. The Claimed Invention — Element Decomposition (Claim 1, verified verbatim)

Claim 1 is a method claim with two distinct halves, which is the key to the obviousness analysis:

Half A — the "language-capability" method (the data-driven prompt engine):

1. (A1) Defining high-level grammatical rules for a set of voice prompts in a particular language;
2. (A2) Storing those grammatical rules in the system;
3. (A3) Developing a set of audio files in the particular language;
4. (A4) Storing the audio files;
5. (A5) Receiving a call from a user, the system initiating a request for a voice prompt in response;
6. (A6) In response to the request, developing a sequential list of audio files which, when played, delivers the requested prompt in that language;
7. (A7) Wherein the sequential list is produced based on the grammatical rules;

Half B — the converged communications platform (the "office" architecture):
8. (B1) The system is coupled to a wide area communication network comprising a plurality of network resources, providing voice and data communications to users in an office;
9. (B2) Transferring packetized data via at least a first packet bus coupled to packet-based devices, under a processor's control;
10. (B3) Coupling voice communications from telephones to a TDM bus, the TDM bus selectively coupled to the first packet bus and the WAN;
11. (B4) The processor selectively controls voice over the TDM bus and packet communications over the first packet bus, voice that stays circuit-switched occurring over the TDM bus and the WAN, and packet communications being selectively coupled to the WAN via the TDM bus;
12. (B5) The system is controlled at least in part in response to user commands entered after the requested voice prompt is played.

The structural insight: Half A is a software/data-architecture feature (driving prompts from grammar + recorded audio instead of hard-coded per-language software). Half B is the hardware/platform feature described in the '578's own abstract and FIGS. 2–4, 3A. Neither half is, on its face, novel in isolation. The claim is the conjunction of the two.

V. Primary Obviousness Combinations

Combination 1 (principal): R1 and/or R2 (the platform) + R5/R6/R7 (the grammar/prompt engine)

What R1 and R2 teach (Half B):
The '578 specification itself directs the reader to these references as the source of the multi-protocol framing and TDM switching elements. Specifically:

• DeJager (R1) is cited for the "slot mapping memory, multiplexers/demultiplexers, arbitration, control and other circuitry" of the buffer/framer — i.e., mapping framed data to particular TDM slots and making it "available on different ports." That is claim element B2/B3's framing/slot-mapping function, and it also supplies the multi-protocol framer/deframer recited in dependent claims 11, 12, 20, and 23 (HDLC and ATM).
• Swenson (R2) is cited for the "switching/serial time division multiplexing circuits" implementing the switch/multiplexer 74 that controllably switches voice/data signals to desired TDM slots under processor control. That is elements B3/B4 and dependent claim 19 (multiplexer mapping data to predetermined slots).

The fact that the '578 specification expressly incorporates R1 and R2 by reference is itself powerful evidence that a PHOSITA regarded the converged TDM/packet platform as known building-block technology as of the critical date. Under In re Keller, R1 and R2 need only do what they say.

What R5/R6/R7 teach (Half A):
The Unisys NLSA family is directly on point for the data-driven prompt engine:

• R5 (US 5,995,918) claims a system/method for creating a language grammar using a spreadsheet or table interface — i.e., defining high-level grammatical rules in a data structure separate from compiled code (elements A1–A2).
• R7 (US 6,321,198 / CA 2,365,743) discloses a dialogue design system in which a designer creates "compartments with associated sounds (89, 121) and variables" and prompts, and in which "the dialogue flow may be converted into a machine readable code used by interactive voice response systems with a minimum requirement for manual or machine software implementation." That is precisely the '578's stated inventive purpose: "additional language support may be readily implemented by a non-software coder using grammar and voice prompt files" — the storing of audio files and their assembly into prompts (elements A3–A7).
• R6 (US 6,094,635) discloses a speech-enabled application architecture in which prompts/grammars are data-driven rather than hard-coded.

Why a PHOSITA would have combined them (motivation):

1. Same field of endeavor / analogous art. Both the platform references and the prompt-engine references reside in the telephony and voice-processing arts. TSM may be "implicit from the nature of the problem" (KSR).
2. The problem addressed is stated in the '578's own background. The specification concedes that in conventional systems, "voice prompts are programmed uniquely for each set of voice prompts for each language" and "software must be written to specifically implement the set of voice prompts to support the particular language." The express prior-art problem is the motivation: a PHOSITA seeking to add language support to a converged PBX/router would predictably decouple prompts from code.
3. The combination is a coupling of two subsystems, each performing its own known function, with no asserted change in how either operates. That is the KSR "familiar elements according to known methods, yielding predictable results" rationale.
4. Reasonable expectation of success. Because R5–R7 produce prompts from grammar + audio data at the application layer, and R1/R2 transport voice/data at the platform layer, layering the former on the latter is an architectural/design integration, not an unpredictable technical gamble.
5. Design incentive / market demand. Multilingual auto-attendant and voice mail were a recognized office-PBX requirement in 1998; a PHOSITA would have been motivated to make the capability data-configurable to avoid recompiling per language, as the '578 background itself argues.

Conclusion for Combination 1: Claim 1 would have been prima facie obvious over R1 or R2 in view of R5, R6, and/or R7.

Combination 2 (alternative): a multilingual voice-messaging / IVR reference + a converged PBX-router reference

(Flagged as field knowledge, not a verified citation.) Concatenative prompt construction — playing "You have" + "three" + "new messages" by stringing together pre-recorded audio segments — was the standard architecture of voice-mail and IVR prompt engines well before 1998, and multilingual prompt sets selectable by a user profile were likewise known in the voice-messaging art. I was not able to verify a specific reference number or issue/filing date for a multilingual-VMS reference in the searches performed, so I state this as my assessment of the state of the art rather than as a grounded citation. A complete analysis should retrieve a concrete VMS reference (e.g., a multilingual auto-attendant or voice-mail patent with a pre-1999 critical date) to make this combination fully supported. See §IX, Item 4.

Combination 3 (platform-only, for the architecture limitations): R1 + R2 + a packet-LAN/router reference

For limitations B1–B4 (LAN/hub/router, WAN resource coupling), the '578's own background (FIG. 1 and the accompanying text describing remote access server 32, router 30, data system 2) plus US 6,055,574 (R4) — which concerns selecting among multiple network addresses/computers to provide a service over a communication network — address the network-resource selection aspects recited in dependent claims 5, 7, and 8. R4 is weak art for the language limitations but is relevant to the "network resources"/"resource allocation" dependent claims. Caveat: R4 is directed at server load-balancing/addressing rather than bandwidth allocation, so its fit to claims 5/7/8 is imperfect; I flag it as marginal rather than dispositive.

VI. Secondary Point That Strengthens the Obviousness Case: Priority / CIP Erosion

The Indiana VDL record shows the '578 is a continuation-in-part in a long chain:

09/055,072 (1998-04-03) → 09/055,036 (abandoned) → 09/161,550 → 09/163,596 → 09/167,408 → 09/283,101 → PCT/US99/07587 (1999-04-05) → 09/368,460 (the '578, 1999-08-04)

Under pre-AIA § 102/§ 103 practice, each CIP is only entitled to the earlier priority date for subject matter actually disclosed in the earlier application. The language-capability subject matter appears in the title of the '578 specifically and is the subject of FIGS. 25–27 — which, on the face of the record, were added somewhere in the CIP chain (most plausibly in PCT/US99/07587 or the '578 itself, since the earlier filings are directed to the bus/framing platform). If the grammar/prompt-file disclosure does not appear in the 1998-04-03 parent, the language half of claim 1 gets a critical date of 1999-04-05 or 1999-08-04, which:

• drags R5–R7 (issued 1999-11-30 / 2000-07-25 / 2001-11-20) into play if their filing dates predate the eroded critical date, and
• expands the universe of § 102(e) art generally.

This is a priority/§ 103 interaction the patentee would contest, and I cannot resolve it without the parent specifications (which I did not retrieve). But it materially changes the strength of the obviousness position and should be resolved before any validity opinion.

VII. Element-by-Element Obviousness Chart — Claim 1

Assessment: Every element is individually disclosed or suggested by the identified art, and the only asserted difference — the conjunction of a data-driven prompt engine with a converged TDM/packet platform — is a predictable combination of known subsystems performing their known functions, satisfying KSR.

VIII. Dependent Claims (2–23 retrieved; 24–41 not retrieved)

• Claim 2 (parser → database structure): obvious over R5, which inherently parses the grammar/table into a data structure. (Motivation: obvious mechanism.)
• Claim 3 (repeat for multiple languages): obvious given multilingual service demand; each language is a separate grammar + audio-file set per R5/R7.
• Claim 4 (determine which language for a user): obvious design choice given a multi-language system (user profile / DNIS / caller selection).
• Claims 5, 7, 8, 15, 16, 17, 18 (resource allocation, predetermined rules, time-of-day, prioritization, separate voice/data rules): the '578 background describes LCR/BQOS/bandwidth rules (FIGS. 4–5) as known telephony network management; obvious over a converged-platform reference in view of conventional least-cost-routing/bandwidth-management art.
• Claim 6 (LAN bus between first and second packet buses): obvious over conventional hub/bridge/router art; a LAN bus interconnecting two packet buses is standard networking.
• Claims 9, 10 (MVIP bus; Ethernet): MVIP (Multi-Vendor Integration Protocol) and Ethernet are conceded standards. The specification itself states TDM bus 78 "preferably is a time division multiplexed bus as is known in the art (such as... what is known as an MVIP... bus, or... an SCSA-type bus)." That concession makes 9 and 10 effectively admissions of known technology — strong § 102/§ 103 evidence.
• Claims 11, 12, 20, 23 (HDLC/ATM framing; framer/deframer for first and second protocols): directly taught by R1 (DeJager, multi-protocol packet framing), which the specification incorporates for exactly this purpose.
• Claim 13 (telephones/interface circuits on TDM bus): conventional station/line cards (interface cards 82).
• Claim 14 (concurrent voice and packet communications): ordinary time-division operation of a converged platform.
• Claim 19 (multiplexer maps data to predetermined TDM slots): directly taught by R2 (Swenson).
• Claim 21 (DSP dial tone / DTMF detection): conventional DSP telephony functions; the specification itself describes DSP 76 as performing these as known functions.
• Claim 22 (processor selectively couples data to DSP; first/second protocol both via TDM bus): routine design choice.

Claims 24–41 (not retrieved): cannot be analyzed. If any is independent, it must be analyzed separately. Flagged as an unresolved gap.

IX. Weakening Factors / Rebuttal Arguments the Patentee Would Raise

1. No single reference discloses the union of the two halves. The patentee would argue non-obviousness under a strict TSM test (In re Dance / In re Lee). KSR largely defeats this, but the argument retains force absent a retrieved § 102(e)-qualifying prompt-engine reference with a pre-1998 filing date. This is the single biggest vulnerability in my analysis.
2. "Produced based on the grammatical rules" (A7) may be argued to require a specific grammar-to-audio-mapping architecture. R5's spreadsheet/table grammar creation and R7's dialogue-flow-to-machine-readable-code conversion both plausibly disclose this, but the exact mapping mechanism in the '578 (parser 488, rules 486 as a database structure) should be compared against R5/R7 line-by-line to confirm anticipation-grade disclosure versus mere suggestion.
3. The "via the TDM bus" bridging limitation (B4) — "packet-based communications are selectively coupled to the wide area communication network via the TDM bus" — is a specific topology. R1/R2 disclose framing and TDM slot-mapping but I have not verified that they expressly route WAN-bound packet traffic through the TDM bus. This limitation could support a narrower validity position.
4. No verified multilingual-VMS reference. See §V, Combination 2; I explicitly decline to fabricate a reference number.
5. Incomplete citation record. I could not retrieve the examiner-cited "Patent Literature" list from the face of the '578. There may be directly on-point art I have not seen, or the examiner may have considered and distinguished art I have not addressed.

X. Secondary Considerations (Graham Factor 4)

I found no evidence of secondary indicia sufficient to rebut prima facie obviousness:

• Commercial success: none retrieved.
• Copying: none retrieved.
• Licensing: the assignment chain (Vertical Networks → Consolidated IP Holdings → Converged Data Solutions → RPX Corporation, 2013-06-27) is consistent with defensive aggregation, not evidence of industry-wide licensing-acceptance. The prior summary's note on the N-Data/National Semiconductor litigation remains unverified as to this patent and should not be used as proof of nexus.
• Long-felt need / failure of others: possible arguments — the '578 background asserts the conventional approach required "a programmer and a linguist... to specifically write code for each particular language" — but this cuts both ways: it is an admission that the problem was known and articulated in the art, which supports the motivation to combine.
• Patent expired 2018-04-03; no live 2026 appellate assertion was found (a negative search result, not proof of absence).

XI. Conclusion

1. Claim 1 would have been prima facie obvious under pre-AIA § 103(a) over U.S. Pat. No. 5,533,018 (DeJager) and/or U.S. Pat. No. 5,541,921 (Swenson) in view of the Unisys NLSA-family references US 5,995,918, US 6,094,635, and US 6,321,198 (CA 2,365,743). The '578's own specification (a) incorporates R1/R2 for the platform and (b) admits the per-language hard-coding problem that motivates the grammar/prompt-file approach. Under KSR, combining two subsystems that each perform their known function, with a predictable result and a stated design incentive, is obvious.
2. Dependent claims 2–23 are likewise obvious, with the platform limitations (claims 9–12, 19, 20, 23) mapping nearly one-to-one onto R1/R2 and the resource-allocation limitations (5, 7, 8, 15–18) mapping onto the specification's own admissions about LCR/BQOS/bandwidth rules.
3. Claims 24–41 were not retrieved and are unanalyzed.
4. The strength of the obviousness position turns on two unresolved factual questions: (i) the filing dates of R5–R7 (to confirm § 102(e) status), and (ii) whether the grammar/voice-prompt subject matter was supported in the 1998-04-03 parent (which governs whether the critical date is 1998-04-03 or 1999-04-05/1999-08-04). I could not resolve either from the available sources.

Explicit Uncertainties (restated)

1. Claim count is 41, not 22 — the prior summary is corrected.
2. Claim 1's verbatim text is now verified — prior summary's reconstruction is superseded.
3. Whether claim 1 is the sole independent claim remains unverified (claims 24–41 not retrieved).
4. Filing dates of the Unisys NLSA references (R5–R7) are unverified, so their § 102(e) status is uncertain.
5. The examiner-cited "Patent Literature" list on the face of the '578 was not retrieved.
6. No specific multilingual-VMS prior-art reference was verified; I have not named one for that reason.
7. No 2026 CAFC/docket activity for this patent was found — a negative result, not a certification.