US 9369779

I'll verify the patent's current status and check for any 2026 litigation activity before summarizing.

Let me get the actual claim set for this patent and check IPR/CAFC status specific to 9369779.

Let me pull the literal claim language for the '779 patent.

US Patent 9,369,779 B2 — Analyst Summary

Important scoping note first: The authoritative full text you supplied (from Google Patents) truncates inside the "Description" section — it ends mid-word in the Profile-provider definition and never reaches the Claims. My searches returned claim text for sibling patents in the same family (US 7,861,260; US 9,813,778; US 10,178,442) and for unrelated patents. I therefore could not verify the verbatim claim language of 9,369,779 in this session, and I flag that explicitly in the claims section rather than passing off sibling claims as the '779's.

1. Bibliographic data

Family/continuity chain (as printed on the patent under "CROSS-REFERENCE TO RELATED APPLICATIONS" and corroborated by the '778 patent face):

• 11/736,544, filed Apr. 17, 2007 → US 7,861,260
• → continuance 12/860,666, filed Aug. 20, 2010 → US 8,281,336
• → division 13/097,885, filed Apr. 29, 2011 → US 8,695,032
• → division 14/246,525, filed Apr. 7, 2014 → US 9,369,779 (this patent)
• → continuations from it include US 9,813,778 (app. 15/179,356), and the later-assigned US 10,178,442, US 10,715,878, US 10,715,878-lineage patents (e.g., US 11,303,973; US 11,564,015; US 11,589,136; US 11,805,300; US 11,974,025; US 12,212,818).

Discrepancy flag (no auto-correction applied): a Unified Patents portal record for the same number shows priority 2007-04-16, application date 2014-04-06, grant 2016-06-13, expiration 2027-04-16 — a consistent one-day shift, most likely time-zone/localization artifact. The docket-copy OCR of the '778 patent also renders the parent as "8,283,336" where the authoritative Google Patents text says 8,281,336; I treat the authoritative text as controlling but note it.

2. Abstract (verbatim)

"In a method for delivering targeted television advertisements based on online behavior, IP addresses indicating online access devices and IP addresses indicating television set-top boxes are electronically associated for a multitude of users. Using user profile information derived from online activity from one of the online access IP addresses, a television advertisement is selected, such as by using behavioral targeting or demographic information, and automatically directed to the set-top box indicated by the set-top IP address associated with that online access IP address. Preferably neither the user profile information nor the electronic association of online access and set-top box IP addresses includes personally identifiable information."

3. Independent claims — plain-language overview

Confidence caveat: the claim set is not in the text I was given, and I did not obtain a verbatim, number-matched copy of 9,369,779's claims. The following is a functional reconstruction from (i) the '779 abstract/specification, (ii) the allowed-claims examiner statement in the same family's file history (Examiner Ekpo, July 27, 2010, quoted in PTAB papers, describing independent claims directed to "electronically associating online access internet protocol (IP) addresses and received set-top box IP addresses for a number of users… use of user profile information derived from online activity from the online access IP addresses… to cause a TV advertisement to be directed to a set-top box indicated by the set-top box IP address associated with the online access IP address"), and (iii) parallel claims of sibling continuations (e.g., US 10,178,442's prosecution-history claim 1). Treat the claim-by-claim labels below as characterization, not verbatim text.

At a functional level, the '779 is a divisional keeping the core method/system subject matter of the 2007 priority application. The independent claims recited in this family generally fall into two groups:

• Independent method claim (the core "targeted TV ad" method). In plain terms, a computer-implemented method with steps of:
  1. Associating addresses in bulk — electronically associating, for each of many users, that user's online access IP address with that user's set-top box IP address. In the important sub-species, the association is based on a common IP address to which network traffic is routed for both the online user interface device (e.g., a computer) and the STB (e.g., a modem/router shared by both). Association by pseudonym from a single combined ISP/TVP is the other disclosed mode.
  2. Deriving profile information from online activity — obtaining user profile information derived from what a user did online from one of those online-access IP addresses (searches, sites/pages viewed, ads clicked, purchases, etc., and/or demographic attributes).
  3. Causing a TV ad to be directed to the matched STB — using that profile information to automatically cause a television advertisement to be directed/transmitted to the set-top box indicated by the STB IP address associated with that online-access IP address (directly, via an intermediate ad server, or by instructing the STB to fetch it), preferably without PII in either the profile or the association.
• Independent "profile-data-delivery" / system-side claim family. Sibling applications in this chain also claim the complementary side of the transaction — computer equipment automatically electronically delivering user profile information (including the online access IP address and the time/date of the activity) to a computer system that performs the association and direction — plus corresponding system (apparatus) claims covering the same combination.

Dependent claims in this family typically add: no-PII limitations; selections based on search requests, or demographic/behavioral attributes, or the television content being viewed; association via pseudonym or via dynamically assigned common modem/router IP; STB identifier verification; time-and-date matching of IP address lists; periodic updating; user-controlled ad filtering ("opt in"/"opt out"); and revenue-sharing limitations.

If you need the literal claim text (e.g., for an infringement or validity chart), I recommend pulling the patent's own XML/full-text claim section from USPTO Patent Center or the Google Patents "Claims" tab — I can do that if you re-run me with a fetch of that section, since the copy provided here stops short of it.

4. Post-grant challenge and litigation status — what I could and could not confirm

Confirmed as not about the '779: the two PTAB proceedings that cite the '779 as a parent patent — IPR2022-01315 (Roku, Inc. v. Intent IQ, LLC) and IPR2022-01505 (Samsung Electronics Co., Ltd. v. Intent IQ, LLC) — challenged US 10,715,878 ("the '878 Patent"), not the '779. Both were denied institution on the merits (IPR2022-01315 Paper 9; IPR2022-01505, Decision filed Mar. 22, 2023, denying institution and denying joinder), on the ground that the petitioner failed to make a sufficient factual showing as to the recited "common local area network."

2026 dockets involving Intent IQ (patent-in-suit not verified as the '779):

CAFC 2026: I found no Federal Circuit appeal specifically identifying US 9,369,779 in 2026. The Federal Circuit activity I surfaced involving Intent IQ patents concerns PTAB final written decisions in other proceedings (e.g., appeals noted for IPR2024-00666/Roku, IPR2024-00701/Samsung, IPR2024-00421/Freewheel, IPR2025-00128 and -00129/Viant) — I could not verify which patent numbers those target, so do not treat them as '779 appeals. There is also a 2026 Federal Circuit amicus brief (Google v. Cellular South, No. 26-1519) challenging the USPTO's "settled expectations" IPR-age bar; that is relevant context for this family (the '779 is >6 years post-grant) but is not a case about the '779.

Bottom line: US 9,369,779 appears active and unexpired (Apr. 17, 2027), owned by Intent IQ, LLC, with no IPR instituted against this number that I could confirm, and no confirmed 2026 Federal Circuit appeal naming it. I could not confirm that the '779 is one of the patents asserted in the 2025–2026 Intent IQ district-court complaints; those complaints list multiple exhibits/patents and the specific numbers were not visible in the results I retrieved.

Uncertainty register

1. Verbatim claim set of 9,369,779 — not verified; overview above is reconstruction.
2. Identity of patents asserted in the 2025–2026 Intent IQ cases — not verified.
3. One-day bibliographic discrepancies (Apr. 6 vs. Apr. 7, 2014; Apr. 16 vs. Apr. 17, 2007; Jun. 13 vs. Jun. 14, 2016) between sources — flagged, not corrected.
4. Parent patent number in an OCR'd court exhibit as 8,283,336 vs. 8,281,336 in the authoritative text — flagged, not corrected.

Sources: Google Patents US9369779B2 (authoritative text provided); Google Patents US9813778B2 PDF face; docketalarm PTAB exhibits in IPR2022-01315 and IPR2022-01505; PTAB Decision Denying Institution, IPR2022-01505 (Mar. 22, 2023); ptacts.uspto.gov petition documents referencing 14/246,525 (now US 9,369,779); Unified Patents portal US-9369779-B2; UniCourt D. Del. 1:25-cv-01084; PacerMonitor E.D. Tex. 2:26-cv-00609; Law360 (May 22, 2026) Snap venue transfer; IP Verse Intent IQ PTAB case list (updated May 24, 2026).

I'll research litigation involving US patent 9369779. Let me search multiple sources.

Let me research the specific cases and which patents they assert.

Let me search specifically for whether the '779 patent was ever asserted in litigation.

Summary: Litigation involving US 9,369,779

Bottom line: Based on my searches, I did not find any litigation in which US Patent No. 9,369,779 ("Targeted television advertisements based on online behavior," Roy Shkedi, current assignee Intent IQ LLC) was specifically identified as an asserted patent. I should be explicit about that conclusion and its limits, per your instruction not to fabricate and not to substitute similar numbers.

What the patent record itself shows

• US 9,369,779 B2 — inventor Roy Shkedi; current assignee Intent IQ LLC; priority date (per Google Patents page) 2007-04-17; filed 2014-04-07; granted 2016-06-14. It is a division in a large family from application 11/736,544 (now US 7,861,260). Google Patents
• The Google Patents page displays a "Family has litigation" marker with a Darts-IP link (family=39872105). That flag denotes that members of the worldwide family have been litigated; it does not, by itself, establish that the '779 patent specifically was asserted in any case. Google Patents

The Intent IQ / AlmondNet enforcement campaigns (context — different patents)

The same assignee family (AlmondNet, Inc. and Intent IQ, LLC) has run a significant multi-front patent campaign. The patents I saw actually asserted in these cases were consistently drawn from the other family members — e.g., 7,822,639; 7,861,260; 7,979,307; 8,244,582; 8,244,586; 8,566,164; 8,595,069; 8,671,139; 8,677,398; 8,959,146; 9,508,089; 9,830,615; 10,026,100; 10,321,198; 10,628,857; 10,715,878; 11,564,015; 11,949,962; 12,457,388 — not 9,369,779. Examples of the enforcement matters I confirmed:

Sources: Justia Dockets, UniCourt, Docket Alarm, PACER Monitor, PTAB/P-TACTS filings.

Notably, the Samba TV AO 120 form (the formal notice to the USPTO of patents in suit) lists only US 8,677,398; US 10,321,198; and US 11,949,962 — again, not '779.

Important caveats

1. I could not complete an exhaustive verification. My research process hit its step/query limit before I could check every docket, every P-TACTS/PTAB filing, and the Darts-IP family record directly. My conclusion is therefore "not found," not "affirmatively confirmed never asserted."
2. Do not assume the "Family has litigation" flag refers to the '779 patent. Family-level litigation flags aggregate all 53 family members (including the heavily litigated '582, '586, '398, '878, etc.).
3. This is a procedural fact search, not legal advice. Per your strict rule, I did not auto-correct any patent number; all identifiers above are reproduced exactly as they appeared in the sources.

Recommended next steps to confirm

• Search Darts-IP for family ID 39872105 to see which specific patents were asserted.
• Search PACER / CourtListener RECAP for the exact string 9,369,779 across all district courts and the CAFC.
• Check the USPTO Patent Trial and Appeal Board ("P-TACTS") for any IPR/PGR/CBM naming 9,369,779 as the challenged patent.
• Check the Unified Patents litigation portal entry for US-9,369,779-B2.

If you'd like, I can attempt these targeted searches individually in a follow-up.

I'll start by searching for any PTAB proceedings on this patent, since the ODP block says none are indexed.

Let me dig specifically for proceedings naming the '779 patent.

Let me search for any PTAB docket specifically naming patent 9,369,779.

PTAB proceedings on US 9,369,779 — analyst report

Proceedings overview

Total AIA trial proceedings on US 9,369,779: 0. The USPTO Open Data Portal returns no IPR, PGR, or CBM proceeding naming this patent, and my independent web searches (PTAB docket aggregators, PTAB petition exhibits, litigation filings, Federal Circuit materials) surfaced no petition, institution decision, final written decision, or appeal naming 9,369,779 — so the breakdown is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denied, and every claim of the '779 patent remains untested at the PTAB. For a defendant, the defensive posture is therefore not "the patent is hardened by surviving IPRs" — it is "the patent has never been through an AIA trial at all, so there is no free roadmap and no estoppel, but also no adjudicated narrowing to lean on."

No proceedings to report — with the methodological caveats stated plainly

I want to be explicit about what I did and did not verify, because in this posture the absence is the finding:

• Canonical source (USPTO ODP): no AIA trial proceedings on file. I treat this as the controlling list, per instructions.
• Web corroboration: Searches for "9369779" / 9,369,779 in combination with IPR, PTAB, Intent IQ, and AlmondNet returned only (a) the patent's own Google Patents / Unified Patents portal pages, (b) prosecution artifacts in sibling applications that cite the '779 patent as a double-patenting reference (see below), and (c) IPRs against other patents in the same family — never against '779 itself.
• Known ingest-lag risk: I did observe that third-party dockets can lead the ODP (e.g., Viant/AppLovin/Mediaocean filed IPR2026-00330 on 2026-04-02 against a different Shkedi patent, US 11,949,962 — not '779). If a petition against '779 was filed very recently, it may not yet be indexed. Verify on PTAB E2E (https://ptacts.uspto.gov/ptacts/) and USPTO Patent Center before relying on the zero-count for a litigation-position statement.

Because there is no proceeding, the per-proceeding template (type / filed / panel / grounds / institution / FWD / settlement / appeal / defensive value) has nothing to populate for '779. I am not going to invent proceeding numbers, panels, or claim-level dispositions. What I can give you — and what actually matters here — is the surrounding context, which is unusually informative.

Family-context IPR activity (these are NOT proceedings on '779 — do not attribute them to it)

The AlmondNet / Intent IQ Shkedi family (priority 2007-04-17, same specification lineage as '779) has been IPR'd repeatedly. These are the proceedings I found; I've noted the target patent for each so the distinction from '779 is unambiguous:

Sources for the above: the Amazon IPR2023-00227 institution decision (https://bannerwitcoff.com/wp-content/uploads/2023/06/IPR2023-00227.pdf), PTAB petition exhibits hosted at ptacts.uspto.gov, and Docket Alarm PTAB dockets (https://www.docketalarm.com/search/?q=patent%3A%288677398%29&court_exact=Patent+Trial+and+Appeal+Board). Where I state a denial or institution, that is what the cited record says; I did not locate a final written decision on the merits for any of these family proceedings, so I am not reporting claim-level outcomes.

Two details worth flagging as strategic signals, not facts about '779:

1. '779 shows up in prosecution, not in PTAB. A sibling application's file history (Appl. No. 15/179,356, which issued as US 9,813,778) contains a provisional nonstatutory double-patenting rejection over "claims 1, 4, 7-9, 11-29 of Patent Application No. 9369779." That tells you the '779 patent has at least independent claims 1 and 11 with dependents through at least claim 29, and that its claims are close enough to the '778 claims to trigger a terminal-disclaimer problem — i.e., the family's claim sets overlap heavily. Treat the claim count as indicative from a prosecution excerpt, not as an authoritative claim listing.
2. '779's own pre-grant publication (US 2008/0259906 A1) is now being used as prior art against later Shkedi patents — Viant listed it as Exhibit 1005 in IPR2026-00330. The '779 family is thus simultaneously the asserted weapon and the prior art against the patent owner's later filings.

Strategic summary

Claim status: everything on '779 is UNTESTED; nothing is canceled, nothing is sustained. There is no final written decision, no certificate canceling claims, and no appeal involving '779. If you have been handed a demand letter or complaint citing the '779 patent, there is no PTAB record you can point to as invalidating any claim — and, symmetrically, there is no PTAB record the patent owner can point to as having survived a validity challenge. The absence of adjudication is a genuine blank slate on both sides. The practical inference is that '779 has not been a front-line assertion patent: the patents AlmondNet/Intent IQ actually asserted in the campaigns I could see (Yahoo/Oath, Roku, Samsung, Meta, Microsoft, Amazon, Snap) were the '582, '822, '398, '164, '198, '878, '307, '139, '146, '574, '486, '904, '249, '089 and '962 patents — '779 does not appear on those lists. I could not exhaustively confirm this for every recently filed case (including the Intent IQ/AlmondNet v. Snap matter transferred to California in 2026), so confirm the asserted-patent list in your specific case.

Estoppel landscape: § 315(e)(2) estoppel attaching to '779 is essentially nil. IPR estoppel is patent-specific — a petitioner's estoppel from IPR2022-00773, IPR2023-01281, IPR2023-00227, etc. runs to the '398, '878, '164, '198 patents it challenged, not to '779. A defendant facing '779 is free to raise any § 102/§ 103 ground it can support, including art that was asserted against sibling patents and even art that a sibling-petitioner raised. The corollary cuts the other way too: because no one has ever petitioned on '779, you also get no pre-made record of how the Board or the district court construes the '779 claims, and no institution-stage signal about which references the panel found compelling. Two cautions: (a) if a defendant in your case was itself a petitioner or real party in interest/privy in a sibling-patent IPR, check whether the same ground was raised against materially identical claims under the family's overlapping claim sets — that is a fact-intensive privity argument, not a per se bar; and (b) the Shkedi family's aggressive use of terminal disclaimers (as reflected in the double-patenting rejection above) means family claims are contractually tied on enforceability, which can affect remedies even where validity is untouched.

Pattern signals: serial petitioners, a diversified but non-aggressive patent owner, and no aggregator. The same defendants appear repeatedly — Roku filed at least three family IPRs (all denied institution), Samsung at least three (all denied; one joined with Microsoft), Meta at least two (one instituted on '398 but then terminated by joint motion; one instituted on '878), Amazon one (denied as time-barred), Yahoo! one (denied), and LiveIntent one (discretionary denial in 2025). That is a classic multi-defendant NPE-defense pattern where each defendant files its own petition to preserve standing rather than a single coordinated challenge. On the patent-owner side, I found no Federal Circuit appeal of any family IPR FWD by Intent IQ — unsurprising, since there appears to be no adverse merits FWD to appeal; the patent owner wins institution-stage denials, which are not appealable by the petitioner (and often not by the patent owner either, since a denial favors it). I saw no Unified Patents or other defensive-aggregator petition in the chain — the challenges are all operating-company defendants, which means '779's apparent non-assertion is doing the work that an aggregator usually would.

Recommended next steps

1. Do not rely on "no PTAB activity" without a same-day check. Pull the Patent Center "PTAB" tab and search PTAB E2E for both the patent number and the assignee (Intent IQ, LLC; AlmondNet, Inc.) — https://ptacts.uspto.gov/ptacts/ — plus Docket Alarm's PTAB search for patent:(9369779). The ODP ingest can trail informal third-party dockets by weeks or months; I could not complete an exhaustive '779-specific docket sweep before running out of search steps, so flag this as verified-zero-as-of-my-search, pending your own confirmation.
2. If you want to file an IPR, understand you'd be first — and that timing matters. There is no § 315(b) bar attributable to a prior '779 petition (there is none), but the one-year clock runs from service of a complaint asserting '779, and the current Director's discretionary-denial framework (settled expectations / patent age, follow-on-petition concerns, parallel-litigation stipulations) is materially harsher than the regime that produced the family's older denials. A 2007-priority patent in force for ~10 years is exactly the profile that has drawn discretionary denials in 2025–2026 (compare the LiveIntent IPR2025-01317 Director denial and the Viant IPR2026-00330 first-to-file posture). Budget for the discretionary-denial briefing, not just the merits.
3. Use the family record, but do not overclaim it. The repeated institution denials on the '398, '164, '198 and '069 patents (Roku, Samsung, Microsoft, Yahoo!, Amazon) are useful persuasive material on the strength of the Shkedi disclosure — they show panels repeatedly declining to find a reasonable likelihood against closely related claims. They are not statutory estoppel, not claim-specific to '779, and not citable as adjudications of '779's claims. If your opponent cites them as "this family is validated," the correct response is that a denied institution is a § 314(a) threshold ruling, not a validity holding, and none of it binds the '779 claims.
4. If claims of '779 ever are invalidated, the anchor will be a certificate, not a decision. There is no FWD to quote today. Until one exists, any statement that "claim X of '779 is dead" would be fabrication — decline to make it, and demand that anyone who does produce the underlying PTAB document.

Bottom line for a defendant today: the '779 patent carries zero PTAB baggage — nothing canceled, nothing upheld, no estoppel, no roadmap. Its sibling patents have been shot at repeatedly and mostly walked away clean at the institution stage, which tells you the family is not an easy 102/103 kill, but it also tells you the Shkedi specification's disclosure is broad enough that later family members have had to fight for distinction over it. Treat '779 as an untested but well-entrenched family member: your invalidity case will be built from scratch, and your best leverage is discretionary-denial-aware scheduling and the family's terminal-disclaimer/overlap structure — not a pre-existing PTAB record.

I'll research the assignment chain and litigation record for US 9369779 before assembling the analysis.

Let me dig into the specific assignment records and litigation involving US 9,369,779.

Let me try to obtain the specific reel/frame records and correspondent of record.

US 9,369,779 B2 — Ownership Reconstruction

Data-provenance note up front, because it affects every citation below: I could not retrieve the underlying USPTO Assignment Center reel/frame numbers or the correspondent of record for this patent with the tools available to me. The three post-issuance/co-pending assignment records below are taken from the Google Patents legal-events panel (which mirrors USPTO assignment data), and Google renders those events with the execution date rather than reel/frame. I am therefore not citing reel/frame numbers — I don't have them, and I won't invent them. Verify at USPTO Assignment Center (search 9369779) and at the Google Patents legal events page.

Inventors

Pattern observations:

• This is a single-inventor patent. There is no co-inventor population to test for post-filing attrition.
• The classic "all inventors left within 12 months of filing" tell is absent and inverted: Shkedi remains the CEO/Chairman of AlmondNet, Intent IQ and Datonics as of the 2026 Amazon settlement press release, and is the named contact of record on USPTO correspondence for the family (the change-of-correspondence filing for related US 8,677,398 lists Intent IQ, LLC at 37-18 Northern Boulevard, Suite 404, Long Island City, NY, with email roy@almondnet.com).
• Family history per IAM's 2024 interview: AlmondNet came close to insolvency in 2013 and survived via a licensing deal Shkedi characterized in court testimony as a take-it-or-leave-it Google deal — relevant to the bankruptcy signal below (no filing occurred).

Original assignee

Two answers, and the distinction matters:

• Assignee on the issued face of US 9,369,779 (issued 2016-06-14): Intent IQ, LLC — Google Patents records "Original Assignee: Intent IQ, LLC," consistent with the AlmondNet→Intent IQ transfer having been recorded before issuance.
• Assignee on the original priority filing (2007-04-17, the '260 parent): AlmondNet, Inc. The 2009 pre-grant publication US20090049468A1 lists AlmondNet, Inc. as applicant, and the parent US 7,861,260 belongs to AlmondNet, Inc.

Entity profile:

• AlmondNet, Inc. — Delaware corporation, founded 1998, HQ 37-18 Northern Boulevard, Suite 404, Long Island City, NY. Self-described in the 2016 complaint as specializing in media/Internet advertising software, now "focused on R&D and the licensing of its patents."
• Intent IQ, LLC — Delaware LLC; majority-owned subsidiary of AlmondNet, Inc. (stated at ¶5 of the 2016 Yahoo complaint; corroborated by Rule 7.1 disclosure statements identifying AlmondNet, Inc. as corporate parent in AlmondNet v. LinkedIn, 1:23-cv-01373 and Intent IQ v. GrowthCode, 1:25-cv-00865).
• Products shipped: Yes. Intent IQ ships a commercial "Dynamic Device Map" cross-device identity-resolution product (cookieless/MAID-less/CTV), HQ New York City with an R&D center in Herzliya, Israel; Datonics ships data-segment products. The group claims 170+ granted patents and licenses Google, Microsoft, Meta, Roku, Samsung, FreeWheel, and (April 2026) Amazon.
• Status: Operating. No bankruptcy, no dissolution, no acquisition of the group.

Assignment timeline

Three recorded reassignments appear on the legal-events panel, all dated 2014-04-07 — the same date application 14/246,525 was filed. I present them in the logical chain order; the recording order on the source panel is the reverse of the transaction order.

• 2014-04-07 (per Google Patents event date) / recorded date not retrievable — Reel/frame not retrieved

  • Conveyance: Assignment of assignors' interest
  • Assignor: SHKEDI, ROY, MR.
  • Assignee: ALMONDNET, INC.
  • Correspondent: not retrievable (no reel/frame surfaced, therefore no correspondent field). Recurrence flag: cannot be assessed.
  • Context: Confirmatory/clean-up inventor-to-company assignment executed contemporaneously with the filing of the divisional application.

• 2014-04-07 / recorded date not retrievable — Reel/frame not retrieved

  • Conveyance: Assignment of assignors' interest
  • Assignor: ALMONDNET, INC.
  • Assignee: INTENT IQ, LLC
  • Correspondent: not retrievable. Recurrence flag: cannot be assessed.
  • Context: Internal corporate reorganization — consolidation of the targeted-advertising patent family under the operating subsidiary. Corroborating independent evidence: a terminal disclaimer filed 2013-07-11 in sibling application 13/167,605 (which issued as US 8,677,398) already lists "INTENT IQ, LLC — 100%" as owner, showing the AlmondNet→Intent IQ consolidation was underway about a year before this record; and the '398 change-of-correspondence filing moved the address of record to Intent IQ, LLC.

• 2014-04-07 / recorded date not retrievable — Reel/frame not retrieved

  • Conveyance: License (see document for details)
  • Assignor: INTENT IQ, LLC
  • Assignee: ALMONDNET, INC.
  • Correspondent: not retrievable. Recurrence flag: cannot be assessed.
  • Context: Intra-group license-back — Intent IQ grants rights back to the parent, the entity that runs group-level R&D and licensing. Characteristically an internal arrangement, not a monetization transfer.

Why these dates cluster: recording three instruments on the same day the divisional was filed is consistent with a single housekeeping package (confirm inventor→company, move family into the operating subsidiary, license back to the parent). It is not a cascading third-party transfer chain.

No other post-2014 assignments appear. The patent has remained with Intent IQ, LLC since, which is consistent with the entire continuation family (US 9,813,778; 10,178,442; 10,715,878; 11,303,973; 11,564,015; 11,589,136; 11,805,300; 11,974,025; 12,212,818 — all listed as Intent IQ, LLC) staying in the same hands.

Timeline diagram

timeline
    title Ownership of US 9369779
    1998 : Roy Shkedi founds AlmondNet
    2007 : Priority application filed
    2010 : Parent US 7861260 issues to AlmondNet
    2013 : AlmondNet nears insolvency
         : Google license deal signed
         : Family consolidating into Intent IQ
    2014 : Divisional application filed
         : Shkedi to AlmondNet assignment recorded
         : AlmondNet to Intent IQ assignment recorded
         : Intent IQ license back to AlmondNet
    2016 : US 9369779 issues to Intent IQ LLC
         : Family first asserted against Yahoo
    2024 : Jury awards 122M against Amazon
    2026 : Amazon dispute settled with license

NPE / troll-pattern signals

1. Shell-entity transfer — not present.
The only operating-assignee-to-LLC movement is AlmondNet, Inc. → Intent IQ, LLC (2014-04-07). This is intra-group: Intent IQ is a majority-owned subsidiary of AlmondNet (¶5, AlmondNet, Inc. v. Yahoo! Inc., 16-cv-1557, E.D.N.Y.; Rule 7.1 disclosures in 1:23-cv-01373 and 1:25-cv-00865). Intent IQ is not licensing-only: it ships the Dynamic Device Map identity product, has an R&D center in Herzliya, and its address of record is a real office (37-18 Northern Blvd, Suite 404, Long Island City) with the CEO's email — not a registered-agent service address. The same-day license-back to the parent is the opposite of a monetization-only transfer. Caveat and honest limit: Intent IQ, LLC is a Delaware LLC, and AlmondNet, Inc. describes itself as "focused on R&D and the licensing of its patents," so the licensing-only characterization does apply to the parent; and I could not verify domicile/registered-agent data or the reel/frame without Assignment Center access.

2. Known asserter in the chain — not present.
No assignee in this chain matches Acacia, Marathon Patent Group, Intellectual Ventures, IPNav, Wi-LAN, Mosaid/Conversant, Vringo, Pendrell, Innovatio IP Ventures, MPHJ, Lumen View, Round Rock, Document Generation Corp, or Spangenberg entities, based on the searches run. What is present is a high-volume self-funded enforcer: the group has sued Amazon, Samsung, Meta, Microsoft/Xandr, Roku, FreeWheel/Beeswax, Viant, Lotame, LiveIntent, LinkedIn, Oath/Yahoo, Oracle and GrowthCode. Unified Patents maintains a page for this patent (portal.unifiedpatents.com/patents/patent/US-9369779-B2) and the family has drawn IPRs from Roku, Samsung, Meta, Amazon and Microsoft. Active enforcer ≠ listed NPE, and I am not collapsing the two.

3. Repeat correspondent across the chain — unclear (data not available).
This is the signal I most wanted and could not close: no correspondent of record for any of the three 2014-04-07 records was retrievable, so I cannot test for recurrence and I will not infer it from names. Two adjacent, non-assignment data points worth carrying forward for whoever has Assignment Center access:

• Prosecution correspondent on sibling application 13/167,605 (2011–2013 filings, including the 2013-07-11 terminal disclaimer listing Intent IQ, LLC as 100% owner) was "Louis J. Hoffman / Donald Hertz." Whether Hoffman also signed the 2014 assignment cover sheets is the open question.
• Litigation counsel is Farnan LLP — Brian E. Farnan and Michael J. Farnan, appearing for the group across AlmondNet v. LinkedIn (1:23-cv-01373), AlmondNet v. Viant (1:23-cv-00174), AlmondNet v. FreeWheel (1:23-cv-00220), Roku v. AlmondNet (1:21-cv-01035) and Intent IQ v. GrowthCode (1:25-cv-00865). Farnan is litigation, not recording, counsel — a single firm handling all assertion work is an enforcement-consolidation tell, not a shell-shelling tell.

4. Cascading transfers — not present.
Three instruments, one day, one corporate group, no chained unrelated LLCs, no <24-month relay. Nothing here resembles the Acacia/Marathon multi-hop pattern.

5. Pre-litigation transfer — not present.
The last transfer is 2014-04-07. The earliest family assertion I can date is AlmondNet, Inc., Datonics, LLC and Intent IQ, LLC v. Yahoo! Inc., No. 16-cv-1557 (E.D.N.Y., 2016) — roughly a 24-month gap, well outside the 6-month window. Note US 9,369,779 itself was not issued until 2016-06-14, so no transfer could have been timed to a pre-issuance suit on this patent.

6. Bankruptcy fire-sale — not present.
No Chapter 7/11 proceeding, no 363 sale. The near-death episode was 2013, and per Shkedi's own court testimony (reported by IAM, 2024) it was resolved by a Google licensing deal that preserved headcount — not by a patent sale. If anything, the 2014-04-07 intra-group reorganization reads as the recovery restructuring, not a distress sale.

7. Privateering — not present.
There is no operating-company parent on whose behalf an NPE is asserting against the parent's competitors. AlmondNet/Intent IQ/Datonics litigate for their own portfolio and their own product businesses.

8. Defensive aggregator — not present.
The chain terminates at Intent IQ, LLC (operating), not at RPX, AST, LOT Network, Unified Patents or OIN. Inverse-signal note, unverified and flagged: Unified's portal associates an AlmondNet-origin patent, US 11,610,226 B2, with "Blue Hills Series 95 of Allied Security Trust I." If accurate, some AlmondNet-family assets have passed to an AST-linked vehicle — but that is a different patent and is not in the '779 chain. Do not carry it into this verdict without confirming the assignment record.

Verdict

Operating-company assertion.

The chain terminates at Intent IQ, LLC, which per the 2014-04-07 AlmondNet→Intent IQ assignment (recorded date, no reel/frame retrieved; issued 2016-06-14 with Intent IQ, LLC as assignee of record) is a majority-owned operating subsidiary of AlmondNet, Inc. — shipping a commercial cross-device identity-resolution product, with a real office and its CEO as contact of record, and litigating the same family against genuine competitors (Amazon, Roku, Meta, Microsoft/Xandr, Viant, Lotame, LinkedIn, Oracle) rather than against end users or shell holding companies. None of the shell-entity, known-asserter, cascading-transfer, bankruptcy, privateering or defensive-aggregator signals is present; the only two live questions are the unretrieved assignment correspondents (signal 3) and the fact that AlmondNet, Inc. is self-described as licensing-focused, with group counsel publicly stating the annual revenue mix "depends on the year" between ad-tech vendor and patent litigant.

Honest qualification: if one applies a purely economic NPE definition (majority of revenue from licensing/enforcement), this family trends toward NPE — moderate confidence, and signal 3 could materially shift that call if the 2014 recording correspondents turn out to recur across the group's other recordings. On the evidence actually in hand, "operating-company assertion" is the defensible call, and I flag signal 3 as the thing to verify first.

Verification: USPTO Assignment Center — search patent number 9369779; cross-check the patent-number record against USPTO Patent Assignment Search and the Google Patents legal events for US 9,369,779.

I'll research the prior art citations for US 9,369,779. Let me search multiple sources.

Let me dig into the specific citations listed on the patent's Google Patents page.

Excellent — I found key sources including the prosecution history and an IPR. Let me get the patent's front-page "References Cited" and the IPR grounds.

I hit my research step limit, so I'll report what I verified and be explicit about what I could not verify. Below I build on — and in one place refine — the earlier litigation section rather than repeat it.

Prior art analysis — US 9,369,779 B2

1. Identifier check (USPTO record)

Source: the authoritative full text supplied in this analysis (Google Patents, https://patents.google.com/patent/US9369779/en), corroborated by the PTAB prosecution-history exhibit titled "Prosecution History of U.S. Patent No. 9,369,779," filed 2022-07-27 as Exhibit 1060 in IPR2022-01315, Roku, Inc. v. Intent IQ, LLC (docketalarm.com/cases/PTAB/IPR2022-01315/...).

Important caveat on scope of this answer: My queries hit the tool step limit before I could pull the certified front page (the "(56) References Cited" block, which typically runs ~50–60 U.S. patents plus foreign references and non-patent literature). The reference list below is the verifiable subset I could tie to a source — the prosecution record and the Unified Patents prior-art index for this exact patent — not a verbatim reproduction of Form PTO-892/1449. I flag each source. I did not auto-correct or normalize any number.

2. Refinement to the earlier litigation section (flagged, not a contradiction)

The earlier section concluded that no litigation asserting the '779 patent could be found, and that the "Family has litigation" flag reflects other family members. My searches are consistent with that, but I found one additional fact worth adding, because it bears on prior art:

• The '779 is not a stranger to the PTAB record. Its prosecution history was filed as an exhibit in Roku, Inc. v. Intent IQ, LLC, IPR2022-01315 (petition filed 2022-07-27). Per the Unified Patents index, IPR2022-01315 concerned the parent/related U.S. Pat. No. 10,715,878 and institution was denied ("insufficient factual showing to support its allegations regarding the recited common local area network" — Samsung IPR2022-01505 institution decision quoting IPR2022-01315 Paper 9). The '779 appears as a related/parent patent whose prosecution history was used as evidence, not as the challenged patent. This refines (does not contradict) the earlier "not asserted" conclusion.
• Several invalidity contentions / IPR appendices (e.g., Roku's "State of the Art References for Asserted Patents," filed in the Delaware and PTAB matters) recite the '779 in the family chain (No. 14/246,525 → '779, "a divisional of application Ser. No. 13/097,885").

3. Prior-art references cited in / against this patent

All dates below are as reported by the sources cited; where a title was not shown in the source I say so rather than guess.

3a. References that appear in the '779's own prosecution history (highest confidence)

Source: PTAB prosecution-history exhibit (docketalarm.com/cases/PTAB/IPR2022-01315/...Exhibit-1060-Prosecution_History_of_US_Patent_No_9,369,779.pdf) and a related IDS/transmittal in the same family (PTACTS/docketalarm exhibit for app. 13/097,885).

3b. Prior art indexed against this exact patent

The Unified Patents portal entry for US-9369779-B2 lists "Patent Art (201)" — an index of references associated with this patent (it may span the family). The following entries were retrieved with titles and dates; each is a § 102(a)/(b)/(e) candidate depending on its effective date relative to the 2007-04-17 priority date. The strongest structural candidates for single-reference (§ 102) anticipation of the IP-association + cross-device-targeting concept are the cross-device / profile-delivery references (items 7, 11, 13, 17); the remainder are more naturally § 103 combination material.

Additional references that Google Patents cross-links to this patent (all candidates for the "(56) References Cited" block; descriptions unverified):

• US 5,809,242 A — Shaw et al. — "Electronic mail system for displaying advertisement at local computer received from remote system…" (patents.google.com/patent/US5809242/en)
• US 5,995,943 A — Bull et al. — "Information aggregation and synthesization system" (patents.google.com/patent/US5995943A)
• US 5,848,396 A (Gerace), US 5,855,008 A (Goldhaber et al.), US 5,933,811 A and US 6,385,592 B1 (Angles et al.), US 5,948,061 A (Merriman et al.), US 6,160,989 A (Hendricks et al.), US 6,216,129 B1 (Eldering), US 6,366,298 B1 (Haitsuka et al.), US 6,377,986 B1 (Philyaw), US 6,446,261 B1 (Rosser), US 6,718,551 B1 (Swix et al.), US 6,771,290 B1 (Hoyle), US 6,845,396 B1 (Kanojia), US 6,868,392 B1 (Ogasawara), US 6,944,585 B1 (Pawson) — plus the Shkedi-family originals US 6,832,207 B1; 6,925,440 B1; 6,973,436 B1; 7,072,853 B2.
• Foreign: EP 1150508 A2 and EP 1058999 (both appear in the References Cited of sibling/related U.S. patents 8,566,164 and 8,607,267).
• Non-patent literature: Battelle, John, The Search (Portfolio, N.Y., 2005), pp. 167–171 — cited as background and, per the specification, the source for the Google/BSkyB cross-media targeting example.

Sources: Unified Patents portal, https://portal.unifiedpatents.com/patents/patent/US-9369779-B2; Google Patents pages for the cross-linked numbers above; front pages of related Shkedi-family patents (e.g., patentimages.storage.googleapis.com/…/US10645438.pdf, …/US8566164.pdf, …/US8607267.pdf).

4. Assessment — § 102 vs § 103

1. No verified single-reference § 102 anticipation. Nothing I could retrieve shows one reference disclosing the full combination claimed — (i) electronically associating an online-access IP address with a set-top-box IP address for a multitude of users, and (ii) selecting and automatically directing a TV ad to the associated STB using online-behavior/demographic profile information. The references cluster into three groups (online behavioral targeting; IPTV/ad-insertion; TV-viewer profiling) that together support § 103 combinations, not standalone anticipation.
2. The documented rejection was § 103, not § 102 — Blasko + McCarthy in the parent application 13/097,885.
3. Self-citation risk. A large fraction of the "patent art" are Shkedi/AlmondNet/Datonics family members (e.g., US 8,239,264; 8,244,583; 8,677,398; US 2011/0040607). These are the same inventive entity/common ownership and normally are not § 102 prior art against the '779 — they were cited for background and family continuity.
4. Claim mapping limitation. The '779's granted claim set was not retrievable within my search budget. The only claim text I could verify is from the parent's amendment (claim 52, added 2014-08-08: "receiving … electronic indicia of a first online user interface device identifier; … derived automatically from online activity …; … automatically facilitating direction of a first television advertisement … via a first set-top box … electronically associated with the first online user interface device identifier by the television service provider, without causing online behavioral profile information … to be transmitted to the television service provider, and without receiving … personally identifiable information"). Accordingly, the § 102 mapping above is directed at the independent method/system claims covering this subject matter, not to specific granted claim numbers. I will not invent claim numbers I could not read.

5. What I could not verify (stated explicitly, per instructions)

• The certified front-page "References Cited" block of US 9,369,779 (exact count, examiner-vs-applicant distinction, and foreign/NPL entries) — not retrieved.
• The exact publication/filing dates of Blasko (US 2001/0049620 A1) and McCarthy (US 2007/0115389 A1), and their full contents.
• Whether any reference is a § 102(b) vs § 102(a)/(e) reference — this turns on the actual invention date, which is not established in the record I have.
• The exact list of patents asserted against Roku/Samsung etc. — the earlier section already flagged this; my only addition is the IPR2022-01315 linkage in § 2 above.

6. Recommended verification steps

1. Pull the front page of US 9,369,779 from USPTO Patent Center / Google Patents PDF (patentimages.storage.googleapis.com/.../US9369779.pdf) to get the definitive (56) list.
2. Pull the file wrapper for 14/246,525 (and parent 13/097,885) via USPTO Patent Center to obtain the PTO-892/1449 and the Blasko/McCarthy rejection on the record.
3. Read IPR2022-01315 Exhibit 1060 in full — it is the actual prosecution history of the '779 and will contain every reference and rejection verbatim.
4. Check IPR2024-00666, -00701, -00709 (Roku-related) and the Delaware contentions (Case 1:21-cv-01035-MN) for any charting that names the '779.

If you'd like, I can run these targeted follow-ups individually to close the gaps above.

I'll research the prior art cited on the '779 patent page and related references before building the obviousness analysis.

I found the prosecution history and an IPR reference. Let me dig into the specific prior art and the claim set.

Let me retrieve the exact claim set of the '779 patent and details on the key prior-art references.

Obviousness Analysis — U.S. Patent No. 9,369,779 (Shkedi / Intent IQ)

0. Scope, law, and confidence notes

Legal standard applied. Because the '779 patent's earliest effective filing date is 2007‑04‑17 (via 11/736,544 → 12/860,666 → 13/097,885 → 14/246,525), the pre‑AIA version of 35 U.S.C. § 103(a) governs. The analysis below follows the Graham v. John Deere, 383 U.S. 1 (1966) factors (scope/content of the prior art; differences from the claims; level of ordinary skill; objective evidence) and the KSR Int'l v. Teleflex, 550 U.S. 398 (2007) rationales — notably (a) known elements combined by known methods to yield predictable results, (b) use of a known technique to improve a similar device in the same way, and (c) a finite set of identified, predictable solutions with a reasonable expectation of success. A motivation to combine must be articulated and supported by the record (which is why the prosecution record and the PTAB expert declarations cited below matter).

Confidence / accuracy caveats (required by my operating rules):

1. I was cut off by my search-step limit before retrieving the verbatim, full claim set of the '779 patent. I therefore analyze by limitation groups reconstructed from (i) the Google Patents abstract and definitions, (ii) the prosecution history of application 14/246,525 (the '779 application) introduced as Roku Exhibit 1060 in IPR2022‑01315, and (iii) the closely related claim text quoted in the AlmondNet v. FreeWheel infringement chart. Where I use a sibling's claim text as a proxy, I flag it.
2. Three of the four references the examiner called "closest prior art" (Ertugrul, Howcroft, Haeuser) have publication dates after 2007‑04‑17. They are only available as prior art under pre‑AIA § 102(e) if their own effective filing dates precede the '779 priority date. I could not confirm those filing dates in this session; treat the § 102(e) status of those three as a to‑verify item.
3. Search results that appear to contradict training data were treated as ground truth; no patent number has been auto‑corrected.

Contradiction check against the earlier Litigation Summary. No contradiction. The earlier section concluded the '779 patent itself was "not found" as an asserted patent. The new material is consistent — but adds an important nuance: the '779's prosecution history is in the PTAB record as an exhibit (Roku Exhibit 1060 in IPR2022‑01315), even though the patents challenged in the petitions I saw were the '198 (10,321,198) and the '878 (10,715,878), not the '779. IPR2022‑01315 is styled Roku, Inc. v. AlmondNet, Inc. & Intent IQ, LLC and resulted in a denial of institution (Paper 9, 2023‑03‑13) on the "common LAN" limitation, followed by a rehearing request and a POP request. Denial of institution is not a § 103 adjudication and carries no preclusive weight.

1. The invention as claimed (limitation groups)

The '779 shares the specification of the family (7,861,260 / 8,281,336 / 8,695,032). The Google Patents page states "Prior art keywords: online, address, television, top box, user" and "Prior art date: 2007‑04‑17." The claims fall into these functional groups:

The examiner's reasons for allowance (application 14/246,525, Primary Examiner Nnenna Ekpo, Art Unit 2425, 2015‑07‑13) expressly relied on the Group D language and stated that Ertugrul, Blasko, Howcroft, and Haeuser "show similar method" but "do not disclose nor suggest" the third‑party‑TVSP/no‑PII‑transmission combination. That is the rejection–allowance battleground a § 103 challenge must attack.

2. Level of ordinary skill

A POSITA here would be a software/network engineer or systems architect with ~2–4 years' experience in networked content delivery, online advertising systems, and/or IP‑based television (IPTV), familiar with DHCP dynamic IP allocation, NAT/router IP addressing, HTTP cookies, ad‑serving infrastructure, and set‑top‑box middleware. The IPR expert declaration in the family (Dr. Samrat Bhattacharjee, ROKU‑1003) and the PTAB's own framing align with this level.

3. The prior art of record (the "Prior Art" materials on the page)

(i) References cited by the examiner in the '779 prosecution (PTO‑892, 2015‑07‑13):

(ii) References on the face of the family (56) lists and Unified Patents "Patent Art (201)" (selected): Merriman, US 5,948,061 (online ad targeting/delivery based on a client‑side profile — DoubleClick); Guyot, US 6,119,098 (targeted ads to a computer using a unique proprietary identifier and subscriber data); Kung, US 6,055,573 (customized advertising); Eldering‑era TV targeting; US 2002/0129368 (Prime Research Alliance E — "Profiling and Identification of Television Viewers"); US 6,073,241 (CBS Interactive — tracking browser requests across distinct domains using persistent client‑state); US 6,385,592 (delivering customized ads in interactive communication systems); Warrier, US 7,949,785 (private/public address mapping via a gateway — supports the "common IP address"/NAT concept); foreign art including JP 2003‑331183 and JP 2006‑004441.

(iii) References used by Roku in the family's IPRs (PTAB record): Madhavan, US 8,180,674; Whitehead, US 2008/0281697; Baum; Damick; Varghese; Vallier, US 9,495,682; and Shkedi's own US 2005/0086112.

4. Combinations that render the claims obvious

Ground 1 — Ertugrul + Blasko (the examiner's own "closest art" pairing)

• Ertugrul supplies Group A/B architecture: collecting network‑activity information from client devices and correlating it to client devices and to service providers, applying rules to identify a user, and providing targeted content to the client device associated with that user — i.e., cross‑device, cross‑provider behavioral targeting.
• Blasko supplies the missing TV linkage and the privacy mechanism: a profiling server that receives profile vectors from set‑top boxes alongside Internet surfing data, with transaction identifiers deliberately chosen to "protect the privacy of the consumer."
• Motivation: Both are in the same field (targeted advertising across devices/networks) and address the same problem — matching ads to users based on prior behavior. A POSITA seeking to extend Ertugrul's correlation to television delivery would look to Blasko's STB‑profile‑vector teaching; the result (profile the online behavior, associate it with the household's STB, deliver a targeted TV ad) is the predictable aggregation of two known systems, and Blasko expressly supplies the non‑PII/privacy motivation that the '779's spec asserts as its advantage.
• This pairing was actually applied by the examiner against dependent claims (claims 61‑62 rejected over Blasko in view of Ertugrul per the family prosecution record), which is powerful evidence that the combination was considered technically sound.

Ground 2 — Ertugrul + Howcroft + Haeuser (IPTV delivery)

• Howcroft teaches the Group B end‑to‑end flow in an IPTV environment: build subscriber profile/behavioral data, select a segment, insert a targeted ad into video, and report delivery to the advertiser.
• Haeuser teaches the STB‑side mechanism: receive targeted ad content with the TV programming from the IPTV video server, detect the ad‑insertion point, and splice the targeted ad.
• Motivation: Both are IPTV targeted‑advertising references; Haeuser supplies the client‑side enabler without which Howcroft's segment‑targeted ads cannot be presented. Combining a server‑side targeting/insertion system with a known STB insertion mechanism is a textbook KSR "known technique improving a similar device in the same way," and the IPTV service provider inherently knows the STB identity/address (the "common IP address"/STB identifier of Groups A and C).

Ground 3 — Madhavan + Whitehead (± Vallier) — the combination Roku actually ran

• Madhavan (US 8,180,674) provides an advertising/content management entity that matches ads to user events/actions/interests, with user profiles containing demographic, geographic, and behavioral data.
• Whitehead (US 2008/0281697) is the key reference for Group C: it expressly teaches that multiple IP‑enabled devices of a residential home network — PCs, laptops, set‑top boxes, mobile handsets — are accessed "through a single IP connection managed by a home‑router," so that "a single IP access session can involve multiple devices," and that devices/IP addresses can be associated using the router's single IP address plus a timestamp (e.g., "1180474.4374" or "2007.06.01.22:05:36" formats), all as non‑PII/anonymous addressability. Whitehead even articulates the motivation: "anonymous marketing techniques can be considered more desirable, from a privacy standpoint, than non‑anonymous marketing techniques."
• Vallier (US 9,495,682) adds TV‑viewing‑habit monitoring and interactive‑ad feedback for more effective targeting (relied on for dependent claims such as timing/feedback limitations).
• Motivation: Whitehead itself supplies the express privacy motivation and the exact LAN/common‑IP‑address association mechanism the claims require, while Madhavan supplies the ad‑selection engine. Roku's expert (Dr. Bhattacharjee) averred a POSITA would combine them for "improved content management… and improved privacy protection without using PII."

Ground 4 — Merriman / Guyot + Blasko or Whitehead (extend known online behavioral targeting to TV)

• Merriman (US 5,948,061) and Guyot (US 6,119,098) establish, well before 2007, that ads can be selected from an online behavioral profile tied to a computer identifier (cookie / "unique proprietary identifier"). The '779 specification itself admits this: "the use of cookies enables an ad server to recognize an Internet site visitor… The ad server can target future advertising to the site visitor based on that previous activity."
• The only remaining problem is transporting that known targeting decision to the household's television. Whitehead (or Blasko's STB profile vectors, or Hauser/Howcroft's IPTV delivery) supplies that transport.
• Motivation: The specification also admits the prior art of cross‑media targeting — the announced Google–BSkyB arrangement is described as using BSkyB portal searches to target that subscriber's DVR with ads. The '779 characterizes the sole shortcoming of those systems as the need to use PII. Once a non‑PII association (Whitehead's common router IP + timestamp; Blasko's privacy‑preserving transaction IDs) is available, the claimed subject matter is the predictable combination of admitted prior art with a known anonymity technique — squarely KSR.

5. Dependent‑claim targets

Dependent limitations on common‑IP‑address identity ("online access IP address and STB IP address are the same"), dynamic DHCP reassignment, periodic/irregular STB reporting, matching by time+date, identifiers/cookies placed on the STB, and mobile/roaming devices are each independently met by Whitehead's IP‑access‑session + timestamp model, Blasko's STB profiler/server architecture, and, for mobile, Whitehead's inclusion of mobile handsets in the same IP session. These are the kinds of "routine" secondary limitations the PTAB treats as obvious once the independent claim falls.

6. Objective evidence and counter‑arguments (why the patent survives today)

A defensible § 103 opinion must also present the weaknesses:

1. The "common LAN / common IP address routed to both the computer and the STB" limitation is the sticking point. In IPR2022‑01315 the Board denied institution, finding an insufficient showing that the asserted combination taught associating devices of a common LAN (it examined Baum's FIG. 8 largely in isolation). Roku moved for rehearing and POP review, arguing the Board ignored Baum's FIG. 5 (multiple IP‑addressed devices on a common LAN) and Madhavan's own teachings. This shows a real, litigated dispute about whether the prior art reaches the specific common‑IP‑address‑routing architecture. However, the decision is preliminary and non‑precedential, and the denial turned on the evidentiary sufficiency of that petition's mapping — not on a holding of validity.
2. Group D (no online‑behavioral‑profile transmission to, and no PII receipt from, the third‑party TVSP) is the limitation the examiner used to allow claims 52‑80. Howcroft and Haeuser are, respectively, an IPTV network operator's and a TVSP's own systems — i.e., the profile naturally sits inside the TV service provider. A challenger must therefore supply a reference in which the ad‑selection profile is held by an entity outside the TVSP (e.g., an online ad server / profile provider, as in Ertugrul's multi‑party correlation or Whitehead's anonymous IP‑session marketer) and explain why no PII needs to cross to the TVSP. This is articulable but is precisely where the prosecution record shows the art falling short.
3. § 102(e) timing. Confirm the effective filing dates of Ertugrul (US 2008/0222283), Howcroft (US 2008/0040742), and Haeuser (US 2007/0283384) against 2007‑04‑17. Their publication dates all post‑date the '779 priority date; only a § 102(e) showing (or reliance on the earlier‑published Blasko, Merriman, Guyot, and Whitehead references) will support a § 103 ground.
4. No secondary‑considerations record has surfaced (no nexus evidence of commercial success, long‑felt need, or industry praise attributable to the claimed subject matter), so objective evidence currently weighs neither way.
5. Design‑around / hindsight risk. Because the specification expressly frames the invention as PII avoidance, a challenger must be careful that the motivation to combine is drawn from the references (Whitehead's express anonymity rationale; Blasko's express privacy transaction IDs) rather than from the '779's own disclosure.

7. Bottom line

On the record available, the '779 patent is vulnerable under § 103 for its broad claims via:

• Ertugrul + Blasko (the examiner's own "closest art," already applied to dependent claims), and
• Madhavan + Whitehead (+ Vallier) — the strongest combination for the common‑IP‑address / common‑LAN limitation, because Whitehead expressly teaches associating a PC, STB, and mobile device of one home network via the single router IP plus timestamp, and expressly motivates doing so anonymously without PII.

Its best defense rests on the Group D privacy limitations (prohibiting transmission of online behavioral profile data to the third‑party TV service provider and receipt of PII from it, claim type allowed as claims 52‑80) and on the untested factual dispute over whether the art teaches the common IP address routing traffic to both the online device and the STB — the same issue on which the PTAB denied institution of Roku's IPR against the sibling '878 patent.

Highest‑value next steps: (a) pull the verbatim claim set of 14/246,525 / US 9,369,779 from the issued patent PDF to lock down every limitation; (b) confirm the effective filing dates of Ertugrul, Howcroft, and Haeuser for § 102(e) purposes; (c) obtain the full ROKU‑1002/ROKU‑1003 declarations and the IPR2022‑01315 institution decision text; and (d) review the AlmondNet v. FreeWheel and AlmondNet v. Roku invalidity contentions, which likely already chart these exact combinations against the family.

Sources: Google Patents US9369779B2; Roku Exhibit 1060 — Prosecution History of US 9,369,779; IPR2022-01315 Petition (’878); IPR2022-01315 Request for Rehearing; P‑TACTS petition 1549592 (Madhavan/Whitehead/Vallier expert declaration, ’198 patent); FreeWheel infringement chart (’260 claim 1); Unified Patents US‑9369779‑B2; US9813778 (sibling, (56) references).