US 10251021

US Patent 10251021 — Summary

Bibliographic data (verified against Google Patents and RPX search results)

Abstract

A method and system for generating by a server, a recommendation for a venue or an event of interest proximally located to an inferred current location of a user of a portable media player device. In an embodiment, a server may determine one or more media preferences for a user based at least in part on media stored on or accessed using the portable media player device of the user. The server may generate information related to the current location or a recommendation for a venue or an event of interest to the user based at least in part on the one or more media preferences and the inferred current location of the user. Once generated, the server then may communicate the recommendation (or the information) to the user, e.g. the portable media player device.

Plain-language overview of the independent claims

Claim 1 (server-side computer-readable medium). A non-transitory computer-readable medium with instructions that cause a server to: (a) receive media (or info about it) stored on or accessed via a portable media player device; (b) infer the user's entertainment preferences from that media/info and from the media content types involved — expressly including textual, image, audio, and video media; (c) access sources describing venues/events near the user's current location; (d) pick a venue/event matching the inferred preferences and location; (e) generate a recommendation naming it; and (f) send the recommendation to the user. In plain terms: a server learns what the user likes from the content on their media player, learns where the user is, and pushes a nearby venue/event recommendation.

Claim 10 (device-side computer-readable medium). Instructions that cause a portable media player device to: (a) run a media player application accessing media from local or remote storage; (b) connect to a server that provides location- and preference-based venue/event recommendations; (c) send its location info so the server can infer the user's current location; and (d) receive a recommendation of an event/venue near that inferred location, where the recommendation is based on inferred preferences (drawn from the accessed media, its content types among a plurality including textual, image, audio, and video, and info about the media) and is selected from sources describing nearby venues/events. In plain terms: the device's side of the arrangement — it reports where it is and receives the nearby recommendation.

Claim 16 (computer server). A server having a communication interface that receives media (or info about it) from remote storage, plus processors that: (a) infer the user's preferences from the received media/info and its content types (textual, image, audio, video); (b) infer the user's current location from location info associated with the device; (c) access sources of info about nearby venues/events; (d) identify a venue/event of interest based on the inferred preferences and location; (e) generate a recommendation; and (f) send it to the device via one of SMS, email, pager, facsimile, or phone call. This is the system claim covering the server hardware configured to perform the claimed functionality.

Litigation / IPR / appellate status (with confidence notes)

• District court: The patent has been asserted in Texas Western District Court (6:21-cv-01245; 6:22-cv-00357) and California Northern District Court (3:22-cv-04928), per the Google Patents record. In the N.D. Cal. action, Google moved for judgment on the pleadings on § 101 ineligibility; a CourtListener document shows the court treating Claim 1 of the related '960 patent as representative of a family of patents (including the '021) directed to using media preferences + location to recommend nearby venues/events.
• IPR: Google LLC and YouTube, LLC filed IPR2023-00244 against this patent (petition dated Nov. 30, 2022). Oral argument was held Feb. 29, 2024 (combined with IPR2023-00246 on the related '960 patent). An inter partes review certificate (K1) was issued Aug. 19, 2024 (per the USPTO legal-events record on Google Patents), confirming a Final Written Decision. I do not have authoritative confirmation of which claims survived vs. were found unpatentable in that FWD — I could not retrieve the FWD text in my searches, so that specific outcome should be treated as unverified.
• CAFC 2026: My search of CAFC 2026 dockets did not locate an appeal specific to the '021 patent. I did find a Federal Circuit decision dated Feb. 13, 2026 in Google LLC v. Wildseed Mobile, LLC (Case No. 24-2178, before Judges Prost, Hughes, and Stark), but that appeal concerns a different patent in the same family — the '169 patent (IPR2023-00247) — and the Board's finding that claim 15 of that patent was not unpatentable. I have no verified evidence of a parallel CAFC appeal involving 10251021 as of the current date, and I will not speculate one exists.

Points of uncertainty

1. The specific outcome (claims canceled vs. upheld) of the IPR2023-00244 Final Written Decision for the '021 patent — not verified from the FWD itself.
2. Whether any CAFC appeal in 2026 concerns the '021 patent specifically — not found in searches; the confirmed 2026 CAFC case (24-2178) involves a sibling patent.
3. Exact current chain of title beyond the recorded assignments noted above (Wildseed Mobile LLC; security interest to Lit-US Chisum 21-A, LLC).

I'll research litigation involving US Patent 10,251,021. Let me search across multiple sources.

I have good initial leads. Let me verify the specific case numbers, filing dates, and outcomes.

I now have the core district court picture. Let me verify the IPR2023-00244 outcome and check for any Federal Circuit appeal specific to the '021 patent.

Based on my research, I can now provide a comprehensive litigation history for US Patent 10,251,021. Note that I reached my search limit before I could retrieve the IPR Final Written Decision text itself, so I flag the specific confidence levels below.

---

Litigation involving US Patent 10,251,021 (as of April 26, 2026)

All known litigation involves a single plaintiff, Wildseed Mobile LLC (the current assignee), asserting the '021 patent together with family members 7,376,414; 9,141,960; 10,959,040; and 10,869,169 against Google LLC and YouTube, LLC. There are three district court dockets (two of which are the same case after transfer) plus one inter partes review.

Summary table

---

Case details

1. Wildseed Mobile LLC v. Google LLC et al., No. 6:21-cv-01245-ADA (W.D. Tex., Waco Division)

• Plaintiff: Wildseed Mobile LLC (a Texas LLC, at 510 Austin Avenue, Waco, TX)
• Defendants: Google LLC; YouTube, LLC
• Filing date: December 1, 2021 (docketed under Judge Alan D. Albright)
• Patents asserted: The First Amended Complaint (Dkt. 45, filed Feb. 28, 2022) asserted five patents: 7,376,414; 9,141,960; 10,251,021; 10,959,040; and 10,869,169. Wildseed's infringement contentions later charted claims 1, 2, 4–7, and 9–14 of the '021 patent (per Patent Owner's Mandatory Notices in the related IPRs).
• Outcome: Wildseed filed a Notice of Voluntary Dismissal Without Prejudice under Rule 41(a)(1)(A)(i) on April 4, 2022 (Dkt. 50). This is confirmed by the dismissal exhibit (Ex. 1014 in IPR2023-00248) and by Wildseed's own Mandatory Notices in IPR2023-00246, which list the case as "Voluntarily Dismissed April 4, 2022."

2. Wildseed Mobile LLC v. Google LLC et al., No. 6:22-cv-00357 (W.D. Tex.)

• Plaintiff: Wildseed Mobile LLC
• Defendants: Google LLC; YouTube, LLC
• Filing date: April 4, 2022 (same day the prior case was dismissed — a dismiss-and-refile)
• Patents asserted: Same five patents, including 10,251,021 (per the AO 120 Report on Filing/Determination of Action)
• Outcome: By order dated August 29, 2022 (Dkt. 66), the court granted transfer to the Northern District of California (the parties stipulated to transfer). The case was re-docketed as 3:22-cv-04928.

3. Wildseed Mobile LLC v. Google LLC et al., No. 3:22-cv-04928 (N.D. Cal.)

• Plaintiff: Wildseed Mobile LLC
• Defendants: Google LLC; YouTube, LLC
• Filing date in N.D. Cal.: August 29, 2022 (upon transfer from W.D. Tex.)
• Judge: William H. Orrick
• Patents asserted: 7,376,414; 9,141,960; 10,251,021; 10,959,040; 10,869,169 (Wildseed asserted 12 claims of the '021 patent)
• Outcome — § 101 invalidation: Google and YouTube moved for partial judgment on the pleadings under Rule 12(c), arguing the '960, '021, and '040 patents were ineligible under 35 U.S.C. § 101. Judge Orrick treated Claim 1 of the '960 patent as representative of the three patents, finding all "directed at using information about the media contained on a user's mobile device and the user's location to recommend nearby venues or events." Applying Alice step one, he found the claims directed to the abstract idea of tailoring recommendations based on user preferences and location (analogizing to Intellectual Ventures I LLC v. Capital One Bank (USA)), and at step two found no inventive concept — the portable media player was "not functionally distinguishable from any other portable device" and the claims "could be performed by humans without the aid of technology." The motion was GRANTED, and Claims II, III, and IV of the First Amended Complaint — corresponding to the '960, '021, and '040 patents — were dismissed with prejudice. (This is the CourtListener opinion at Dkt. 110, gov.uscourts.cand.399836; also summarized by Studicata and Casetext.)

4. Google LLC v. Wildseed Mobile LLC, IPR2023-00244 (PTAB)

• Petitioners: Google LLC and YouTube, LLC
• Patent Owner: Wildseed Mobile LLC
• Patent: US 10,251,021 B2
• Filing: Petition filed November 30, 2022 (institution of inter partes review)
• Status: A Final Written Decision issued (PTAB), and an inter partes review certificate (K1) was issued August 19, 2024, per the USPTO legal-events record on Google Patents. A contemporaneous Fenwick report (June 5, 2024) describes "a notable pair of decisions invalidating patents held by Wildseed Mobile LLC as obvious over art asserted by Google," including a rare dissent. Caveat: I could not retrieve the FWD text itself, so I cannot state with certainty which specific claims of the '021 patent were canceled versus upheld. The prior summary in this analysis flagged the same uncertainty; it remains unresolved.

---

Points to flag (confidence notes)

1. CAFC appeal specific to the '021 patent: I did not locate a Federal Circuit appeal specifically involving the '021 patent. The confirmed 2026 Federal Circuit decision in Google LLC v. Wildseed Mobile, LLC (No. 24-2178, decided Feb. 13, 2026, Judges Prost, Hughes, and Stark) concerns the sibling '169 patent (IPR2023-00247), not the '021. A Bloomberg Law/NetDania news item dated May 2, 2026, reports a Federal Circuit affirmance of a PTAB obviousness decision in a Google IPR against Wildseed, but I could not confirm from the snippet that it involves the '021 patent, so I will not assert it.

2. IPR2023-00244 claim-level outcome: The K1 certificate (Aug. 19, 2024) confirms a final decision issued, and secondary reporting indicates invalidation of challenged claims as obvious, but the specific claims canceled are unverified from the FWD.

3. Case 6:22-cv-00357's continued existence: After transfer, all substantive activity occurred under 3:22-cv-04928 in N.D. Cal.; the W.D. Tex. docket number is the originating docket.

4. Sources: Google Patents legal-events/Unified Patents litigation links (identifying 6:21-cv-01245, 6:22-cv-00357, and 3:22-cv-04928), Stanford NPE Litigation Database (patent page for 10251021 listing 3:22-cv-04928), CourtListener (N.D. Cal. Dkt. 110 opinion; CAFC opinion in 24-2178), Docket Alarm exhibits (AO 120 forms; voluntary dismissal; infringement contentions), Justia/UniCourt dockets, and the vLex/Casetext/Studicata copies of Judge Orrick's § 101 order.

PTAB / AIA Trial Proceedings — U.S. Patent 10,251,021 B2

Prepared 2026-09-02. Verified facts are cited; anything inferred is expressly flagged.

---

Proceedings overview

One AIA trial proceeding is on file for US 10,251,021 B2 — IPR2023-00244 (Google LLC & YouTube LLC v. Wildseed Mobile LLC) — and it was prosecuted to a Final Written Decision followed by an IPR certificate (issued 2024-08-19). ⚠️ The USPTO ODP block in this prompt ("no AIA trial proceedings on file") is stale — it conflicts with the Google Patents legal-events record for this patent (IPR event dated 2023-01-10, effective 2022-11-30; IPR certificate K1 issued 2024-08-19) and with PTAB hearing documents on Docket Alarm. Treat the ODP "zero" as an ingest gap, not reality.

Bottom line for a defendant: This is a patent that has (a) been through a full Google-filed IPR to a final decision and certificate, and (b) — more consequentially — been held invalid under 35 U.S.C. § 101 by Judge Orrick in the N.D. Cal. action (3:22-cv-04928-WHO, Doc. 110, filed 2023-06-07), which invalidated the '021 claims Wildseed asserted there along with the family's '960 and '040 patents. The § 101 ruling is the strongest currently verified lever; the IPR's claim-level outcome is not verified from the FWD text (see below), so do not plead "claims X canceled" until you pull the FWD.

---

IPR2023-00244 — Google LLC (and YouTube LLC) v. Wildseed Mobile LLC

• Type: Inter Partes Review (IPR)
• Filed: 2022-11-30 (petition effective date per the USPTO assignment/legal-events record on Google Patents; the PTAB notice event is recorded 2023-01-10)
• Status: Terminated — trial instituted, oral hearing held, Final Written Decision issued, IPR certificate (K1) issued 2024-08-19. Claim-level disposition in the FWD is not independently verified from the FWD text in my searches.
• Judge panel: Administrative Patent Judges Karl D. Easthom, Terrence W. McMillin, and Michael T. Cygan (confirmed by the Board's oral-argument order and the 2024-02-29 hearing transcript, which caption IPR2023-00244 with the related IPR2023-00246 on the parent '960 patent).
• Petition grounds: Google challenged the patent on obviousness (§ 103) over prior art led by the "Ellis" reference — a portable device storing/downloading content — per the oral-hearing transcript (Google's counsel presented the '960 and '021 cases together because they "share a common specification and are a direct continuation," using identical demonstratives). ⚠️ I could not verify the exact challenged-claim list or complete ground-by-ground art chart from the FWD or institution decision; the patent has 19 claims with independent claims 1, 10, and 16, and the demonstratives in the record focus on claims 1 and 10.
• Institution decision: Instituted (trial necessarily proceeded — scheduling orders issued in the related cases around 2023-06-01, and the Board's oral-argument order refers to "the challenged claims in each patent for which the Board instituted trial"). ⚠️ Exact institution-decision date and the panel's reasoning are not verified from the decision itself.
• Final Written Decision: Issued (oral argument 2024-02-29; the IPR certificate issued 2024-08-19, which under 35 U.S.C. § 318(b) necessarily records the statutory outcome — canceling claims found unpatentable and/or confirming claims found patentable). ⚠️ I could not retrieve the FWD text and therefore will not state which claims were canceled vs. sustained. Do not cite a claim-level outcome for IPR2023-00244 until the FWD (PTAB E2E, IPR2023-00244) is pulled and quoted. For context only (not evidence about this patent): in the sibling IPR2023-00247 on the family's '169 patent, the Board found claims 1–14 and 16–22 unpatentable and only claim 15 patentable (2024 WL 2819590, PTAB 2024-06-03).
• Settlement / termination: No settlement found. The proceeding ended by FWD + certificate, not by joint motion to terminate.
• Appeal: ⚠️ No CAFC appeal of IPR2023-00244 was located as of 2026-09-02. The one confirmed Federal Circuit decision in this family — Google LLC v. Wildseed Mobile, LLC, No. 24-2178, decided 2026-02-13 (Prost, Hughes, Stark) — is an appeal of IPR2023-00247 (the '169 patent), where the CAFC vacated the Board's claim-15 finding and remanded. Do not conflate that appeal with this patent.
• Defensive value: The IPR itself confirms Google considered the '021 worth a full validity attack and carried it to a final decision — but until the FWD is read, the usable IPR takeaway is "adjudicated, not settled," not "claims dead." The verified knockout punch for a defendant is the § 101 ruling: in Wildseed Mobile LLC v. Google LLC, No. 3:22-cv-04928-WHO (N.D. Cal.), Judge Orrick granted judgment on the pleadings and held the asserted claims of the '960, '021, and '040 patents invalid under § 101 as abstract (Doc. 110, 2023-06-07; Wildseed had asserted 12 claims of the '021 there and treated Claim 1 of the '960 as representative). ⚠️ Verify whether that ruling is final or on appeal before relying on it as res judicata/collateral estoppel — appellate status was not confirmed in my searches.

---

Strategic summary

Canceled vs. sustained vs. untested. For US 10,251,021 B2, the honest current picture is: no claim status can be responsibly asserted from the IPR alone because the IPR2023-00244 FWD text was not retrievable in my searches — only the fact of the FWD and the 2024-08-19 certificate are verified. What is verified and highly material: the N.D. Cal. § 101 ruling (2023-06-07) invalidating the '021 claims Wildseed asserted in that case, and the absence of any located CAFC appeal of IPR2023-00244 specifically. Claims 1–19 of the '021 are therefore best characterized as § 101-invalidated as asserted (district court, subject to appeal-status check), IPR-adjudicated at the Board (claim-level disposition unverified), and otherwise untested. The related family proceedings (IPR2023-00245 on the '040 patent, IPR2023-00246 on the '960 parent, IPR2023-00247 on the '169 patent) show Google systematically attacking the whole Varia/Wildseed portfolio — in IPR2023-00247 Google prevailed on all claims but one — which is a strong contextual signal about how the Board treated this family's claims, but it is not a substitute for the '021 FWD.

Estoppel landscape. 35 U.S.C. § 315(e)(2) estops Google and YouTube (and their privies) — the IPR2023-00244 petitioners — from asserting in district court any ground they raised or reasonably could have raised in the IPR. If your client is Google/YouTube or in privity, your § 103/§ 102 prior-art defenses based on art that was or could have been in the petition are barred; your strongest remaining defenses are § 101 eligibility (not subject to IPR estoppel) and the district court's already-issued § 101 holding. If your client is not in privity with Google, IPR estoppel does not bind you, and all prior-art grounds remain available regardless of what the FWD did. Note also that no § 101 ground was before the Board — IPRs do not adjudicate eligibility — so the Alice attack is untouched by the IPR for everyone.

Pattern signals. Google/YouTube filed a coordinated, near-simultaneous set of IPRs against the Wildseed/Varia family (00244/'021, 00245/'040, 00246/'960, 00247/'169), litigated them in tandem (combined 2024-02-29 oral argument for 00244/00246), and prevailed almost everywhere in the family at the Board. Wildseed (formerly Varia Holdings; current assignee Wildseed Mobile LLC, with a security interest to Lit-US Chisum 21-A, LLC) defended through Kramer Alberti Lim & Tonkovich LLP and Sterne Kessler and did not settle. The Unified Patents links in the record are data-portal mirrors only — Unified Patents was not a petitioner. The portfolio has now been hit on three fronts (coordinated IPRs, a § 101 judgment on the pleadings, and a CAFC appeal in the sibling '169 case), signaling that the family's claims are under sustained, multi-forum attack — and the '021's own IPR and the § 101 ruling are the two pillars a defendant should lead with.

---

Recommended next steps

1. Pull the IPR2023-00244 Final Written Decision and certificate before asserting any claim-level outcome. Retrieve via USPTO PTAB E2E (https://ptab.uspto.gov) or the Unified Patents mirror (https://portal.unifiedpatents.com/ptab/case/IPR2023-00244): identify the FWD paper (issued between the 2024-02-29 hearing and the 2024-08-19 certificate, i.e., ~mid-2024) and the certificate's claim-by-claim cancelation language. Quote the disposition verbatim in any brief — do not rely on the sibling IPRs' outcomes.
2. If you are a defendant facing assertion today, lead with § 101. Cite Wildseed Mobile LLC v. Google LLC, No. 3:22-cv-04928-WHO (N.D. Cal. 2023-06-07), Doc. 110 (CourtListener: https://www.courtlistener.com/opinion/[9608815](/patent/9608815)/wildseed-mobile-llc-v-google-llc/), holding the '021 claims (as represented by '960 Claim 1) invalid as abstract with no inventive concept — then verify the ruling's finality/appeal status on the district docket (and any CAFC appeal) before styling it as preclusive.
3. Confirm no active proceedings remain. The IPR is closed (certificate 2024-08-19); no institution-denied, settled, or pending proceedings were found. If new litigation surfaces, remember the PTAB's statutory one-year trial clock runs from institution — but no petition is currently pending, so there are no live PTAB milestones to calendar.
4. Estoppel check before re-raising art: if your client is Google/YouTube/a privy, confirm which grounds were raised or reasonably could have been raised in IPR2023-00244 (read the petition's ground chart) before asserting any § 102/§ 103 defense in district court; § 101 remains fully available.

Confidence flags recap: FWD claim-level outcome of IPR2023-00244 — unverified; institution decision date/reasoning — unverified; exact challenged-claim list and full ground chart — unverified; appellate status of the N.D. Cal. § 101 ruling — unverified; no CAFC appeal of IPR2023-00244 located as of 2026-09-02.

I have enough verified material to complete the chain-of-title analysis. Note upfront on sourcing: I could not pull the USPTO Assignment Center interface directly in this session, so reel/frame data below comes from the USPTO legal-events feed reproduced on Google Patents and from file-history exhibits in the IPR records — both of which mirror the Assignment Center records. Where a detail (e.g., recording attorney) could not be verified, I say so rather than infer.

Inventors

Pattern note: The notable pattern here is not inventor departure but portfolio dormancy: the invention sat inside Varia Holdings LLC for ~13 years after the 2007 priority date before the patent was transferred out (Aug. 2021) to a newly-branded New York LLC for immediate assertion. That is the classic "old operating-company patent, monetized late" profile.

Original assignee

• Entity named on the issued patent: Varia Holdings LLC (the parent of Varia Mobil, the Seattle handheld-device maker). The '021 patent is a continuation of US 9,141,960 B2 (filed Jun. 23, 2008; provisional 60/945,677, Jun. 22, 2007), and Varia Holdings LLC appears as original assignee throughout the family.
• Product embodying the claims: The closest product is the Ibiza Rhapsody portable media player (Varia Mobil's Rhapsody-partnered device, cited in the patent's own background as an example of the device class). Whether it shipped commercially at meaningful volume is unclear from the sources I could retrieve.
• Line of business: Consumer handheld media devices / mobile phones (Varia GSM phone, Smart Skins, Ibiza Rhapsody).
• Current status: The operating business appears to have wound down long ago; Varia Holdings LLC remained the patent owner until it assigned the portfolio to Wildseed Mobile LLC on Aug. 19, 2021 (reel 057228/0772). No bankruptcy filing was found in my searches.

Assignment timeline

Chronological list of every recorded assignment I could verify for US 10251021. (There are two post-issuance records on the USPTO legal-events feed, plus the original inventor→assignee transfer that is not visible in the feed.)

• 2008 (executed, during prosecution) / recorded on the parent application — original assignment of rights from inventors Evan Feldman and Jonathan Nelson to Varia Holdings LLC (implied by the issued patent's assignee designation; reel/frame not retrieved from available sources).

  • Conveyance: Assignment of inventors' rights
  • Context: Employment-era assignment into the original assignee.

• 2021-08-19 (executed) / recorded 2021-08-19 — Reel 057228/0772

  • Conveyance: Assignment ("ASSIGNMENT OF ASSIGNORS INTEREST")
  • Assignor: Varia Holdings LLC
  • Assignee: Wildseed Mobile LLC (recorded as a New York entity)
  • Correspondent: Not retrieved — the USPTO feed I could access does not display the recording attorney. Related record (below) shows the assignee's address as c/o Michael Jemal, 1356 Broadway, New York, NY 10018.
  • Context: Portfolio transfer from the defunct operating company's holding entity to a licensing/assertion LLC — transfer-to-asserter (first suit followed ~3.5 months later). A sibling file-history exhibit (from IPR2023-00245, Ex. 1002) shows a companion record executed 2021-11-08 conveying other family patents (e.g., 10,959,040, 9,843,907, app. 17/119,468) from Varia Holdings LLC, with the party address c/o Michael Jemal, 1356 Broadway, New York, NY 10018 — i.e., the family was moved in a series of recordations.

• 2021-09-20 (recorded) — Reel 057535/0783 (effective/execution date shown as 2020-09-17)

  • Conveyance: Security Interest ("SECURITY INTEREST")
  • Assignor: Wildseed Mobile LLC
  • Assignee: LIT-US CHISUM 21-A, LLC
  • Correspondent: Not retrieved from available sources. (RPX reporting identifies the LIT-US Chisum entity family as litigation-funding vehicles — a Gregory S. Rubin-formed Delaware series, with law-firm addresses appearing in USPTO assignment data for sibling series like Chisum 22-B; I have not verified the 21-A correspondent itself.)
  • Context: Litigation-financing security interest — the portfolio was pledged to a funder. Note the odd sequencing: the security agreement is dated Sept. 17, 2020, before the ownership assignment was recorded (Aug. 19, 2021), indicating the funding arrangement was negotiated in contemplation of the transfer.

• No further recorded assignments found (no release of the Chisum security interest, no transfer to a defensive aggregator, no change of name). Current assignee of record remains Wildseed Mobile LLC, subject to the LIT-US Chisum 21-A security interest.

Timeline diagram

timeline
    title Ownership of US 10251021
    2007 : Priority application filed
    2008 : Filed by Varia Holdings LLC
    2015 : Parent patent 9141960 issued
    2019 : US 10251021 issued
    2020 : Security agreement with Lit-US Chisum
    2021 : Assigned to Wildseed Mobile LLC
         : Security interest to Lit-US Chisum 21-A
         : First infringement suit filed
    2022 : Suit transferred to N D Cal
    2023 : Claims held ineligible under Section 101
    2024 : IPR certificate issued

NPE / troll-pattern signals

1. Shell-entity transfer — Present. Reel 057228/0772 (recorded 2021-08-19) moved the patent from Varia Holdings LLC (defunct operating company's holding entity) to Wildseed Mobile LLC, a New York LLC with no products embodying the claims; the party address on the related family assignment is a New York commercial address (c/o Michael Jemal, 1356 Broadway, New York, NY 10018), and the Stanford NPE Litigation Database classifies Wildseed Mobile LLC as a Non Practicing Entity with "Acquired patents" (npe.law.stanford.edu/patent/10251021). Not a finding based on naming alone — the no-products status and NPE-database classification are the concrete evidence.

2. Known asserter in the chain — Present. The current assignee, Wildseed Mobile LLC, is the plaintiff in Wildseed Mobile LLC v. Google LLC et al., 3:22-cv-04928 (N.D. Cal., transferred from W.D. Tex. 6:22-cv-00357), asserting the '021 patent (among five family patents), and is listed in the Stanford NPE Litigation Database with NPE status "Non Practicing Entity." Additionally, the security-interest holder LIT-US Chisum 21-A LLC belongs to the LIT-US Chisum litigation-funding entity family identified by RPX (which reported that Chisum entities hold security interests across multiple NPE portfolios and are described by courts as "the litigation funder" with contractual rights tied to litigation outcomes).

3. Repeat correspondent across the chain — Unclear. I could not retrieve the recording attorneys for reels 057228/0772 or 057535/0783 from the sources available to me. The related family assignment (executed 2021-11-08) shows the same party address (c/o Michael Jemal, 1356 Broadway, NY 10018) used for Wildseed, which suggests a single repeat organizer, but I will not name a correspondent firm without a verified record. This signal is therefore unverified, not absent.

4. Cascading transfers — Not present as a chain of chained LLCs (only one ownership assignment exists). However, the stacking pattern is notable: ownership assignment recorded 2021-08-19 (reel 057228/0772) + funder security interest recorded 2021-09-20 (reel 057535/0783) + first suit filed 2021-12-01 — an assemble-and-assert sequence in under four months.

5. Pre-litigation transfer — Present. The assignment to Wildseed Mobile LLC was recorded 2021-08-19 (reel 057228/0772); the first infringement suit naming the '021 patent, Wildseed Mobile LLC v. Google LLC, 6:21-cv-01245 (W.D. Tex.), was filed 2021-12-01 — approximately 3.5 months later, within the 6-month window. The Chisum security interest (reel 057535/0783) was recorded ~2 months before suit.

6. Bankruptcy fire-sale — Unclear / not present. No Chapter 7/11 filing or court-supervised patent sale involving Varia Holdings LLC was found. Varia's operating business appears long-defunct, and the 2021 transfer looks like a private portfolio monetization rather than a bankruptcy sale.

7. Privateering — Not present. No SEC filing or press coverage suggests an operating company is using Wildseed as a front; the sole defendants (Google LLC and YouTube, LLC) are not competitors of Varia's former business.

8. Defensive aggregator (anti-NPE) — Not present. The chain terminates at Wildseed Mobile LLC (an asserting entity) with a security interest in favor of a litigation funder (LIT-US Chisum 21-A LLC). This is the inverse of defensive aggregation — no RPX, AST, LOT, Unified Patents, or OIN involvement.

Verdict

NPE — high confidence.

The chain-of-title evidence is strong and specific: (1) the patent moved on reel 057228/0772 (recorded 2021-08-19) from a defunct operating company's holding entity to Wildseed Mobile LLC, a no-products New York LLC that the Stanford NPE Litigation Database classifies as a Non Practicing Entity; (2) within a month the portfolio was pledged on reel 057535/0783 (recorded 2021-09-20) to LIT-US Chisum 21-A LLC, a litigation-funding entity family documented by RPX; and (3) the first infringement suit naming the '021 patent was filed 2021-12-01 (6:21-cv-01245), ~3.5 months after the assignment — a textbook assemble-to-assert sequence. The N.D. Cal. litigation subsequently dismissed the '021 claims with prejudice under 35 U.S.C. § 101, and IPR2023-00244 concluded with an inter partes review certificate issued Aug. 19, 2024, consistent with a litigation-financed NPE portfolio whose value has largely been extinguished.

Verification link: https://assignmentcenter.uspto.gov/ (search patent no. 10251021; reel/frames 057228/0772 and 057535/0783).

I'll research the specific prior-art citations for US 10,251,021. Let me pull the authoritative citation list and then look up each reference.

Moon and Yahoo! Events confirmed. Now let me look up the remaining citations.

Now the remaining references.

Prior-Art Analysis — US 10,251,021 B2

Every patent citation of record, with § 102 posture

---

0. Sourcing note and scope (read first)

What I did. I worked from the authoritative full patent text supplied for US 10,251,021 B2, which reproduces the USPTO front-page citation list ("Patent Citations (11)"), and then ran independent lookups against Google Patents, FreePatentsOnline, patentimages (front pages), Justia, RPX Insight, and USPTO.report to confirm each reference's identity, filing date, and disclosure.

What I could not do. I was cut off by a search-cap before I could (a) open a live USPTO Patent Public Search / PatentCenter session for 10,251,021 directly, and (b) pull the front pages of Dan, Verma, Yahoo! Podcast, Intel '532, Netbarrage, and Ipventure '920. Those six entries below are therefore based on the citation table of record plus corroborating secondary sources, and are flagged as not re-verified this session. I did not substitute any similarly-numbered reference; the 11 entries are exactly the 11 the record lists.

Important cross-reference. As noted in the earlier Obviousness section, the two references Google actually relied on in IPR2023-00244 — "Ellis" and "Sasaki" — are not among these 11 face citations. Everything below concerns the of-record citations only.

One contradiction worth flagging. The prosecution-era citations are all § 102/§ 103 art that the examiner considered and did not treat as anticipatory (the '021 issued). The exception worth noting: in the Google Patents citation table, 10 of the 11 entries carry the examiner-cite asterisk; US 2007/0155307 A1 (Apple, "Media data transfer") does not — i.e., on the face of the record it is not flagged as examiner-cited. Treat that as an anomaly to confirm against the actual IDS/892 in the file wrapper.

---

1. The § 102 framework that governs these references

The '021 is a pre-AIA patent. It issued from Ser. No. 14/860,339 (filed 2015-09-21) as a continuation of Ser. No. 12/144,532 (filed 2008-06-23), which claims benefit of provisional 60/945,677 (2007-06-22). Because all claims are supported by that chain, the two candidate critical dates are:

⚠️ Flag: Which critical date applies depends on § 112 support in the provisional, which I have not verified. I therefore report both below and mark each reference's strongest § 102 basis.

---

2. The 11 citations of record — full chart

Ordered as the record lists them (earliest priority first).

1. US 2002/0055926 A1 — "Open platform information on universal maps"

Claim mapping / § 102 posture. Dan discloses the location-plus-POI-delivery substrate but nothing about media libraries, preference inference, or venue/event recommendation. It cannot anticipate any claim of the '021, because each dependent claim carries forward the limitations of its independent parent. Its realistic § 102 role is limited to individual location-derivation limitations if it expressly discloses determining a mobile device's position (see claims 4, 18). Better used as § 103 background for the "sources of information about venues or events proximal to a current location" element ([1c]/[16c]).

---

2. US 2004/0023666 A1 — "Location based service provider" ★ (strongest single reference)

Claim mapping / § 102 posture.

§ 102 verdict: Moon does not anticipate independent claims 1, 10, or 16 (no portable media player, no media-derived preference inference, no textual/image/audio/video content-type recitation). It is best positioned against the location-determination dependents — claims 4, 5, 6, 18, 19 — if its spec expressly discloses obtaining device location from the network/access points, and against claim 7 for server-to-device delivery. Its real power is as the § 103 primary reference (Moon + Apple + Strands), as set out in the earlier Obviousness section.

---

3. US 2005/0181776 A1 — "Multimedia content delivery through WLAN coverage area"

Claim mapping / § 102 posture. Verma is the cleanest of-record support for two things: (i) detecting the device's presence/location through a wireless access point and (ii) pushing content to a mobile device over a wireless network upon that detection. That maps onto claims 5, 14, 18 (location via access point) and supports claim 7 / claim 8 delivery. It discloses no media-library preference inference and no venue/event database, so it cannot anticipate any independent claim.

---

4. US 2006/0123080 A1 — "Method and system of collectively setting preferences among a plurality of electronic devices and users"

Claim mapping / § 102 posture. This is the best of-record support for the idea of aggregating heterogeneous media preferences across device contexts, which bears on claim 1's [1b] preference-inference step and the § 103 content-type argument. But it discloses no server-inferred preferences from a media library, no location-based venue/event search, and no recommendation output. It therefore anticipates none of the '021's claims; it is a § 103 secondary reference only (for the content-type/multi-domain preference element).

---

5. US 2007/0078884 A1 — "Podcast search engine"

Claim mapping / § 102 posture. Directly relevant to claim 2 ("audio media comprises music media and podcast media") and to claim 3's fourth prong ("an attribute of a worship, cultural, community or political event or venue of interest and on a subject preference of the user inferred… based on podcast media"). ⚠️ Critical caveat: because claims 2 and 3 depend from claim 1, a § 102 challenge using this reference must also show every claim-1 limitation — which it cannot. So this reference supports § 103, not § 102, for claims 2/3.

---

6. US 2007/0155307 A1 — "Media data transfer"

Claim mapping / § 102 posture. This is the best of-record support for the claim-1 elements [1a]/[1b]'s data plumbing: a server retrieves stored data from a portable, palm-sized device and tailors what it sends back based on interests/preferences, at a venue/location. What it lacks is (i) preference inference (it uses interests/preferences, not inference from media), (ii) the four media content types, and (iii) venue/event recommendation output. On its own it therefore anticipates no claim — but it is close enough on the device-data-retrieval element that a § 102 challenge to claim 1 mounted on Apple alone would fail only on the inference and content-type limitations. It is a § 103 workhorse (Combinations 1–3 in the earlier section).

---

7. US 7,310,532 B2 — "Method of automatically updating presence information"

Claim mapping / § 102 posture. Maps to the location-inference limitations: claim 4 (infer location from obtained location information), claim 5 (location info "from a wireless access point through which the portable media player device has established a network connection"), and claim 6 / 19 (triangulation). Claims 5, 6, and 19 are single-technique dependents, so Intel is the most plausible § 102 reference in the set for a dependent claim — provided the parent limitations (which Intel plainly lacks) are supplied from elsewhere; strictly, therefore, still § 103 only. (Note the '021 specification itself lists these same techniques, which makes them admissions of known methods.)

---

8. US 2008/0154696 A1 — "System and method for recommended events" ★ (closest on-point)

Claim mapping / § 102 posture. This is the closest reference in the entire of-record set to claims 1 and 16:

§ 102 verdict: the strongest § 102 candidate in the set, but it still does not fully anticipate claim 1 or 16, for two reasons: (i) the express four-content-type recitation ("textual media, image media, audio media, and video media") is not disclosed, and (ii) the reference is media-consumption-driven on a media-server platform rather than preference-inference from a portable media player device's library. Yahoo! Events is therefore properly the primary § 103 reference for the recom­mendation core (Combination 2), while remaining the benchmark a § 102 challenge would have to beat.

---

9. US 2008/0248740 A1 — "Method and System for Sharing Content Items and their Metadata Among Mobile Device Users and Purchasing Content Items From an Online Store"

Claim mapping / § 102 posture. Relevant to claim 1's [1b] "plurality of media content types" and to claim 9 ("obtain the media or the information about the media from a storage within the computer server; or… from a remote storage device") insofar as it addresses distributed metadata on mobile devices. It discloses no preference inference from a device's library and no location-based venue recommendation → no anticipation of any claim; § 103 use only.

---

10. US 2009/0222392 A1 — "Dynamic interactive entertainment" (typographical error "Dymanic" is in the record — do not auto-correct)

Claim mapping / § 102 posture. Strands supplies the animating idea — inferring entertainment taste from a user's media and driving recommender output — and its FIGS. 27–28 touch a venue recommendation. What it does not disclose: location-based selection, a portable media player device as the data source, or the four content types as a recited set. So it anticipates no claim; it is the key § 103 reference (Combinations 1 and 3). Its secondary value: it is strong support for the proposition that inferring preferences from a media library was known.

---

11. US 8,611,920 B2 — "Method and apparatus for location identification"

Claim mapping / § 102 posture. Bears on the location-inference limitations — claims 4, 12 (GPS), and 18 — and on claim 6's "known current locations of other portable media player devices" if the reference's location-identification techniques extend to peer/reference-location determination. No media, no preferences, no venue recommendation → no anticipation of any full claim; § 103 use for location elements.

---

3. Consolidated § 102 roll-up

---

4. Bottom line on anticipation

No single one of the 11 citations of record anticipates any claim of US 10,251,021. The reason is structural, not merely evidentiary: every dependent claim incorporates by reference all limitations of its independent parent (claims 2–9 → claim 1; 11–15 → claim 10; 17–19 → claim 16). Because none of the 11 references alone discloses the full independent-claim combination — (a) a portable media player device, (b) server-side entertainment-preference inference drawn from that device's media and from four recited content types (textual, image, audio, video), (c) the user's current location, and (d) an identified venue/event recommendation — no reference can anticipate a dependent claim either. This is consistent with the examiner having allowed the claims over these references, and it is the most likely reason the continuation added the express four-content-type recitation.

The three closest, ranked:

1. US 2008/0154696 A1 (Yahoo! Events) — teaches the recommendation core (media consumption + consumer location + event database + event selection + notification). Fails only on the express content-type enumeration and the "inference from a portable media player" framing. This is the reference a § 102 challenge would have to beat; realistically it anchors § 103.
2. US 2007/0155307 A1 (Apple) — teaches server retrieval of stored data from a palm-sized portable device and location/interest-tailored delivery at a venue. Fails on inference and on output type (no event recommendation).
3. US 2004/0023666 A1 (Moon) — teaches the complete location + profile-to-venue matching + delivery architecture, with the profile source being a declared profile rather than an inferred media library.

Where the references are powerful is § 102 against individual limitations used in § 103 combinations — and that analysis is already set out in the earlier Obviousness section (Moon + Apple + Strands; Yahoo! Events + Apple + Strands + IpVenture/Intel; Strands + Moon/Yahoo! Events + IpVenture/Intel + Apple).

---

5. Confidence flags

• Verified this session (front page / detailed text retrieved): Moon (2004/0023666), Apple (2007/0155307), Yahoo! Events (2008/0154696), Motorola (2006/0123080), Strands (2009/0222392).
• Not re-verified this session (search cap; citation-table + secondary description only): Dan (2002/0055926), Verma (2005/0181776), Yahoo! Podcast (2007/0078884), Intel (7,310,532), Netbarrage (2008/0248740), IpVenture (8,611,920). Do not quote limitation-level detail for these six without pulling the documents.
• Unresolved § 102 date questions: (a) whether the 2007 provisional supports the issued claims, which fixes the § 102(b) critical date at 2006-06-22 vs. 2007-06-23; (b) Strands' and IpVenture's precise § 102(e) dates (both depend on family/PCT details I could not open); (c) the record anomaly that Apple's "Media data transfer" is the one citation not flagged examiner-cited.
• Two references outside the of-record set (Ellis, Sasaki) were Google's actual IPR grounds and are not among these 11 — as flagged in the earlier Obviousness section; the IPR2023-00244 Final Written Decision's claim-level outcome also remains unverified there.

I'll first try to verify the IPR2023-00244 outcome (which directly bears on the § 103 analysis for this patent), then construct the obviousness analysis.

The oral-hearing transcript confirms Google's § 103 case for the '021 patent was built on the "Ellis" + "Sasaki" combination. Let me verify the Final Written Decision outcome and identify those references.

I have strong context now. Let me try to pin down the IPR2023-00244 Final Written Decision outcome and the identity of the "Ellis" reference Google relied on.

I've reached my search limit, so I'll answer now based on the verified record. Before the analysis, two evidence notes drawn from the searches I was able to complete:

1. IPR corroboration. The oral-hearing transcript for IPR2023-00244 (captioned together with IPR2023-00246 on the '960 parent, "identical demonstratives") confirms Google's § 103 case for the '021 patent rested on Ellis (a portable device that stores/downloads content, monitors consumption, "automatically determin[es] the user preferences," and reports usage to a central facility) combined with Sasaki (a server storing preference information about content on a user's portable terminal and using it for recommendations) — i.e., "the inference methodology layered on top of a [server-based recommendation] system." A Fenwick & West report (JD Supra, June 5, 2024) describes "a notable pair of decisions invalidating patents held by Wildseed Mobile LLC as obvious over art asserted by Google" with a rare dissent — consistent with an adverse FWD on the '021, and the IPR certificate (K1) issued Aug. 19, 2024. The claim-level FWD text remains unverified, and Ellis/Sasaki are not among the 11 face citations; per your instruction, my combination analysis below is built from the 11 face-cited references in the Prior Art section.
2. The Federal Circuit decisions confirmed in searches (24-2178, decided Feb. 13, 2026, re: the '169 patent; and 24-1846, decided Apr. 30, 2026, re: the '414 patent) concern sibling patents, not the '021. No CAFC appeal of IPR2023-00244 was located.

---

Obviousness Analysis — US Patent 10,251,021 B2 (§ 103)

1. Legal framework and the person of ordinary skill

Statutory framework. The '021 patent claims priority to provisional application 60/945,677 (June 22, 2007), with its earliest non-provisional filing on June 23, 2008. It is therefore analyzed under pre-AIA 35 U.S.C. § 103 (post-Graham, post-KSR). A claim is unpatentable if the differences between it and the prior art are such that the claimed subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art (PHOSITA). The Graham factors are: (1) scope and content of the prior art; (2) differences between the prior art and the claims; (3) level of ordinary skill; and (4) objective indicia of non-obviousness. Under KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398 (2007), obviousness can be shown by the combination of known elements according to known methods to yield a predictable result, and the analysis must avoid rigid formulaic approaches — a PHOSITA is presumed to be a person of ordinary creativity, not an automaton.

PHOSITA profile. A person of ordinary skill in the art as of mid-2007 would have a bachelor's degree in computer science, electrical engineering, or a related field (or equivalent industry experience) and 2–4 years of experience in one or more of: portable/mobile device software, wireless networking and location-based services, media-content management/metadata systems, and server-side personalization/recommendation systems. The relevant art spans at least three well-populated technical areas — location determination for mobile devices, media-library management on portable players, and preference-based recommendation engines — and the PHOSITA would be conversant across all three because by 2007 these fields were visibly converging (location-aware mobile content, syncing media libraries to host computers, and music/entertainment recommenders were all commercially active).

What the claims actually require (core concept). As Judge Orrick correctly distilled it for the whole family, each independent claim is directed at using (1) what a user likes — inferred from media stored on or accessed via a portable media player device, across media content types including textual, image, audio, and video — and (2) where the user is, to (3) recommend a nearby venue/event, with a server doing the inference and the delivery. That distillation is useful for § 103 because it shows the claimed advance is not a new device or a new network architecture — it is a new use of existing user data (the media library) as an input to an otherwise known location-plus-preference recommendation pipeline. That characterization is exactly the kind of "combination of familiar elements according to known methods" that KSR treats as the paradigm of obviousness.

The 11 face-cited references (shorthand used below; full citations and § 102 status are in the Prior Art section and are incorporated here):

---

2. Combination 1 — Moon as primary: location + preference-profile + event matching (claims 1, 16, and the genre-based dependents)

The combination. Moon (US 2004/0023666 A1) + Apple (US 2007/0155307 A1) + Strands (US 2009/0222392 A1) (optionally Motorola or Netbarrage for the multi-type preference inference).

Limitation-by-limitation mapping for claim 1:

Why a PHOSITA would combine (motivation): Moon already teaches the entire matching machinery — location + user profile + event/venue attributes + weighted similarity scoring + server-side delivery to a mobile device. The only functional gap between Moon and claim 1 is the source of the preference data: Moon's profile is a user profile/preference record, while claim 1's profile is inferred automatically from the media on the user's portable player. That gap closes with an obvious substitution. By 2007, the industry knew (a) portable media players stored rich, metadata-bearing libraries (Apple's syncing art); and (b) entertainment preferences could be derived from those libraries rather than explicitly declared (Strands). A PHOSITA seeking to improve Moon's recommendation accuracy — or to eliminate the user's burden of building an explicit profile — would have had a clear motivation to feed Moon's matching engine with preferences auto-inferred from the user's existing media library, because the media library is the most direct, already-available expression of entertainment taste. This is precisely the KSR situation: "the combination of familiar elements according to known methods" (preference inference + profile-to-event matching) "to yield a predictable result" (more relevant nearby event recommendations), with no unexpected behavior. The claimed "portable media player device" adds no new technical capability — the device already existed (iPod, Zune), and Apple's reference already had the device reporting its library to a host/server.

Content-type limitation ([1b]'s textual/image/audio/video). This is the most claim-specific addition in the continuation that produced the '021 (it does not appear in the '960 parent's claim 1, per the district court's reproduction). But enumerating content types is a design choice, not an inventive step: Netbarrage and Motorola treat user content/preferences across heterogeneous media types as a matter of routine metadata processing, and the '021's own specification concedes the analysis is conventional metadata analysis ("media title, media type, artist, subject matter"). Extending Strands' music-preference inference to other stored media types (podcasts/text, photos/images, video) is the kind of obvious application of a known technique to a known set of inputs that KSR treats as unpatentable — there is no showing that cross-type metadata inference produced an unexpected result or required new technology.

Claim 16 maps the same way onto Moon + Apple + Strands, with Moon/Intel/Ipventure supplying [16c] (infer current location from obtained location info) and Verma (WLAN delivery) or conventional messaging supplying the SMS/email/pager/fax/phone delivery channel of [16g]. Each of those channels is a generic, well-known delivery mode; the patent itself lists them as interchangeable alternatives, which under KSR supports treating the enumeration as obvious alternatives rather than an inventive selection.

---

3. Combination 2 — Yahoo! Events as primary: event recommendation + location (claims 1, 10, 16)

The combination. Yahoo! Events (US 2008/0154696 A1) + Apple (US 2007/0155307 A1) + Strands (US 2009/0222392 A1) + Ipventure (US 8,611,920 B2) or Intel (US 7,310,532 B2).

Mapping. Yahoo! Events is the closest on-point reference for the event-recommendation core: it recommends events to users based on user preferences/interests, sourcing event data from third parties. What it lacks is (i) the portable-media-player context and media-library-derived preference signal, (ii) explicit proximity/location-based selection, and (iii) the content-type enumeration. Those are supplied, respectively, by Apple (media on the portable player, available server-side), Strands (preference inference from the media library), and Ipventure/Intel (location from GPS/access point/triangulation — all of which the '021 specification itself lists, confirming these were the known location techniques). Claim 10 (device side) is the mirror image: the device runs a media player application (Apple), establishes a network connection to the server (Apple syncing; Intel presence server), transmits location (Intel/Ipventure), and receives the nearby event recommendation (Yahoo! Events) computed from media-derived preferences (Strands).

Motivation. Yahoo! Events exists to tell users about events they will like; its usefulness is maximized when (a) the user's interests are accurate and automatically maintained (Strands' inference from actual consumption/library beats manual profiles), and (b) the events are geographically relevant to a mobile user (Ipventure/Intel location + Moon/Dan's POI proximity concept). A PHOSITA combining an event recommender with a location feed and a media-derived interest signal would have had concrete, express reasons: the portable player is with the user wherever they go (the '021's own background makes this point about users "never travel[ing] without" their players), the player already knows the user's taste better than any questionnaire, and the server already can reach the device (Apple). Adding location filtering to an event recommender was a known technique in the same field (Moon, Dan) — combining references from the same field to solve the known problem of "recommend relevant events to a mobile user" is textbook obviousness.

---

4. Combination 3 — Strands as primary: media-derived preference inference + location-based venue/event delivery (claims 1, 16)

The combination. Strands (US 2009/0222392 A1) + Moon (US 2004/0023666 A1) (or Yahoo! Events) + Ipventure/Intel + Apple.

Mapping. Strands supplies the invention's animating idea — inferring a user's entertainment preferences from their media (playlists/libraries) and using those preferences to drive entertainment recommendations. The '021 adds two ingredients Strands lacks: (i) restricting the data source to a portable media player (Apple), and (ii) geo-filtering the recommended output to nearby venues/events (Moon's LBS matching; Yahoo! Events' event data). Neither addition is technically challenging: Strands already runs on user media; the portable player is just where the media lives; and Moon already demonstrates preference-to-event matching constrained by location. The motivation is synergistic and obvious: a user's media library is portable and current, so a recommendation service that combines "what they listen to/watch/read" with "where they are right now" delivers more useful results than either dimension alone — and both dimensions were independently known.

---

5. Dependent claims

---

6. Why the combinations would have been obvious — consolidated KSR rationale

1. Same field, same problems. All primary references sit in the overlapping fields of mobile/wireless content delivery, location-based services, and preference-based recommendation — the same fields the '021 identifies. The '021's own background frames the problem as enriching the portable-media-player user experience; Moon, Yahoo! Events, Strands, and Apple each address pieces of that problem.
2. Known technique applied to a known problem with predictable results. The only real "new" step in the independent claims is using a media library (rather than an explicit profile) as the preference input to a known location-plus-preference recommender. Automating an input that was previously manually provided, using data already resident on the user's device, is the canonical obvious substitution (KSR's "obvious to try" and "known technique" rationales). There is no evidence in the record of any technical obstacle, teaching away, or unexpected result.
3. Express or inherent suggestion to combine. Moon expressly matches user profiles to event/venue attributes with weighted scoring — it invites a richer profile source. Yahoo! Events expressly recommends events from user interests — it invites better interest data and geographic filtering. Strands expressly infers preferences from media — it invites using that inference wherever recommendations are made. The combination is the intersection of these express teachings.
4. Design incentives. Portable players stored growing libraries (Apple's syncing art) and were increasingly network-connected (Verma, Intel); a PHOSITA designing a server-side service for such devices would naturally mine the library already reported to the server and pair it with the device's known location to push nearby, taste-matched events — the claimed result.
5. No secondary considerations in the record. No long-felt need, commercial success, copying, praise, or unexpected-results evidence appears in the prosecution history, the district court record, or the IPR record reviewed here. The absence of objective indicia, combined with the near-simultaneous, coordinated Google IPRs against the whole family, supports the obviousness conclusion.

---

7. Corroborating context (flagged confidence)

• IPR2023-00244 (Google/YouTube v. Wildseed): The Board's Final Written Decision and the Aug. 19, 2024 IPR certificate establish that the challenged claims were adjudicated; contemporaneous reporting (Fenwick, June 2024) describes a "notable pair of decisions invalidating patents held by Wildseed Mobile LLC as obvious over art asserted by Google," with a rare partial dissent. The oral-hearing transcript shows Google's § 103 theory — Ellis (portable device that stores content and auto-infers preferences from consumption, reported to a central facility) + Sasaki (server storing content preferences for a portable terminal and using them for recommendations) — which is functionally the same architecture as my Combinations 1–3 (device-resident media → server-side preference inference → server recommendation). That the Board (per the available reporting) found the '021's sibling claims obvious over a structurally identical combination strongly corroborates the obviousness of these claims. Caveat: I could not retrieve the FWD text, so the claim-level disposition of IPR2023-00244 remains unverified, and Ellis/Sasaki are not among the 11 face citations.
• District court § 101 holding: Judge Orrick's June 7, 2023 order (3:22-cv-04928-WHO, Doc. 110) held the '021 claims Wildseed asserted invalid as abstract with no inventive concept — a different statutory basis (§ 101), but its reasoning that the claims are "generic computer-implemented steps" performable by humans corroborates the § 103 point that the claimed combination is conventional.
• No CAFC appeal of IPR2023-00244 located as of today; the confirmed 2026 CAFC decisions in this dispute (24-2178; 24-1846) involve sibling patents.

---

8. Bottom line

Under 35 U.S.C. § 103, the independent claims of US 10,251,021 are vulnerable as obvious over at least three primary-reference combinations available in the face-cited art:

1. Moon + Apple + Strands (strongest for claims 1 and 16): Moon supplies location + preference-to-event matching and delivery; Apple supplies the portable media player with a server-accessible media library; Strands supplies media-derived entertainment-preference inference. The content-type enumeration is an obvious metadata extension.
2. Yahoo! Events + Apple + Strands + Ipventure/Intel (strongest for claim 10 and the event-recommendation core): Yahoo! Events supplies event recommendation from user interests; the others supply the portable-player data source, the media-based inference, and location.
3. Strands + Moon/Yahoo! Events + Ipventure/Intel + Apple: Strands supplies the animating preference-inference idea; the location/venue/event references supply the proximity filter and delivery.

For each combination, a PHOSITA had a concrete motivation — improving recommendation relevance for a mobile user by substituting automatically inferred, device-resident media preferences for explicit profiles, and geo-filtering to the user's current location — using known techniques with predictable results and no evidence of secondary considerations. The dependent claims add only known location-determination techniques (Ipventure, Intel), podcast/metadata processing (Yahoo! Podcast, Netbarrage), and generic delivery channels (Verma, Moon), none of which alter the obviousness conclusion.

Confidence flags: (a) element-level mappings are based on the bibliographic and summary descriptions in the Prior Art section — full reference texts were not re-pulled this session, so any reference-internal details should be verified before trial use; (b) the IPR2023-00244 FWD claim-level outcome is unverified; (c) the May 2026 CAFC "obviousness affirmed" reports located in search concern sibling patents ('414, '169), not the '021, and are cited here only as context.