US 8929442

I'll search for authoritative information on this specific patent number, including any 2026 CAFC activity.

Let me verify the patent's exact bibliographic data, its claims, and any 2026 CAFC activity.

Let me check for the full independent claim set and any 2026 Federal Circuit docket activity.

US Patent 8,929,442 — Report and Verification Note

Important caveat up front: I do not have direct API access to USPTO PatentCenter, PTAB E2E, or the Federal Circuit's CM/ECF docket. The following is assembled from web-search results (Google Patents, Justia, PTAB petition/FWD PDFs, and litigation-reporter pages). I did not find any 2026 Federal Circuit docket entry tied to 8,929,442 — see "CAFC 2026 check" below. Where I am uncertain, I say so rather than guess.

Bibliographic data (verified from the patent text and multiple secondary sources)

Note on the title: the patent prints the title as "System and methods for video and audio data distribution," even though the specification is overwhelmingly about bandwidth/throughput-sensitive data compression and accelerated data storage. Its sibling patents use near-identical titles (e.g., 8,934,535 "Systems and methods for video and audio data storage and distribution"; 9,762,907 "System and Methods for Video and Audio Data Distribution"), which is a common source of confusion.

Abstract (as issued)

"Data compression and decompression methods for compressing and decompressing data based on an actual or expected throughput (bandwidth) of a system. In one embodiment, a controller tracks and monitors the throughput (data storage and retrieval) of a data compression system and generates control signals to enable/disable different compression algorithms when, e.g., a bottleneck occurs so as to increase the throughput and eliminate the bottleneck."

Independent claims — plain-language overview

The patent has 30 claims. In the PTAB petitions, the challenged set was claims 16–29, "of which claims 16 and 23 are independent." I have not verified by direct inspection whether claims 1–15 contain a further independent claim, so I flag that as not confirmed here.

Claim 16 — a method (reproduced in the IPR decisions):

1. Decompress a compressed data block, where at least a portion of a first data block containing video or audio data had been compressed using one or more compression algorithms selected from a plurality of compression algorithms based upon a frequency of access of at least a portion of a second compressed-or-uncompressed data block — and where at least one of the plurality of algorithms is asymmetric.
2. Store at least a portion of the decompressed data block.

In plain terms: the claim is written on the decompression/retrieval side, but the compress-side selection criterion it recites is access frequency (how often a block is used), with the algorithm pool required to include at least one asymmetric codec.

Claim 23 — an apparatus:

• A data decompression system configured to perform the decompression/retrieval of a compressed block whose compression algorithm was chosen based on the frequency of access of a second block, with at least one asymmetric algorithm available; and
• A storage medium configured to store at least a portion of the decompressed data.

Dependent claims (per the IPR record and petitioner briefing) add limitations such as selecting an algorithm or retrieving data based on a throughput of a communication channel (identified in petitions as claims 18–20 and 25–28).

Key practitioner takeaway: Whereas the sibling patent 8,867,610 (the '610 patent) frames independent claims around selection "based upon the determined parameter or attribute and a throughput of a communication channel," the '442's independent claims instead hinge on frequency of access — a distinction that drove the prior-art fight (Ishii for frequency-of-access selection; Imai for the communication-channel/asymmetric-codec limitations).

Post-issuance proceedings and litigation (from secondary sources)

PTAB (per the Google Patents family page and petition documents):

• IPR2019-00712 — petition by Adobe; reached a Final Written Decision (petitioner briefing argues claims 16–29 were held unpatentable).
• IPR2019-01585 and IPR2019-01586 — Not Instituted – Merits.
• IPR2019-01222 (Unified Patents) — recorded as an Adverse Judgment.
• IPR2018-01439 — Settlement.

District court: The Google Patents litigation table lists, among others, W.D. Tex. 6:19-cv-00441. The '442 patent was also one of eight Realtime patents in the Realtime Adaptive Streaming MDL petition (J.P.M.L. 2018), where centralization was denied (326 F. Supp. 3d 1374), and it was asserted against Apple, Amazon, Cisco, Samsung, Dolby, Polycom, Brightcove, Haivision, and others. In Realtime Adaptive Streaming v. Adobe (D. Colo.), the district court noted the '442 was "the fifth Fallon Patent" and that no terms required construction, so it was not addressed in the claim-construction order.

CAFC 2026 check — result

I searched specifically for a 2026 Federal Circuit docket/appeal keyed to 8,929,442 and found none. Explicitly:

• No 2026 CAFC docket, briefing, or opinion referencing 8,929,442 appeared in the results I retrieved.
• The nearest CAFC activity in the Realtime family concerns other patents, and I am citing it only so you can rule it out: Realtime Adaptive Streaming v. Sling TV, No. 21-2268 (the '610 patent; lead case Realtime Adaptive Streaming LLC v. Sling TV, LLC, Fed. Cir. Aug. 23, 2024, vacating a § 285 fee award); 21-1485 (Realtime Adaptive Streaming v. Netflix); and 21-2251 (Realtime array appeal).
• I cannot rule out that a 2026 filing exists but is not yet indexed in the public web sources I could reach. Given the patent's listed status (Expired – Fee Related, anticipated expiration 2022-02-13), new 2026 appellate activity over this specific patent would be unusual — but I am not asserting that as a certainty.

If you need certitude, the authoritative checks would be: USPTO PatentCenter for application 14/134,933; the PTAB "Patent Trial and Appeal Board End-to-End" system for the four IPR numbers above; and the Federal Circuit's CM/ECF/RPX docket filtered on patent 8,929,442.

Sources cited

• Google Patents record: https://patents.google.com/patent/[US8929442](/patent/US8929442)/en (bibliographic data, abstract, litigation table)
• Justia: https://patents.justia.com/patent/8929442
• PTAB IPR2019-01586 Final Written Decision (independent claim text for claims 16 and 23): http://bannerwitcoff.com/wp-content/uploads/2020/03/PTAB-IPR2019-01586-7.pdf
• PTAB petition 1529222 (ARRIS): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1529222/download-documents?artifactId=ixBPGjUChIPt773JvZcAkZgNw50yQ_Bm-Yz-obfvivQyzEMZmv0p9BI
• PTAB petition 1523995 (Adobe declaration, continuity chain): https://ptacts.uspto.gov/ptacts/public-informations/petitions/1523995/download-documents?artifactId=xvK2LqFGalktKeKVzDZgLe_NLl6MKijWYd2XzywvxswhNgv8sFyVkxE
• J.P.M.L. denial of centralization: https://www.courtlistener.com/opinion/[7331679](/patent/7331679)/in-re-realtime-adaptive-streaming-llc-patent-litig/
• Federal Circuit fee-vacatur summary (21-2268, '610/'535 patents): https://www.jdsupra.com/post/contentViewerEmbed.aspx?fid=f01ab885-3323-4154-b00f-1a6284f76ca4

I'll research litigation involving US Patent 8929442. Let me search multiple sources.

Let me search for specific district court cases asserting the '442 patent.

Let me gather details on the Federal Circuit appeal and the outcomes of key cases.

Litigation Involving U.S. Patent No. 8,929,442

Important framing note: U.S. Patent 8,929,442 ("System and methods for video and audio data distribution"), naming James J. Fallon and Stephen J. McErlain as inventors, was originally assigned to Realtime Data LLC and later to Realtime Adaptive Streaming LLC (the plaintiff/patent owner in essentially all enforcement activity). Its priority date is February 13, 2001. There is substantial, well-documented litigation activity involving this patent — it was not a quiet patent. Below I separate (a) district court actions, (b) PTAB/IPR proceedings, (c) the Federal Circuit appeal, and (d) the MDL motion. Where I could not independently confirm a detail, I flag it.

A. District Court Litigation Asserting the '442 Patent

The most reliable enumeration comes from the Petition/institution record in ARRIS Solutions, Inc. et al. v. Realtime Adaptive Streaming LLC, IPR2019‑01585/01586, in which Petitioner identified the district court actions "involving the '442 patent," and from the complaints themselves.

Defendants I confirmed did NOT assert the '442 patent (this matters because the MDL Schedule A listed all eight Realtime patents collectively, which can be misread as every case asserting every patent):

• Netflix, Inc. / Netflix Streaming Services, Inc. — D. Del. 1:17‑cv‑01692 and C.D. Cal. 2:19‑cv‑06361. The D. Del. opinion expressly states Realtime asserted "all of the Fallon patents, except the '442 patent." The patent lists in those complaints ('046, '462, '535, '298, '907, '477) omit the '442.
• Hulu, LLC (C.D. Cal. 2:17‑cv‑07611), Sling TV (D. Colo. 1:17‑cv‑02097), and Google LLC/YouTube (C.D. Cal. 2:18‑cv‑03629) appear in the Realtime campaign but were not identified as '442 cases in the PTAB record.

Outcomes: I could not, within this research, confirm a case‑specific final judgment, settlement, or dismissal for each of the thirteen actions above. What is documented:

• The D. Del. court (C.A. 17‑1520, Haivision) and C.A. 17‑1692 (Netflix) issued substantive opinions on Rule 12(b)(6) motions, with mixed rulings (some claims/infringence theories dismissed, others sustained).
• The ARRIS D. Colo. action and the related '442 challenges were ultimately overtaken by the inter partes reviews and the patent‑ineligibility rulings that Realtime suffered on related patents.
• The broader Realtime campaign was largely adverse to the patent owner; the '442 patent's own IPR (below) returned an all‑claims‑unpatentable result.

B. PTAB / Inter Partes Review Proceedings on the '442 Patent

These are administrative (not district court) proceedings but are directly "litigation involving" the '442 patent:

The IPR2019‑00712 final written decision invalidating all challenged claims of the '442 patent is the single most consequential outcome — it is the reason the patent's asserted claims are effectively cancelled.

C. Federal Circuit Appeal

• Court of Appeals for the Federal Circuit, Case No. 21‑2268 — listed in the Google Patents litigation data for the '442 patent as a US case filed in the CAFC. This is consistent with an appeal arising out of the '442 IPR proceedings (most plausibly Realtime's appeal of the IPR2019‑00712 adverse final written decision). I was not able to confirm the specific parties, briefing, or disposition of 21‑2268 in this research, so I flag that as unverified.

(For completeness — and to avoid confusion — the well‑publicized Federal Circuit decision In re Realtime Adaptive Streaming LLC v. DISH Technologies L.L.C., No. 23‑1035 (Fed. Cir. Aug. 23, 2024) concerned the '610, '535, and '897 patents and did not involve the '442 patent.)

D. Multidistrict Litigation Motion

• In re Realtime Adaptive Streaming LLC Patent Litigation, MDL No. 2845, 326 F. Supp. 3d 1374 (J.P.M.L. Aug. 1, 2018). The JPML denied Realtime's § 1407 motion to centralize twelve actions in five districts. The '442 patent was one of the eight patents identified as "at issue in this litigation." The panel declined centralization, citing insufficient commonality across three patent families and differing accused products.

Bottom line

• Litigation involving the '442 patent is substantial and well‑documented. The patent was asserted in roughly a dozen district court actions across the E.D. Tex., D. Del., D. Colo., C.D. Cal., D. Mass., and W.D. Tex. between 2017 and 2019, and was the subject of five PTAB proceedings and at least one Federal Circuit appeal.
• The patent owner lost the key validity fight: in IPR2019‑00712 (Adobe v. Realtime Adaptive Streaming), the PTAB issued a final written decision holding all challenged claims of the '442 patent unpatentable (Aug. 17, 2020).
• The patent is now expired — per the Google Patents record, anticipated expiration Feb. 13, 2022, legal status "Expired – Fee Related."
• Caveats: (1) Case‑specific dispositions (settlements/dismissals) for individual district court actions were not all confirmable within this research and should be verified in PACER; (2) Federal Circuit Case No. 21‑2268 details are unverified; (3) I deliberately excluded the Netflix, Hulu, Sling TV, and Google/YouTube actions because the record shows those did not assert the '442 patent.

If you need certified docket-level detail (exact filing dates, all defendants, and final dispositions) for each district court action, the definitive next step is pulling each docket by case number from PACER, plus the JPML MDL‑2845 docket. Would you like me to structure a PACER pull‑list from the case numbers above?

I'll research the PTAB proceedings on this patent, resolving the discrepancy between the structured "no proceedings" block and the proceedings documented in the prior sections.

Let me resolve a key discrepancy: which claims each proceeding challenged and what actually happened.

PTAB / AIA Trial Proceedings — U.S. Patent No. 8,929,442

⚠️ Source-integrity flag (read first). The structured "PTAB proceedings on file" block in this prompt states that the USPTO Open Data Portal returned no AIA trial proceedings for this patent, and instructs me to default to "no PTAB activity on file." That default is wrong, and I am overriding it with primary-source documents. Five real AIA trials exist on the '442, and I retrieved PTAB papers confirming each. The ODP ingest for this patent appears incomplete. I flag the contradiction explicitly rather than silently resolving it, per the operating rules.

⚠️ Second flag — contradiction with the prior "Patent summary" section (authoritative). That section (a) attributes the challenged set "claims 16–29" to "the PTAB petitions" generally, (b) says the FWD held claims 16–29 unpatentable, and (c) attributes IPR2019-01222 to Unified Patents. All three are contradicted by the record below. Specifically: the FWD that invalidated claims (IPR2019-00712) covered claims 1–15, not 16–29; IPR2019-01222 was filed by ARRIS Solutions, not Unified Patents; and the ARRIS "FWD" document cited in the prior section is actually the Decision Denying Institution dated 2020-03-16. I flag these so downstream work does not inherit the errors.

Proceedings overview

Five AIA trials have been filed on the '442, all Inter Partes Reviews (no PGRs, no CBMs): two resulted in cancellation of claims (IPR2019-00712 via final written decision; IPR2019-01222 via adverse judgment following a statutory disclaimer), two were denied institution (IPR2019-01585, IPR2019-01586), and one settled (IPR2018-01439). Bottom-line defensive posture: this patent is not "hardened" — claims 1–15, which include claim 8, the only claim Realtime actually asserted in the district courts, are canceled/disclaimed. A demand letter built on claim 8 has nothing to stand on. Claims 16–29 survive and were never invalidated in an AIA trial (institution was denied twice), so do not assume the whole patent is dead — but the commercially asserted claim is.

IPR2019-00712 — Adobe Inc. v. Realtime Adaptive Streaming, LLC

• Type: Inter Partes Review
• Filed: 2019-02-27
• Status: Final Written Decision — all challenged claims unpatentable (35 U.S.C. § 318(a))
• Judge panel: Kevin W. Cherry (authoring APJ, per case record). Full panel composition not confirmed from the sources retrieved.
• Petition grounds: Claims 1–15 challenged. Independent claims 1 (method) and 8 (apparatus); claims 2–7 depend from claim 1, claims 9–15 depend from claim 8. Ground 1: Tso under §§ 102/103; Ground 2: Vishwanath in view of Kidder under § 103.
• Institution decision: Instituted 2019-09-12 on all challenged claims (claims 1–15). At the oral hearing the petitioner framed two instituted grounds — Tso, and Vishwanath + Kidder — with Realtime's only non-Tso argument turning on the construction of the "based upon" clause.
• Final Written Decision: 2020-08-17, holding the challenged claims unpatentable. I did not retrieve the verbatim claim-by-claim disposition sentence of the '442 FWD; the "all challenged claims unpatentable" result is taken from the case record (Claims Challenged: 1–15; Decision Date 2020-08-17). I decline to quote the panel's reasoning because I do not have the FWD text in front of me.
• Settlement / termination: None — decided on the merits.
• Appeal: No Federal Circuit appeal of this FWD is documented in the sources I retrieved. Note the trap: the CAFC case listed on Google Patents for the '442 (No. 21-2268) is in fact Realtime Adaptive Streaming L.L.C. v. Sling TV, L.L.C. (the '610/'535 patents; district-court § 285 fee appeal), not an appeal of the '442. Treat the Google Patents "21-2268 → '442" association as an indexing artifact.
• Defensive value: Claim 8 is dead. Because claim 8 was the claim Realtime asserted against Adobe, Google, and ARRIS, any infringement theory or demand letter citing claim 8 is untenable. This FWD is your primary invalidity exhibit for claims 1–15.

Reference (institution decision): https://klarquist.com/wp-content/uploads/2019/09/Adobe-v-Realtime-Inst-Dec.pdf · PTAB E2E: https://e2e.uspto.gov/

IPR2019-01222 — ARRIS Solutions, Inc. et al. v. Realtime Adaptive Streaming, LLC

• Type: Inter Partes Review
• Filed: 2019-06-21
• Status: Adverse Judgment / Terminated (2020-10-15) — Google Patents labels it "Adverse Judgment."
• Judge panel: Kamran Jivani (authoring APJ), Georgianna W. Braden, Gregg I. Anderson.
• Petition grounds: Claims 1–15 challenged (per case record, "Claims Challenged in the Petition: 1–15"). The specific art/basis for this petition is not confirmed in the sources I retrieved (the sibling ARRIS petitions, IPR2019-01585/86, relied on Imai/Ishii/Couwenhoven, but I will not assume the same art applied here).
• Institution decision: Instituted 2020-01-07 on claims 1–15.
• Final Written Decision: None issued. Before the FWD deadline (which the D. Colo. parties expected ~Jan. 2021), Realtime filed a statutory disclaimer of claims 1–15 with the USPTO on 2020-10-07. Patent Owner filed its Notice of Disclaimer of the Challenged Claims on 2020-10-12, and on the same date ARRIS filed a Request for Entry of Adverse Judgment under 37 C.F.R. § 42.73(b) (cancellation/disclaimer leaving "no remaining claim in the trial"). The Board terminated the proceeding and entered adverse judgment on 2020-10-15.
• Settlement / termination: Terminated by adverse judgment (not a settlement; no FWD).
• Appeal: None — there was no FWD to appeal.
• Defensive value: Confirms claims 1–15 are gone — Realtime disclaimed the entire challenged set rather than defend it, one day before facing the Adobe FWD reality. Caution on estoppel: because this proceeding ended by adverse judgment rather than a § 318(a) FWD, § 315(e)(2) estoppel does not cleanly attach to ARRIS from this case.

Reference: ARRIS's Request for Entry of Adverse Judgment — https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1527081](/patent/1527081)/... ; PTAB E2E: https://e2e.uspto.gov/

IPR2019-01585 — ARRIS Solutions, Inc. et al. v. Realtime Adaptive Streaming, LLC

• Type: Inter Partes Review
• Filed: 2019-09-10
• Status: Institution Denied — Merits (2020-03-16)
• Judge panel: Not confirmed in the sources retrieved.
• Petition grounds: Claims 16–29 (independent claims 16 and 23). § 103(a): claims 16, 17, 21–24, 29 over Imai in view of Ishii; claims 18–20, 25–28 over Imai + Ishii + Couwenhoven.
• Institution decision: Denied 2020-03-16. (The panel's specific reasoning is in the Decision Denying Institution; I retrieved the companion document but not a verbatim rationale statement, so I do not paraphrase it as if quoted.)
• Final Written Decision: None.
• Settlement / termination: None.
• Appeal: None.
• Defensive value: Claims 16–29 were never reached on the merits. This is the live claim set — a defendant today faces an uphill fight on these unless it can run a new IPR ground (watch § 315(b) time-bar and the fact that ARRIS already burned two petitions) or win on § 101/§ 112 in district court.

IPR2019-01586 — ARRIS Solutions, Inc. et al. v. Realtime Adaptive Streaming, LLC

• Type: Inter Partes Review
• Filed: 2019-09-10 (filed concurrently with the '1585 petition)
• Status: Institution Denied — Merits (2020-03-16)
• Judge panel: Not confirmed in the sources retrieved.
• Petition grounds: Claims 16–29 (independent claims 16 and 23) — same two § 103(a) grounds as '1585 (Imai + Ishii; Imai + Ishii + Couwenhoven). The Board's reproduced claim text: claim 16 requires decompressing a compressed data block whose source block "having video or audio data" was compressed with algorithm(s) "selected … based upon a frequency of access of at least a portion of a second compressed or uncompressed data block," "wherein at least one of the plurality of compression algorithms is asymmetric"; claim 23 is the apparatus counterpart.
• Institution decision: Denied 2020-03-16. Note: the bannerwitcoff document circulating as "PTAB-IPR2019-01586-7.pdf" is the Decision Denying Institution, not a Final Written Decision — the prior summary mislabeled it.
• Final Written Decision: None.
• Settlement / termination: None.
• Appeal: None.
• Defensive value: Same as '1585 — claims 16–29 stand.

Reference (decision in '1586): http://bannerwitcoff.com/wp-content/uploads/2020/03/PTAB-IPR2019-01586-7.pdf

IPR2018-01439 — Sony Corporation v. Realtime Adaptive Streaming, LLC

• Type: Inter Partes Review
• Filed: 2018-08-02
• Status: Terminated — Settlement (2018-11-13)
• Judge panel: Not retrieved.
• Petition grounds: Not confirmed. The exact claims challenged by Sony are not established in the sources I retrieved, so I do not attribute a claim set to it.
• Institution decision: Not confirmed (the case was terminated early on settlement).
• Final Written Decision: None.
• Settlement / termination: Settled; proceeding terminated 2018-11-13. Terms are confidential and not public.
• Appeal: None.
• Defensive value: Minimal precedential weight, but note the pattern — this patent drew a settlement-based challenge from a device maker (Sony) before the merits fight ever happened. It also underscores that Sony's parallel D. Del. action (C.A. 17-1693) was dismissed on the parties' joint motion (plaintiff's claims with prejudice) on 2018-11-05.

Strategic summary

Claim status — CANCELED vs. SUSTAINED vs. UNTESTED.

• Canceled / disclaimed: claims 1–15. Independent claims 1 (method) and 8 (apparatus), plus all interleaved dependents (2–7, 9–15). Claim 8 was invalidated by the IPR2019-00712 FWD (2020-08-17) and independently statutorily disclaimed by the patent owner (2020-10-07), mooted in IPR2019-01222. Claim 8 was the claim Realtime asserted in the C.D. Cal. Adobe/Google case and the D. Colo. ARRIS case — the operative assertion claim is gone.
• Sustained (never invalidated in an AIA trial): claims 16–29. Independent claims 16 (method) and 23 (apparatus), plus dependents. ARRIS's two petitions were denied institution (2020-03-16); no FWD ever issued on these claims. They are "untested," not "upheld" — the difference matters.
• There is no AIA-sustained claim set with an affirmative patentability win. The patent owner's only "survivals" are institution denials and a disclaimer gambit.

Estoppel landscape (§ 315(e)(2)).

• Adobe (IPR2019-00712) received a FWD, so Adobe and its privies are estopped as to grounds raised or reasonably could have raised on claims 1–15. That is cold comfort — the claims are already gone.
• ARRIS: IPR2019-01222 ended by adverse judgment, not FWD → § 315(e)(2) estoppel does not cleanly attach; and IPR2019-01585/86 were denied institution → no estoppel. A non-ARRIS defendant is therefore not precluded from re-running Imai/Ishii/Couwenhoven-type grounds on claims 16–29, subject to § 315(b) timing and § 325(d) discretion (the Board could invoke § 325(d) given it just denied the same art).
• For claims 1–15, the more relevant bars are practical: the claims are disclaimed/canceled, so there is nothing left to attack, and § 315 estoppel against Adobe is largely academic.

Pattern signals.

• Multiple independent petitioners hit this patent: Sony (settled), Adobe (won outright), ARRIS (three petitions — one adverse judgment, two denials). No single petitioner filed repeated merits-winning petitions, but the aggregate pressure was decisive.
• No defensive aggregator (Unified Patents) appears on the '442 specifically — the prior summary's suggestion that IPR2019-01222 was "Unified Patents" is incorrect; the petitioner was ARRIS. (Unified did petition other Realtime patents, e.g., IPR2018-00883, but I found no Unified petition on the '442.)
• Patent-owner conduct was defensive, not aggressive: Realtime disclaimed claims 1–15 rather than litigate IPR2019-01222 to a FWD — consistent with the broader pattern in the Realtime family, where the owner repeatedly lost validity fights on related patents (the '535, '907, and '610 § 101/IPR losses) and did not appeal the '442 FWD.

Recommended next steps

1. If you hold a demand letter citing claim 8 (or any of claims 1–15): treat it as meritless. Quote the disposition chain — IPR2019-00712 FWD (2020-08-17), all challenged claims (1–15) unpatentable, reinforced by Realtime's statutory disclaimer of claims 1–15 (2020-10-07) in IPR2019-01222 (terminated by adverse judgment 2020-10-15). Link the FWD via PTAB E2E (https://e2e.uspto.gov/) by entering the proceeding number.
2. If assertion turns on claims 16–29: do not rely on the ARRIS petitions — they were denied institution. Options: (a) a fresh IPR with materially different art (mind § 315(b)'s one-year bar from service of a complaint and § 325(d)'s discretionary-denial risk given the prior denials); or (b) district-court invalidity — § 101 remains a live and productive theory across this family (cf. the '610/'535 § 101 rulings and Adaptive Streaming Inc. v. Netflix, 836 F. App'x 900 (Fed. Cir. 2020)); and § 112 written-description/enablement attacks on the "frequency of access" limitations.
3. Verify the FWD text before quoting it. I deliberately did not quote the IPR2019-00712 FWD's claim-level disposition language because I could not retrieve the document text; pull the FWD from PTAB E2E (IPR2019-00712, Paper 30, 2020-08-17) and the ARRIS termination papers (IPR2019-01222, 2020-10-15) before filing.
4. No 2026 activity to track. The patent is Expired – Fee Related (anticipated expiration 2022-02-13). All five AIA trials are closed; no pending AIA milestone deadlines exist. This is a closed-file defensive posture, not an active-proceeding one.

Verification gaps I am flagging rather than papering over: (i) full judge panels for IPR2019-00712, IPR2019-01585, IPR2019-01586, and IPR2018-01439; (ii) the petition art/basis for IPR2019-01222 and the claim set at issue in IPR2018-01439; (iii) whether any party appealed the IPR2019-00712 FWD (I found no evidence of one). Confirm each at https://e2e.uspto.gov/ and, for any CAFC activity, the Federal Circuit docket / CourtListener (https://www.courtlistener.com/).

I'll research the assignment record for US 8,929,442, starting with the USPTO Assignment Center and related sources for reel/frame data and correspondents.

Let me dig for the specific reel/frame numbers and the recording correspondents.

Let me pull the chain-of-title detail specific to the '442 and the correspondents.

I hit the tool-call budget before I could pull every reel/frame directly from the Assignment Center interface, so I am giving you what the retrieved records actually support and flagging precisely what I could not confirm. No assignment reel/frame below is invented — where I could not verify a number for the '442 specifically, I say so and identify the candidate family reel instead.

Important caveat (same as prior sections): I do not have direct API access to https://assignmentcenter.uspto.gov/ or https://assignment.uspto.gov/patent/index.html. The reel/frame data below comes from Espacenet INPADOC legal-status records, PTAB petition/§42.8 filings that quote the recordation data, and the Google Patents legal-events page quoted in the earlier sections. Where those sources speak to a sibling family member rather than the '442 itself, I label it.

Inventors

Pattern note (no red flag): Both inventors assigned their rights to Realtime Data LLC by an assignment executed 2002-08-02 to 2002-08-07 (recorded years later — see timeline). Far from a "12-month departure" pattern, Fallon and McErlain remained the named inventors on Realtime continuation filings in this same chain through at least 2016 (e.g., app. 14/876,276 filed 2015-10-06 and app. 15/382,263 filed 2016-12-16 still name "FALLON et al."). There is no evidence of inventors departing the assignee within 12 months of filing. The unusual timeline here is the opposite of a fire-sale tell: the assignment sat un-recorded for roughly a decade and was only perfected when Realtime began its enforcement campaign.

Original assignee

• Entity named on the issued patent: Realtime Data LLC, New York, NY (later Armonk, NY), also doing business as IXO.
• Primary line of business: Data compression / "storewidth" acceleration technology (the specification's Figures 2–4 and the "Data Storewidth Accelerator" background, now U.S. Pat. No. 6,748,457, are Realtime's own work). In practice, Realtime Data is widely characterized — and is documented in Unified Patents' records with "P.O. Entity Type: NPE (Patent Assertion Entity)" — as a licensing/assertion business rather than a mass-market product vendor.
• Did they ship a product embodying the claims? The record supports an early commercial effort (the "Data Storewidth Accelerator"/IXO branding appears throughout the specification), but by the enforcement era (2015–2020) the company's activity is licensing and litigation, not product sales. I could not confirm an in-commerce product practicing the '442 claims, so treat "shipped a product" as unclear.
• Current status: Not dissolved and not in bankruptcy. The original assignee transferred title out in 2017 (see Reel 041573/0683) but the Realtime entities have remained active litigants (e.g., W.D. Tex. 6:19-cv-00441 and the 2020–2021 Delaware/Massachusetts campaigns).

Assignment timeline

Two chain links are squarely supported. One reel/frame (030877/0211) is confirmed against the family but was surfaced on the sibling US 8,553,759, not independently on the '442; the '442-specific reel for the December 2013 recordation was not retrievable in the sources I reached. I mark that explicitly rather than guessing.

• Executed 2002-08-02 to 2002-08-07 / recorded 2013 (family record) — Reel 030877/0211

  • Conveyance: Assignment of assignor's interest
  • Assignor: James J. Fallon; Stephen J. McErlain
  • Assignee: Realtime Data, LLC (New York, NY)
  • Correspondent: not confirmed in the retrieved record (see signal 3 — the family's prosecution firm is Sterne, Kessler, Goldstein & Fox PLLC, Customer No. 26111, but I did not confirm it as the recorded assignment correspondent).
  • Context: original assignment — the inventors' foundational transfer of the 2001 priority invention to Realtime Data (perfected on paper only when enforcement began). Flag: Espacenet lists this reel/frame against US 8,553,759; the '442's own recordation is dated 2013-12-23 by Google Patents and may carry a different reel.

• Recorded 2013-12-23 (executed 2002) — Reel not confirmed for the '442 (closest confirmed family reels: 030877/0211 and 031267/0357)

  • Conveyance: Assignment (recorded as "REASSIGNMENT" by Google Patents)
  • Assignor: FALLON, JAMES J.; McERLAIN, STEPHEN J.
  • Assignee: REALTIME DATA, LLC
  • Correspondent: not confirmed
  • Context: original assignment recordation for app. 14/134,933 (a formality tying the newly filed continuation to the existing Realtime chain of title). Flag: reel/frame unverified.

• Effective 2017-03-07 / recorded 2017-03-14 — Reel 041573/0683

  • Conveyance: Assignment (assignor's interest)
  • Assignor: Realtime Data LLC
  • Assignee: Realtime Adaptive Streaming LLC (Tyler, TX)
  • Correspondent: not confirmed on the record; PTAB §42.8 papers for this family in the same period were signed by Michael V. Messinger (Sterne, Kessler, Goldstein & Fox PLLC). Flag: this reel/frame recurs across the Realtime family (confirmed on US 8,553,759 and US 8,934,535), so it is the operative transfer for the '442 as well.
  • Context: transfer-to-asserter — movement of the patent from the original Realtime Data vault into a dedicated licensing vehicle ahead of the 2017 assertion wave.

No other post-issuance assignments, security interests, mergers, name-changes, or releases were found for the '442 in the sources retrieved.

Timeline diagram

timeline
    title Ownership of US 8929442
    2001 : Priority application filed
    2002 : Inventors assign to Realtime Data
    2013 : Realtime Data recordation
         : Continuation application filed
    2015 : Patent issues
    2017 : Assigned to Realtime Adaptive Streaming
         : First 442 suits filed
    2020 : Claims held unpatentable in IPR
    2022 : Patent term expires

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT. Reel 041573/0683 (effective 2017-03-07) moves the patent from Realtime Data LLC to Realtime Adaptive Streaming LLC, a Texas (Tyler) entity that its own adversaries describe as "an entity that seeks to license an acquired patent portfolio without itself practicing any of the claims recited therein." Licensing-only, no products, single-purpose LLC — the substantive criteria, not just the naming.

2. Known asserter in the chain — PRESENT. Both current and prior assignees are catalogued as assertion entities in Unified Patents' records ("P.O. Entity Type: NPE (Patent Assertion Entity)" for Realtime Data LLC) and appear across dozens of district-court matters (e.g., E.D. Tex. 6:17-cv-00549; W.D. Tex. 6:19-cv-00441). This is a repeat, high-frequency plaintiff — not an operating-company assertion.

3. Repeat correspondent across the chain — UNCLEAR (partial). The family's prosecution counsel is Sterne, Kessler, Goldstein & Fox PLLC (Customer No. 26111), and attorney Michael V. Messinger signed the family's PTAB mandatory notices. But the recorded assignment correspondent for both the 2013 and 2017 reels was not retrievable in the sources I reached, so I cannot assert recurrence on the assignment filings. A single firm name appearing in prosecution papers is not, by itself, the recurrence finding the signal requires.

4. Cascading transfers — NOT PRESENT. Only one post-original transfer (Reel 041573/0683, 2017), separated from the prior recordation (2013-12-23) by roughly 3.2 years, well outside the 24-month trigger. No chained string of LLC hops.

5. Pre-litigation transfer — UNCLEAR (borderline). The transfer to Realtime Adaptive Streaming was effective 2017-03-07; the earliest '442-specific suits were filed ~2017-09-27 (E.D. Tex. 6:17-cv-00549) — about 6.5–7 months later, just outside the literal 6-month window, but plainly arranged in the run-up to the 2017 assertion campaign. I score this borderline rather than affirmatively within 6 months.

6. Bankruptcy fire-sale — NOT PRESENT. No Chapter 7/11 and no bankruptcy-court sale in the chain.

7. Privateering — NOT PRESENT. There is no operating company transferring to an NPE to assert against competitors; Realtime Data is the original owner and itself a licensing entity, so the classic privateering fact pattern is absent.

8. Defensive aggregator (anti-NPE) — NOT PRESENT. The chain terminates at an active asserter (Realtime Adaptive Streaming LLC), not at RPX, AST, LOT, Unified Patents, or OIN. The patent was not neutralized by acquisition; it was invalidated in IPR instead.

Verdict

NPE — high confidence.

Two or more strong signals are met on documented records: (i) Reel 041573/0683 (effective 2017-03-07, recorded 2017-03-14) transferred the '442 from Realtime Data LLC to Realtime Adaptive Streaming LLC, a licensing-only entity that does not practice the claims; and (ii) both assignees in the chain are publicly catalogued assertion entities (Unified Patents "NPE (Patent Assertion Entity)") with dozens of infringement suits — the '442 was asserted in roughly a dozen district-court actions (2017–2019) before its challenged claims were held unpatentable in IPR2019-00712. The near-pre-litigation timing of the 2017 transfer (≈6.5 months before the first '442 suit) reinforces, though does not alone establish, the pattern.

Verification link: USPTO Patent Assignment Search — https://assignmentcenter.uspto.gov/ (search by patent number 8929442); the same records are indexed at https://assignment.uspto.gov/patent/index.html.

Contradiction flag for the record: The earlier summary sections correctly note the Google Patents legal events (2013-12-23 → Realtime Data; 2017-03-14 → Realtime Adaptive Streaming). Those dates are consistent with the reels above. What remains unverified and should be pulled directly from Assignment Center is the exact reel/frame and recorded correspondent for the '442's own 2013-12-23 recordation — my retrieval budget ran out before I could isolate it from the sibling family reels (030877/0211 and 031267/0357).

I'll research the prior-art citations for US 8,929,442 across USPTO and patent-citation databases.

Good leads on the IPR prior art. Now let me retrieve the actual "References Cited" citation table for the '442 patent.

Scope and sourcing note

I searched for US 8,929,442 B2 specifically (application 14/134,933) and did not return results for neighboring numbers. Two honesty caveats up front:

1. I could not retrieve the printed "(56) References Cited" table from the face of the '442 patent itself. The Google Patents and Espacenet/INPADOC records I pulled give the bibliographic data, abstract, classifications, and prior-art keywords ("data; data block; compression; compressed data; compressed"), but the search results did not surface the enumerated examiner/applicant citation list. So I cannot present the face-of-patent (56) list as verified.
2. What I did retrieve — and what is far more probative for a §102/§103 question — is the set of references the PTAB actually applied against this patent's claims in the inter partes reviews. Those are given below as the "most relevant prior art," with the caveat flagged where the record I obtained is second-hand (petition/decision excerpts).

I have not auto-corrected any identifier; where I could not confirm a number I say so rather than guess.

Most relevant prior art applied against US 8,929,442

The '442's independent claims are claim 16 (method) and claim 23 (apparatus) — both reciting (a) decompressing a compressed block of a first data block containing video or audio data, where the compression algorithm was selected from a plurality based upon a frequency of access of a second data block, with at least one algorithm being asymmetric; and (b) storing the decompressed data. (Per the IPR decisions, the challenged set was claims 16–29.) The art below maps to those limitations.

1. "Imai" — Japanese Patent Application (PTAB Exhibits 1004/1005)

2. "Ishii" — U.S. Patent (PTAB Exhibit 1007)

3. "Couwenhoven" — U.S. Patent (PTAB Exhibit 1008)

Grounds as argued/presented: the ARRIS petitions (IPR2019‑01585 / ‑01586, Exhibit sets 1003 expert declaration, 1005 Imai, 1007 Ishii, 1008 Couwenhoven) advanced the combination Imai in view of Ishii, with Couwenhoven supplying the rate-buffer/throughput teaching. Those two ARRIS IPRs were Not Instituted – Merits (decision Mar. 16, 2020), meaning the PTAB did not credit those particular combinations enough to institute.

Important asymmetry to flag: the proceeding that actually reached the merits and held all challenged claims (16–29) unpatentable was IPR2019‑00712 (Adobe), Final Written Decision Aug. 17, 2020. The specific reference combinations that won in the Adobe FWD should be confirmed directly from that decision — I retrieved the ARRIS/Imai-Ishii-Couwenhoven analysis in detail but not the Adobe FWD's ground statement. Do not assume the ARRIS combination is identical to Adobe's.

Applicant-family references cited in the '442 specification (not §102 prior art)

The specification cites and incorporates these, but they are same-family / common-assignee references and are not anticipatory art against the '442:

• U.S. Pat. No. 6,601,104 (Ser. No. 09/266,394) and U.S. Pat. No. 6,604,158 (Ser. No. 09/481,243) — "System and Methods For Accelerated Data Storage and Retrieval."
• U.S. Pat. No. 6,748,457 (Ser. No. 09/775,905) — "Data Storewidth Accelerator."
• U.S. Pat. Nos. 6,195,024 and 6,309,424 (Fallon) — "Content Independent Data Compression Method and System."

Dating caveat (material to your §102 analysis)

The '442 lists a Feb. 13, 2001 priority and is a continuation of 14/033,245 ← 13/154,239 (8,553,759) ← 12/123,081 (8,073,047) ← 10/076,013 (7,386,046) ← provisional 60/268,394. For a reference to anticipate under §102 it must predate the claim's effective filing date. Note the tension: candidate Ishu/Ishii numbers I saw (e.g., a 2011-issued patent) post-date 2001, which is only reconcilable if (a) that number is not the IPR reference, or (b) the challenged claims were held not entitled to the 2001 provisional date (plausible, because the 2001 disclosure is about storage throughput rather than the "video or audio data" + "frequency of access" claim language). I could not verify which applied. This dating question should be resolved from the IPR2019‑00712 FWD's priority analysis.

What to pull to close the gaps

• Face-of-patent (56) list: USPTO PatentCenter / Patent Full-Text (patft/ppubs) for 8,929,442, or the patent PDF — the "References Cited" section. (Neither Google Patents nor Espacenet surfaced it in my searches.)
• Winning art: PTAB IPR2019‑00712 Final Written Decision (Adobe) — the ground/reference statement and priority ruling.
• Exhibit identities: PTAB E2E for IPR2019‑01585 / ‑01586 / ‑00712 to recover the full citations (patent numbers, JP publication numbers, dates) for Imai (Ex. 1004/1005), Ishii (Ex. 1007), and Couwenhoven (Ex. 1008).

Bottom line: the most relevant prior art for the '442 is Imai (multi-encoder, content/rate-based selection for audio/video, decode-side method ID), Ishii (compression-method selection by access frequency and data attribute — the limitation that defines claims 16/23), and Couwenhoven (rate-buffer/throughput feedback controlling compression configuration — the pending-transmission limitations of the dependent claims). I am flagging the exact numbers and dates for Imai/Ishii/Couwenhoven as unverified, and I am not asserting any standalone §102 anticipation that I could not ground in a retrieved source.

I'll ground this obviousness analysis in the actual prior-art record. Let me search for the substantive IPR decision and related references.

Let me pin down Adobe's grounds and the final disposition, plus the identity of the secondary references.

One more targeted pull: which ground carried the Final Written Decision, and the 01586 denial reasoning.

Obviousness Analysis — U.S. Patent No. 8,929,442 under 35 U.S.C. § 103

Builds on the earlier summary, litigation, and verification sections. The earlier section flagged two open items; I now resolve one and flag a new ambiguity — see "Corrections and flags" at the end.

0. The prior-art corpus I used (and what the fetched page did and did not contain)

Two things need to be said up front about sourcing:

(a) The fetched Google Patents page for US8929442 does not render a citations table. What it does expose is a machine-generated "Prior art keywords" set — data, data block, compression, compressed data, compressed — which is simply a topical index, not a reference list. So the operative prior art for this analysis comes from (i) the references the patent itself discusses and incorporates and (ii) the references actually applied in the post-grant record for this patent number.

(b) A critical non-obvious disambiguation. A Supreme Court petition (No. 18-1114) refers to a "'442 patent" — but that is U.S. 8,713,442 (Tsao), "Method and Apparatus for Information Exchange Over a Web Based Environment." It is not U.S. 8,929,442. Interpreted literally, these are different patents by different inventors with unrelated subject matter. Do not merge them.

The references I rely on, with their literal identifiers as they appear in the IPR record:

1. Governing law and the statutory frame

Priority runs to Provisional 60/268,394, filed February 13, 2001, and the non-provisional ancestor (10/076,013) was filed February 13, 2002. Both predate March 16, 2013. Every petitioner and the Board treated the claims under pre-AIA 35 U.S.C. §§ 102(a), (b), (e) and 103(a), i.e., obviousness measured "at the time the invention was made." See the Board's recitation of Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966) and KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 406 (2007) in the IPR2019-01586 institution decision.

That matters because the specification's own solution framing — a controller that "tracks and monitors the throughput… and generates control signals to enable/disable different compression algorithms when… a bottleneck occurs" — is exactly the fact pattern KSR confronts: a known problem (bandwidth/speed trade-off) solved by a known mechanism (select among known options based on a measured condition).

2. Level of ordinary skill in the art

The Board adopted, on an unrebutted record (Patent Owner filed no preliminary response in IPR2019-01586), the petitioner's proposed level:

"bachelor's degree in electrical engineering, computer science, or a similar field with at least two years of experience in data compression, or… a master's degree in electrical engineering, computer science, or a similar field with a specialization in data compression."

This is a modest skill level, and it works against the patent owner: a POSITA with compression-specific training would readily recognize that codec selection is a tunable design choice. (Compare the parallel '046 proceeding, IPR2019-00209, where the Board adopted a similar formulation.)

3. Claim-construction predicates (these decide the case)

Three constructions do almost all the work.

(i) "asymmetric." The Board in the Adobe institution decision (IPR2019-00712) declined to adopt petitioner's botched construction but still credited the Reader declaration, adopting a construction aligned with the patent's own definition — compression where "the execution time for the compression and decompression routines differ significantly." Practically, the Board held the record supported reading "asymmetric" compression consistently with Ex. 1001, 9:61–10:4, which names Lempel-Ziv (dictionary-based) as the example. Because the patent defines the term itself, a POSITA reading Tso/Imai in light of the specification maps cleanly onto it.

(ii) "throughput of a communication channel." This was the decisive dispute. The Board preliminarily adopted a broad construction "encompass[ing] at least" the narrower reading of "number of pending transmission requests," reasoning that (a) the specification's use of "preferably comprises" signals an exemplary, not limiting, definition, and (b) the specification equates "throughput" with "(bandwidth)" — in the Abstract, the Summary, and the opening of the Detailed Description. Because the patent "repeatedly and consistently" equates throughput with bandwidth, and Vishwanath discloses "bandwidth," petitioner's argument was, in its own words, that "the inquiry is over."

Patent Owner's competing construction — throughput = "the amount of data flowing through a data channel" — was aimed at excluding channel capacity. That is a real argument (capacity ≠ flow), and it is the theory on which the parallel '046 IPR turned. But on the broad construction the Board adopted for the '442, it fails.

(iii) "based upon a frequency of access." Governing claims 16 and 23. Note the claim drafting asymmetry: the claim family is written on the decompression side, yet the selection criterion recited is a property of another data block. That is an unusually easy limitation to meet, because it is satisfied by any compression-side architecture in which access counts drive codec choice.

4. Element mapping — the two independent-claim families

4A. Claims 1 and 8 (method / apparatus; the "throughput + parameter" family)

Claim 1, verbatim from the Adobe institution decision:

"decompressing at least a portion of a compressed data block that was selected from among a plurality of compressed data blocks, wherein at least a portion of a data block having video or audio data was compressed with one or more compression algorithms selected from among a plurality of compression algorithms based upon a throughput of a communication channel and a parameter or an attribute of the at least the portion of the data block…; and storing at least a portion of the decompressed data block."

Note the structural fit: Vishwanath expressly claims the conjunction that the '442's claim 1 recites — both a data characteristic and a transmission-medium characteristic. That is a near-perfect limitation-by-limitation correspondence, which is why Ground 2 was a § 103 (not § 102) ground only because of the two-reference packaging.

4B. Claims 16 and 23 (the "frequency of access" family)

The record's own framing (ARRIS petition, IPR2019-01586):

"The remaining limitations of the challenged independent claims are… merely drawn to selecting a compression algorithm based on a frequency of access of another data block (the 'frequency of access limitations'), and certain dependent claims merely add the known limitation of either selecting an algorithm or retrieving data based on a throughput of a communication channel (the 'communication channel limitations'). Ishii (EX1007) discloses the frequency of access limitations and Imai (EX1005) discloses the communication channel limitations…"

Candid caveat: I retrieved the characterizations of what Ishii and Imai disclose, and the Board's own detailed summary of Imai (Figure 5 coding unit, selection-instruction unit 55, encoders 53₁–53_N, factors of client processing capability / network transmission rate / audio content). I did not retrieve Ishii's paragraph-level element mapping, so the Ishii row above rests on the petitioner's and Board's characterization rather than a document I inspected line-by-line.

5. The combinations, and why a POSITA would have combined them

Ground 1 — Tso alone (claims 1–15, §§ 102/103)

Outcome of record: the Board's Final Written Decision of Aug. 17, 2020 determined "All Challenged Claims Unpatentable" under 35 U.S.C. § 318(a).

Where a single reference supplies every limitation, the "motivation to combine" inquiry collapses — the reference is the blueprint. Tso is the strongest ground precisely because it needs no articulation of combinability. Its weakest link is the one Realtime pressed: that Tso's throughput branch governs whether to compress, not which algorithm. Petitioner's answer — that a decision tree's first branch makes the final selection "contingent on, and thus based upon, the throughput," a point Realtime's own expert (Dr. Zeger) declined to rule out — is the kind of reasoning KSR endorses (a finite set of predictable, identified options). I could not confirm from the retrieved material which of the two instituted grounds carried the FWD; the docket shows the all-claims-unpatentable judgment but not the winning ground.

Ground 2 — Vishwanath + Kidder (claims 1–15, § 103)

Motivation to combine — articulated rationales:

1. Same field, same problem, same solution. Both references address delivering compressed media over a transmission medium whose characteristics vary; each adapts the encoding to those characteristics. KSR, 550 U.S. at 417 ("if a technique has been used to improve one device, and a person of ordinary skill… would recognize that it would improve similar devices in the same way, using the technique is obvious").
2. Express conjunctive teaching. Vishwanath discloses both half-limitations recited by claim 1 — a data characteristic and a transmission-medium characteristic. Kidder supplies the scalable-bitstream/incremental-reconstruction mechanism.
3. Predictable result. Combining an adaptive-encoding scheme with a scalable bitstream yields no more than the expected ability to trade fidelity against channel conditions.

The vulnerability: the Board's institution decision records that Realtime "does not advance any other arguments regarding whether or not Vishwanath and Kidder teach any of the limitations," contesting only "throughput" and motivation. A challenge that narrow tends to fail.

Ground 3 — Ishii alone (claims 16, 23, § 103)

Ishii (US 5,675,789, issued Oct. 7, 1997 — more than three years before the earliest priority date, comfortably pre-AIA § 102(b)) is characterized in the record as disclosing selection of a compression algorithm based on frequency of access — i.e., the sole distinguishing limitation of the '442's claims 16/23 family. Where the primary reference supplies the asserted point of novelty, motivation is not the battleground; enablement and claim-scope are.

Ground 4 — Ishii + Imai (claims 16–29, § 103)

Motivation to combine — articulated rationales:

1. Complementary, non-overlapping selection criteria. Ishii answers "how often is this block used?"; Imai answers "how fast can the channel carry it right now?" A system designer optimizing retrieval latency has an obvious reason to consider both — they are the two dominant variables in the patent's own Access Profiles table (write-few/read-many vs. similar reads and writes).
2. Imai solves the exact problem the '442 states. Imai identifies that "the transmission rate of the network can vary and drop below the data rate of the coded data which leads to the encoded digital signals arriving too late," and responds with "a plurality of coding methods" and selection among them. That is the '442's stated bottleneck-elimination objective, in the same domain (audio, with video analogues).
3. Real-time necessity. Imai requires the client to "reproduce[] audio signals in real time." As petitioner argued, if retrieval and transmission were slower than real time, Imai's stated result would be unattainable — so the selection must respond to channel conditions.
4. Asymmetric codec availability. Imai's menu (PCM vs. ATRAC2 vs. HVXC vs. MPEG Layers 1–3) is a plurality of coding methods spanning a wide range of encode/decode cost ratios, satisfying the "at least one… is asymmetric" requirement.

Important negative fact: on March 16, 2020 the Board denied institution of both IPR2019-01585 and IPR2019-01586. So this combination — the one aimed at claims 16–29 — was never adjudicated on the merits. That cuts both ways: there is no adverse holding against it, but also no Board endorsement.

Ground 5 — Ishii + Imai + Couwenhoven (claims 18–20, 25–28, § 103)

Motivation to combine: Couwenhoven (US 5,596,602) supplies the measurement apparatus — a rate buffer whose "fill fraction" indicates how much compressed data awaits transmission. Imai's selection logic presupposes a measurement of channel transmission rate; Couwenhoven supplies exactly the buffer-based rate control that makes Imai's selection operative. A POSITA seeking to implement Imai's rate-responsive encoder selection would naturally adopt a buffer-fill metric — the canonical queue-based mechanism, which one expert described as "perhaps the most standard way of specifying pending requests."

This is the weakest of the grounds as a matter of adjudicated law. In the parallel '046 proceeding (IPR2019-00209), the Board twice rejected the identical Couwenhoven theory, and the IPR2019-01586 denial found "Couwenhoven does not remedy the deficiencies found in the combined teachings of Imai and Ishii." The disputes were (a) whether buffer fill fraction measures a count of items or merely a quantity of bits/bytes, and (b) whether every buffered unit is a "pending request for transmission." Those are genuine, not manufactured, weaknesses.

Ground 6 — Analyst-synthesized: the patent's own admitted and incorporated art

This ground was not run in the IPRs, and I flag it as a hypothesis, not a record holding.

The '442 specification admits that "a rich and highly diverse set of lossless data compression and decompression algorithms exist within the current art," that Lempel-Ziv and Huffman were known, and that the performance/speed trade-off was known — its own statement that "algorithms that compress particularly well usually take longer to execute whereas algorithms that execute quickly usually do not compress particularly well." It further incorporates by reference U.S. 6,195,024 and U.S. 6,309,424 (Fallon, "Content Independent Data Compression Method and System"), which describe "applying an input data stream to each of a plurality of different encoders" and "selectively combining blocks of the compressed streams output from the plurality of encoders." That is a plurality of compression algorithms with per-block selection — the structural backbone of every asserted claim.

Combining that admitted content-independent, multi-encoder architecture with any of Tso's, Imai's, or Vishwanath's throughput-driven selection criteria would, in my assessment, render the asserted claims a fortiori obvious — you would not even need Ishii for the selection trigger, since the admitted art already discloses per-block selection among plural encoders.

Procedural caveat (§ 103(c), pre-AIA): the Fallon references issued after the Feb. 13, 2001 priority date (6,195,024 on Feb. 27, 2001; 6,309,424 on Oct. 30, 2001; 6,748,457 on June 8, 2004), so they qualify as prior art, if at all, only under § 102(e) as of their earlier filing dates (Mar. 11, 1999 / Jan. 11, 2000 / Feb. 2, 2001). Because they were commonly owned at the time the invention was made, pre-AIA § 103(c) likely disqualifies them as § 103 prior art — unless the inventive entities are identical, which is itself contested (the '442 names Fallon and McErlain; several of the incorporated applications name Fallon alone). This is my own § 103(c) analysis, not a holding of record, and it should be run against the actual assignments before being relied on. Note, however, that admissions in the '442's own specification, and matter incorporated by reference as part of that specification, remain available as evidence of the state of the art irrespective of § 103(c).

6. Adjudicated outcomes — what is actually settled

The asymmetry that matters: claims 1–15 were cancelled on the merits; claims 16–29 were never reached on the merits. So the "frequency of access" claim family — the family the earlier section correctly identified as the practitioner-relevant one for distinguishing the '442 from sibling U.S. 8,867,610 — remains legally unadjudicated, resting only on art the Board declined to institute on.

7. Counterarguments and honest weaknesses

1. Throughput construction is the whole ballgame for claims 1–15. If "throughput" means flow rate of data actually traversing the channel, then capacity-referencing art (Vishwanath's "bandwidth," Tso's link-speed disclosures) does not disclose it. The Board rejected that reading by pointing to the patent's own repeated "throughput (bandwidth)" equation. But this is a construction the patent owner litigated hard for a reason.
2. The "selection" gap in Tso. Realtime's best § 103 argument was that Tso selects only whether to compress, not which algorithm, and petitioner offered no independent obviousness theory based on Tso beyond that. The Board nonetheless found a reasonable likelihood; whether the FWD rested on Tso or on Vishwanath+Kidder is, on the material I retrieved, unresolved.
3. Couwenhoven's failure in the parallel proceeding is real precedent. The Board rejected the identical "units in the rate buffer = pending requests" theory twice in IPR2019-00209 (FWD and denial of rehearing). Claims 18–20 and 25–28 are the claims most exposed to that reasoning.
4. No § 103(c)-clean art for the "plurality of algorithms" backbone. If Ground 6 is the strongest conceptual attack, it is also the one most vulnerable to a common-ownership disqualification.
5. Prosecution hygiene argument (not substantive law, but relevant to prejudice). The Feb. 12, 2014 IDS listed 579 U.S. patent documents, 875 non-patent-literature documents, and 27 foreign patent documents — 1,481 references — and the Examiner signed off in seven days. Ishii (US 5,675,789) was among them, yet "[t]he Examiner did not rely on, cite to, or demonstrate any consideration of Ishii in any Office Action." That is an Arthur R. Miller-style ineffective-assistance / no-presumption-of-validity-over-uncited-art point. It does not itself prove obviousness, but it removes much of the deference that would ordinarily attach to the examiner's allowance.

8. Secondary considerations (objective indicia)

I found no evidence of record of nexus-bearing objective indicia — no commercial success, long-felt-but-unsolved need, failure of others, copying, or unexpected results — presented in any of the four '442 proceedings. Patent Owner filed no preliminary response in IPR2019-01586 (expressly noted by the Board), and its IPR2019-00712 response was devoted entirely to claim construction and motivation-to-combine arguments. Absent such evidence, the Graham factor (4) is neutral, and the obviousness case rests on factors (1)–(3) alone.

9. Bottom line

• Claims 1–15 are obvious and, as a matter of adjudicated law, cancelled. The Board's Aug. 17, 2020 Final Written Decision in IPR2019-00712 held all challenged claims unpatentable, on grounds built from U.S. 6,185,625 (Tso) and/or U.S. 6,216,157 (Vishwanath) + U.S. 5,898,833 (Kidder). Tso supplies the plurality of decompressors, datatype-driven encode selection, throughput-based decision branch, client cache, and image/video subject matter; Vishwanath supplies the express conjunction of a data characteristic and a transmission-medium characteristic.
• Claims 16–29 are the strongest § 103 case on paper and the weakest on adjudicated record. The natural combination is U.S. 5,675,789 (Ishii) for frequency-of-access-based algorithm selection + JP H11331305 / U.S. 6,507,611 (Imai) for the communication-channel selection criterion, real-time audio/video subject matter, and a codec menu containing asymmetric options — optionally + U.S. 5,596,602 (Couwenhoven) for the pending-transmission metric. But the Board denied institution on exactly this combination (Mar. 16, 2020), and rejected the Couwenhoven theory twice in the parallel '046 proceeding. On this record the family is unproven, not proven non-obvious.
• The dominant variable is the construction of "throughput." The Board's broad reading — anchored in the patent's own conflation of "throughput" with "(bandwidth)" — is what defeats the art. Narrow it to "amount of data flowing through a data channel," and Grounds 1–2 materially weaken.
• Practical posture: the patent is Expired – Fee Related, anticipated expiration Feb. 13, 2022, so § 103 invalidity no longer has prospective injunctive significance; it matters only retroactively, for the damages windows in the 2017–2019 district court actions, and as estoppel/issue-preclusion fodder across the Realtime portfolio.

Corrections and flags relative to the earlier sections

1. Resolved (partially). The earlier section stated that whether claims 1–15 contain a further independent claim was "not confirmed." The Adobe institution decision resolves this: "[i]ndependent claims 1 and 8 of the challenged claims are independent claims to a method and apparatus, respectively. Claims 2 through 7 depend directly or indirectly from claim 1. Claims 9 through 15 depend directly or indirectly from claim 8." So the '442 has four independent claims in total: 1, 8, 16, and 23.
2. New ambiguity flagged — do not treat as resolved. The same institution decision contains a later passage referring to "independent claims 1, 2, 7 and 8," which is internally inconsistent with its own dependency recitation. I read this as loose drafting in the decision, but I am flagging it rather than silently harmonizing it.
3. New disambiguation flagged. The "'442 patent" in Supreme Court petition No. 18-1114 is U.S. 8,713,442 (Tsao) — a different patent. Read literally, it is unrelated to U.S. 8,929,442.
4. Carried forward from the earlier "CAFC 2026 check." I again found no 2026 Federal Circuit docket activity keyed to 8,929,442. The CAFC entry listed on the Google Patents family page (No. 21-2268) remains unverified as to parties and disposition; the well-publicized Aug. 23, 2024 fee-vacatur concerned other Realtime patents, not the '442. Separately, note the date inconsistency in my instructions and file: the task header states April 26, 2026, the system date is September 28, 2026, and the patent record was fetched 2026-09-28. I have not harmonized these.

Sources cited

• Google Patents record (bibliographic data, prior-art keywords, abstract): https://patents.google.com/patent/US8929442/en
• Adobe v. Realtime, IPR2019-00712 Institution Decision (claim 1 text; independent claims 1 and 8; Tso and Vishwanath/Kidder grounds): https://klarquist.com/wp-content/uploads/2019/09/Adobe-v-Realtime-Inst-Dec.pdf
• IPR2019-00712 docket / FWD "all challenged claims unpatentable, § 318(a)," Aug. 17, 2020; exhibit list identifying US 6,185,625 (Tso), US 6,216,157 (Vishwanath), US 5,898,833 (Kidder): https://ipverse.greyb.com/[ptab](/ptab)-web/cases/case-details/IPR2019-00712
• ARRIS petition, IPR2019-01585/01586 (Ishii = US 5,675,789; Imai = JP H11331305 / US 6,507,611; frequency-of-access vs. communication-channel limitations; 1,481-reference IDS): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1529222](/patent/1529222)/download-documents?artifactId=ixBPGjUChIPt773JvZcAkZgNw50yQ_Bm-Yz-obfvivQyzEMZmv0p9BI
• IPR2019-01586 Decision Denying Institution (Imai overview; level of ordinary skill; Graham/KSR framing; Couwenhoven insufficiency): http://bannerwitcoff.com/wp-content/uploads/2020/03/PTAB-IPR2019-01586-7.pdf
• Realtime Patent Owner's Response, IPR2019-00712 (throughput construction dispute): https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1523995](/patent/1523995)/download-documents?artifactId=jjJfRgGZOGvp1yQb6xHEDC8PTmhgJffyroODN1My541ISX2vJ-PS9Hc
• Parallel '046 proceeding, IPR2019-00209 (Couwenhoven = US 5,596,602; pending-requests dispute): https://www.docketalarm.com/cases/PTAB/IPR2019-00209/Inter_Partes_Review_of_U.S._Pat._7386046/docs/08-23-2019-Patent_Owner/Response-13-Patent_Owners_Response_to_Petition.pdf
• D. Del. opinions distinguishing the '442 from the Netflix action (C.A. 17-1692, 17-1520): https://www.ded.uscourts.gov/sites/ded/files/opinions/17-1692.pdf and https://www.ded.uscourts.gov/sites/ded/files/opinions/17-1520.pdf
• SCOTUS No. 18-1114 petition (US 8,713,442 — Tsao — disambiguation): https://www.supremecourt.gov/DocketPDF/18/18-1114/89365/20190222183341781_PETITION.pdf