US 6337858

I'll search for this specific patent number and check for any 2026 litigation activity.

Let me check specifically for litigation and Federal Circuit docket activity.

US Patent 6,337,858 (rendered "6337858") — Summary

Scope note on method. I searched live public web sources (Google Patents full text, EPO search reports, district‑court docket aggregates, courtlistener/Justia CAFC opinion files). I do not have direct API access to USPTO PatentCenter or to the CAFC docket system, so I cannot claim an exhaustive docket sweep. I found no CAFC 2026 docket entry, opinion, or Rule 36 judgment involving this patent. Findings below are grounded in the sources cited.

Bibliographic data (high confidence — consistent across sources)

Source: https://patents.google.com/patent/US6337858/en

Naming uncertainty — stated literally, not corrected: the assignment record in the Google Patents file lists the assignor as "KAPOL, VIVEK," whereas the printed inventor list reads "Vivek Kapil." One EPO search report cites the reference as "FRANK ALAN STUART ET AL," while the patent's own front page leads with Douglas Petty. I flag these as they appear; I cannot resolve which is authoritative.

Abstract (verbatim)

"A method and apparatus for providing voice communications between two parties using computer controlled telephony hardware which is separate from the PSTN is described. The voice communications may be voice over Internet or PSTN voice connections, or any combination of the two. The apparatus includes a WEB server, a computer telephony server, a Voice over IP gateway and an operations, administration and maintenance server. The apparatus is suitably installed and operated by a service provider such as an Internet Service Provider. The advantage is unexcelled control over call setup, transfer and conference functions. A further advantage is the rapid, unfettered development of new services without compatibility issues with the PSTN."

Independent claims — plain language

Claim 1 (method; two‑leg origination architecture). The core concept: a party with data‑network access clicks a "voice button" on a web page, which sends a voice communication request (containing information needed to set up both call legs). Computer‑controlled telephony (CTI) hardware outside the PSTN, driven by a computer telephony server, then (a) originates a first voice connection, (b) originates a second voice connection using the information in the request, and (c) bridges the two together so the parties can talk. Both call legs originate from the CTI hardware — the PSTN is not the originator.

Claim 9 (method; server/rule‑base orchestration, steps a–h). A 10‑step method between a service subscriber and a data‑network user of an interactive page:

• (a) a voice button server accepts a voice communication request initiated by the user via a request button on the page;
• (b) the server analyzes the request or a preference file (cookie) from the user's terminal to determine the preferred voice medium;
• (c) it consults a rule base keyed to a subscriber‑identifying code to find at least one rule governing how the subscriber leg is established;
• (d) the CTI server controls CTI hardware outside the PSTN to place a first call to the user by the preferred medium;
• (e) the CTI server controls the same hardware to place a second call to the subscriber per the rule;
• (f) the CTI server controls the hardware to bridge the calls;
• (g) the system monitors the data network for a request (from user or subscriber) to modify the call configuration and reconfigures if received;
• (h) it monitors both calls and tears down the surviving call when either is disconnected (switch‑hook disconnect propagation).

Claim 29 (system). A combination claim: at least one computing machine programmed with (a) a voice button server for accepting requests from a user who selects a voice request button on a page; (b) an operations, management and maintenance (OAM) server to establish/maintain rule bases governing call completion to the subscriber; and (c) a computer telephony server for controlling CTI hardware outside the PSTN — plus (d) a voice‑over‑data‑network gateway converting voice data packets to telephony format, and (e) CTI hardware outside the PSTN that can initiate calls on a switched telephone network or a data‑network voice gateway, bridge calls, and respond to predefined control commands originating from either the subscriber or the user.

Claim 30 (depends on 29): the outside‑PSTN CTI hardware comprises a controller and a switch fabric.

Dependent claims (compressed)

• 2–4: medium combinations — VoIP + PSTN; PSTN + PSTN; VoIP + VoIP. 10–11: user's preferred medium specifics.
• 5 / 12–17: multiple simultaneous second‑leg attempts ("simultaneous notification groups"), cancelling all but the first answered; different agents versus the same agent at multiple numbers/IP addresses ("agent locator"); prompting the answerer to enter an acceptance code; work/home/cellular/pager number list; rule that the pager number is dialed only after the others fail.
• 6–8 / 23–25: no‑answer handling — hold the first call and download alternate options (voice mail, text message, callback request); for voice mail, originate a second call to the subscriber's voice mailbox and bridge it.
• 18: the voice button offers a menu of targets that the rule base maps to call destinations.
• 19–22: a second, access‑code‑protected voice button, with rules treating those calls differently (different termination, or different numbers to the same address producing distinct ringing patterns to identify call origin).
• 26–28: paging option — hold the user, page the subscriber with a call identification number; when the paged subscriber calls the CTI hardware and enters the ID, they are bridged to the held caller ("automated page return").

Drafting observation (not a correction): as printed, claim 26 recites dependency on claim 22 ("A method as claimed in claim 22 wherein the menu includes a selection for paging…"), although claim 22 concerns ringing patterns rather than menus. I report the text literally; whether this is a typographical error in the printed patent is something I cannot verify authoritatively.

Litigation / prior‑art significance (grounded)

• Delaware D.Del. 1:07‑cv‑00507 (GMS) — listed on the Google Patents record as the case for this patent. In that action (Vonage Holdings Corp. v. Nortel Networks Inc. et al., filed 2007‑08‑17), Nortel's counterclaim asserted infringement of U.S. Patent No. 6,337,858 ("the '858 Patent") as its Claim 9. Sources: https://dockets.justia.com/docket/delaware/dedce/1:2007cv00507/38754 and the counterclaim text (Case 1:07‑cv‑00507‑GMS, Doc. 28, filed 2007‑12‑14).
• Prior art citation: cited as "Petty '858" in an invalidity claim chart for U.S. 7,822,188 (Exhibit A‑21, dated 2014‑06‑23), and as an X‑category reference in EPO search reports for EP 1398945 A1 and EP 1395027 A1.
• CAFC 2026: No 2026 Federal Circuit activity located for this patent. Given the patent's expiration on 2017‑10‑10 and the age of the enforcement activity, this is expected. I state this as a negative finding from the searches performed, not as a certified docket clearance.

Overall takeaway

The '858 patent is an early "click‑to‑call from a web page" patent. Its distinguishing contribution, as claimed, is that both call legs are originated by computer‑controlled telephony hardware located outside the PSTN (rather than by a PSTN switching node, as in the applicant's earlier application Ser. No. 08/652,659), with a rule base and OAM server configurable per subscriber, cookie‑based medium preference, and agent‑initiated transfer/conference control — with the hardware, not the PSTN, bridging the legs. The patent is expired, so it now functions principally as prior art and as licensing/portfolio history through the Nortel → GenBand → Ribbon chain.

If you need a certified status confirmation or a definitive CAFC docket check, that should be run directly against USPTO PatentCenter/Assignment and the CAFC PACER docket — I cannot verify those authoritatively with the tools available here.

I'll research litigation involving US 6337858 using web searches.

Let me dig into the specific litigation references and the Unified Patents portal.

Let me search for the specific case number and related details.

I'll verify the canonical "no PTAB proceedings" signal against live sources before finalizing.

Let me run a few more targeted checks for any AIA trial or appeal tied to this patent.

Proceedings overview

Total AIA trial proceedings on US 6337858: zero (0). The structured "PTAB proceedings on file" block sourced from the USPTO Open Data Portal reports no AIA trial proceedings for this patent, and my independent web searches returned no IPR, PGR, or CBM trial number referencing 6,337,858 — no proceedings are active, none invalidated claims, none were settled, and none were denied institution; the defensive posture this gives a defendant is therefore not "the patent has been hardened by surviving IPR," but the unusual and more favorable posture of a fully untested patent that is also expired (anticipated expiration 2017-10-10), meaning no claim has ever been adjudicated at the Board and no § 315(e)(2) estoppel has attached to anyone.

Because the count is zero, there are no per-proceeding sections to populate. I am not going to invent proceeding numbers to fill the template. What follows is what I verified, what I could not verify, and the strategic picture that the absence of PTAB activity creates.

What I checked (and what came back empty)

I could not complete additional docket-specific searches (Delaware case 1:07-cv-00507 party identification, and a PTAB E2E / Patent Trial docket lookup) before running out of tool steps. Treat the "zero proceedings" finding as the canonical answer but as a negative finding, not a positive confirmation from PTAB's own trial docket. If you need it airtight, a Patent Trial and Appeal Board End-to-End (PTAB E2E) or Docket Alarm "patents/6337858" query is the confirming step.

A caution on the one litigation datapoint on the patent's own page

The Google Patents record for US 6337858 carries a litigation notice pointing to a 2007 case in the District of Delaware (case link 1:07-cv-00507) and lists Genband US LLC as current assignee (now Ribbon Communications Operating Company, Inc.). I could not retrieve the complaint or confirm which claims were asserted, and the previously generated litigation summary section is a placeholder — so do not treat that Delaware case as established detail about asserted claims. The one large Genband patent case I could verify in search results — Genband US LLC v. Metaswitch Networks Ltd., No. 2:14-cv-00033 (E.D. Tex.), jury verdict 2016-01-15 — asserted U.S. Patents 6,772,210; 6,791,971; 6,885,658; 6,934,279; 7,995,589; 7,047,561; 7,184,427; and 7,990,984. US 6,337,858 is not on that list. Do not conflate the Metaswitch litigation with this patent.

Strategic summary

Claim status: everything is UNTESTED and everything is EXPIRED. US 6337858 has 30 claims. Claims 1–8 are the "originate, originate, bridge outside the PSTN" method claims; claim 9 is the rule-base/voice-button method independent claim; claims 10–28 depend from it; claim 29 is the system independent claim; claim 30 depends from it. No AIA trial ever challenged any of them, so there is no claim that has been cancelled, disclaimed, or confirmed by the Board, and there is no FWD to quote. The patent's Google Patents legal status is "Expired – Lifetime," with anticipated expiration recorded at 2017-10-10 — the 20-year date from the 1997-10-10 filing. That date is the single most important fact in this analysis: the patent has been past its term for roughly nine years, so the Board route and the district-court route have entirely different payoffs (see below). The Canadian family member, CA 2249821 (granted as CA2249821C), shares the 1997-10-10 priority and is likewise long expired.

Estoppel landscape: there is none — which is a gift. Because no IPR/PGR/CBM was ever instituted, 35 U.S.C. § 315(e)(2) estoppel has never attached to any party. No petitioner, no privy, no real party in interest is barred from raising any § 102 or § 103 ground, including grounds that "reasonably could have been raised." Practically that means an accused infringer today keeps the entire prior-art universe available: every printed publication, patent, and system prior art predating the 1997-10-10 priority date, on any combination, in any forum. Contrast that with the usual situation on an asserted 1990s VoIP/web-to-phone patent, where earlier challengers have burned through the best art and left the survivors shielded by estoppel. Here nothing has been burned. This patent is a clean slate defensively.

Two procedural closers to keep in mind before anyone budgets for a Board challenge:

• CBM is closed. The AIA transitional covered-business-method program sunset for new petitions on 2020-09-16, so a § 18 CBM attack is no longer available regardless of how "financial" the click-to-call subject matter looks.
• IPR against an expired patent is possible but awkward. The Board retains jurisdiction over expired patents, but claim terms are construed under Phillips rather than the broadest reasonable interpretation, and the practical remedy is a cancellation that anyone can already achieve in district court — with the added problem that the patent expired 2017-10-10, capping recoverable damages to pre-expiry conduct by the § 286 six-year lookback in any event.

Pattern signals: no petitioner, no aggregator, no appeal. There is no repeat-petitioner pattern to analyze because no one has ever petitioned. There is no Unified Patents or other defensive-aggregator proceeding in the chain — the only defensive-aggregator reference in the record is the litigation-data provenance (Unified Patents' Delaware docket entry), not a filing. Ribbon/GENBAND has never had to defend a Board challenge to this patent and therefore has never had to appeal an adverse FWD to the Federal Circuit on it; the line of Genband/Ribbon litigation I could verify (the Metaswitch case and the later Metaswitch antitrust counterattack in S.D.N.Y.) runs through other patents. What this tells you: '858 was never worth attacking — it is a 1997-filed, 2002-issued click-to-dial/voice-button patent whose term ran out in 2017, and the market for "voice button on a web page" was overtaken by the industry well before its expiry.

Recommended next steps

1. If you have a demand letter citing US 6337858, lead with expiration, not invalidity. The patent's recorded anticipated expiration is 2017-10-10 and its Google Patents status is "Expired – Lifetime" (https://patents.google.com/patent/US6337858/en). Any damages theory is limited to pre-2017-10-10 conduct and, via 35 U.S.C. § 286, to the six years before the complaint. Ask for the accused products, the accused time window, and the pre-suit notice date — most modern VoIP/UC deployments postdate the term entirely and cannot infringe at all.
2. Because zero IPRs exist, do the invalidity work yourself and keep every ground in play. There is no § 315(e)(2) estoppel and no Board record to work around. The prior art that matters is the 1996–1997 window: voice-button/click-to-call web telephony, CTI call origination, and bridging of independently originated legs. Note the EP search report on EP 1 398 945 A1 already treated US 6,337,858 as X-category art against claims 1–21 of another application, which tells you the family of features it covers is broad — useful context, but not invalidity art against '858 itself.
3. If you nonetheless want a Board challenge, price it correctly: no CBM (program closed 2020-09-16), an IPR is legally available against the expired patent but construed under Phillips, and you should expect the § 315(b) one-year bar to be measured from service of the earliest complaint. The statutory 1-year institution-to-FWD clock (35 U.S.C. § 316(a)(11)) gives you no strategic advantage against an already-expired patent.
4. Confirm the zero count before relying on it. A PTAB E2E search on application 08/948,975 / patent 6,337,858 and a Docket Alarm patent-level query are the two authoritative negative checks. Flag the result if either surfaces a proceeding the ODP ingest missed.

Explicit uncertainty disclosure: I am reporting a verified absence. The ODP structured block is the canonical source and it says no proceedings; my independent searches found none. I could not complete a PTAB E2E / Docket Alarm confirmation or identify the parties in the Delaware 1:07-cv-00507 matter, and I have not verified which claims (if any) were asserted in that case. Do not attribute any claim-level outcome, judge panel, FWD, settlement term, or Federal Circuit appeal to this patent — none exists in the record I could reach.

I'll research the assignment record for US 6337858, focusing on reel/frame numbers, correspondents, and the Nortel→Genband→Ribbon chain.

I have strong leads. Let me drill into reel/frame specifics and correspondents.

Assignment & Ownership Analysis — US 6,337,858 B1

Methodology note / limitation. The authoritative source for this patent (the Google Patents full-text you supplied) reproduces the recorded assignment events (dates, conveyance types, parties) but not reel/frame numbers or correspondents of record. I was able to recover some reel/frame numbers for the Nortel→Genband link from secondary primary sources (a Genband 37 CFR 3.73(b) statement of chain of title and a Nortel-bankruptcy court filing), but I could not confirm which specific frame entry carries US 6,337,858, and I found no correspondent-of-record data at all. Where a reel/frame is given below, I mark it as chain-wide / not patent-specific unless confirmed. Nothing here is inferred from names alone.

Inventors

• Employer determination is documentary, not inferred. The first recorded assignment (1997‑10‑10) conveys the inventors' rights to BELL‑NORTHERN RESEARCH LTD. (BNR), Northern Telecom/Nortel's research subsidiary. The assignor list on that record reads: FRANK, ALAN STUART; KAPOL, VIVEK; MISTRY, DHANSUKH; PETRAS, MICHAEL WILLIAM; PETTY, DOUGLAS; SAN MARTIN, RAUL.
• Name-discrepancy flag (literal reading): the assignment record spells inventor #5 "KAPOL, VIVEK", whereas the issued patent and Espacenet both read "KAPIL, VIVEK." Treated literally, these are not the same string; in practice this is almost certainly a transcription/spelling variant in the recorded assignment, but it is worth a quiet corrective check against the original Recordation Cover Sheet.
• Unusual-pattern check: I found no evidence of all-six-inventors departing within 12 months of filing, and no evidence of any inventor-side re-assignment of the granted patent. Google Patents shows a single inventor→BNR transfer and nothing else involving an inventor. Nothing here precedes a fire-sale.

Original assignee

• Entity on the face of the issued patent: Nortel Networks Limited (Google Patents lists Original Assignee: Nortel Networks Ltd; Espacenet lists Applicant: NORTEL NETWORKS LTD). The 1997 filings were made under the BNR/Northern Telecom name and the title migrated through two renames before issuance (1999 → Nortel Networks Corporation; 2000 → Nortel Networks Limited).
• Primary line of business: Nortel (formerly Northern Telecom) was a large publicly traded telecom network equipment manufacturer — optical, carrier switching, and VoIP/softswitch platforms. It is an operating company, not a holding vehicle.
• Did it ship a product embodying the claims? Indirectly, yes. The patent describes a carrier/ISP-hosted "voice button" service built on computer-telephony (CTI) hardware and a VoIP gateway. This functionality maps directly onto Nortel's CVAS (Carrier VoIP and Application Solutions) product line — the softswitch/gateway/SIP business later sold as a unit. I could not confirm a named Nortel SKU reading on the claims; treat "product embodiment" as unclear-to-likely rather than proven.
• Current status: Dissolved through bankruptcy. Nortel filed for creditor protection on 2009‑01‑14 (Nortel Networks Inc., Bankr. D. Del., Case No. 09‑10138). The CVAS business — including its patents — was sold to GENBAND Inc. under an Asset Sale Agreement dated 2009‑12‑22, approved in a joint U.S./Canadian hearing, and closed in 2010. The residual Nortel patent portfolio (≈6,000 patents) went to the Rockstar Consortium in 2011; US 6,337,858 did not — it rode out inside the CVAS asset package.

Assignment timeline

Google Patents is the authority I have for dates/types; reel/frame and correspondent are supplied only where a separate primary source provides them, and then flagged. The Assignment Center's per-patent record was not retrievable in this session, so I cannot vouch for a frame-level hit on this specific patent.

• 1997‑10‑10 (executed) / recorded 1997‑10‑10 — Reel unverified

  • Conveyance: Assignment
  • Assignor: Petty; Petras; San Martin; Frank; Kapol; Mistry (inventors)
  • Assignee: Bell-Northern Research Ltd.
  • Correspondent: not captured in available sources
  • Context: initial employment assignment of the invention to Nortel's R&D subsidiary.

• 1998‑06‑01 (executed) / recorded 1998‑06‑01 — Reel unverified

  • Conveyance: Assignment (intra-group)
  • Assignor: Bell-Northern Research Ltd.
  • Assignee: Northern Telecom Limited
  • Correspondent: not captured
  • Context: internal corporate reorganization of the BNR/Northern Telecom group.

• 1999‑12‑23 — Reel unverified

  • Conveyance: Change of Name
  • Assignor: Northern Telecom Limited → Assignee: Nortel Networks Corporation
  • Correspondent: not captured
  • Context: name change only (no change in beneficial ownership).

• 2000‑08‑30 — Reel unverified

  • Conveyance: Change of Name
  • Assignor: Nortel Networks Corporation → Assignee: Nortel Networks Limited
  • Correspondent: not captured
  • Context: name change only. This is the holder that asserted the patent in 2007.

• 2010‑06‑02 — Reel likely 024468/0507 (from the Genband 3.73(b) chain-of-title statement, which lists "GENBAND Inc. → GENBAND US LLC, Reel 024468, Frame 0507"; not confirmed as the entry for this patent)

  • Conveyance: Change of Name (GENBAND Inc. → GENBAND US LLC)
  • Assignor: GENBAND Inc. → Assignee: GENBAND US LLC
  • Correspondent: not captured
  • Context: entity conversion of the acquirer, recorded prophylactically across the acquired portfolio.

• 2010‑06‑18 — Reel unverified

  • Conveyance: Security Agreement
  • Assignor: GENBAND US LLC → Assignee: One Equity Partners III, L.P., as Collateral Agent
  • Correspondent: not captured
  • Context: securitization — the PE backer's lien securing the debt used to fund the Nortel CVAS purchase. Not an ownership transfer.

• 2010‑08‑25 — Reel likely within {024879/0475, 024879/0519} (both are cited in the Genband 3.73(b) statement / related filings as "Nortel Networks Limited → GENBAND US LLC, Reel 024879, Frame 0475" and "GENBAND US LLC, Reel 024879, Frame 0519"; the Nortel bankruptcy record further lists "reel numbers: 27992‑443; 24879‑519; 24879‑475" for the "Nortel Transferred Patents." None of these is confirmed as the US 6,337,858 entry.)

  • Conveyance: Assignment of Interest (bankruptcy asset sale, not a distress dumping)
  • Assignor: Nortel Networks Corporation (per Google Patents) → Assignee: GENBAND US LLC (note: the 3.73(b) statement names "Nortel Networks Limited" as transferor — a recorded-name discrepancy between the Google event and Genband's own chain statement)
  • Correspondent: not captured
  • Context: bankruptcy fire-sale of an operating business unit to an operating-company buyer (CVAS).

• 2010‑11‑09 — Reel unverified

  • Conveyance: Security Agreement
  • Assignor: GENBAND US LLC → Assignee: Comerica Bank
  • Correspondent: not captured
  • Context: securitization / replacement lender lien. Not an ownership transfer.

• 2012‑03‑30 — Reel unverified

  • Conveyance: Corrective Assignment
  • Assignors: Nortel Networks Corporation; Nortel Networks Limited → Assignee: GENBAND US LLC
  • Correspondent: not captured
  • Context: correction to perfect the 2010 chain (fixes the transferor-name discrepancy above). Ownership-neutral.

• 2014‑01‑10 — Reel unverified

  • Conveyance: Release by Secured Party
  • Assignor: One Equity Partners III, L.P. → Assignee: GENBAND US LLC
  • Correspondent: not captured
  • Context: release of the 2010 collateral lien. Ownership-neutral.

• 2016‑07‑06 — Reel unverified

  • Conveyance: Patent Security Agreement
  • Assignor: GENBAND US LLC → Assignee: Silicon Valley Bank
  • Correspondent: not captured
  • Context: securitization (new lender). Not an ownership transfer.

• 2016‑07‑07 — Reel unverified

  • Conveyance: Release and Reassignment of Patents
  • Assignor: Comerica Bank → Assignee: GENBAND US LLC
  • Correspondent: not captured
  • Context: release of the 2010 Comerica lien. Ownership-neutral.

• 2017‑01‑03 — Reel unverified

  • Conveyance: Corrective Assignment (to correct a patent-number error in the SVB security agreement)
  • Assignor: GENBAND US LLC → Assignee: Silicon Valley Bank
  • Correspondent: not captured
  • Context: administrative correction of the 2016 security record. Ownership-neutral.

• 2017‑10‑10 — (legal event, not an assignment) Anticipated expiration; status Expired – Lifetime.

• 2017‑12‑29 — Reel unverified

  • Conveyance: Termination and Release of Patent Security Agreement
  • Assignor: Silicon Valley Bank → Assignee: GENBAND US LLC
  • Correspondent: not captured
  • Context: lien release. Ownership-neutral.

• 2018‑01‑02 — Reel unverified

  • Conveyance: Security Interest
  • Assignors: GENBAND US LLC; SONUS NETWORKS, INC. → Assignee: Silicon Valley Bank
  • Correspondent: not captured
  • Context: securitization following the Genband/Sonus combination into the Ribbon group. Not an ownership transfer.

• 2020‑03‑03 — Reel unverified

  • Conveyance: Security Interest
  • Assignor: RIBBON COMMUNICATIONS OPERATING COMPANY, INC. → Assignee: Citizens Bank, N.A., as Administrative Agent
  • Correspondent: not captured
  • Context: securitization. Note: the assignor here is already named Ribbon — i.e., the Genband US LLC entity had been renamed. There is no separately recorded assignment from GENBAND US LLC to RIBBON COMMUNICATIONS OPERATING COMPANY, INC.; the rename is evidenced instead by the 2021 release reciting "f/k/a GENBAND US LLC and SONUS NETWORKS, INC."

• 2021‑12‑06 — Reel unverified

  • Conveyance: Termination and Release of Patent Security Agreement
  • Assignor: Silicon Valley Bank → Assignee: RIBBON COMMUNICATIONS OPERATING COMPANY, INC. (f/k/a GENBAND US LLC and Sonus Networks, Inc.)
  • Correspondent: not captured
  • Context: lien release. Ownership-neutral. This entry is the clearest documentary proof of the Genband→Ribbon name continuity.

• 2024‑06‑24 — Reel unverified

  • Conveyance: Release by Secured Party
  • Assignor: Citizens Bank, N.A. → Assignee: RIBBON COMMUNICATIONS OPERATING COMPANY, INC.
  • Correspondent: not captured
  • Context: final lien release. Ownership-neutral.

Bottom line on the record: the patent has never left operating-company hands. Every non-lien event is (i) an intra-Nortel reorg/rename, (ii) a bankruptcy sale of a business unit to a strategic buyer, or (iii) a name change of that buyer.

Timeline diagram

timeline
    title Ownership of US 6337858
    1997 : Inventors assign to Bell-Northern Research
    1998 : Reorg to Northern Telecom Limited
    1999 : Rename to Nortel Networks Corporation
    2000 : Rename to Nortel Networks Limited
    2007 : Nortel asserts patent vs Vonage in Delaware
    2009 : Nortel files Chapter 11
    2010 : CVAS patent sale to GENBAND US LLC
         : GENBAND Inc converted to GENBAND US LLC
         : One Equity lien recorded
    2012 : Corrective assignment to GENBAND US LLC
    2016 : Silicon Valley Bank security agreement
    2018 : SVB security interest with Sonus
    2020 : Citizens Bank security interest
    2024 : Citizens Bank lien release

NPE / troll-pattern signals

1. Shell-entity transfer — NOT PRESENT. The 2010‑08‑25 conveyance (Nortel → GENBAND US LLC, reel family 024879) runs operating company → operating company. The only non-operating party in the chain, One Equity Partners III, L.P. (2010‑06‑18), is a collateral agent under a security agreement, not a titleholder — a lien, not an acquisition. No "IP/Holdings/Licensing/Ventures" LLC appears as an owner anywhere.

2. Known asserter in the chain — NOT PRESENT. No assignee matches Acacia, Marathon, IV, IPNav, Wi‑LAN/Mosaid/Conversant, Vringo, Pendrell, Innovatio, MPHJ, Lumen View, Round Rock, DGC, or any Spangenberg vehicle. Nuance worth stating: the Rockstar Consortium — an NPE-style vehicle — did acquire much of Nortel's 2011 patent estate, but this patent was carved out via CVAS in 2010 and never went to Rockstar.

3. Repeat correspondent across the chain — UNCLEAR / NOT CAPTURED. I obtained no correspondent-of-record data for any event. Because the tell for this signal is recurrence of a filing attorney, I cannot score it either way. This is the single biggest evidence gap in the analysis and the one item worth pulling directly from Assignment Center (the "Correspondent" field on each reel/frame).

4. Cascading transfers — NOT PRESENT. The title-moving events are spaced 12–13 years apart (1998 → 2010; 2010 → never again). Multiple lien records cluster in 2010 and 2016–2018, but liens are financing events recorded against portfolios wholesale and carry no indicium of assertion planning.

5. Pre-litigation transfer — NOT PRESENT (inverse, in fact). The one infringement assertion involving this patent, Vonage Holdings Corp. v. Nortel Networks Inc., No. 1:07-cv-00507 (D. Del., filed 2007‑08‑17), predates the Nortel→Genband transfer by ~3 years. The assignment came after the litigation, not before it — the opposite of the set-up-for-suit pattern. Nortel's December 2007 counterclaim pleaded willful infringement of the '858 patent among 12 asserted patents (see the Answer at 1:07‑cv‑00507‑GMS, Dkt. 28, ¶¶ 61–66).

6. Bankruptcy fire-sale — PRESENT. Original assignee Nortel filed Chapter 11 on 2009‑01‑14 (Bankr. D. Del. 09‑10138). The CVAS unit — "softswitching, gateways and SIP applications … all patents and intellectual property predominantly used in the CVAS business" — was sold to Genband Inc. under an Asset Sale Agreement dated 2009‑12‑22, with no competing bidders, closing in 2010 (recorded as the 2010‑08‑25 assignment). The Nortel bankruptcy record expressly defines the "Nortel Transferred Patents" as "all patents transferred from Nortel to Genband, including those … on reel numbers: 27992‑443; 24879‑519; 24879‑475." Caveat: this is a genuine bankruptcy sale to a strategic operating buyer, not a sale into an assertion vehicle — so the signal is present but benign here.

7. Privateering — NOT PRESENT. No evidence in SEC filings, the bankruptcy docket, or press coverage that Genband asserted these patents on Nortel's behalf or remitted recoveries to Nortel. Genband bought the business to keep selling the products (and later sued Metaswitch as a market competitor — see the Genband/Metaswitch bench-trial findings — which is ordinary competitive enforcement, not privateering).

8. Defensive aggregator — NOT PRESENT. The chain terminates at Ribbon Communications Operating Company, Inc., an operating supplier, not at RPX/AST/LOT/Unified/OIN. No evidence this patent was neutralized by a defensive aggregator; to the contrary, it expired naturally at its 2017‑10‑10 term.

Verdict

Operating-company assertion.

The title chain runs Nortel → GENBAND US LLC → Ribbon Communications Operating Company with no shell-entity or known-asserter link, and the sole litigation episode is the original assignee Nortel asserting US 6,337,858 against competitor Vonage in the 2007 Delaware counterclaim (reel-side record: the pre-2010 Nortel events; the transfer-to-buyer event recorded 2010‑08‑25 in the 024879 reel family). The one strong signal here — the Nortel Chapter 11 sale of 2009‑12‑22 / recorded 2010 — is a bankruptcy disposal of a product business to a strategic acquirer (Genband), which is why it does not tip the patent into NPE territory; every post-2010 record is a financing lien or a name change, and the patent simply expired in 2017. The verdict is given with one material caveat: reel/frame and correspondent fields for the specific frame carrying US 6,337,858 are unverified in this session, so the "repeat correspondent" signal (3) is unscored rather than negative.

Verification links:

• USPTO Assignment Center (search by patent number 6337858): https://assignmentcenter.uspto.gov/
• Legacy assignment search: https://assignment.uspto.gov/patent/index.html — e.g. adv=patAssignorName:Nortel + patAssigneeName:Genband (the query form cited in the Nortel bankruptcy docket)
• Litigation cross-reference (prior section): Vonage Holdings Corp. v. Nortel Networks Inc. et al., 1:07‑cv‑00507 (D. Del.) — https://dockets.justia.com/docket/delaware/dedce/1:2007cv00507/38754

Cross-reference check vs. previously generated sections: consistent — the earlier Litigation summary points at the Unified Patents entry for the Delaware District Court case, which is the same 1:07‑cv‑00507 matter. One clarification to carry forward: that case is Vonage (plaintiff, DJ action) v. Nortel (defendant/counterclaimant), and US 6,337,858 appears on the Nortel counterclaim side (Dkt. 28, Claim 9), not as a plaintiff's asserted patent.

I'll research the prior art cited for US 6337858, searching for the patent's front-page references and related prosecution history.

Let me retrieve the front-page "References Cited" list for US6337858 from alternate patent databases.

Prior-Art Analysis — US 6,337,858 B1

Important scope note (read first)

I must flag a limitation honestly, per the ground rules: the authoritative full text supplied for this analysis (the Google Patents "en" document) does not include the front-page (56) "References Cited / U.S. Patent Documents" list, and my web searches did not return the examiner-cited reference list for US 6,337,858 itself. I therefore cannot enumerate the examiner-cited U.S. patent references by number, and I will not invent them.

What I can do reliably is analyze the prior art that is explicitly cited inside the authoritative specification text you provided, plus the closely related family member. Every citation below is grounded in the supplied document; where I am uncertain, I say so.

Also note a definitional point: the "Cited By (251)" table on the Google Patents page is a list of forward citations (patents that later cite US 6,337,858). Those are not prior art to US 6,337,858 and are outside §102. I exclude them.

1. Confirmed identity of the patent (do not confuse with similar numbers)

• Patent: US 6,337,858 B1
• Title: "Method and apparatus for originating voice calls from a data network"
• Application: US 08/948,975
• Filing / priority date: 1997-10-10
• Grant date: 2002-01-08
• Inventors: Petty, Douglas; Petras, Michael William; San Martin, Raul; Frank, Alan Stuart; Kapil, Vivek; Mistry, Dhansukh
• Original assignee: Nortel Networks Ltd (later GenBand US LLC → Ribbon Communications Operating Company)
• Family: CA 2,249,821 (CA 2249821 C)
• Classifications: H04M 7/00; H04M 7/003 ("Click to dial services"); H04M 3/42323 ("PBX's with CTI arrangements"); H04M 3/56 (conferencing)
• Status: Expired – Lifetime (anticipated expiration 2017-10-10)

2. References actually cited in the specification (authoritative text)

The specification names only two discrete prior-art items, plus one non-patent commercial system. These are the references I can analyze with confidence.

Reference A — U.S. Pat. No. 4,191,860 to Weber

• Full citation: U.S. Pat. No. 4,191,860, R. P. Weber (AT&T), "Method and apparatus for performing customer-dialed telephone calls."
• Date: issued 1980 (patent text describes it as having "enabled" 800 toll-free calling services). I did not independently re-verify the exact issue date in this session; treat the year as approximate and the patent number as given verbatim in the specification.
• Brief description (as characterized in the patent): The foundational enabling invention behind modern "800" toll-free calling, i.e., network-based translation/routing of a dialed toll-free number to a destination using a database in a PSTN switching node.
• §102 anticipation analysis: Weak/background only — does not anticipate any claim.
  • Claim 1 requires originating a first voice connection AND a second voice connection from "computer controlled telephony hardware which is outside the Public Switched Telephone Network," and bridging them at that hardware. Weber teaches routing a single PSTN call using a database lookup inside the switched network. It lacks (a) two independent call origination legs from outside-PSTN CTI hardware, (b) the data-network-originated call request ("voice communication request ... received from the data network"), and (c) bridging outside the PSTN. It is cited as background art establishing the call-center/toll-free context — not as anticipatory art.
  • At most it is relevant to the "rule base"/routing concept (Claims 9(c), 12, 18), but it does not disclose a data-network voice-button request or the CTI bridging architecture.

Reference B — Applicant's co-pending U.S. patent application Ser. No. 08/652,659

• Full citation: U.S. Patent Application Ser. No. 08/652,659, filed May 28, 1996, entitled "METHODS AND APPARATUS FOR ORIGINATING VOICE CALLS" (same applicant family — Bell-Northern Research / Nortel).
• Filing date: 1996-05-28 — before the 1997-10-10 priority date, so it is temporally prior art (potential §102(e) art if later published/granted).
• Brief description (as characterized in the patent): Methods and apparatus for originating voice calls between voice terminals using a data terminal and a data service node, without requiring human attendants or IVR systems, in which an apparatus connected with a data network sends commands to a PSTN switching node to initiate a call between the voice terminals (so that a Web browser can call the business owning the page).
• §102 anticipation analysis: This is the closest self-cited prior art and the most important one to consider.
  • It discloses the data-network-initiated call concept and the "no human attendant" voice-button objective — overlapping the purpose of Claims 1 and 9.
  • But it does not anticipate Claim 1: the reference's call is initiated by sending commands to a PSTN switching node, whereas Claim 1 expressly requires the two legs to be originated from "computer controlled telephony hardware which is outside the PSTN" and bridged there. That "outside the PSTN" limitation is the express point of novelty the patent touts (see Abstract: "computer controlled telephony hardware which is separate from the PSTN"; and Summary: "no compatibility issues with the PSTN").
  • It also does not disclose the Claim 9 limitations of (i) a rule base determining subscriber-leg setup via a subscriber-identifying code, (ii) preference-cookie / preferred-medium analysis, (iii) monitoring the data network for configuration-change indications, or (iv) automatic disconnection of the other leg when one leg drops.
  • Net: potentially relevant to the preambles of Claims 1 and 9 and to dependent background, but not anticipatory of any independent claim as written. It is best treated as §102(e)/§103 art for the general "call-back from a Web page" concept, not for the claimed outside-PSTN CTI bridging architecture.

Reference C — Lucent Technologies' Web-page/800 call-center integration (non-patent)

• Full citation: No patent number given; described in the specification as a Lucent Technologies® method/apparatus for integrating Web pages with 800 call centers (launching VoIP or voice-terminal connections to an agent, with a screen "pop" of the page the browser is viewing).
• Date: prior to the 1997-10-10 filing (characterized as existing/developed prior art).
• Brief description: Web-page-launched call to an 800 call center with agent screen-pop showing the caller's page.
• §102 analysis: Non-patent prior art (if documented), not anticipatory on the record here.
  • It overlaps the "voice button → agent, with the agent seeing the caller's URL/page" feature (relevant to the FIG. 5 URL screen-pop described in the specification and to the general concept of Claim 9).
  • It does not disclose the claimed architecture: independent origination of both legs by outside-PSTN CTI hardware, the rule-base OAM server, simultaneous-notification groups, agent locator, or automated page return (Claims 5, 12–17, 26–28).
  • Like the Lucent system, it also routes through 800/call-center infrastructure — the very inflexibility the patent sets out to escape.
  • Priority: because no publication/patent number is identified, its precise §102(a)/(b) date cannot be verified from the supplied record.

3. Closely related family matter (for completeness, not a prior-art citation)

• CA 2,249,821 (CA 2249821 C) — the Canadian counterpart, filed 1998-10-06, claiming priority to the same 1997-10-10 US filing. This is a family member, not prior art, but it is relevant to the priority/§102(b) date analysis of any third-party art near the Oct. 1997 / Oct. 1998 window.

4. What I could NOT verify (explicit gaps)

1. The examiner-cited (56) "References Cited" list for US 6,337,858 (U.S. and foreign patent documents) — not present in the supplied text and not retrieved by search. Any examiner-cited anticipatory art (which is where the strongest §102 references usually sit) therefore remains unidentified in this report.
2. Issue date of US 4,191,860 — number is quoted verbatim from the specification; I did not re-verify the exact date.
3. Publication/grant disposition of application 08/652,659 — cited only as a co-pending application in the text; I did not confirm whether/when it issued, which affects whether it qualifies as §102(e) art.
4. The Houghton et al. ("A Packet Telephony Gateway for Public Network Operators," ISS, 21 Sept. 1997) and US 5,838,665 (Kahn) references that appear in a separate PCT search report (WO 2001/093553) surfaced during searching are not confirmed to be cited against US 6,337,858, so I do not attribute them to this file.

5. Bottom line

On the record available, no cited reference anticipates any of the independent claims (1, 9, 29) of US 6,337,858. The patent's claimed point of novelty — two voice connections originated and bridged by computer-controlled telephony hardware outside the PSTN, driven by a data-network voice-button request and a subscriber rule base — is not disclosed by:

• Weber, US 4,191,860 (single PSTN-network-routed call; background),
• US App 08/652,659 (data-originated call, but via a PSTN switching node — lacks the outside-PSTN limitation), or
• the Lucent Web/800 call-center system (call-center infrastructure; no rule-base CTI bridging).

To complete a §102 analysis to examiner standard, the front-page (56) reference list and the file wrapper (Office actions / cited art) for application 08/948,975 must be retrieved from USPTO PatentCenter / Global Dossier, which I was unable to do within this session. If you can supply that (56) list, I will map each citation to specific claims 1–30.

I'll research the prior art references available for this patent before building the §103 analysis.

Let me pin down dates and US counterparts for the key references I've surfaced.

Obviousness Analysis — U.S. Patent 6,337,858 Under 35 U.S.C. § 103

Note on scope and on the "Prior Art" section of this page. The Google Patents record for US6337858B1 contains a Prior art keywords field (voice, call, subscriber, user, server) and a Prior art date field (1997‑10‑10), plus a "Cited By (251)" list. I want to be precise about what can and cannot be used:

• The "Cited By (251)" entries are NOT prior art to the '858 patent. They are later patents (priority dates 1999–2003 in the excerpt provided) that cite the '858 patent. They postdate the '858 filing date and, with the possible exception of a few §102(e) siblings, cannot be combined in a §103 ground against '858. Anyone reading that list as an anticipatory art set would be making an error; I flag it because the instruction was to use "the Prior Art section of this page." The genuinely usable prior art on this page consists of (i) the Prior art date establishing the cutoff, and (ii) the references named inside the '858 specification itself, which are admitted prior art.
• I therefore build the grounds below on references the '858 specification itself discusses (Weber '860; applicant's own Ser. No. 08/652,659; the Lucent web/800 system) plus third‑party art I located that is dated before 1997‑10‑10. Where I could not confirm a date, I say so.

The '858 patent is pre‑AIA (filed 1997‑10‑10, never granted the benefit of the earlier 08/652,659 application). Pre‑AIA § 103(a) governs, and § 102(a)/(b) plus § 102(e) define the art. The analysis below applies the Graham v. John Deere framework as refined by KSR Int'l v. Teleflex (2007).

A. Level of ordinary skill (POSITA)

A person having ordinary skill in the art as of October 1997 would be a telecommunications/computer‑telephony engineer with roughly 2–4 years of experience designing CTI or call‑center systems and web/Internet telephony, and would be familiar with: HTML/CGI/JAVA web programming and browser "cookies"; PBX/ACD features (hold, transfer, conference, hunt groups, distinctive ringing, voice mail, music‑on‑hold); PSTN trunk signalling (DTMF and PRI); open CTI hardware (e.g., the Dialogic boards the '858 specification names) and switch fabrics; and emerging H.323 packet‑voice gateways (e.g., the Brooktrout/Analogic H.323 board the '858 specification names). This is the skill level against which the "motivation to combine" is assessed.

Claim-construction note bearing on § 103. The crux of independent claim 1 (and echoed in claims 9 and 29) is the negative limitation that both call legs are originated and bridged by "computer controlled telephony hardware … outside the Public Switched Telephone Network (PSTN)." The '858 specification frames the whole invention as an improvement over the applicant's earlier '659 application, in which the PSTN switch itself originated both legs. So the dispositive obviousness question is not "was click‑to‑call known?" (it plainly was), but "would it have been obvious to locate the two‑leg call origination and the bridge in a service‑provider CTI platform rather than in a PSTN switching node?" The answer, on this record, is yes.

B. Reference set and dates

C. Grounds of rejection and motivation to combine

Ground 1 — Claims 1–4, 29, 30: Haserodt '836 + applicant's '659 application + Weber '860

Coverage.

• Claim 29’s system is largely met by Haserodt '836: a "voice button server" = the WWW server that serves the feature‑form page and interprets the marked‑up selection (Haserodt, claim 2 & ¶¶0015–0017); an OAM/rule‑base function = the telephony feature server/provider and its directory/database interactions (¶¶0013, 0017, 0024); a "computer telephony server" controlling CTI hardware outside the PSTN = the telephony feature server acting as H.323 gatekeeper for the telephony gateway 107 (¶¶0013, 0022); the voice‑over‑data‑network gateway = gateway 107; and controllability "by either the … subscriber or the user" = the feature‑form model in which the client selects and can reactivate features while the provider "can send conventional Internet network commands to client 101" (claim 7; ¶0017). Claim 30's "controller and a switch fabric" is met by the gatekeeper/gateway pairing and the MCU 108, or by the conventional Dialogic‑class CTI board the '858 specification itself acknowledges.
• Claim 1's two‑leg origination and bridging comes from the '659 application: the data service node relays a call‑origination request that causes two calls to be originated and connected (one to the requestor's voice terminal, one to the destination) in response to a request from a data terminal that identifies both terminals. The '858 patent does not claim benefit of '659, so '659 is available as admitted prior art (it is described in the '858 background) — and I note that, even if it were treated as § 102(e) art, its common ownership with '858 could trigger the pre‑AIA § 103(c) exclusion for § 103 purposes, so the safer footing is the applicant's own admission.
• Claim 1's "outside the PSTN" limitation is the only gap: '659 locates origination in the switch. That gap is closed by Haserodt '836, which expressly teaches placing the telephony‑feature intelligence in a service‑provider platform that controls a telephony gateway rather than in the PSTN switch, and by the Lucent call‑back center admitted in the '858 background.
• Claims 2–4 (VoIP+PSTN, PSTN+PSTN, VoIP+VoIP) are simple media selections, each of which was an available, known connection type in the references (Haserodt's gateway 107 and Internet phones 114; the '858 specification's own Brooktrout H.323 board).

Motivation to combine (articulated, not assumed).

1. Same field, same problem. Haserodt, '659 and the Lucent system all address the identical problem: letting a data‑network user invoke a real‑time voice connection without human switchboard intervention. KSR holds that where references address the same problem, combination is likely obvious.
2. Express technical lead from Haserodt. Haserodt states that telephony features cannot be accessed over "open" networks and that gateways were needed to interface the Internet with the telephone network. A POSITA implementing '659's call‑origination service for Web users — rather than for a LAN/PBX — would naturally put the origination logic in the provider's CTI/gateway platform, because that is precisely what Haserodt teaches.
3. Recognized design incentives that the '858 specification itself concedes. The specification states the advantages of origination outside the PSTN are (a) eliminating "compatibility issues with PSTN switching equipment" and (b) enabling "rapid service development." Those are exactly the kinds of "market pressures… design incentives… and market demand" KSR treats as motivations. Placing origination in the applicant's own ISP‑side platform (using commercially available Dialogic‑class CTI hardware and H.323 boards) was a routine engineering relocation of a known function, with a predictable result.
4. Common ownership of the underlying platform. Both the '659 and '858 inventions were owned by the Nortel/Bell‑Northern family, so the '659 solution was in the same hands; the '858 specification itself identifies implementing the service on an ISP‑hosted CTI platform as the object. Combining was not merely obvious ex post — it was the stated program.

Conclusion. Claims 1–4, 29 and 30 would have been obvious over Haserodt '836 in view of the applicant's '659 application, further in view of Weber '860 and the Lucent call‑back system.

Ground 2 — Claims 9, 10, 11: Haserodt '836 + '659 + Weber '860 + cookie reference + conventional CTI

The step‑by‑step method of claim 9 maps almost element‑for‑element onto known art:

• (a) voice button server accepts the request → Haserodt's WWW server 104 and feature‑form page 115 (¶¶0015–0017).
• (b) analyze request or a preference file to determine the user's preferred medium → the '659 application expressly discloses reading the requestor's identifier from a stored "mail profile" to avoid re‑typing; browsers attaching a stored preference file is the well‑known cookie mechanism (the '858 specification calls cookies "commonly called"). Motivation: the '659 application states the purpose — eliminate repeated data entry. The choice between VoIP and PSTN was dictated by whether the user has one line or two, which the '858 specification itself treats as ordinary design logic.
• (c) consult a rule base keyed to a subscriber‑identifying code → Weber '860 translates a dialed code into call‑completion instructions via a database; subscriber‑provisioned routing databases were old. Motivation: let each subscriber control its own call‑handling, a result Weber '860 and the '858 background both describe as desirable.
• (d)–(f) CTI server controls CTI hardware to place and bridge both calls → '659 (two‑leg origination/bridging) + Haserodt (provider‑controlled gateway/MCU).
• (g) monitor the data network for a reconfiguration request and reconfigure → Haserodt's provider keeps a control connection open to the client while redirecting the media connection so it "can send conventional Internet network commands to client 101" (claim 7; ¶0017), and the feature form permits activating/deactivating features and setting parameters. Transfer/conference/release are standard PBX features that Haserodt lists (¶0002).
• (h) monitor both calls and tear down the survivor when either disconnects → ordinary switch‑hook/release‑link behaviour of any bridged PBX call; nothing more than the predictable operation of a bridge the POSITA would build. No non‑obvious contribution is recited.

Motivation. The same field and the same goal (dynamic control of an Internet‑initiated voice session) run through all references; the final two steps are the inherent, expected behaviour of a bridged call, which KSR treats as within the skill of the artisan. Claims 10–11 merely recite which of the two known media (VoIP vs. PSTN), a result the references make inevitable.

Ground 3 — Claims 5, 12–17 (simultaneous notification groups; agent locator; accept code; number lists; pager last): Weber '860 + conventional ACD/hunt features + '659

• Claim 5 / claim 12–13: launching several second‑leg calls and cancelling all but the first answered is the classic ACD "broadcast/hunt‑group ring" or ring‑all/ring‑one‑answer behaviour, and Weber '860 already teaches distributing a call across a plurality of answering points with alternate routing. Motivation: maximize the probability that someone answers quickly — the '858 specification states this reason verbatim.
• Claims 14–16: attempting the same agent at multiple numbers/IP addresses ("agent locator") is the well‑known telephone "find‑me/sequential‑and‑simultaneous ringing" pattern; the '659 application already contemplates plural distinct call‑request commands mapped to distinct destination identifiers, and lists of work/home/cellular/pager numbers were conventional.
• Claim 15 (enter a code to accept): prompting an answering party to enter an acceptance code is conventional automated‑attendant/IVR practice (see the Advanced‑800 Call Prompter reference, which recognizes caller DTMF/speech to select among destinations). Motivation: reduce mis‑routed acceptances, as the '858 specification explains.
• Claim 17 (pager dialed only after the others fail): Weber '860's alternate routing on busy/no‑answer teaches the ordering principle; paging last is an obvious consequence of the fact (which the '858 specification acknowledges) that a pager answers on the first ring. This is a classic "arrange known elements in a known way" case under KSR.

Ground 4 — Claims 6–8, 23–25 (no‑answer handling; options menu; voice‑mail bridging): Weber '860 + Advanced‑800 Call Prompter + conventional voice mail

• Claims 6 and 23 (hold the first call; download options; menu of voice mail / text message / callback): music‑on‑hold and automated attendant menus presenting recorded options and recognizing caller selections were old (Call Prompter; music‑on‑hold in PBX/ACD art). Motivation: don't lose the call — the '858 specification's stated purpose.
• Claims 7, 24: the specific option set (voice mail, text, callback time) is a recitation of the then‑standard "leave a message" alternatives offered by answering services and ACDs.
• Claims 8, 25 (originate a second call to the voice mailbox and bridge it): bridging a caller to a voice‑mail system is conventional PBX/ACD call‑coverage behaviour (the same two‑leg bridging skill already supplied by '659/Haserodt). Motivation and result are predictable.

Ground 5 — Claims 18–22 (menus; second, access‑controlled voice button; different termination or distinctive ringing): '659 application + Weber '860 + conventional distinctive‑ringing/PIN‑access

• Claim 18 (voice button with a menu of targets mapped to destinations): the '659 application expressly teaches plural distinct voice‑call‑request commands that "may explicitly state the type of information available," each mapped to a distinct destination identifier. This is essentially the same teaching.
• Claims 19–21 (password/PIN‑protected second button treated differently by the rule base, routed to a different termination): PIN‑controlled access and separate routing for privileged callers was commonplace; and the '858 specification itself presents these as alternatives, one of which (a second button on a password‑protected subordinate page) is no more than a web‑design choice.
• Claim 22 (same address, different numbers, distinctive ringing to identify call origin): distinctive ringing is a long‑known PSTN/PBX feature; Weber '860 supplies the multiple‑number‑to‑one‑service concept. Motivation: distinguish established clients from prospects — the '858 specification's own stated purpose.

Ground 6 — Claims 26–28 (automated page return with a call‑identification number): conventional paging + '659/Weber call handling

• Claim 26/27 (hold the user; page the subscriber; send a call ID): paging a subscriber and transmitting a callback number/ID so the subscriber can call in and be bridged to a held call is ordinary paging and "meet‑me" conferencing practice. The Advanced‑800 Call Prompter and Weber '860 provide the announcement/selection and call‑distribution pieces; the '659/Haserodt two‑leg bridging provides the connection.
• Claim 28 (subscriber dials in, enters the ID, and is bridged to the held caller): this is a routine application of an interactive‑response port plus a bridged call — the same mechanism as the IVR/feature‑server art. The '858 specification's asserted benefits (discount rates; returning a page into a VoIP leg) are commercial motivations for using the known mechanism, not evidence of non‑obviousness in the mechanism itself.

Drafting defects to keep in mind (reported literally, not corrected). Claim 26 is printed as depending on claim 22 while reciting "the menu," although the menu is introduced in claim 23; claim 12 refers to "rule b" where the rule base is step (c). I flagged the claim‑26 dependency in the prior section and cannot authoritatively confirm whether these are typographical errors in the printed patent. They affect how the claims are grouped, not the substance of the grounds.

D. Cross‑cutting "motivation to combine" summary

Four independent motivations each independently support the combinations above:

1. Identity of field and problem — all references address enabling a data‑network user to obtain a real‑time voice connection to a business party. Same problem → combination is presumptively obvious (KSR).
2. Express suggestions in the references — Haserodt states that Internet clients need "open network" access to telephony features and that a gateway and gatekeeper are required; Weber '860 supplies code‑to‑destination translation; '659 supplies two‑leg origination and command menus.
3. Design incentives the '858 specification concedes — avoiding PSTN compatibility issues and enabling rapid feature development. The patent's own background recites these as the reasons for the invention, i.e., as the motivations a POSITA would have had.
4. Contemporaneous market activity — the Lucent and Aspect web‑call products described in the German trade coverage of March 1998 confirm the industry was converging on web‑page‑initiated voice, which is evidence of a known, expected solution rather than an unbounded inventive leap.

E. Anticipated rebuttals and secondary considerations

• "Teaching away": the only possible argument is that '659 locates origination in the PSTN switch. But that is a design choice, not a teaching that origination outside the switch would be inferior or unworkable. There is no statement in any reference disparaging a provider‑side CTI implementation; Haserodt in fact points the opposite way. No teaching away.
• Unexpected results: the '858 specification asserts "unexcelled control" over setup/transfer/conference and "rapid, unfettered development of new services." These are the expected consequences of relocating a known bridge function into a programmable platform, and the specification supplies no comparative data or surprising result. Under KSR, predictable advantages known to flow from the combination do not rebut obviousness.
• Nexus: no evidence of a nexus between any secondary consideration (commercial success, licensing, copying) and the alleged invention has been presented here; the expired‑patent licensing history through the Nortel → GenBand → Ribbon chain is not probative of non‑obviousness. I have not found and do not assert any secondary‑consideration evidence.
• The § 103(c) nuance on '659: as noted, if '659 is treated as § 102(e) art, its common ownership with '858 could remove it as § 103 art. This is why the grounds above rest '659 as applicant's admission (described in the '858 background), which is not subject to § 103(c). Grounds 1 and 2 are constructed so that Haserodt '836 — genuine third‑party art — carries the "outside the PSTN" and web‑interface limitations even if '659 were excluded.

F. Caveats and limits (stated explicitly)

1. Date uncertainty for two references. I could not confirm the priority/publication date of EP 0 843 454 A2 (Lucent) with the tools available, and therefore I rely on the '858 specification's own admission of the Lucent web/800 call‑center system rather than on a confirmed date. Similarly, I could not confirm a US patent number issuing from Ser. No. 08/652,659; the text I located is the Canadian counterpart CA 2,228,661 C, and I have not verified whether it is a US § 102(e) reference. Treat these two as "admitted prior art by the '858 specification," not as date‑verified third‑party art.
2. Haserodt '836 is the strongest dated third‑party reference (US filing 1996‑09‑13, before the 1997‑10‑10 filing), so it is § 102(e) art subject to a possible Rule 1.131 swear‑behind and independent corroboration — a POSITA‑level question I cannot resolve from the record.
3. This is a technical analyst's preliminary opinion, not a legal conclusion. It does not apply the specific claim constructions a court or the PTAB would adopt, and it does not perform a formal invalidity contention chart. Claims 1 and 9 in particular are "means‑plus‑function‑free" method/system claims that would be construed on their literal steps, which strengthens the mapping above.
4. The 251 "Cited By" documents are not usable as § 103 art against '858; I state this so the record is not misread.

Bottom line: every independent claim (1, 9, 29) and every dependent claim of US 6,337,858 would have been obvious under pre‑AIA § 103 as of 1997‑10‑10 over US 6,031,836 (Haserodt) in view of applicant's own Ser. No. 08/652,659, further in view of Weber US 4,191,860, the Lucent web/800 call‑center system admitted in the patent's background, and the conventional cookie and CTI/PBX feature sets — with the specific ancillary features (SNGs, agent locator, no‑answer options, voice‑mail bridging, distinctive ringing, automated page return) each supplied by long‑known ACD, IVR and paging techniques.