US 5608786

I'll search for authoritative information on this patent and check for any 2026 CAFC activity.

Let me run a couple of more targeted searches to confirm whether there is any 2026 Federal Circuit activity tied to this patent number.

Search scope and caveats

I searched the open web (which surfaces USPTO/Google Patents data, PTAB/IPR dockets via Docket Alarm, and law‑firm Federal Circuit weekly digests). I could not query PatentCenter or the CAFC/PACER docket system directly, so the "no 2026 CAFC activity" conclusion below is based on searches of indexed sources rather than a primary docket check. I did not auto‑correct any identifier; "US5608786", "5,608,786", application 08/387,162, and the child 6,067,350 are all treated literally as they appear.

CAFC 2026 docket check — result: no hits. Searches for the patent number paired with 2026 Federal Circuit terms returned nothing. The closest 2026 Federal Circuit material was a general weekly summary (Patterson Sheridan, week ending April 24, 2026) covering unrelated patents (RE38,806 / 6,352,879; NantWorks v. Niantic; Network‑1 v. Google), which does not mention 5608786.

This is consistent with the legal status: Google Patents records "Anticipated expiration 2015‑02‑13; Status: Expired – Lifetime." The '786 patent's 20‑year term ran from its 1995‑02‑13 filing, so it has been expired for roughly eleven years. An expired patent cannot be asserted for prospective infringement, which makes a 2026 appeal concerning the '786 patent highly improbable. I flag one residual uncertainty: Google Patents shows a "Family has litigation / First worldwide family litigation filed" link (Darts‑IP family ID 4154930), which indicates historical litigation somewhere in the family, but I could not verify any specific case names or that any of it reached the CAFC in 2026.

Important search-noise warning: numerous hits containing "5608786" are not litigation over this patent. They are third‑party citations of it as prior art, e.g. in the file histories/citation lists of US 7,257,201; US 6,711,154; US 7,149,509; US 8,099,046; and in IPR exhibits (IPR2022‑00948, IPR2022‑01073, IPR2022‑01181, IPR2016‑01261). Do not misread these as suits against the '786 patent.

---

Bibliographic summary (US 5,608,786 A)

Recorded chain of title (Google Patents reassignment entries): Alphanet Telecom Inc. (assignment recorded 1995‑06‑21) → Alphanet Hospitality Systems Inc. (2000‑02‑09, assignor of record PricewaterhouseCoopers Inc.) → Alpha Group Holdings LLC (2000‑02‑09) → Alpha Holdings Group LLC (2007‑11‑08, name change) → eBay Inc. (2008‑02‑01) → PayPal, Inc. (2015‑07‑22). Note the current-assignee field is an assumption per Google Patents' own disclaimer, and I have not independently verified the recorded assignments.

---

Abstract (verbatim from the source)

"The present invention relates to a unified messaging system and method which is more convenient and advantageously combines or makes use of existing communication channels or networks. Part of the system and method relies on a data communication network forming an intermediate leg of the distribution network. Telephone communication is typically used for initial or final legs. Voice mail, E-mail, facsimiles and other message types can be received by the system for retrieval by the subscriber. Communications may be centralized and retrieval of messages can be accomplished using one of a number of separate and distinct approaches. Thus, data communication networks such as the Internet can become global voice mail and facsimile mail systems."

---

The independent claims, in plain language

The patent has 18 claims. Three are independent: claims 1, 9, and 11. Claim 8 (long‑distance voice over INTERNET) is dependent on claim 1, despite reading like an independent feature list.

Claim 1 — System claim

A communication system made of multiple commercial access providing computers (CAPs), each tied to INTERNET and each hosting many registered subscribers who are "collectively unrelated" (i.e., an open, multi‑tenant subscriber base, not one company's closed user group). Every subscriber is given three addresses: an E‑mail address, a fax telephone (telephone) address, and a voice mail telephone address. A message sent to any of the three addresses is received and stored by that subscriber's associated CAP for retrieval. The subscriber can retrieve messages (or a summary) either by (a) a telephone set calling the CAP, or (b) a computer and modem calling one of the CAPs. The CAPs exchange information with each other over INTERNET, and a subscriber may contact any CAP — messages stored at the subscriber's home CAP are transferred over INTERNET to whichever CAP the subscriber actually contacted. That cross‑node retrieval, mediated by INTERNET, is the heart of the claim.

Claim 9 — Method claim

A method of receiving voice, E‑mail, and facsimile messages addressed to a subscriber. The messages arrive at a computer system connected to a data communication network that carries E‑mail. That computer system is composed of a plurality of CAPs interconnected by the high‑speed data network, each CAP having many direct‑in‑dial (DID) telephone lines to the PSTN for receiving and sending fax and voice messages. Subscribers may retrieve stored messages in either of two ways: (i) using the PSTN together with the data communication network "directly associated with said computer system," or (ii) using the PSTN to form a direct connection with any of the CAPs. In both cases, the CAPs communicate the messages among themselves over the data network so the subscriber can pull messages from a node other than the home node.

Claim 11 — Unified messaging system claim

A unified messaging system built as a computer system of multiple CAPs interconnected by a high‑speed data communication network for digital data transmission. Each CAP has a bank of DID telephone lines into the PSTN and acts as a commercial access provider for that high‑speed network through which E‑mail is delivered. The system gives each subscriber an E‑mail address, a fax telephone address, and a voice mail telephone address; a message to any address is received and stored for retrieval by that subscriber. The system is reachable by any subscriber via the PSTN and/or the high‑speed data network to retrieve messages or a summary of messages; subscribers may contact any of the CAPs for voice mail, E‑mail, or facsimile messages received on their behalf — and those messages, "if necessary, are exchanged between any of said commercial access providing computers over the data communication network." (The source text of claim 11 contains a typographical/OCR-style artifact — "for of voice mail. E-mail or facsimile messages" — which I have preserved rather than corrected.)

Dependency map of the remaining claims

• Claims 2–8 depend from claim 1: facsimile-machine retrieval (2); broadcast alert signal to subscriber equipment (3); automatic E‑mail→facsimile conversion for some subscribers (4); fax/voice addresses being telephone numbers reached over DID lines (5); audible/voice summary via telephone (6); CAPs in different geographic areas where INTERNET transfer is cost-effective (7); using INTERNET for the subscriber's long-distance voice transmission (8).
• Claim 10 depends from claim 9: subscriber‑originated messages are received for more cost‑effective transmission, addressed to a telephone address, and routed over the data network to a CAP closer to that address, which then completes the call over the PSTN.
• Claims 12–18 depend from claim 11: (12) busy/no‑answer call forwarding into the CAP for store‑and‑retrieve; (13, dep. 12) same number for voice and fax, with the system distinguishing the two and issuing appropriate prompts; (14) conversion of messages to E‑mail or facsimile with automatic delivery to subscriber equipment; (15) automatic E‑mail↔facsimile conversion and forwarding; (16) retrievable voice summary; (17) summary retrievable as voice mail, E‑mail, or facsimile; (18, dep. 17) conversion arrangement so a received facsimile can be stored/retrieved as a text file (e.g., OCR) and a text file can be retrieved as a facsimile.

---

Plain-language overview of the disclosure (beyond the claims)

The specification markets the system under the mark "UniPost" and its nodes as UniPost Access Nodes (UANs), distributed across geographies and linked by INTERNET. A subscriber receives an Internet address (example given: gordon@toronto.unipost.com, or jsmith@acmefireworks.com), a Personal Mailbox Telephone Number (example: 1‑416‑555‑1234), and a PIN/password (example: 63265). Voice and fax messages arrive on the same telephone number, and the UAN distinguishes the two. The disclosure also covers: message conversion (E‑mail→fax, fax→E‑mail, E‑mail→speech), OCR of faxes to text, mailbox structure (fax in‑box, voice in‑box, E‑mail box, fax out‑box on a subdivided hard disk), voice/fax/E‑mail summaries, forced-delivery alerts including pager broadcasts and outbound voice notification, fax and live voice routing where a high-speed data network forms the intermediate leg and the PSTN forms the first/last legs (FIG. 4 and FIG. 5), packet-ordering protocols for digitized voice, encryption between UniPost nodes and optionally on the first/last telephone legs, and operation over cellular or other two‑way arrangements. The priority document CA 2139081 (Dec. 23, 1994) is the Canadian counterpart in the same family.

---

Where I am not certain

• CAFC/PACER ground truth: I confirmed no 2026 appeal via indexed web sources, not via a live docket query. Given the 2015 expiration, I assess 2026 appellate activity as very unlikely, but I cannot certify a complete negative.
• Litigation history: the Darts‑IP family link indicates litigation exists somewhere in family ID 4154930, but I have no verified case name, court, or outcome, and nothing tying it to 2026.
• Assignment accuracy: the assignee chain above is quoted from Google Patents' reassignment records, which explicitly disclaim legal accuracy and note that PricewaterhouseCoopers Inc. is the recorded assignor to Alphanet Hospitality Systems Inc. — a detail I could not verify independently.
• Claim 13 wording and the claim 11 artifact should be verified against the official USPTO printed patent if verbatim quotation is needed; the fetched text contains at least one obvious typographical artifact.

Authoritative source used for the patent text: https://patents.google.com/patent/US5608786/en (family/child data: US 6,067,350, application 08/659,672).

I'll research litigation involving US patent 5608786. Let me run several targeted searches.

Let me search more specifically for litigation databases and the patent's assertion history.

Let me try patent-specific litigation databases and court records.

Summary of Findings

Bottom line: I could not confirm any specific, verifiable litigation asserting US 5,608,786. My searches did not surface a district court complaint, ITC investigation, or PTAB proceeding naming this patent as an asserted patent. I am reporting that gap explicitly rather than supplying case details I could not verify.

Per your strict rule, I am treating "5608786" and "US5608786" literally and not substituting similar numbers.

---

Patent identification (from the authoritative full text you supplied)

What the searches did and did not show

1. No litigation records for '786 appeared in the searches I ran. Querying the bare number and variations returned only:

   • Prior-art citation lists (e.g., the '786 number appearing as a cited US family member in PCT search reports, and in a PTAB exhibit list in IPR2022-00948 concerning an unrelated Apple patent, US 9,774,406);
   • The Google Patents page for '786 itself.

   None of these are litigation against the '786 patent.

2. A "family has litigation" flag exists on the Google Patents page you provided — it shows "Family has litigation / First worldwide family litigation filed" with a Darts-IP link (patents.darts-ip.com/?family=4154930). Important caveats:

   • This flag is attributed to Darts-IP and covers the whole family (ID 4154930), which includes the continuation US 6,067,350 — so it may not correspond to '786 specifically;
   • Google Patents itself states the legal-status/assignee information is an assumption and not a legal conclusion;
   • I could not open or verify the underlying Darts-IP record, so I cannot state what case(s), parties, or courts it reflects. I decline to guess.

3. No ITC investigation (Section 337) referencing '786 was identified in my searches.

4. No PTAB IPR/PGR/CBM naming '786 as the challenged patent was identified. The number appeared only as a cited reference in unrelated proceedings.

Practical / legal-posture notes

• The patent expired on 2015-02-13 (anticipated expiration, "Expired – Lifetime"). Under 35 U.S.C. § 286, the damages window is capped at six years before suit; any suit now would face a lapsed-patent and limitations bar for prospective relief. This narrows the universe of realistically assertable cases to pre-2015 conduct.
• The 2008 assignment to eBay and the 2015 assignment to PayPal are significant: a 2008–2015 assertion would have been by an operating-company affiliate; the record as shown does not reflect a classic NPE campaign on this number.
• The 18-claim set (independent claims 1, 9, 11) is the classic unified-messaging claim family, so it is a plausible candidate for a family-level assertion tied to the continuation or foreign siblings — which is consistent with (but does not prove) the Darts-IP family flag.

What I did not establish (stated rather than fabricated)

• Any plaintiff, defendant, jurisdiction, case number, filing date, or outcome for a suit on '786.
• Whether the Darts-IP "family litigation" flag relates to '786, '350, or a foreign family member.

I do not know these details with high confidence, and I have not invented any.

Recommended verification steps (authoritative sources)

• Darts-IP family record 4154930 (the exact link on the '786 page) — confirm which family member and case(s) the flag covers.
• RPX Insight patent page for US 5,608,786 (shows assertion/litigation history).
• Unified Patents Portal litigation & PTAB case lists (the URL you cited, plus the PTAB tab).
• PACER / CourtListener RECAP docket searches on "5,608,786" and on the assignees (Alphanet, Alpha Group Holdings, eBay, PayPal).
• USPTO Patent Center / Assignment records for any supplementary filings, and USPTO PTAB for post-grant proceedings.

If you can authorize another round of searches against those specific sources, I can attempt to resolve the Darts-IP family flag and produce a case-by-case table.

I'll start by verifying the structured block's "no PTAB activity" signal against public sources.

Let me dig deeper on whether the '786 patent itself was ever challenged.

Proceedings overview

Total AIA trial proceedings on file for US 5,608,786: 0. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming the '786 patent as the challenged patent, and my independent web searches found none either — not a single petition, institution decision, FWD, or termination. The defensive posture this gives a defendant is therefore unusual and worth stating precisely: the '786 patent was never tested at the PTAB, not because it is "hardened," but because it expired on 2015-02-13 and never attracted a post-grant challenge during the short window in which one was available. No claim has been canceled; no claim has been sustained in an AIA trial, because no AIA trial ever happened.

No proceedings to report (and how I verified it)

Because the structured block is authoritative and says the count is zero, I did not invent proceeding numbers. What follows is what I checked and what I did find, so you can see the negative result is real and not an artifact.

Verification steps taken (2026-10-01):

• Structured PTAB block from the USPTO ODP: no AIA trials indexed against US 5,608,786.
• Web search for "{patent} + IPR / PTAB" and for CBM review of the '786: no hits with the '786 as the patent under challenge.
• Web search for the '786 in litigation: no complaint, MDL, or judgment naming the '786 as a patent-in-suit surfaced.

False positive you should not be misled by:

• IPR2022-01073 (Google LLC) — the search hit is an exhibit list in which "US 5,608,786 A" appears as prior art exhibit 1001 inside a Google IPR directed at a different patent. The '786 is the sword here, not the target. Do not cite this as a proceeding against the '786. (Source surfaced as a Docket Alarm exhibit page; the challenged patent is not the '786.)
• Examiner citation of Gordon '786 as § 102(b) art — in the file history of US 6,333,973 (see IPR2015-00201, Exhibit 1002), the Examiner rejected claims 1–28 and 30–33 as anticipated by "Gordon (U.S. Patent No. 5,608,786)." Again, the '786 is prior art, not the patent on trial.
• VOIP-Pal / Netlist-style hits — the '786 appears only inside prosecution-history citation dumps (e.g., "5469497 '5608786' 5712907 …"), i.e., as a reference of record.

Adjacent litigation that is not about the '786 (clarifying so you don't conflate it):

• In re Unified Messaging Solutions LLC & Advanced Messaging Technologies, Inc. Patent Litigation, MDL 2371 (N.D. Ill., Judge Joan H. Lefkow). The UMS/Acacia campaign asserted patents from the Bobo/Advanced Messaging family — e.g., US 6,857,074 and US 7,836,141 in UMS v. Wells Fargo; US 6,067,350-adjacent family members are different patents. I found no indication the '786 was asserted in MDL 2371.
• The consolidated Federal Circuit appeal in that MDL, Nos. 2014-1611, 2016-1065, 2016-2150, 2016-2212, was AFFIRMED per curiam (Lourie, Chen, Hughes) on 2017-12-08 under Fed. Cir. R. 36 (nonprecedential). CourtListener: https://www.courtlistener.com/opinion/[4449765](/patent/4449765)/unified-messaging-solutions-v-google-inc/ . That disposition followed the district court's 2014-06-13 final judgment of non-infringement in favor of defendants. It does not affect the '786.

Strategic summary

Claim status of the '786. All 18 claims — independent claims 1, 9, and 11, and dependent claims 2–8, 10, and 12–18 — are UNSUSTAINED-AND-UNTESTED: no PTAB or PGR/CBM panel ever construed or invalidated any of them. "Untested" is not "valid." The patent is in the public-domain tail of its life: Google Patents records the anticipated expiration as 2015-02-13 (20 years from the 1995-02-13 filing), 27 days short of the filing's "expired - lifetime" status shown after the 2015-02-13 anniversary. Its continuation, US 6,067,350 ("Long distance telephone communication system and method," filed 1996-06-05), expired on the same schedule.

Estoppel landscape — § 315(e)(2) is a non-issue here. There are no petitioners, so there is no estoppel carve-out and no IPR-based estoppel defense to worry about. Conversely, you get no free ride from a prior PTAB record either: there is no FWD to point to, no claim that has been canceled, and no Board finding of invalidity you can borrow. Every invalidity theory you want to run has to be built and litigated from scratch, in district court or in a new IPR — but see the timing problem below before you spend money on either.

Pattern signals. The '786 was heavily cited but never attacked. Its assignee chain ran Alphanet Telecom Inc → Alphanet Hospitality Systems → Alpha Group Holdings LLC → Alpha Holdings Group LLC → eBay Inc. (2008-02-01) → PayPal, Inc. (2015-07-22). eBay/PayPal acquired it late, right as it was expiring, and never monetized it through assertion or defended it at the PTAB. There is no defensive aggregator (Unified Patents, RPX, etc.) anywhere in the chain — consistent with the absence of a threat to aggregate against.

Why the well-asserted-patents-get-IPRs heuristic breaks down here. IPR became available 2012-09-16. The '786 expired roughly 29 months later. A would-be petitioner in that window faced an expired patent, meaning no injunction and a shrinking damages tail — the classic reason no one files. That is the most likely explanation for the empty docket, not robustness. Note also that CBM review would have been a poor fit even if it had been available: CBM's "financial product or service" gate is awkward for a generic unified-messaging patent, and the CBM program sunset on 2020-09-16.

Recommended next steps

• If a demand letter cites US 5,608,786, lead with expiration, not with PTAB. The patent expired 2015-02-13 and it is now 2026-10-01 — more than a decade past expiry and more than four years past the end of the § 286 six-year damages lookback measured from the last possible infringement (2015). Any royalty theory is effectively dead on arrival; there is no injunctive exposure at all. No claim was ever canceled, so do not tell a court or adversary that "the PTAB killed claims 1–5" — that is false for this patent and would be sanction-bait against you.
• Do not budget for an IPR as a defensive play. You can, in principle, still file an IPR against an expired patent (the Board applies Phillips claim construction to expired claims under Wasica), but the cost/benefit is poor when there is no live infringement case and no damages tail. The realistic defense is a Rule 12 motion built on expiration, laches/§ 286, and the absence of any asserted claim within the damages period.
• If you nevertheless need the invalidity record for a counterclaim or declaratory judgment, the strongest citation lineage is the Examiner's own use of Gordon '786 as § 102(b) anticipatory art against US 6,333,973 (see IPR2015-00201, Ex. 1002), plus the dense third-party citation traffic (the '786 shows up as an X/Y reference in EP 1 261 179 and other unified-messaging applications). Those give you a ready-made obviousness story that does not depend on any PTAB proceeding.
• If you want to double-check the negative before relying on it, the canonical sources are the USPTO PTAB E2E / Patent Trial and Appeal Board End-to-End system (https://ptab.uspto.gov) and the PTAB Decisions page at USPTO, searched by application number 08/387,162 and patent number 5,608,786; the Google Patents page for the '786 (https://patents.google.com/patent/[US5608786A](/patent/US5608786A)/en) lists litigating family flags but no PTAB proceedings. I could not find any settled-and-unindexed or sealed proceeding, and I found no ex parte reexamination either — but I flag that a sealed settled CBM or a very old unindexed filing is the one category of proceeding my searches cannot conclusively rule out.

I'll research the assignment record for US 5,608,786. Let me run several targeted searches.

Let me dig into the specific assignment records and the Alphanet chain.

Important limitation up front

I could not retrieve the actual reel/frame numbers or correspondent-of-record fields for US 5,608,786. My searches did not surface a USPTO Assignment Center / Assignment Abstract of Title extract for this patent number, and the assignment records I did find belonged to unrelated patents. Because fabricating reel/frame values would be the single most damaging error in an ownership-chain analysis, I have left those fields marked "not retrieved" rather than filling them in. Everything below that is stated is grounded in the authoritative full text you supplied (which reproduces the Google Patents legal-events block) and is labeled as such.

The dates in the authoritative text are Google Patents legal-event ("Assigned to…") dates, which are normally the USPTO recording dates, not necessarily the execution dates. I flag that distinction where it matters.

---

Inventors

Pattern notes:

• Single-inventor patent. The inventor assigned his rights to the company (1995-06-21, ~4 months after the 1995-02-13 filing). That is an ordinary employment assignment, not the "all inventors departed within 12 months" precursor to a fire-sale.
• The commercial embodiment described in the patent is the "UniPost" service led by Gordon (the specification uses his own name in the working example — gordon@toronto.unipost.com), so the inventor appears to have been the operating founder, not a detached assignor.
• No inventor-departure signal can be corroborated with the data I could retrieve; I am not inferring one.

---

Original assignee

Alphanet Telecom Inc. (also styled "Alphanet Telecom Inc." in the priority/foreign family; the Canadian priority CA 2139081 C lists the same family, WO 1996020553 A2).

• Product embodying the claims: The patent describes, and the specification's commercial example names, the UniPost unified-messaging service with UniPost Access Nodes (UANs) — a voice/fax/e-mail unified mailbox service. So on the face of the document the original assignee was an operating service provider, not a licensing shell.
• Primary line of business: Telecommunications / unified messaging (Internet access provider with PSTN front end).
• Apparent fate: The 2000-02-09 recording in which PricewaterhouseCoopers Inc. appears as the assignor transferring to Alphanet Hospitality Systems Inc. is consistent with a receivership / insolvency transfer (PwC is an accounting and insolvency firm, not a technology operating company). This is a strong contextual lead that Alphanet Telecom failed or was wound down around 1999–2000 — but I could not confirm a bankruptcy case number, so I state it as a lead, not a finding.
• Current status: Not determinable from the sources I reached. The name disappears from the chain after 2000.

---

Assignment timeline

Reel/frame and correspondent fields: not retrieved. I could not obtain the Assignment Center abstract for '786; the entries below give the recorded chain and dates only, sourced from the authoritative legal-events block.

1. Executed ~1995 (recorded event 1995-06-21) — Reel not retrieved

   • Conveyance: Assignment of assignors interest
   • Assignor: Gordon, Alastair T.
   • Assignee: Alphanet Telecom Inc.
   • Correspondent: not retrieved
   • Context: Inaugural inventor-to-company assignment (normal employment assignment by the founder).

2. 1995-02-13 — Application filed by Alphanet Telecom Inc. (filing event, not a recorded assignment; listed for completeness)

3. 2000-02-09 — Reel not retrieved

   • Conveyance: Assignment of assignors interest
   • Assignor: PricewaterhouseCoopers Inc.
   • Assignee: Alphanet Hospitality Systems Inc.
   • Correspondent: not retrieved
   • Context: Receiver/trustee or secured-party transfer — the PwC-as-assignor pairing suggests the asset moved as part of an insolvency or enforcement process rather than a negotiated corporate sale (lead, not confirmed).

4. 2000-02-09 (same date as #3) — Reel not retrieved

   • Conveyance: Assignment of assignors interest
   • Assignor: Alphanet Hospitality Systems Inc.
   • Assignee: Alpha Group Holdings LLC
   • Correspondent: not retrieved
   • Context: Two-hop cascade in a single recording date — the asset passes from the receivership transferee straight into a holding LLC.

5. 2007-11-08 — Reel not retrieved

   • Conveyance: Change of name (see document for details)
   • Assignor: Alpha Group Holdings LLC
   • Assignee: Alpha Holdings Group LLC
   • Correspondent: not retrieved
   • Context: Change of name only — same entity, reordered words in the LLC name; no change in beneficial ownership.

6. 2008-02-01 — Reel not retrieved

   • Conveyance: Assignment of assignors interest
   • Assignor: Alpha Holdings Group LLC
   • Assignee: eBay Inc.
   • Correspondent: not retrieved
   • Context: Sale into a large operating company — exit from the holding structure to eBay.

7. 2015-07-22 — Reel not retrieved

   • Conveyance: Assignment of assignors interest
   • Assignor: eBay Inc.
   • Assignee: PayPal, Inc.
   • Correspondent: not retrieved
   • Context: Internal corporate reorganization — eBay/PayPal separation; PayPal spun off from eBay in 2015.

8. 2015-02-13 — Anticipated expiration ("Expired – Lifetime"). The patent's enforceable term lapsed, and note the PayPal recording (2015-07-22) post-dates expiry.

No additional post-issuance assignments beyond the above are shown in the authoritative text. If the Assignment Center holds further records (e.g., security interests), I could not see them.

---

Timeline diagram

timeline
    title Ownership of US 5608786
    1995 : Filed by Alphanet Telecom Inc
         : Gordon assigns to Alphanet Telecom
    2000 : PwC transfers to Alphanet Hospitality
         : Then to Alpha Group Holdings LLC
    2007 : Change of name to Alpha Holdings Group LLC
    2008 : Assigned to eBay Inc
    2015 : Assigned to PayPal Inc
         : Patent expires

---

NPE / troll-pattern signals

1. Shell-entity transfer — Unclear.
   Two intermediate entities, Alpha Group Holdings LLC (2000) and Alpha Holdings Group LLC (2007–2008), carry "Holdings"-style names, but (a) "Holdings" is not the "IP/Patents/Licensing/Ventures" tell, and (b) I could not retrieve their addresses, formation records, or product record, and (c) the chain exits to eBay (an operating company) rather than resting in the LLC. Per the rule that naming alone is not a finding, this stays unclear. Reel/frame not available to test for a registered-agent address.

2. Known asserter in the chain — Not present (on available evidence).
   The chain's terminal assignees are eBay Inc. (2008) and PayPal, Inc. (2015) — both large operating companies, neither on the named NPE lists (Acacia, Marathon, IV, Wi-LAN/Conversant, Vringo, Pendrell, Round Rock, etc.). "Alpha Group Holdings LLC" / "Alpha Holdings Group LLC" do not match any entity I could confirm on a Unified Patents / RPX high-frequency-plaintiff list.

3. Repeat correspondent across the chain — Not determinable.
   This is the signal the exercise most wants, and it is exactly the field I could not retrieve. I will not guess. Recommend pulling each reel/frame's correspondent from Assignment Center; if the 2000 and 2008 recordings share one attorney/firm, that would materially change the analysis.

4. Cascading transfers — Present (weak).
   Two consecutive assignments on the same date, 2000-02-09 (PwC → Alphanet Hospitality Systems Inc. → Alpha Group Holdings LLC) is a two-hop cascade within a single recording date. The remaining hops are spread out (2007, 2008), so this is a single clustered cascade, not a sustained chain of chained LLCs under 24 months. Weak signal only.

5. Pre-litigation transfer — Not present / unclear.
   My prior litigation review found no confirmed infringement suit naming '786. Without an assertion date, a "within 6 months before suit" test cannot be run. The 2008 eBay transfer and 2015 PayPal transfer are corporate events, not shown to precede any suit.

6. Bankruptcy fire-sale — Unclear, with a concrete lead.
   The 2000-02-09 transfer in which PricewaterhouseCoopers Inc. is the assignor is the textbook signature of a receivership/insolvency transfer. However, I could not confirm a Chapter 7/11 filing, a Canadian receivership, or a court file. Flagged as a strong lead requiring verification, not a finding.

7. Privateering — Not present.
   No SEC-filing, Patent Progress, or EFF evidence surfaced showing an operating company funding assertions through this patent. No assertion was found at all.

8. Defensive aggregator (anti-NPE) — Not present.
   The chain does not terminate at RPX, AST, LOT, Unified Patents, or OIN. It terminates at PayPal, Inc.

---

Verdict

Insufficient data.

The recorded chain is complete in outline — Alphanet Telecom Inc. → (PricewaterhouseCoopers Inc.) Alphanet Hospitality Systems Inc. → Alpha Group Holdings LLC → Alpha Holdings Group LLC (name change) → eBay Inc. → PayPal, Inc. — and it ends at large operating companies, with no confirmed assertion and no entity on any published NPE list. That actively disfavors an NPE finding. However, I could not retrieve the reel/frame or correspondent fields for any link (so signals 1 and 3 cannot be tested), and no litigation is confirmed, so I cannot responsibly place this in any of the assertive buckets. The one live lead — PricewaterhouseCoopers Inc. as assignor on 2000-02-09, suggesting an insolvency transfer — is precisely the fact pattern the bankruptcy fire-sale bucket was built for, and it should be verified next.

Verification links:

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search Patent Number 5608786)
• Legacy Assignment Search: https://assignment.uspto.gov/patent/index.html
• Google Patents legal events (source of the dates above): https://patents.google.com/patent/US5608786/en

Contradiction flag: none with the previous litigation section — that section likewise could not confirm a case and could not verify the Darts-IP family flag (family=4154930), which covers the continuation US 6,067,350 as well as '786. If the Darts-IP flag is resolved, it may attach to '350 rather than '786, and it should be re-checked against the chain above before any NPE conclusion is drawn.

I'll research the prior-art citations on US 5,608,786. Let me run targeted searches.

Let me try to retrieve the front-page references cited on US 5,608,786 from specific patent databases.

I have partial data but not the patent's own front-page "References Cited" list. Let me target that specifically.

Bottom line up front

I confirmed US 5,608,786 through the authoritative text you supplied (Google Patents mirror of the USPTO grant): US 5,608,786 A, "Unified messaging system and method," Alastair T. Gordon, App. 08/387,162, filed 1995-02-13, priority 1994-12-23, granted 1997-03-04, original assignee Alphanet Telecom Inc. I did not substitute any similar number. Numbers that merely appear elsewhere (e.g., 5,608,786 showing up as a cited reference inside US 7,149,509, US 8,135,791, US 6,385,202, US 6,989,129, etc.) are forward citations to '786, not prior art to '786, and I have kept that distinction — see §5.

Important limitation, stated rather than papered over: the authoritative full text I have does not contain the front-page "References Cited" (U.S. Patent Documents / Non-Patent Citations) list, and my live searches could not retrieve that examiner list. The rendering I hold goes Info → Classifications → Definitions → Landscapes → Abstract → Description → Claims → Family → Country Status → "Cited By (759)" — i.e., it shows only the forward citations. So I cannot truthfully claim to have enumerated "each patent citation for 5608786" as they appear on its face. What I can give you, grounded and labeled, is: (A) the prior art expressly identified inside the '786 specification itself, and (B) the contemporaneous unified-messaging references that share '786's exact art space and dates, with a §102 claim-mapping. I flag provenance for each.

---

1. Legal framework that governs the mapping

'786 has a 1994-12-23 priority date and a 1995-02-13 filing date, so pre-AIA 35 U.S.C. §102 applies. Relevant critical dates:

Anticipation under §102 requires all elements of a claim in a single reference, arranged as claimed. Because I could not pull full texts of each reference in this session, the claim designations below are subject-matter-level candidates, not finished claim charts.

---

2. Art expressly identified inside the '786 specification (verified from the authoritative text)

The '786 description states, verbatim:

"Our U.S. Pat. Nos. 4,713,837, 4,922,518, 4,942,599 and 4,969,184 disclose systems for more efficient transmission and/or retrieval of facsimile communications, which can involve the use of dedicated data transmission networks. Facsimile transmissions have been rerouted in the case of long distance transmissions to a computer which uses a data communication to provide a more effective link to another computer which then uses the public switched telephone network to phone out and complete the transmission."

Caveat: I verified these four numbers because they appear in the '786 text itself. I did not verify whether they also appear on the '786 front page, and I could not pull their exact titles/issue dates in this session, so I have not asserted titles. (Two further Gordon patents, US 4,994,926 and US 5,291,302 — cited in US 5,777,754 as "SAFF" art — appear in the same Gordon family art space, but I found them cited in a different patent, so I do not list them as '786 citations.)

---

3. Contemporaneous prior art in '786's exact art space (dated; §102 candidates)

These references were surfaced from the citation table of US 6,711,154 B1 (Google Patents mirror, via web.archive.org: https://web.archive.org/web/20200501105133/https://patents.google.com/patent/US6711154B1/en#6) and cross-checked in the forward-citation tables you supplied. Provenance note: they are not confirmed as '786 front-page citations; they are the unified-messaging references that the art and later examiners consistently treat together. Dates shown are as they appeared in search results.

Strongest single-reference anticipation candidates: US 5,742,905 (Bellcore) for independent claims 1/9/11 on §102(e) grounds, and US 4,837,798 (AT&T) for the unified-mailbox/summary claims (6, 16–17) on §102(a)/(b) grounds. Both would need full claim charts to confirm — I have not done element-by-element charts here.

---

4. What '786's claims actually require (so you can see why mapping is constrained)

Independent claim 1 requires: a plurality of commercial access providing computers associated with INTERNET; many collectively unrelated registered subscribers; E-mail + fax + voice-mail addresses; storage on any address; retrieval by telephone set OR computer+modem; CAPs exchanging information over INTERNET; and subscriber access to any CAP with messages transferred over INTERNET. Independent claim 9 is the method counterpart; claim 11 is the unified-messaging system counterpart. Dependent claims add: fax-machine retrieval (2), broadcast alert (3), auto-conversion of E-mail→fax (4), DID-line addresses (5), audible summary (6), geographically distributed/cost-effective CAPs (7), Internet long-distance voice (8), rerouting to a closer computer + PSTN delivery (10), call forwarding (12), same fax/voice address with fax-vs-voice discrimination (13), conversion to E-mail/fax (14–15), summaries (16–17), and fax↔text conversion (18).

Because the independent claims hard-require INTERNET-connected CAPs exchanging messages over the Internet, no pre-Internet reference (e.g., US 4,837,798) can anticipate claims 1/9/11; such art is confined to the dependent claims or to the sub-combinations. That is the key structural takeaway.

---

5. Documents that are NOT prior art to '786 (guard against mis-classification)

Everything in your authoritative text's "Cited By (759)" table is a forward citation, and its dates prove it: e.g., WO 1997027692 A1 (Firetalk, priority 1996-01-23), WO 1997028628 A1 (priority 1996-01-31), WO 1997029581 A1 (1996-02-09), US 5,728,851? / US 5,724,410 (1995-12-18), US 5,768,513 (1996-06-27), US 5,777,754 (1996-06-21). Likewise US 5,675,507 and US 5,870,549 (Bobo, filed 1995-04-28), US 5,805,587 (1995-11-27), US 5,991,291 (1995-12-19), US 5,946,386 (1996-03-11) all post-date the 1994-12-23 priority and therefore cannot be §102 prior art to '786 — they cite '786, not the reverse. My earlier litigation section already noted '786 appears only as a cited reference in unrelated proceedings (e.g., IPR2022-00948 exhibit lists) — same category.

---

6. Explicit gaps (not fabricated)

1. I could not retrieve the examiner's front-page "References Cited" list for '786 — neither the supplied authoritative text nor my live searches produced it (https://patents.google.com/patent/US5608786A/en; https://uspto.report/patent/grant/5608786). It is possible '786 has a very short or absent backward-citation list, but I cannot confirm that.
2. I did not perform element-by-element anticipation charts against the full texts of the references in §3; my claim designations are subject-matter candidates.
3. I could not verify exact titles/issue dates for US 4,713,837 / 4,922,518 / 4,942,599 / 4,969,184 beyond the numbers quoted in the '786 specification.

To close the gap, the authoritative sources to pull next are: USPTO Patent Center / Patent Full-Text (patft/Patent Center) for the '786 front page; Espacenet "Cited documents" for US 5,608,786 (the EPO GPI pages I saw treat US 5,608,786 as an [A]/[XY] citation — http://data.epo.org/gpi/EP1261179A3, https://data.epo.org/gpi/EP0971299A3.pdf); and the file wrapper/IDS for App. 08/387,162. If you authorize another search round against those, I can produce a true examiner-citation table with confirmed dates and a per-claim §102 chart.

I'll research the prior art applicable to US 5,608,786, focusing on the references cited on its Google Patents page and the field's state of the art as of the 1994 priority date.

Let me verify the specific examiner-cited prior art references on the '786 front page and a few additional candidates relevant to the store-and-forward fax and IP-voice aspects.

Obviousness Analysis of US 5,608,786 under 35 U.S.C. § 103

0. Framing, sources, and caveats

What I used from the page's prior-art content. The Google Patents "Prior Art" material supplied with this patent consists of three things:

1. Prior-art keywords: "subscriber, messages, mail, computer, facsimile."
2. References the applicant itself cited in the '786 Background as "our earlier patents": US 4,713,837; US 4,922,518; US 4,942,599; US 4,969,184 (all Gordon / Gordon et al.). These are applicant-admitted prior art and all predate the 1994-12-23 priority date (granted 1987, 1990, 1990, 1990).
3. "Cited By (759)" — because these postdate '786, they are not § 102/§ 103 prior art against '786. Any office action that cited them did so as later art citing '786. I flag this because it is a common analytical trap.

Verified external prior art. In addition I confirmed the existence and content of US 4,837,798 (Cohen et al., "Communication system having unified messaging," AT&T, filed 1986-06-02, granted 1989-06-06) via Google Patents, Espacenet, and secondary sources. It is squarely prior art (pre-1994) and is the seminal "unified messaging" reference.

Explicit limitation on what I could establish. I was unable to retrieve the patent's front-page examiner-cited US patent list during this session (search limits). I therefore do not claim to have the complete 102(b)/102(e) reference set. My analysis is built only on references I verified (above) plus the applicant's own admissions in the specification. Where I reach the limits of that set, I say so rather than inventing a citation. Note also that under the pre-AIA framework applicable here (filed 1995), § 103 uses Graham/KSR/TSM analysis, not the AIA's Helsinn formulation.

Cross-reference to the earlier litigation section. Because '786 is Expired – Lifetime (anticipated expiration 2015-02-13), any § 103 analysis now is retrospective/hypothetical — relevant only to a backward-looking validity challenge over pre-2015 conduct, not to prospective enforcement. That does not change the technical analysis below.

---

1. The independent claims to be analyzed

• Claim 1 (system): a plurality of commercial access providing computers (CAPs) associated with INTERNET; many collectively unrelated registered subscribers; each subscriber assigned an E-mail address, a fax telephone address, and a voice mail telephone address; message to any address is received and stored by the associated CAP; retrieval by telephone set or computer+modem; the CAPs use INTERNET to exchange information; a subscriber may access any CAP; messages stored in the associated CAP are transferred over INTERNET to the CAP the subscriber contacted.
• Claim 9 (method): receipt of voice/E-mail/fax at a computer system connected to a data network; plurality of CAPs interconnected by a high-speed data network; each CAP has direct-in-dial (DID) lines to the PSTN; subscriber retrieves via PSTN and data network, or via PSTN direct connection with any CAP; computers communicate the messages between them through the data network.
• Claim 11 (system): computer system = plurality of CAPs interconnected by a high-speed data network; each CAP has a bank of DID lines and acts as a commercial access provider for the data network through which E-mail is delivered; provides E-mail address, fax telephone address, voice mail telephone address; stored for retrieval; accessible via PSTN and/or the high-speed data network; messages exchanged between CAPs over the data network.

The three independent claims are the same inventive concept (unified mailbox reachable from phones/fax/PC, distributed across Internet-connected access nodes). Dependent claims 2–8 and 12–18 add: fax retrieval (2), broadcast alert (3), E-mail→fax conversion (4), DID telephone numbers (5), audible summary (6), geographically separated CAPs (7), long-distance voice over the data network (8), PSTN-forwarding of routed messages (10), call forwarding (12), single fax/voice address with discrimination (13), media conversion + automatic delivery (14, 15), summary in voice / any form (16, 17), fax↔text conversion (18).

---

2. The two-reference core: Cohen '798 + Gordon '184 (with Gordon '837/'518/'599)

2.1 US 4,837,798 (Cohen et al.) — teaches the unified mailbox and media conversion

Verified disclosures (from the Google Patents/Espacenet text retrieved this session):

• "a single electronic mailbox for different types of messages… The mailbox can be on a user's host computer, PBX, PC."
• The "unified messaging system (UMS 10) is the one access point for all messages regardless of the message type and regardless of the message origination."
• A user has "unified access to any messaging service such as… electronic mail 110, voice mail 109, private data system 111, local area network 112, message coverage 113 or fax 114."
• A user "is able to retrieve messages from their chosen unified messaging mailbox using any of several terminal types… (101–108), from any location, local or remote."
• Retrieval is described for a multi-frequency (MF) voice telephone (dial a number, enter login/password, hear header: "Thomas Smith x3887 called on April 4th at 10:15 am, the message is 20 seconds long"), and for data terminals.
• "Text-to-speech converter 13 uses the well-known text-to-speech technology for media conversion so that most types of messages can be retrieved in voice form from a conventional voice telephone"; messages are "translated from one media to another for reception"; a single message may be composed of parts using different native media.
• Alerting is provided (message waiting lamp; "a user-designated receptor receives a new message, be it text, voice or facsimile, [and] that user is alerted").
• "Message coverage 113" is expressly listed as a service — i.e., coverage of unanswered calls into the mailbox.

What Cohen '798 does not contain (this is where the Gordon references are needed): no public commercial access provider for a data network serving unrelated subscribers; no Internet/CAP E-mail address combined with fax and voice-mail telephone addresses; no DID line bank; no reliance on a public data network as the transport between distributed access nodes; no PSTN-based final delivery to a destination fax.

2.2 US 4,969,184 (Gordon et al.) — teaches DID addressing + store-and-forward over a digital data network + multi-node routing

Verified disclosures:

• A "process intermediary" / local node leases "a plurality of telephone number addresses" that are "sometimes referred to as DID numbers in the case of a private company leasing such addresses with DID standing for 'direct in-dial' numbers… these numbers do not represent dedicated telephone lines, but merely telephone addresses which when used, result in a communication with the local node."
• Each facsimile machine (or "other data transmitting/receiving device") has its own TNA/TNA number; the originating fax dials the TNA, which routes into the local node rather than to a dedicated line.
• "The local node 8 is connected to a digital data network 14 whereby the data… can be transmitted between local nodes 8 of the digital data network. In this way, long distance charges associated with a conventional telephone communication are replaced by a significantly reduced expense incurred for transmitting over the digital data network." (i.e., multiple geographically separated nodes interconnected by a data network — claims 7, 9, 10, 11.)
• The intermediary "based on the telephone number address, creates a non-telephone signal to actuate or otherwise advise the destined… device of the receipt of an addressed data message", which then calls in to retrieve (support for claims 3 and 12's "outdial/notify" concept).
• Note the express negative limitation: Gordon defines "digital data network" as "not includ[ing] networks capable of transmitting real time continuous voice communications." This is why claim 8 (real-time voice over the data network) is the one element not supported by this reference (see § 5).

2.3 US 4,713,837 / 4,922,518 / 4,942,599 (Gordon) — teach non-telephone broadcast alert and forced delivery

• '837 ("Communication network"): a broadcast channel (e.g., FM paging) carries an authorizing/instruction signal identifying a subscriber station by code; on receipt the station automatically seizes an available telephone line and dials up the central process intermediary to receive the information. It also lists electronic mail and electronic billing among the services, and uses a central intermediary serving "a host of subscribers."
• '518 (selective dissemination): same non-telephone authorizing-signal + telephone-channel delivery, with FM paging, and an acknowledgement signal causing retransmission if not properly received.
• '599: subscriber devices automatically complete a telephone call to a predetermined computer, which then supplies location-specific information.

These directly supply claim 3 (broadcast alert to subscriber equipment).

2.4 Additional Gordon facsimile store-and-forward art (medium confidence on exact citations)

Reference lists in later Gordon family patents (e.g., US 7,365,884) recur on US 4,994,926 and US 5,291,302 (Gordon et al., facsimile telecommunications / store-and-forward). Both predate 1994-12-23. I annotate these medium confidence because I did not independently open the documents this session; I do not rely on any specific passage from them.

---

3. Grounds of rejection

Ground 1 — Cohen '798 in view of Gordon '184 → claims 1–7, 11, 13–18

Element mapping (representative):

Why a POSITA would combine them. Both are in the same field (stored/forwarded, telephone-accessible messaging). Cohen presents the unified multimedia mailbox but leaves open how to host it for a large, dispersed, service-provider subscriber base and how to reach it from the installed base of ordinary fax machines and telephones. Gordon '184 supplies precisely that missing piece: a shared, centrally located intermediary that hands out DID telephone addresses to a host of unrelated subscribers, stores addressed faxes on their behalf, and moves the payload over a data network (PSTN only on the local legs) to cut long-distance cost. The rationale is the classic KSR one: a known problem (cost of long-distance messaging + reaching non-computer terminals) met by a known solution (DID-addressed store-and-forward over a data backbone), with a reasonable expectation of success because both are proven messaging technologies. Both references even share the motivation the '786 Background admits — combining message types and serving "hundreds of millions of telephone and facsimile devices."

Claim-by-claim result under Ground 1:

• Claim 1 — obvious. Cohen supplies the unified multi-media mailbox and multi-terminal retrieval; Gordon '184 supplies DID fax/voice telephone addresses, a commercial intermediary serving unrelated subscribers, and the data-network backbone with transfer between nodes. The "E-mail address for INTERNET" limitation is supplied by the applicant-admitted prior art (the '786 Background describes Internet CAPs giving "each subscriber… an Internet address and password number for retrieval of E-Mail"); no reference is needed beyond the admission.
• Claim 2 (facsimile-machine retrieval) — obvious; Gordon '184's terminal is a fax machine and Cohen teaches multi-terminal retrieval.
• Claim 3 (broadcast alert) — obvious over Cohen+Gordon, and directly taught by Gordon '837/'518 (non-telephone broadcast authorizing signal that triggers the subscriber unit to call in).
• Claim 4 (E-mail auto-converted to fax, forwarded to subscriber's fax address) — obvious; Cohen's cross-media translation + Gordon's outbound fax-to-a-fax-address.
• Claim 5 (fax/voice addresses are telephone numbers over DID lines) — obvious; literally Gordon '184's DID/TNA teaching.
• Claim 6 (audible summary by telephone) — obvious; Cohen's MF-telephone header playback + TTS.
• Claim 7 (CAPs in different geographic areas; Internet between them) — obvious; Gordon '184's geographically separated local nodes on the data network.
• Claim 11 — obvious for the same reasons as claim 1, with DID banks (Gordon) and E-mail delivery over the data network (admitted CAP art).
• Claim 13 (single fax/voice address; system distinguishes) — obvious. Cohen's "single controllable point of contact" for all message types, combined with the well-known auto-detection of fax vs. voice calls (the patent's own class Y10S379/902, "Auto-switch for an incoming voice data, or fax telephone call, e.g. comp/fax/tel," confirms this was a recognized art class), makes the discrimination routine.
• Claims 14, 15 (conversion + automatic delivery) — obvious; Cohen's translation plus Gordon's automatic outbound/alert-driven delivery.
• Claims 16, 17 (summary as voice / voice, E-mail, or fax) — obvious; Cohen's header/abstract retrieval in voice (TTS) and on data terminals.
• Claim 18 (fax↔text conversion) — obvious; Cohen's express "translated from one media to another," including TTS and document attachment handling; OCR/text-file handling of fax images is a conventional design choice.

Ground 2 — Ground 1 + conventional call forwarding/coverage → claims 12 (and reinforces 9, 10)

Claim 12 requires that a call to the subscriber's equipment, if unsuccessful, is automatically call-forwarded to the provider. The '786 specification itself states: "This method uses the call forwarding technology of existing public switched telephone networks." Cohen '798 expressly lists "message coverage 113" — coverage of an unanswered call into the mailbox. Combining a standard PSTN call-forward-on-busy/no-answer feature (admitted art) with the Cohen/Gordon mailbox is a predictable, result-driven substitution with a reasonable expectation of success. Claim 12 is obvious.

Claims 9 and 10 are also met by the Ground 1 combination: Gordon '184's multiple nodes interconnected by a data network, with the PSTN used only for local origination/delivery legs, is exactly claim 10's "route via the data network to the computer closer to the telephone address, then forward via the PSTN."

Ground 3 — for the "E-mail over the data network + Internet exchange" limitations (claims 1, 9, 11)

The Internet CAP, the subscriber's Internet E-mail address/password, and E-mail transport over the data network are applicant-admitted prior art (the '786 Background ¶ on Internet CAPs). Read together with Gordon '184's data-network backbone between nodes, the "CAPs use INTERNET to exchange information" limitation is a substitution of one known data network (the Internet) for the generic "digital data network" taught by Gordon — a predictable substitution of a known equivalent, further supported by the admitted CAP art.

---

4. Motivation to combine — the KSR/Graham case

A person of ordinary skill at the 1994 priority date would be a telecommunications/messaging engineer familiar with PSTN switching, DID/service-node architecture, voice mail, fax store-and-forward, LAN/WAN e-mail, and multimedia mailboxes. The motivation to combine Cohen and Gordon is strong on each KSR axis:

1. Same field, overlapping problem. Both address stored, telephone-accessible messaging; both aim to reach non-computer terminals and reduce cost.
2. Known need, articulated in the art. The '786 Background itself concedes E-mail "popularity does not nearly approach the popularity of voice and facsimile messaging," and states "the present invention has recognized the need for a system… for combining these different message types." An admitted, long-felt industry need for message-type convergence cuts toward obviousness.
3. Predictable combination of known elements, no change in principle. Cohen's conversion/retrieval engine + Gordon's DID/store-and-forward/network backbone each perform their known functions with a reasonable expectation of success.
4. Market/design incentives. Reducing long-distance charges and giving E-mail-only correspondents access to the telephone/fax installed base are exactly the cost/utility drivers that made the combination obvious (KSR: "design incentives… market forces").

---

5. Where the obviousness case is weakest: claim 8

Claim 8 ("ability to… use INTERNET for long distance voice transmission") is not established by Cohen + the Gordon references, because Gordon '184 expressly excludes real-time continuous voice from its "digital data network," and Cohen '798 concerns stored messaging, not real-time packet voice. Rendering claim 8 obvious would require an additional reference teaching digitized/packetized voice transported over a data network with sequencing/routing logic (the '786 specification describes exactly this: protocols "which ensure[] that the digitized voice packets remain in sequence"). I was unable to verify such a reference from the page's prior-art content in this session, so I do not cite one. I do not assert claim 8 is obvious on a two-reference basis; a challenge to claim 8 would need separately-verified § 102/§ 103 art on packet voice.

---

6. Summary table

---

7. Honest limits of this analysis

• I did not obtain the '786 front-page examiner-cited US patent list this session; my reference set is Cohen '798, the four applicant-admitted Gordon patents, and the admitted Internet CAP art. A complete § 103 challenge should be run against the full ITS/IDS and the examiner's citation set.
• I did not verify US 4,994,926 / 5,291,302 by opening the documents; they are noted as medium-confidence corroborating art only, and nothing above depends on them.
• The "Cited By (759)" references are later art and cannot support § 103 against '786; I have not used them for that purpose.
• Claim 8 is the identified gap; I have deliberately not manufactured a reference for it.
• Because '786 expired 2015-02-13, this is a retrospective validity analysis with no prospective-enforcement consequence.

If you can authorize another search round, the highest-value follow-ups are: (a) the actual examiner-cited references on the '786 front page, and (b) verified packet-voice/packetized-voice prior art (pre-1994) to address claim 8.