US 7254605

I'll search for this specific patent number, including litigation/docket context.

Initial queries returned nothing. Let me try alternative search formulations.

Let me search for the claims text and any litigation or Federal Circuit activity.

Search report — US 7,254,605

Scope note: I ran the requested lookups against the patent corpus and general web index. The two initial targeted queries ("US patent 7254605" and "7254605 patent Federal Circuit 2026") returned zero results, which I report rather than fill in. Broader queries surfaced the patent document itself and assignment/docket-adjacent records. I did not obtain the verbatim granted claim set (the fetched patent text the user supplied is truncated in the "Detailed Description," before the claims section), so the claim overview below is flagged as reconstructed, not quoted.

---

1. Bibliographic data (per the authoritative patent text)

Ownership chain of record: speakout.com (assignment recorded 2000‑10‑26, from Strum) → name change to Ntercept Communications, Inc. (recorded 2004‑11‑22) → AGENCY.COM LTD. (recorded 2004‑09‑10) → Austen Services LLC (recorded 2004‑09‑10) → Zarbaña Digital Fund LLC via merger (recorded 2015‑12‑11).

Continuations claiming priority to this patent: US 7,809,788 B2 (from app. 11/826,363, filed 2007‑07‑13) and US 7,806,273 B2 (from app. 11/878,119, filed 2007‑07‑20).

Cross-referenced applications in the specification: 60/177,704; 09/544,624 ("the '624 application"); 09/614,862; 60/185,117; 09/544,337; 09/544,623; and 09/695,929, filed concurrently ("Method, System and Computer Program Product for Controlling Data Loss in a Distributed Opinion Research Network").

Minor discrepancies, flagged not corrected:

• Google Patents gives the title as "Method of modulating the transmission frequency…" whereas the USPTO assignment record (Plainsite, Assignment 15109/396) describes the application as "Method, System And Computer Program Product For Modulating The Transmission Frequency In A Real Time Opinion Research Network." The issued title controls per the patent text supplied.
• The "Definitions" block on the Google Patents page is an auto-generated terminology extraction, not part of the specification.

---

2. Abstract (authoritative text)

"A computer-implemented transmission scheme is provided to control client-server interchanges within a distributed communications network, such as a real time opinion research system. Interchanges include transmitting media streams between one or more clients to a server over a computer network, including the global Internet. A polling management unit sets and manages the transmission mode that includes event-driven and periodic interchanges. Periodic interchanges can be simultaneously or staggeredly transmitted to a sampling pool of active clients. A transmission mode unit implements the transmission scheme set by the polling management unit. A parameter selector establishes the transmission interval and transmission period which are used to trigger each communication interchange. A client assignor creates one or more sampling classes from the sampling pool by applying a sampling quotient that is generated by the parameter selector. A schedule editor produces a transmission schedule for the active clients. If more than one sampling class has been created, each sampling class would receive a separate transmission schedule for providing staggered transmissions at designated transmission intervals. The transmission schedule can include other data preparation and formatting instructions for compression, aggregation and packetization."

---

3. Plain-language overview of the claims

Important caveat — read this first. The patent text I was given is cut off mid‑sentence in Section III of the "Detailed Description" ("…poll/media manager 142 uses media management server 134 to conduct opinion research for an event. In an embodiment"). I do not have the granted claims in hand, and I found no reliable source reproducing the claim set verbatim. I therefore cannot tell you the exact number of independent claims, their preambles, or their precise limitations. Treat the following as a functional reconstruction from the specification's stated invention and the preferred embodiments, not as quoted claim language. If verbatim claim text matters for your analysis, it must be pulled from the USPTO PatentCenter/PatFT record for 09/695,932 (or the printed patent columns).

Based on the specification, the claimed subject matter centers on:

1. A method (and corresponding system / computer program product) of modulating transmission frequency in a real-time opinion research network — i.e., controlling how often each client and the server exchange opinion data and composite-result/display updates, rather than merely what they exchange. The specification frames the transmission scheme as an operational mode set by the polling management unit (314), distinct from synchronization, auto-reset, and time-out modes.

2. Establishing a "sampling pool" of active clients — the clients with live sessions for an event, identified via the registering unit (310), with a transmission interval and a transmission period selected/established by a parameter selector (414).

3. Applying a "sampling quotient" to divide the sampling pool into one or more sampling classes by a client assignor (416). In the disclosed embodiments the sampling quotient is the inverse of the transmission period, or a multiple thereof, or independently selected.

4. Generating a transmission schedule by a schedule editor (420), assigned per client (simultaneous scheme) or per class (staggered scheme), instructing each client/class when to commence and repeat interchanges. The specification's worked examples (FIGS. 7–8) show a four-interval transmission period with classes offset by one interval, and an embodiment with a primary class on a shorter period than secondary classes.

5. Optionally, the transmission schedule carries data-preparation instructions: frame aggregation into a single packet, compression, encryption, and synchronization of client clocks to an absolute time standard (e.g., Greenwich Mean Time).

The stated technical problem and advantage: IP/UDP networks cannot reserve bandwidth or guarantee QoS, so buffering/aggregation adds latency and lost packets add retransmission delay. Staggering the interchange times reduces instantaneous bandwidth demand and lets the servers "process and respond to client feedback" between intervals, enabling larger panels.

---

4. Litigation and CAFC docket search

• USPTO database: I did not access a USPTO-hosted docket directly; the confirmation above comes from the authoritative patent record plus a USPTO assignment record (Assignment 15109/396) mirrored at Plainsite.
• CAFC 2026 dockets: No results. Targeted queries for this patent number in a 2026 Federal Circuit context returned nothing. I found no evidence of any CAFC appeal, IPR, or district-court action involving US 7,254,605 in the material returned.
• This is consistent with the patent's status: it is recorded as Expired – Lifetime with an adjusted expiration of September 15, 2023 (i.e., ~20 years from the October 26, 2000 filing plus roughly 1,050 days of adjustment). A 2026 appeal from a live infringement/validity dispute is possible in principle only for past damages, but nothing in my searches indicates one exists.
• Note the unrelated patents that cite 7,254,605 as prior art (e.g., US 8,132,200 "Intra-video ratings"; US 11,483,610), and that the "7254605" string collides with Zarbaña's unrelated RF/amplifier portfolio — those are not this patent and should not be conflated.

---

5. Explicit uncertainties

• Claim set: unavailable to me; the overview in §3 is a reconstruction from the specification and is not a substitute for the granted claims.
• Assignee accuracy: the Google Patents page itself disclaims accuracy of assignee listings; the chain above is per the recorded assignments, but I have not verified current ownership beyond the 2015 merger record.
• Term adjustment: the 2023‑09‑15 adjusted expiration is taken from the patent record; I did not independently verify the PTA calculation (which would require the printed face of the patent).
• Docket coverage: a null search result is not proof of absence. Commercial docket databases (Docket Alarm, Lex Machina, RPX) were not directly queried and may hold records my queries did not surface.

I'll research litigation involving US patent 7254605. Let me run several targeted searches.

Initial results are inconclusive. Let me search the patent owner's litigation activity, since the patent was acquired by a fund that appears to be a monetization entity.

Let me search for the exact patent number in litigation contexts and check the litigation portal.

Summary: No litigation specifically involving US Patent 7,254,605 was found

I was unable to identify any court case — asserted-infringement suit, declaratory-judgment action, or PTAB proceeding — in which US 7,254,605 was the patent-in-suit. Here is what my searches did and did not establish.

What I searched

• Direct searches on the patent number: "7254605" patent lawsuit defendant, "7,254,605" infringement complaint, "7254605" Zarbana OR "Austen Services" patent case
• Owner-based searches: "Zarbana Digital Fund" patent litigation, "Austen Services" patent infringement lawsuit
• Sources: general web results pointing to PACER Monitor, CourtListener/RECAP, Unified Patents portal, PTAB filings, and USPTO records.

Result: no hit tying US 7,254,605 to any docket. The patent-number-driven queries returned unrelated patents (e.g., U.S. 5,894,554, U.S. 9,659,285, U.S. 12,103,576) or unrelated defendants. I am reporting this as "not found," not as a verified negative — see caveats below.

Relevant ownership history (from the USPTO assignment record)

The '605 patent changed hands several times, which is the context in which any enforcement activity would arise:

Source: Google Patents "US7254605B1" assignment/reassignment and legal-status data (https://patents.google.com/patent/US7254605/en).

Important caution about the "Zarbana" hit

Current searches do show active 2026 litigation naming Zarbana Digital Fund LLC as a defendant — e.g., Hartford Fire Insurance Co. et al. v. Intellectual Ventures I LLC et al, 1:26-cv-00392 (D. Del., filed Apr. 7, 2026); The Travelers Indemnity Co. v. Intellectual Ventures I LLC et al, 1:26-cv-00397 (D. Del., 2026); and Robinhood Markets, Inc. v. Intellectual Ventures I LLC et al, 1:26-cv-00774 (D. Del., 2026) (PACER Monitor; NewsBreak/Claims report dated 2026-04-09).

However, these are declaratory-judgment actions involving other IV/Zarbana patents, not US 7,254,605. For example, the Hartford complaint describes Zarbana as assignee of "the '183 Patent." The docket excerpts I retrieved do not list 7,254,605 among the patents at issue. So the presence of Zarbana in these suits should not be reported as litigation over the '605 patent.

Caveats (please verify before relying on this)

1. My searches hit a step limit and did not exhaust PACER, Docket Navigator, Lexis CourtLink, or Westlaw KeyCite — the authoritative tools for a complete litigation history.
2. The patent expired on 2023-09-15, which substantially reduces the likelihood of recent enforcement, but does not rule out suits filed during the ~2000–2023 term.
3. There may be litigation from the 2000–2007 Speakout.com / Ntercept / Austen Services era (a period when real-time polling technology was actively commercialized) that surfaced only in sources I could not reach.

Recommendation

To confirm a negative, run a docket-level check: (a) Lexis CourtLink or Westlaw KeyCite "Litigation History" on U.S. Patent 7,254,605; (b) PACER full-text search for "'7254605' or '7,254,605'" and for "U.S. Patent No. 7,254,605"; and (c) USPTO Patent Public Search / Patent Assignment Search for any reexamination or PTAB proceeding referencing the patent. If those return nothing, the correct statement is: "No litigation involving U.S. Patent 7,254,605 is known."

Bottom line: On the record available to me, I cannot identify any litigation involving US 7,254,605, and I will not attribute the 2026 Zarbana/IV declaratory-judgment cases to it, since those cases concern different patents.

Proceedings overview

Zero AIA trial proceedings on file. The structured "PTAB proceedings on file" block supplied in this prompt — the canonical list from the USPTO Open Data Portal — returns no IPR, no PGR, and no CBM for US 7,254,605 as of the most recent ingest, so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution-denied. My independent web searches (patent-number queries; petitioner/owner queries; CBM-specific and reexamination-specific queries) surfaced nothing tying 7,254,605 to any PTAB docket — see the caveats below on the limits of that verification.

Bottom-line defensive posture for a defendant: the patent is untested at the PTAB, not hardened. There is no prior FWD, no institution decision, no claim construction, and therefore no § 315(e)(2) estoppel against anyone. Every defense that has ever existed against this patent is still available — but see §"Strategic summary," because the more important fact is that the patent expired on 2023-09-15 and there is no live enforcement to defend against.

---

Proceedings

None. There are no proceeding numbers to report, and I will not invent any.

For completeness — the three AIA vehicles and why this patent's posture is what it is:

---

Strategic summary

Claim status: every claim is UNTESTED — none canceled, none sustained. This is an unusual posture and worth stating plainly: in the ~11 years between the AIA's IPR provisions taking effect (2012-09-16) and the patent's adjusted expiration (2023-09-15), no third party ever filed a post-grant challenge. The absence is itself a signal, and the correct reading is not "the patent is strong." The correct reading is "the patent was never worth attacking." Three structural reasons explain the null result, and each is more probative than any merits analysis: (1) the patent is a 2000-vintage pre-AIA business-method-flavored software patent in a niche — real-time opinion research / dial-testing — where the commercial players were few and the damages exposure for any given accused product was small relative to IPR fees; (2) the asserted-rights entity Zarbaña Digital Fund LLC acquired the patent by merger in 2015-12-11, and its IV-affiliated portfolio activity that did attract 2026 declaratory-judgment litigation involved other patents (the "'183 Patent" per the Hartford complaint), not the '605; and (3) the patent simply ran out of term — the economic incentive to invalidate an asset whose damages window was closing was never strong enough to justify a petition.

Estoppel landscape: clean slate. Because no IPR reached a final written decision, § 315(e)(2) estoppel has never attached to any party — not to a petitioner, not to their real parties in interest, and not to privies. There is no estoppel bar preventing any defendant from raising any § 102/§ 103 ground in district court, and there is no IPR-driven narrowing of the claims to navigate around. A defendant therefore has the full menu: § 102/§ 103 invalidity on any art (including art a hypothetical petitioner "could have raised" in a never-filed IPR — that "could have raised" bar simply does not exist), § 112 written-description/enablement, and — most promising for a 2000-filed, pre-Alice-era claim set directed at scheduling data exchanges over a network — § 101 patent eligibility. § 101 was unavailable in IPR anyway; it is now a district-court-only defense, and it was never tested here. If a demand letter arrives citing this patent, § 101 and § 112 are the natural first lines, precisely because no PTAB record exists to pre-empt or estop them.

Pattern signals. No petitioner filed multiple IPRs on this patent (nobody filed one at all). No defensive aggregator appears in the chain — Unified Patents, RPX, or any other entity is absent from the recorded ownership history (speakout.com → Ntercept Communications → Agency.com Ltd./Austen Services LLC → Zarbaña Digital Fund LLC), which is consistent with the patent never having been asserted. The patent owner has never pursued a PTAB appeal because it has never had a PTAB loss to appeal. Cross-reference / flag: the prior "Patent summary" section notes two continuations claiming priority to this patent — US 7,809,788 B2 (app. 11/826,363, filed 2007-07-13) and US 7,806,273 B2 (app. 11/878,119, filed 2007-07-20). Those are separate patents with separate terms and separate PTAB exposure; I found no PTAB activity on them either, but my searches were light-touch and should not be read as a verified negative for the family. Also flagged, not corrected: the earlier "Litigation summary" correctly warns against conflating the 2026 Delaware DJ actions naming Zarbaña (e.g., Hartford Fire Insurance Co. et al. v. Intellectual Ventures I LLC et al, 1:26-cv-00392 (D. Del., filed 2026-04-07)) with this patent. Those cases concern different patents. Nothing in the Zarbaña litigation record I retrieved puts 7,254,605 at issue — and nothing puts it in front of the PTAB.

---

Recommended next steps

1. If you are a defendant and a demand letter cites US 7,254,605: note first that the patent expired on 2023-09-15 and is recorded as Expired – Lifetime. Enforcement today can only reach past infringement — and only within the § 286 six-year lookback from the filing of a complaint, which from a 2026 filing still reaches 2020-01-01 onward (i.e., pre-expiration conduct may remain actionable). There is no forward-looking royalty or injunction exposure.
2. There is no FWD to cite, because none exists. I cannot link you to a disposition — the honest statement is: "No PTAB proceeding involving U.S. Patent 7,254,605 is known." Do not represent to a court or an adversary that claims were canceled or sustained; nothing has been adjudicated.
3. Rebuild an IPR-based defense from scratch or don't bother. Any IPR you file would be a first-in-time petition with no General Plastic follow-on problem and no § 315(b) problem if you have not been served — but you would be paying six figures to cancel claims in a dead patent. The rational play is a § 101 / § 112 invalidity posture in district court, plus a damages-side attack on the expired-term issue.
4. Verify the null result before relying on it — my searches hit a step limit and did not exhaust the authoritative sources. Run, at minimum: (a) PTAB E2E / USPTO ODP case search on "7,254,605" and on the application number 09/695,932; (b) PTAB E2E "Patent Number" search for any ex parte reexamination or inter partes reexamination (pre-AIA vehicle, 2000-era filings sometimes triggered these — the ODP structured block covers AIA trials, not the full reexam history); (c) USPTO Patent Public Search for a reexamination certificate or certificate of correction on the '605; and (d) the same checks on the family members 7,809,788 and 7,806,273.
5. If, contrary to this report, you locate a proceeding, the analysis changes materially. A single instituted IPR reaching FWD would have (i) created § 315(e)(2) estoppel for that petitioner and its privies, and (ii) either canceled claims (making any theory built on them untenable) or sustained them (hardening the patent against the exact art presented). Flag it to me and I will re-run the claim-level analysis — but it must be a real, verifiable proceeding number from E2E. I will not fabricate one.

Verification caveats (read before quoting this): the "no PTAB activity" conclusion rests on (a) the structured USPTO ODP block in this prompt, which is the canonical source and is the strongest evidence I have, and (b) web searches that returned no hit. A null web-search result is not proof of absence. I did not directly query PTAB E2E's case search interface, PACER, Docket Navigator, Lexis CourtLink, or Lex Machina — a full reexam history in particular is outside the ODP AIA-trial dataset and is the most likely place a 2000–2010-era challenge would hide.

I'll reconstruct the assignment record. Let me first attempt to pull the actual USPTO Assignment Search records (reel/frame and correspondents), since the prior sections did not obtain them.

Found one reel/frame (15109/396). Let me pull the correspondent and the other links in the chain.

Let me try to get the correspondent for the known reel and find the other links.

Inventors

William E. Strum — sole named inventor (per the issued patent front page and the Google Patents record for US 7,254,605 B1 / application 09/695,932, filed 2000‑10‑26).

Employer at time of filing: speakout.com, inferable from the same‑day assignment of Strum's rights to SPEAKOUT.COM (recorded 2000‑10‑26, the filing date). This is the classic employment/assignment‑as‑a‑condition‑of‑hire pattern: rights were never held by the inventor for any meaningful period.

Unusual patterns — no red flag on inventor conduct:

• Only one inventor; no co‑inventor fragmentation.
• I found no evidence of the inventor departing the original assignee within 12 months of filing, and no second career assignment that would indicate a pre‑sale inventor exit.
• However, the specification cross‑references a dense cluster of Strum‑invented, commonly assigned applications filed the same year — 60/177,704; 09/544,624 ("the '624 application"); 09/614,862; 60/185,117; 09/544,337; 09/544,623; and 09/695,929 (concurrently filed). That pattern — a single inventor populating an entire portfolio at one company — is worth noting because it means the entire asset family moved together when the company was wound down (see Timeline).

Caveat: the employment relationship is an inference from the same‑day assignment, not from an employment agreement I retrieved. I did not locate a signed inventor assignment instrument or an employment contract.

---

Original assignee

Two records conflict; flagging per instructions.

• USPTO assignment record (controlling): the first recorded conveyance (filed/recorded 2000‑10‑26) is from Strum → SPEAKOUT.COM. The chain of record therefore begins with speakout.com.
• Google Patents "Original Assignee" field: lists Austen Services LLC as "Original Assignee." This contradicts the recorded first assignment. Google's field appears to be a dataset artifact (likely its earliest non‑inventor assignee in its own crawl), and the page itself disclaims assignee accuracy. For ownership‑chain purposes the first recorded assignment to SPEAKOUT.COM controls.

Speakout.com, Inc. — profile:

• Line of business: dot‑com‑era internet opinion‑research / real‑time online polling company. The patent's commercial embodiment is the real‑time opinion research system described in the incorporated '624 application ("System, Method and Computer Program Product Producing Real Time Polls of Media Events," Strum, filed 2000‑04‑06).
• Did it ship a product embodying the claims? The specification describes the system as implemented in a commercial real‑time polling platform, and speakout.com is the entity that took the inventor's assignment on filing day — consistent with a product‑bearing operating company. I could not independently verify commercial shipment of a product reading on the granted claims in this session; treat "shipped a product" as likely but unverified.
• Current status: defunct. It was renamed Ntercept Communications, Inc. (change of name recorded 2004‑11‑22), and its patents passed out of the family to Agency.com Ltd. and then Austen Services LLC in 2004 — the timing coincides with the dot‑com collapse and the sell‑off of speakout.com's assets. I did not retrieve a bankruptcy filing record for speakout.com.

---

Assignment timeline

Sourcing note (read first). I reconstructed this from the Google Patents reassignment log plus one Plainsite‑mirrored USPTO assignment record. I was only able to confirm one reel/frame — reel 15109 / frame 0396. I could not retrieve the reel/frame numbers, execution dates, or the correspondent of record for the other four links, and I will not invent them. Where a field is blank below it is because the source did not expose it, not because the field is empty in the USPTO record. Verify the gaps directly at the Assignment Center (below).

Presented in logical chain order (which is not the recording order — see note under 2004 entries).

---

• 2000‑10‑26 (recorded 2000‑10‑26) — Reel [not retrieved]

  • Conveyance: Assignment (assignment of assignors' interest)
  • Assignor: William E. Strum (inventor)
  • Assignee: SPEAKOUT.COM
  • Correspondent: [not retrieved]
  • Context: Initial employment assignment — inventor's rights transferred to the operating company on the filing date.

• 2004 (recorded 2004‑11‑22) — Reel [not retrieved]

  • Conveyance: Change of Name
  • Assignor: SPEAKOUT.COM, INC.
  • Assignee: NTERCEPT COMMUNICATIONS, INC.
  • Correspondent: [not retrieved]
  • Context: Change of name only — no change in beneficial ownership; the operating company rebrands post‑dot‑com.

• 2004 (recorded 2004‑09‑10) — Reel [not retrieved]

  • Conveyance: Assignment
  • Assignor: NTERCEPT COMMUNICATION, INC.
  • Assignee: AGENCY.COM LTD.
  • Correspondent: [not retrieved]
  • Context: Asset sale / distressed divestiture — the renamed speakout.com entity sells the patent (and, per the 2007‑issued '605 record, the application) to the interactive‑agency operating company Agency.com Ltd.

• Executed 2004‑07‑09 / recorded 2004‑09‑10 — Reel 15109 / Frame 0396 (confirmed)

  • Conveyance: Assignment ("Assignment Of Assignors Interest")
  • Assignor: Agency.com Ltd.
  • Assignee: Austen Services LLC — Suite #271, Los Altos, CA 94022
  • Correspondent: [not retrieved] — the record exists and exposes a correspondent field, but my retrieved snippet did not capture the name. This is the single most valuable field to pull manually (see Recommendations). I found no recurrence evidence for any correspondent on this chain, because I obtained none.
  • Context: Transfer out of the operating company into a serviced LLC at a mail‑drop address — the first shell‑entity‑type link. Source: https://www.plainsite.org/patents/assignment.html?id=[2846687](/patent/2846687) (USPTO Patent Assignment 15109/396).

• 2015‑12‑11 (recorded 2015‑12‑11) — Reel [not retrieved]

  • Conveyance: Merger
  • Assignor: AUSTEN SERVICES LLC
  • Assignee: ZARBAÑA DIGITAL FUND LLC
  • Correspondent: [not retrieved]
  • Context: Merger into a monetization fund — the terminal holder of record is a "Digital Fund," the current assignee per Google Patents legal events. (Recorded merger date per the Google Patents reassignment log.)

Recording‑order oddity (flagged, not corrected): the two 2004 asset transfers were recorded 2004‑09‑10, i.e., before the 2004‑11‑22 recording of the Speakout.com → Ntercept change of name. Recording dates are not execution dates; the underlying name change logically precedes the transfers. Anyone reading only the Google Patents event log (which sorts by recording date) will get the chain backwards. This is a genuine trap in this record.

If the Assignment Center returns no records: it does return records — five links are visible across Google Patents and Plainsite. The problem is field‑level completeness, not absence.

---

Timeline diagram

timeline
    title Ownership of US 7254605
    2000 : Filed by William E Strum
         : Assigned to Speakout.com
    2004 : Speakout.com renamed Ntercept Communications
         : Ntercept assigned to Agency.com Ltd
         : Agency.com assigned to Austen Services LLC
    2007 : Patent issued as US 7254605
    2015 : Austen merged into Zarbana Digital Fund LLC
    2023 : Patent term expired

(Diagram uses the unaccented "Zarbana" for parser safety; the recorded name is ZARBAÑA DIGITAL FUND LLC.)

---

NPE / troll-pattern signals

---

Verdict

NPE — moderate confidence.

The chain shows a documented transfer from an operating company into a service LLC at a mail‑drop address — Agency.com Ltd. → Austen Services LLC, Suite #271, Los Altos, CA 94022, executed 2004‑07‑09 and recorded 2004‑09‑10 at reel 15109/0396 — followed by a 2015‑12‑11 merger into Zarbaña Digital Fund LLC, a "fund" that appears as a patent‑holding defendant in unrelated 2026 declaratory‑judgment actions. That is one strong signal (shell‑entity transfer) plus weak corroboration (2004 transfer cascade; fund‑type terminus), which lands at moderate rather than high confidence. Confidence is capped because the correspondent of record was not retrieved for any link (so the classic "same lawyer runs every LLC" test is untested), no product‑or‑no‑product evidence for Austen Services LLC was independently verified, and no assertion of this particular patent exists on the record — indeed it expired 2023‑09‑15.

Verify at:

• USPTO Assignment Center — search by patent 7254605 or application 09695932: https://assignmentcenter.uspto.gov/ (and the legacy interface https://assignment.uspto.gov/patent/index.html)
• The one confirmed record: https://www.plainsite.org/patents/assignment.html?id=2846687 (Assignment 15109/396)

Two‑step follow‑up to close the open items: (a) pull the correspondent of record for 15109/0396 and the Strum→Speakout, Ntercept→Agency.com, and Austen→Zarbaña records, then run each attorney/firm name against Unified Patents and RPX asserter directories to test signal 3; (b) pull the execution dates for those four records to test whether the 2004 cluster was a single orchestrated transaction.

---

Flagged contradictions vs. prior sections

1. "Original assignee": the prior Patent summary listed Austen Services LLC as original assignee (following Google Patents). The USPTO assignment record shows the first assignee was SPEAKOUT.COM, with the Strum→Speakout assignment recorded 2000‑10‑26. The Google field is a dataset artifact; the recorded first assignment controls.
2. Chain ordering: the Google Patents legal‑events list is sorted by recording date and therefore shows the 2004 transfers before the Speakout→Ntercept change of name, implying a backwards chain. Logical order is name change → Ntercept→Agency.com → Agency.com→Austen.
3. Reel/frame availability: the prior summary characterized reel/frame data as unavailable; I confirmed one (15109/0396). The other four remain unretrieved — consistent with the prior caveat, now partially narrowed.

I'll research the prior-art citations for US 7,254,605, starting with the patent's own front-page "References Cited" (USPTO/Google Patents citation data), then verify each reference independently.

I have citation-of-this-patent data. Now I need the patent's own front-page "References Cited." Let me pull the Justia and FreePatentsOnline records, which reproduce the cited references.

The Justia record reproduces the front-page citations. Let me pull the fuller list (foreign documents + non-patent literature) and attempt to retrieve the granted claims.

Prior-Art Analysis — US 7,254,605 B1

Analyst note on scope and honesty constraints (read before using the §102 section):

Two things materially limit what I can responsibly state, and I am flagging them up front rather than papering over them:

1. I could not retrieve the granted claim set. The patent text supplied to me truncates mid-sentence in Section III of the Detailed Description, and my searches did not surface a verbatim reproduction of the claims of US 7,254,605. My searches on claim language ("What is claimed", "sampling quotient"/"transmission interval" in claim context) hit a tool step-limit and returned nothing usable. I therefore cannot produce a literal claim-by-claim §102 mapping (e.g., "US 6,xxx,xxx anticipates claim 3"). Any such sentence-level mapping would be fabricated. What I can do is map references to claim groups reconstructed from the specification, clearly labelled as such.
2. I did not query a USPTO API directly. I retrieved the front-page citation data via the Google Patents record for US7254605B1 and the Justia mirror (https://patents.justia.com/patent/7254605), both of which reproduce USPTO front-page data. I did not open the printed patent PDF or PatentCenter image file.

A third caveat specific to the task: "cited on the front page" ≠ "anticipatory." Front-page references may have been cited under §102, §103, or merely as background art. Nothing in the record I retrieved identifies which reference the examiner applied against which claim, and I found no office-action/PTAB record stating an anticipation rejection.

---

1. Confirmation of the patent record

This matches the previously generated bibliographic section; no contradiction found. Note that the term "frequency" in the title refers to interchange frequency (rate/schedule), not RF carrier frequency — this matters because the string "7254605" also collides with unrelated RF/amplifier patents, which are not this patent.

---

2. References cited on the face of US 7,254,605 (U.S. patent documents)

Retrieved from the Justia/Google Patents reproduction of the front page. All are U.S. patent documents; dates are publication (grant) dates.

Anomaly flagged, not auto-corrected: item 25 (US 6,403,610, "Malleron et al.") sits in the middle of a sequence of network/computing references and the name Malleron is associated in my knowledge with pharmaceutical chemistry, not networking. I could not verify its title in this session. It may be a transcription error in the citation list, or it may simply be an unrelated reference the examiner cited for a narrow point. It must be checked against the printed patent before being relied on. Do not silently renumber or substitute it.

Non-patent literature cited: one item was visible:

"Digital Equipment Corporation Provides First Interactive Participation in Presidential Debates," PR Newswire, p. 1003NEM024, Oct. 3, 1996, plus excerpts from Digital Equipment Corporation's InPulse Web site.

This is the only NPL item I could confirm; the list may contain others (e.g., additional press/web excerpts) that did not render. Foreign patent documents: I did not retrieve any for the '605 front page (unlike the forward-citing IBM patents, whose lists do include EP/JP items). Absence in my retrieval is not proof of absence — verify on the printed face.

Not prior art (for completeness): the two continuations claiming priority to this application — US 7,809,788 B2 and US 7,806,273 B2 — and the co-pending sibling applications incorporated by reference ('624 application Ser. No. 09/544,624; Ser. No. 09/695,929 filed the same day). Same-inventor/same-day material is not "by another" under §102(a)/(e) and cannot anticipate, though it bears on obviousness-type double patenting and on what the specification discloses.

---

3. Relevance assessment mapped to claim groups

Because I lack the claim text, I map to claim groups reconstructed from the specification (see the earlier summary, §3). Treat each row as a hypothesis to test against the actual claims, not as an established anticipation.

Direct answer to "which claim(s) does each reference potentially anticipate": on the record available to me, I cannot name a specific reference as anticipating a specific numbered claim. What I can say with confidence is the distribution above: the audience-response/opinion-research references bear on claim group A; the network-scheduling references bear on claim groups B and E; and the sampling-quotient/class-scheduling subject matter (group C) has no obvious anticipation candidate in the cited set. Any claim-level §102 statement requires the granted claims in hand.

---

4. What must be verified before this analysis is relied upon

1. Pull the granted claims from USPTO PatentCenter / Patent Public Search for application 09/695,932 (or the printed patent's claim columns). This is non-negotiable — the §102 mapping is otherwise unfalsifiable. (Note the front page lists 39 lines of "methods" text and 196 mentions of "claims/description"-neighbouring terms in Google's term-frequency block; that is a term-frequency artifact, not a claim count.)
2. Pull the printed "References Cited" block to (a) recover the full NPL list and any foreign documents, and (b) resolve the US 6,403,610 "Malleron et al." anomaly.
3. Pull the file wrapper (Patent Public Search → "Global Dossier"/"Documents" for 09/695,932) to see which references were actually applied, under §102 or §103, and against which claims. This converts my hypotheses into facts.
4. Forward citations are not prior art. Patents citing US 7,254,605 (e.g., US 8,812,716; US 10,601,884; US 11,483,610; US 7,779,140-family, all listed as citing "7254605 | August 7, 2007 | Strum") post-date it and cannot anticipate it. Do not conflate the "Cited By" list with the "References Cited" list.

---

5. Explicit uncertainties

• Claim text unavailable to me → no claim-by-claim §102 table is possible; §3 is a claim-group hypothesis.
• Titles/disclosures of most cited references unverified in this session — I deliberately did not supply one-line descriptions for references whose subject matter I could not confirm (e.g., Shibao, Carpenter, Homma, Hogan, Caccavale, Worthy, Wronski, Shaw, Solimene, Havens, Callaghan, Shatto, Link, Nielsen, Park, Harter, Young, Linsk, Berstis, Takeda, Quach, Jorgensen). Supplying descriptions from memory would risk fabrication; these should be read off the printed abstracts.
• Only one reference I will characterize with high confidence: US 5,848,396 (Gerace, Dec 8, 1998) — behavioral profiling of computer users for targeted delivery. It is relevant to the audience-measurement/demographic side of the system (the spec's user-demographic database 130) but is not a strong candidate for the transmission-frequency claims.
• Front-page citation ≠ anticipation, and no results ≠ no rejection: I found no office action or PTAB record applying a cited reference under §102 to a specific claim.
• Source limitation: Google Patents and Justia are secondary sources mirroring USPTO front-page data; I did not query USPTO's API directly. The task's instruction to "search the USPTO database" is satisfied only at second hand and should be repeated against PatentCenter for the printed face.

I'll first flag a structural issue: no "Prior Art" section was included in the material you gave me (I have the Patent summary and Litigation summary only). Let me attempt to retrieve the actual prior-art citations for this patent.

I now have the actual granted claims (16 claims) and the start of the cited-references list. Let me try to pull the full prior-art citation list.

Obviousness Analysis — US 7,254,605 under 35 U.S.C. § 103

---

0. Threshold corrections and input gaps (read first)

0.1 The "Prior Art section of this page" was not supplied to me. The material I was given contains only a Patent summary and a Litigation summary. There is no prior-art section, no reference list, and no claim chart. Per your instruction I attempted to retrieve it from the live record. Results of that attempt, and what I could and could not obtain, are in §3. If you paste the actual prior-art section, I will convert §4–§6 from a reference-class analysis into a reference-by-reference chart.

0.2 Correction to the previously generated Patent summary — the claims are now available. That section stated: "I do not have the granted claims in hand… Treat the following as a functional reconstruction." I have now located a third-party mirror of the granted claim set (16 claims) at Justia (https://patents.justia.com/patent/[7254605](/patent/7254605)). The reconstruction in that section should be superseded by §1 below. Caveat: this is a third-party mirror, not the USPTO printed patent; verify against the face of the patent before relying on it in a filing.

0.3 Date discrepancy. The task header gives the current date as April 26, 2026; the previously generated sections were produced against a system date of 2026‑09‑30. This affects nothing in the §103 analysis (which is keyed to the 2000 priority date), but it does affect the statement of the patent's legal status.

0.4 Governing law: pre-AIA § 103(a). Application 09/695,932 was filed October 26, 2000, before the March 16, 2013 AIA date. So this is pre-AIA 35 U.S.C. § 103(a), and pre-AIA §§ 102(a)/(b)/(e)/(g) supply the prior-art categories. Post-KSR case law (which applies retroactively) governs the legal standard.

0.5 Practical posture. The patent is recorded Expired – Lifetime, adjusted expiration 2023‑09‑15. An obviousness attack today is therefore relevant to (a) a declaratory-judgment or invalidity defense directed at past damages, (b) IPR (available even for expired patents, though without claim amendment), and (c) freedom-to-operate/landscape work. It is not a live injunction question.

---

1. The claims as issued (basis for the analysis)

Claim 1 is the only independent claim; claims 2–16 depend from it (13–16 via claim 3).

1. A method of managing communications between a server and one or more clients over a computer network, comprising the steps of:
– selecting a transmission period to define a time frame for sampling one or more sampling classes created from the clients, wherein the transmission period is divided into a plurality of transmission intervals, wherein said sampling occurs at the last transmission interval within the transmission period;
– dividing the clients by the transmission period to derive a sampling rate, wherein the sampling rate designates a sampling size of each sampling class; and
– interchanging communications based on a transmission schedule, wherein the transmission schedule defines the transmission period for interchanging communications between the server and each sampling class.

The limitations that carry any §103 weight are only three: (i) period-divided-into-intervals with sampling at the last interval; (ii) dividing client count by the period to get a class size; (iii) an interchange schedule keyed to the period. Everything else (opinion data, composite results, Java applet distribution, clock sync, aggregation/compression/encryption, "over the global Internet") appears in the dependent claims and is directed at conventional, admitted-known subject matter — which matters enormously to the analysis.

---

2. Level of ordinary skill (POSA)

Not stated in the patent. On this record the POSA is reasonably defined as a person with a bachelor's degree in computer science/electrical engineering and 2–4 years of experience in networked client–server or real-time multicast/polling applications, familiar with TCP/IP and UDP (RFC 791 is cited in the patent itself), bandwidth/QoS limitations of the public Internet, and time-slotted ("scheduled access") transmission protocols. That is a relatively high-skill, low-invention-pressure profile that favors the obviousness outcome.

---

3. Prior art actually on the record

3.1 The strongest anchor: admissions inside the '605 specification itself

The Background sets out, as known art:

• IP/RFC 791 and its inability to guarantee QoS; bandwidth availability varying by time of day and locality; packet overhead inflating bandwidth needs.
• "One possible solution to reducing packet overhead is to buffer and aggregate the frames into a single data packet (e.g., IP datagram or protocol data unit (PDU))" — i.e., aggregation/buffering is admitted known, and its latency trade-off is admitted known.
• UDP environments losing/delaying packets, and retransmission adding delay.
• Real-time opinion gathering via dial polls and surveys and the '624 system — i.e., networked real-time polling is admitted known.

Under In re Constant / Riverwood Int'l v. R.A. Jones (Fed. Cir. 2003), statements in the specification about what is old may be treated as admitted prior art. These admissions alone cover: aggregation (claims 13–16), the inter-network environment (claim 12), and the existence of real-time polling networks (claim 3's context).

3.2 References of record — what I could retrieve

Google Patents / Justia list the patent's cited references; Patsnap reports "34 Cites, 165 Cited by." I captured only the first entries before hitting a retrieval limit:

I could not retrieve the remaining ~32 cited references, or the non-patent literature list. The obviousness case below therefore rests on reference classes plus the specification's own admissions, not on a verified reference-by-reference mapping.

3.3 Candidate references to pull (found in a third party's IDS, not the '605 record)

Searching surfaced a later wireless audience-response patent (US 8,223,709, Fleetwood Group) whose cited art includes pre-October-2000 items in the same conceptual space — time-slotted collection of responses from many remote terminals: US 5,379,213 (Derks, Jan. 1995); RE 5,449 (Derks, Feb. 1997); WO 96/18982 A1 (June 1996); US 5,724,357. I have not verified their disclosures; treat these as leads to pull, not as established art.

3.4 What is not prior art (a trap to avoid)

• The '624 application (Ser. 09/544,624) — same inventor (Strum), and expressly incorporated by reference. Pre-AIA §102(e) requires art "by another," and pre-AIA §103(c) would disqualify commonly owned §102(e)/(f)/(g) art from §103 anyway. It is not prior art against '605; it is part of the '605 disclosure. A naive chart that uses it is defective.
• Ser. 09/695,929 (filed the same day, "Controlling Data Loss…") — same effective date, no §102(e) benefit (In re Hilmer).
• US 7,809,788 B2, US 7,806,273 B2, and US 2007/0260680 / US 2007/0266086 — later-filed continuations/publications claiming priority to '605. Not prior art. Note that both siblings replicate the same "sampling occurs at the last transmission interval" limitation in their claim 1, which is itself a useful signal about what the examiner treated as the point of novelty.

---

4. Claim 1, limitation by limitation, against the relevant art classes

Net: claim 1 recites a time-slotted, round-robin polling schedule applied to an Internet client–server polling system. Nothing in it requires opinion data, aggregation, compression, encryption, or even that the clients be polling respondents.

---

5. Combinations and the motivation to combine

Combination A — Admitted art + conventional slotted polling (attacks claims 1, 5, 7, 8, 11)

Combination: The '605 Background (IP/UDP QoS limits, bandwidth contention, aggregation of frames into a datagram) + a conventional time-slotted polling protocol that assigns staggered transmission slots to grouped remote terminals.

Motivation (KSR prong 1 — known technique to improve a similar device in the same way; prong 4 — predictable result): Spreading transmissions across slots is the textbook remedy for exactly the problem the patent admits — bandwidth contention and server overload from simultaneous traffic ("thundering herd"). The patent's own stated advantage ("increase the number of users that a server can manage"; give servers "more time to process and respond") is the predictable result of staggering, not an unexpected one.

Combination B — A + networked real-time polling system (attacks claim 3)

Combination: Combination A + an Internet-based real-time polling/opinion system of the type the patent admits exists (dial-poll/survey systems; the '624-type architecture).

Motivation: The problem to be solved is stated in the patent as managing large online polling panels. Where the object is real-time opinion gathering, applying a known stagger schedule to the polling traffic is the obvious design choice, and the only claimed difference is the content of the interchanged data ("opinion data streams," "composite results"). Data content does not make an otherwise old scheduling method patentable.

Combination C — B + downloadable client-side schedule (attacks claim 2)

Java applet delivery of client-side instructions is admitted in the specification as conventional ("the instructions are included within a Java applet that is routed to the clients," with JavaScript/C/C++ as alternatives). Combining is a simple substitution of one known distribution mechanism for another (KSR prong 2). The patent cites no unexpected result from applet-based delivery.

Combination D — C + absolute-time clock synchronization (attacks claim 4)

Remote clock setting to an absolute standard (Greenwich Mean Time) is admitted in the specification, and NTP/GMT time-code distribution was long-standing network practice. Motivation: a staggered schedule is only useful if the parties agree on the clock — which the patent itself states as the reason for synchronization.

Combination E — D + round-robin class assignment and interval-as-a-function-of-client-count (attacks claims 9, 10, 12)

Sizing a polling interval as a function of the number of active nodes, and assigning each node to a group, is the essence of load-partitioning — and the patent itself names Microsoft's load-balancing architecture as usable in this system. Claim 12 ("over the global Internet") is a pure field-of-use/intended-use recitation, which is generally given no patentable weight where the method is otherwise old.

Combination F — E + aggregation/compression/encryption (attacks claims 13–16)

Aggregation into a single datagram is expressly admitted in the Background. Compression and encryption using then-standard techniques are not merely conventional; the specification enumerates DES, ANSI DEA, Diffie–Hellman, El Gamal, RSA, and DSA as known. Motivation: overhead reduction and confidentiality in a public-network opinion system — the patent's own stated rationale.

Claim 8 (intervals in seconds / milliseconds / nanoseconds / minutes)

A mere change of scale, with no new result. In re Miller, 164 F.2d 786 (CCPA 1947). Give this claim essentially no weight.

The KSR rationales, restated

---

6. Where the § 103 case is genuinely weak

I will not overstate this. Three honest vulnerabilities:

1. The "last transmission interval" limitation is the whole ballgame. It is unusual, it appears in the granted claim and in both sibling patents' claim 1, and its placement reads as the prosecution-time differentiator. If the art polls clients at the start of the period, continuously, or event-driven (as the patent's own "event-driven" alternative mode does), the limitation is not met. Expect this to be the sole battleground. Counter: it is a two-option design choice (first vs. last slot) with a predictable result, squarely within KSR's "obvious to try."

2. Evidence gap. I lack the ~32 unrecovered cited references, the non-patent literature, and any expert declaration. A §103 holding needs an articulated, reference-grounded reason — not merely "these steps are generic." A petitioner/defendant must supply the mapping.

3. Secondary considerations are unaddressed. My searches surfaced no litigation (see the previously generated Litigation summary) and therefore no nexus-bearing evidence either way. Absence of litigation is not a secondary consideration; but absence of any proof of commercial success, copying, licensing, or industry praise also means the patent owner has little to point to. Net neutral, leaning toward the challenger.

---

7. A companion theory worth flagging (§ 112, not § 103)

The claim requires sampling "at the last transmission interval within the transmission period," but the specification's only worked examples (FIG. 7 / FIG. 8, discussed in the Detailed Description) start class 1 at interval 4, class 2 at interval 5, and class 3 at interval 6, against a four-interval period — i.e., only class 1 samples at what is plausibly the last interval of an aligned period; classes 2 and 3 sample one and two intervals later. On its face there is a written-description/enablement question (pre-AIA §112 ¶1) and possibly a new-matter question, because the figure's staggered offsets do not cleanly map onto "the last transmission interval within the transmission period" for every class. This should be run as a parallel invalidity theory, and it also weakens the patent owner's ability to read the limitation broadly to avoid the prior art.

---

8. Bottom line

• Claim 1 is, functionally, a time-slotted round-robin polling schedule applied to an Internet client–server polling system. Every hardware/software element is a known element performing its known function.
• Claims 2, 3, 5, 6, 7, 8, 11, 12 are the weakest: Java applet distribution, polling and report timing, staggered first/subsequent class polling, units of measure, and "over the Internet" are all either expressly admitted in the specification or routine.
• Claims 9, 10, 13, 14, 15, 16 are also vulnerable, since aggregation is expressly admitted and compression/encryption algorithms are enumerated as known.
• Claim 4 falls to conventional absolute-time synchronization, also admitted in the specification.
• The single contested limitation is "sampling occurs at the last transmission interval within the transmission period" — and it is a two-option design choice, which KSR resolves against the patent owner absent unexpected results.
• Conclusion (preliminary, reference-class level): a prima facie case of obviousness under pre-AIA § 103(a) is available against claims 1–16, anchored on the patent's own admissions plus conventional slotted/staggered polling art — but it is not yet a completed case, because the prior-art section you asked me to use was not in my inputs and I could not reconstruct the full 34-reference list.

To convert this into a defensible chart, supply: (a) the cited-references list from the patent's front page (all ~34 entries) and the "Cited By" list; (b) the non-patent literature; (c) if available, the prosecution history / any IDS and examiner's reasons for allowance; and (d) the full text of two or three candidate primary references (a slotted-polling reference and an end-of-window batch-reporting reference). With (a)–(d) I can produce a limitation-by-limitation §103 chart with pinpoint citations and express KSR motivations for each combination.