US 10078623

I'll search for authoritative information on US patent 10078623.

Let me retrieve the actual claims and check for any 2026 CAFC activity.

Let me get the exact claim text from a claims-specific source.

I was able to retrieve substantial bibliographic and descriptive data, but I could not directly query the USPTO PatentCenter/public-search API or the CAFC docket system from my available tools — I relied on web search (Google Patents, Justia, FreePatentsOnline, litigation aggregators). I found no 2026 CAFC (Federal Circuit) appeal for this specific patent number. Details below, with explicit uncertainty flags.

US Patent 10,078,623 B2 — "Hyperlink with graphical cue"

Priority chain (as recited in the patent's "Related Applications"): 15/721,939 is a continuation of Ser. No. 14/224,002 (filed Mar. 24, 2014), which descends from Ser. No. 12/334,068 (filed Dec. 12, 2008; issued as US 8,682,961), which is a continuation of Ser. No. 11/384,957 (filed Mar. 20, 2006; issued as US 7,529,795), which in turn references provisional applications 60/784,141 and 60/784,140 (both filed Mar. 20, 2006).

Abstract (verbatim)

"A system, method, and computer program product are provided for causing display of a first set of representations of a first set of hyperlinks. In operation, a first input from a user is received indicating a selection of one of the first set of hyperlink representations and causes, in response to receiving the first input, display of a second set of representations of a second set of hyperlinks. Further, a second input from the user is received indicating a selection of one of the second set of hyperlink representations, and causes, in response to receiving the second input, navigation to a destination specified by the selected one of the second set of hyperlink representations."

Plain-language overview of the independent claims

The patent is directed to a two-tier hyperlink UI. In plain terms, the independent claims (which appear in method, apparatus/system, and computer-program-product form) cover:

1. First-tier display — causing display of a first set of hyperlink representations (e.g., textual/category links) in a web page.
2. First selection — receiving a first user input selecting one of those first-tier representations (the spec says this can be a click or a hover), and in response, displaying a second set of hyperlink representations (e.g., graphical company/site logos).
3. Second selection — receiving a second user input selecting one of the second-tier representations, and in response, navigating/displaying the destination (the hyperlink's anchor). In the spec's FIG. 5A example, hovering a "Stock Research" category reveals logos such as the bigcharts.com logo, which links to that site.

The specification emphasizes that the second-tier graphical links are pre-downloaded with the page and revealed without a further server access (the spec describes this as achievable via AJAX-style preloading), and that the invention addresses the "visual real-estate" drawback of purely graphical hyperlinks versus text links.

Important caveat on the claims: The full patent text supplied to me was truncated before the printed claim set, and my searches surfaced the pre-grant publication claims of US 2018/0024973 A1 (e.g., apparatus claim 1 with dependents 16, 17, 22; apparatus claim 23; method claims 29, 30) rather than a verified copy of the granted claims of US 10,078,623 B2. Those publication claims recite essentially the steps above plus limitations such as: the second set of representations being initially hidden when the first set is displayed; the representations taking the form of hypertext/textual representations; and the reveal occurring without requiring an additional access to the server. Because claims are frequently amended during prosecution, I cannot confirm the exact granted independent-claim wording or the total claim count from the available material. This should be verified against the USPTO PatentCenter "Claims" tab or the granted B2 PDF before being relied upon.

Litigation / CAFC status

• No 2026 Federal Circuit (CAFC) appeal for US 10,078,623 was found in my searches of 2026 CAFC materials. I cannot rule out an unindexed docket; this is a "not found," not an affirmative "none exists."
• District court: Google Patents lists a case in the U.S. District Court for the Western District of North Carolina, case 3:22-cv-00513 (Unified Patents litigation data). This is a 2022 filing, not 2026.
• The patent had a broad litigation/assignment history involving the "Message Board Aggregator" family and alleged inequitable conduct (a complaint in the record describes assignment chains through Stragent Technologies, Stragent, Aloft Media, Boadin Technology, and Jenam Tech). Treat those allegations as reported claims, not adjudicated facts.
• Expiration: Consistent with the March 20, 2006 priority date, the anticipated expiration is March 20, 2026, and the recorded legal status is "Expired – Lifetime." As of the current date (April 26, 2026), the patent would presumptively have lapsed.

Uncertainty notes

• I did not independently access USPTO PatentCenter or the CAFC docket; findings are based on secondary aggregators (Google Patents, Justia, FreePatentsOnline, Unified Patents, Darts-IP).
• The exact granted independent claim language is unverified (see caveat above).
• One search result returned an unrelated entity using "10078623" as a UK school URN (school URN 120466, UKPRN 10078623) — that is not this patent and was excluded.

If you want, I can next attempt to pull the verbatim granted claims from the B2 PDF (patentimages/Justia) so the independent-claim summary can be quoted exactly rather than paraphrased.

I'll research litigation involving US patent 10078623. Let me run several targeted searches.

Let me dig into the specific litigation lead from the patent record and search for the assignee's known campaigns.

I found a key lead. Let me verify the specific cases and check whether the '623 patent itself was asserted.

I hit my search-step limit before I could exhaustively verify every docket, but I have enough to give you a grounded answer. Here is what I can confirm, with sources and explicit caveats.

Bottom line

One litigation is directly and specifically tied to U.S. Patent No. 10,078,623 ("Hyperlink with graphical cue," Gal Arav, Aloft Media LLC):

• Schaefer Systems International, Inc. v. Aloft Media, LLC, et al. — W.D.N.C., Case No. 3:22-cv-00513, filed September 29, 2022.

The Google Patents record for US10078623B2 carries a "Family has litigation" flag and a specific litigation link to a U.S. case in the North Carolina Western District Court, case 3:22-cv-00513 (Unified Patents litigation portal). That is the same case, and the docket filings include an exhibit of U.S. Patent No. 10,078,623 and "Claim Charts for the '623 Patent."

I did not find any other case in which the '623 patent itself is identified as a patent‑in‑suit. In particular, the well‑known 2008–2010 Aloft Media E.D. Tex. campaign predates the '623 patent's 2018 issuance, so it cannot involve the '623 patent (it asserted earlier family members such as the '795 patent). Recent Aloft Media DJ actions (2025–2026) involve different patents and should not be attributed to the '623 patent.

The case involving US 10,078,623

How the '623 patent fits: The dispute began with a March 4, 2022 demand letter from Aloft (signed by Gordon, identifying George Street Partners/Schmidt as Aloft's agent) to Schaefer, attaching a claim chart and a proposed license. Schaefer's suit challenges the Aloft portfolio patents referenced in that assertion. The docket exhibits include Exhibit 2 — U.S. Patent No. 10,078,623, along with Exhibit 4 — U.S. Patent No. 10,482,164 and Exhibit 5 — U.S. Patent No. 11,308,260, plus separate "Claim Charts for the '164 Patent" and "Claim Charts for the '623 Patent."

Sources:

• Google Patents US10078623B2 (family‑litigation flag → Unified Patents NCWD case 3:22-cv-00513): https://patents.google.com/patent/US10078623B2/en
• Court order (Doc. 79, W.D.N.C. 2023): https://law.justia.com/cases/federal/district-courts/north-carolina/ncwdce/3:2022cv00513/[109822](/patent/109822)/79/ and https://cases.justia.com/federal/district-courts/north-carolina/ncwdce/3:2022cv00513/109822/79/0.pdf
• Docket summary/party list and exhibit list: https://unicourt.com/case/pc-db5-schaefer-systems-international-inc-v-aloft-media-llc-[1303276](/patent/1303276)
• CourtListener opinion: https://www.courtlistener.com/opinion/[9793216](/patent/9793216)/schaefer-systems-international-inc-v-aloft-media-llc/

Important cautions and unresolved points

1. Outcome is not final. I could not verify a final judgment or termination date from PACER. One aggregator lists the case as "Closed" (ai‑lab‑cl‑prod.azurewebsites.net), while another shows "Pending – Other Pending" with entries through late 2023 (UniCourt). Treat the current status as unverified absent a live PACER check of case 3:22-cv-00513.

2. One secondary source is internally inconsistent. An AI‑generated "complaint analysis" page purporting to analyze 3:22-cv-00513 describes the patent‑in‑suit as U.S. Patent No. 10,042,823 and the accused product as Microsoft's MSN Money website. That does not match the Schaefer docket (which involves warehouse/logistics technology, not MSN Money) and appears to conflate separate matters. I did not rely on it, and you should not either. (Note: 10,042,823 and 10,078,623 are different numbers; per your strict rule I have not "corrected" either.)

3. Aloft Media's other litigation is NOT '623 litigation — do not attribute it. For context:

   • Aloft Media, LLC v. Microsoft Corp. et al., 6:08-cv-00051 (E.D. Tex., filed Feb. 22, 2008) — dismissed (final judgment Mar. 16, 2010 as to SAP defendants).
   • Aloft Media LLC v. Nokia Inc. et al., 2:08-cv-00099 (E.D. Tex., filed Mar. 6, 2008) — Nokia dismissed July 13, 2009; Sony Ericsson dismissed Sept. 9, 2009.
   • Aloft Media LLC v. Brookstone, Inc. et al., 2:08-cv-00153 (E.D. Tex., filed Apr. 8, 2008).
   • Aloft Media LLC v. Palm, Inc. et al., 2:08-cv-00292 (E.D. Tex., filed July 29, 2008).
   • Aloft Media, LLC v. Yahoo!, Inc. et al., 6:08-cv-00509 and 6:09-cv-00291 (E.D. Tex.).
   • Aloft Media, LLC v. Oracle Corp. et al., 6:09-cv-00304 (E.D. Tex.).

   These asserted earlier family patents (e.g., U.S. 7,529,795, U.S. 8,682,961, U.S. 8,683,003 — the '795 patent's Stanford NPE database entry shows the Aloft v. Yahoo! case). The '623 patent issued Sept. 18, 2018, so it could not have been asserted in this 2008–2010 campaign.

   • Horace Mann Insurance Co. v. Aloft Media, LLC, 3:25-cv-03262 (C.D. Ill., filed Sept. 5, 2025) — Aloft‑related DJ action.
   • Nouryon USA LLC v. Aloft Media, LLC et al., 1:26-cv-00193 (D. Del., filed Feb. 23, 2026) — patents at issue include 10,372,793, not the '623 patent. Defendants moved to dismiss (May 2026).

   These are recent Aloft Media assertion targets but the patents identified differ from the '623 patent.

4. Search coverage caveat. Google Patents and Unified Patents both flag the '623 patent litigation family as the NCWD case. However, I was unable to complete a full crawl of PACER/CAFC, the Unified Patents case list, or RPX Empower before reaching my tool limit. If Aloft sent demand letters to other targets who filed declaratory judgment actions naming the '623 patent, those could exist beyond what I found. A live PACER/Unified Patents/RPX search on "10,078,623" is the definitive check.

5. No CAFC appeal found. I found no Federal Circuit appeal involving the '623 patent. (The case I did find in search results, Avid Technology v. Harmonic, CAFC No. 15‑1072, concerns U.S. Patent 5,495,291 and is unrelated — noted only to avoid confusion.)

Suggested verification steps

• PACER Docket: W.D.N.C. 3:22-cv-00513 (confirm current status, any final judgment, and whether the '623 patent was formally adjudicated vs. settled/dismissed).
• Unified Patents litigation portal and RPX Empower: search "10,078,623" and "Aloft Media" to catch additional DJ actions.
• USPTO Patent Center / Assignment: confirm any post‑2023 ownership transfers (e.g., to Stragent LLC, which appears as a related assignee in Unified Patents records for the '003/'961 family members).

Primary answer: The only litigation I can specifically tie to US 10,078,623 is Schaefer Systems International, Inc. v. Aloft Media, LLC, et al., 3:22-cv-00513 (W.D.N.C., filed Sept. 29, 2022) — a declaratory judgment / NC APAA action in which the '623 patent appears among the asserted‑portfolio exhibits, with the court denying the defendants' motions to dismiss on June 16, 2023.

I'll start by verifying the structured data against live sources — checking for any AIA trial proceedings, related litigation, and family history.

PTAB Proceedings — US 10078623 ("Hyperlink with graphical cue"), Aloft Media LLC

Canonical source consulted: the structured "PTAB proceedings on file" block (USPTO Open Data Portal). It returns no AIA trial proceedings. I supplemented with web search across PTAB dockets, CourtListener, Justia, Unified Patents, and NPE-focused commentary. No IPR, PGR, or CBM naming US 10078623 or Aloft Media as patent owner surfaced in any source.

Proceedings overview

Total AIA trial proceedings on file: 0 — 0 active, 0 with claims invalidated, 0 with claims sustained, 0 settled, 0 institution denials; the PTAB has never construed a single claim of the '623 patent, which means a defendant today gets no ready-made invalidity judgment and no FWD to lean on, but also inherits no § 315(e)(2) estoppel and no adverse claim-construction record — the practical exposure on this patent is defined by its litigation and prosecution history (and its term), not by any Board outcome.

Because there are no proceedings, the per-proceeding template is not populated below. I have not invented proceeding numbers, petitioner names, panels, or dispositions.

No proceedings identified — verification trail

• Structured data (canonical): USPTO ODP returns zero AIA trial proceedings for US 10078623 as of the most recent ingest.
• Independent search: queries targeting "Aloft Media" IPR, "10078623" inter partes review, US10078623 PTAB, and Aloft Media + the sibling '793/'823 patents returned no petition, institution decision, FWD, or appeal naming this patent. The IPR hits I did surface (IPR2022-01060/01061, IPR2023-00559/00561 on the Jawbone '691 patent; the Netlist/Longhorn IPR clusters) are unrelated patent owners and unrelated patents.
• PTAB E2E / Patent Trial and Appeal Board End-to-End: https://ptacts.uspto.gov/ptacts/ — no trial record tied to the '623 patent was reachable through search.
• Caveat: I hit a search-step ceiling before running the final round of confirmatory queries (sibling-patent IPR check, Nouryon/Krueger DJ check, PTAB E2E direct). The finding of zero proceedings is consistent across the canonical ODP data and every search that did run, but it should be treated as high-confidence, not certifiable, until PTAB E2E and Docket Alarm are checked directly by the defendant.

Related non-PTAB activity (context, not AIA trials)

Strategic summary

Claim status — all claims UNTESTED at the PTAB. No claim of US 10078623 — independent or dependent — has been canceled, confirmed, or construed by the Board. Nothing has been narrowed through IPR. If the assertion you are facing cites claims 1–20 (or any subset), the claim set is exactly as issued on 2018-09-18, except that it has not been hardened by a win either — an untested claim set is not the same as a validated one. The only claim-level narrowing available is whatever the district courts and the examiner did, and the relevant examiner-level story is laid out in the DJ complaints (which allege an inequitable-conduct / failure-to-disclose theory running through the continuation chain from the 2006 '957 application through the '068, '002, '487, and '044 applications). Treat those allegations as unadjudicated — a complaint is not a finding.

Estoppel landscape — wide open. Because no IPR/PGR was ever instituted, § 315(e)(2) estoppel is a non-issue: there is no petitioner, no privy, and no "ground raised or reasonably could have been raised" bar. A defendant today may file an IPR on any § 102/§ 103 ground, using any prior art, including art that was before the examiner — subject only to the Board's § 325(d) discretion. There is also no § 315(b) one-year bar running against you unless you have already been served with a complaint alleging infringement of this patent. Note the clock asymmetry: if you have been sued on the '623, your § 315(b) window may already be spending down, and there is no earlier petitioner's estoppel to piggyback on.

Pattern signals — this is an NPE campaign, not a PTAB war. Aloft Media (managed by George Andrew Gordon; described by Unified as an NPE and an entity of Oso IP, LLC) runs a serial-continuation family off a 2006 priority date: '823, '623, '793, '164, '260, and '874 are all continuations sharing essentially the same disclosure. The enforcement pattern is demand letters and form license agreements with claim charts (the Schaefer record: a 2022-03-04 letter addressed to the North Carolina entity, followed by at least ten more contacts) answered by declaratory-judgment suits, not by IPR petitions. The absence of PTAB activity against a patent asserted in that many letters is itself informative: targets have overwhelmingly chosen DJ actions, abusive-assertion statutes, and jurisdictional fights over Board proceedings — plausibly because the patents are close to, or past, the end of their term and because validity is not the only lever (unenforceability and state-APAA exposure are in play). Unified Patents is working the siblings, not this patent.

Term — the elephant in the room. The structured Google Patents data states legal status "Expired - Lifetime," with anticipated expiration 2026-03-20 (20 years from the 2006-03-20 priority date claimed through the continuation chain). As of 2026-09-27, that date has passed. If the term calculation holds (i.e., no material PTA extension), there is no prospective injunctive relief on the '623 and damages are limited to past infringement inside the § 286 six-year lookback — i.e., acts on or after 2020-09-27 — subject to any laches/notice arguments. Flag: this is a Google Patents assumption field and a priority-date calculation, not a USPTO term determination. Verify the actual expiration (including any PTA) against Patent Center before you build a defense on it.

Recommended next steps

1. Confirm the zero-proceeding finding directly, and do it once, in writing. Pull PTAB E2E (https://ptacts.uspto.gov/ptacts/) and Docket Alarm for both this patent number and "Aloft Media" as patent owner, plus the sibling patents ('823, '793, '164, '260, '874). My search returned nothing, but I did not exhaustively clear the docket. If it holds, you have a clean, documented "no PTAB history" record you can put in front of the client and, if useful, the court.
2. Run the expiration math first — it may moot the whole engagement. Order the file history / Patent Center term calculation for US 10078623 and confirm whether 2026-03-20 is the true expiration and whether any PTA pushes it later. If the patent has expired, your client's exposure collapses to past damages only, six-year lookback (acts from 2020-09-27 forward), and any demand letter dated after expiration is a much weaker instrument.
3. If a § 315(b) clock is running on you, decide fast. With no prior petitioner, whoever you are is the first mover — you get no estoppel benefit and no free FWD, but you also get the entire prior-art universe. The DJ complaints identify a specific pool of references that Aloft allegedly knew of and did not disclose (Benson US 7,913,185; Kavner US 6,366,947; Solotko US 2008/0103889; Buchmiller US 2003/0187706; Huynh US 2002/0198909; Cohen US 2003/0184582; Thakur US 7,895,644; Soroca US 2010/0094878; Weber US 2008/0052372; Shuster US 2010/0287368; plus a Microsoft FrontPage "JavaScript Drop-Down Menu" NPL reference). Those are the first references to chart — and if they were never before the examiner on the '623, they come with no § 325(d) baggage. If you are a defendant already sued, you can raise these in district court without any § 315(e) estoppel risk, since no IPR has ever been instituted against this patent.
4. Coordinate with the sibling-patent fights. Unified Patents' PATROLL contest and the DJ complaints by Schaefer, Brotherhood Mutual, Nouryon USA, and Krueger International target the same family and the same disclosure. Prior-art charts and inequitable-conduct evidence developed there are transferable to the '623 — the specifications are portions of one another. Consider pooling rather than duplicating.
5. If no PTAB activity exists and you rely on that as a signal — qualify it. The absence is meaningful (well-asserted, high-value patents usually attract IPRs), but here it most likely reflects expired/near-expired term plus a preference for DJ/APAA strategies, not a judgment that the claims are strong. Do not read the empty PTAB docket as validation of validity. The claims are simply untested — and, on the current term calculation, may soon be unenforceable in practice.

Confidence and sourcing notes

• High confidence: zero PTAB proceedings on US 10078623 per the canonical ODP block and every web query that completed.
• Medium confidence: patent expiration 2026-03-20 — derived from the Google Patents "anticipated expiration" field and the 2006-03-20 priority date; verify against Patent Center, including PTA.
• Unadjudicated allegations, not findings: the inequitable-conduct and detection-shifting theories and the reference lists in the DJ complaints. They are pleadings.
• Not verified (search ceiling reached): direct PTAB E2E lookup; whether any joinder-target proceeding exists; final disposition/docket of Schaefer 3:22-cv-00513 and Brotherhood Mutual 1:2025-cv-00273.
• No fabricated proceeding numbers appear in this report. If the verification steps in Next Steps 1–2 return a live IPR/PGR, this report must be regenerated with the actual proceeding number, panel, grounds, institution decision, FWD claim-level disposition, and any CAFC appeal.

I'll research the assignment record, litigation, and ownership history for US 10078623 before reconstructing the chain.

Strong leads: Aloft Media is identified as an NPE/Oso IP entity, and there's an Instantbull → Strategent → Aloft chain plus Stragent in the family. Let me dig into the specific assignment records and litigation.

Assignment & Ownership Analysis — US 10,078,623 B2 ("Hyperlink with graphical cue")

Scope check / provenance note: The USPTO Assignment Center (https://assignmentcenter.uspto.gov/, mirrored at https://assignment.uspto.gov/patent/index.html) is a JavaScript query application; my retrieval tools returned no indexed record text from it, and no source in the results reproduced reel/frame numbers or the recorded correspondent-of-record field for this patent. What follows is therefore reconstructed from (a) the patent's own front matter, (b) Google Patents legal events, (c) the E.D. Tex. pleadings in the Aloft Media campaign, (d) the Brotherhood declaratory-judgment complaint (saved at iniplaw.org) which tabulates this family's assignment history, and (e) Unified Patents / Stanford NPE database entries. Reel/frame and recorded correspondent values are marked as unverified rather than guessed. Verify at the Assignment Center link above before relying on any date here.

Inventors

• Only one inventor is named on US 10,078,623. This is a single-inventor family, which makes the entity chain — not inventorship — the entire ownership story.
• Pattern flag: the inventor's operating company divested the family to a patent-holding vehicle early. Filing 2006‑03‑20 → assignment to Stragent Technologies, LLC on 2008‑11‑11 (~32 months post-filing, one month after the intervening continuation was filed). The inventor does not appear as an assignor on any later link; no inventor-retention interest is recorded. This is the classic "startup monetizes/divests its IP before wind-down" shape, not an operating-company retention.

Original assignee

Aloft Media, LLC — a Texas limited liability company, principal place of business 211 W. Tyler Street, Suite C‑1, Longview, Texas 75601 (address admitted by Aloft in its own E.D. Tex. pleadings). Google Patents lists Aloft Media LLC as both original assignee/applicant (published application US2018/0024973 A1) and current assignee.

• Primary line of business: patent licensing/assertion. Aloft Media does not ship a product embodying the claims. Nothing in the record shows Aloft (or its successors) making, using, or selling anything; its only visible activity is filing infringement suits.
• Status: operating as an asserting entity. Unified Patents describes Aloft Media, LLC as "an NPE and entity of Oso IP, LLC" (PATROLL contest on sibling patent 10,372,793).
• Upstream provenance: the invention came from Instantbull, LLC (operating company, ~2006), which divested the family to Stragent Technologies, LLC in 2008. Instantbull's current corporate status is not documented in the sources reviewed; no bankruptcy filing was found.

Assignment timeline

Reel/frame: not retrieved. The entries below are the recorded assignments of this patent family as tabulated in the Brotherhood DJ complaint (which draws on the USPTO record) and Google Patents legal events. Execution dates are as recorded there; recording dates, reel/frame, and correspondent-of-record were not independently verifiable with the tools available. The complaint's table expressly covers applications 11/384,957 (→7,529,795), 12/334,068 (→8,682,961) and 14/224,002 (→10,042,823); US 10,078,623 (App. 15/721,939, filed 2017‑10‑01) is a continuation of 14/224,002 and therefore sits inside the same chain.

• 2008‑11‑11 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Instantbull, LLC
  • Assignee: Stragent Technologies, LLC
  • Correspondent: not retrieved. (Attorney associated with Strategent on prosecution: Kevin J. Zilka, who is also named President of Strategent Technologies, LLC and President of Strategent, LLC — see Signal 3.)
  • Context: initial divestiture by the operating startup to a patent-holding entity.

• 2008‑12‑31 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Stragent Technologies, LLC
  • Assignee: Stragent, LLC
  • Correspondent: not retrieved (Kevin J. Zilka is the recurring attorney/principal of record in this chain — flagged here).
  • Context: internal reorg within the same Stragent control group; 50 days after the prior link.

• 2009‑04‑21 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Stragent, LLC
  • Assignee: Aloft Media, LLC
  • Correspondent: not retrieved. Patrick E. Caldwell is the named attorney for Aloft Media, LLC, Strategent Technologies, LLC and Boadin Technology, LLC in the family litigation — flag as repeat counsel across the chain.
  • Context: transfer to the asserter that fires the 2008–09 E.D. Tex. campaign (Aloft Media v. Palm / Yahoo! / AOL / Google / Oracle).

• 2009‑08‑21 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Stragent, LLC
  • Assignee: Aloft Media, LLC
  • Correspondent: not retrieved.
  • Context: second, closely spaced Aloft acquisition covering the '795 line.

• 2011‑11‑12 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Aloft Media, LLC
  • Assignee: Boadin Technology, LLC
  • Correspondent: not retrieved. (Patrick E. Caldwell again named as Boadin's counsel in litigation.)
  • Context: onward transfer to another anonymous licensing vehicle.

• 2014‑11‑25 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Boadin Technology, LLC
  • Assignee: Stragent, LLC
  • Correspondent: not retrieved.
  • Context: assets cycled back to Stragent — a re-transfer between entities already in the same control group.

• 2016‑12‑02 (executed) / recorded date unverified — Reel not retrieved / not retrieved

  • Conveyance: Assignment
  • Assignor: Stragent, LLC
  • Assignee: Jenam Tech, LLC
  • Correspondent: not retrieved.
  • Context: further cascading transfer to another shell-style LLC.

• 2017‑10‑01 — not an assignment: continuation application 15/721,939 filed naming Aloft Media, LLC as applicant, issuing 2018‑09‑18 as US 10,078,623. The 2017 applicant identity (Aloft) is inconsistent with the last recorded link on the family table (Jenam Tech, 2016), indicating further transfers that are not documented in the sources reviewed. Treat this gap as an open item, not a finding.

• 2022 — Aloft Media case filed in N.C. Western District, 3:22‑cv‑00513 (Google Patents litigation record for this patent). Parties and assignment chronology relative to this filing not verified.

• 2026‑03‑20 — Patent expired (20-year term from the 2006‑03‑20 priority date). As of today the '623 is expired.

Timeline diagram

timeline
    title Ownership of US 10078623
    2006 : Priority application filed
         : Inventor Gal Arav at Instantbull LLC
    2008 : Assigned to Stragent Technologies LLC
         : Assigned to Stragent LLC
    2009 : Assigned to Aloft Media LLC
    2011 : Assigned to Boadin Technology LLC
    2014 : Assigned to Stragent LLC
    2016 : Assigned to Jenam Tech LLC
    2017 : Continuation filed by Aloft Media
    2018 : US 10078623 issued
    2022 : Suit filed in North Carolina
    2026 : Patent expired

NPE / troll-pattern signals

1. Shell-entity transfer — PRESENT.
The patent moved from an operating company (Instantbull, LLC) through a daisy-chain of licensing vehicles: Stragent Technologies, LLC (2008‑11‑11) → Stragent, LLC (2008‑12‑31) → Aloft Media, LLC (2009‑04‑21 / 2009‑08‑21) → Boadin Technology, LLC (2011‑11‑12) → Stragent, LLC (2014‑11‑25) → Jenam Tech, LLC (2016‑12‑02). Concrete evidence beyond naming: Aloft's own pleading admits a suite address (211 W. Tyler St., Suite C‑1, Longview, TX); no assignee in the chain is documented as manufacturing or selling anything; and Unified Patents independently labels Aloft Media "an NPE and entity of Oso IP, LLC."

2. Known asserter in the chain — PRESENT.
Aloft Media LLC appears in the Stanford NPE Litigation Database as an asserter, category "5 Individual-inventor-started," with Aloft Media LLC v. Palm, Inc. et al, 2:08‑cv‑00292 (E.D. Tex., filed 2008‑07‑29) against Palm, Nokia, RIM, Motorola, Sony Ericsson, and AT&T. Aloft also brought 6:08‑cv‑255 (Yahoo!/AT&T/AOL), 6:08‑cv‑440 (Google), and 6:09‑cv‑304 (Oracle et al.). Stragent, LLC is a long-recognized high-frequency patent plaintiff. This is a listed-entity match, not a naming inference.

3. Repeat correspondent across the chain — PRESENT (attorney level), reel/frame UNVERIFIED.
The family litigation identifies a small, recurring cast controlling every link: Kevin J. Zilka — counsel and President of Strategent Technologies, LLC and President of Strategent, LLC; and Patrick E. Caldwell — counsel for Aloft Media, LLC, Strategent Technologies, LLC, and Boadin Technology, LLC. A third name, Robert Poltkin, is the filer of the '957 application. The signal is the recurrence across multiple links and multiple differently named LLCs (Zilka spans the Stragent side; Caldwell spans Aloft/Strategent/Boadin). Caveat: I could not confirm these as the USPTO recorded correspondent-of-record, nor their firms/addresses, because the Assignment Center records were not retrievable.

4. Cascading transfers — PRESENT.
Six recorded transfers between 2008 and 2016 across five differently named assignees, two pairs inside 24 months: 2008‑11‑11 → 2008‑12‑31 (50 days) and 2009‑04‑21 → 2009‑08‑21 (4 months). The 2014 transfer moves assets back to Stragent, which is itself evidence of a common control group cycling paper rather than arm's-length sales.

5. Pre-litigation transfer — UNCLEAR for this patent.
The family's first Aloft suits (July 2008) actually predate the 2009‑04‑21 and 2009‑08‑21 Aloft assignments, so no clean "assignment within 6 months before suit" can be shown from this record. For US 10,078,623 specifically, the only known case (N.C. W.D. 3:22‑cv‑00513, 2022) postdates the last recorded link (2016‑12‑02) by roughly five years, and no intervening recorded assignment was found. Call: unclear.

6. Bankruptcy fire-sale — NOT PRESENT on the record reviewed.
No Chapter 7/11 filing or court-supervised patent sale was found for Instantbull, LLC or any assignor in the chain. Instantbull's public operation appears to have ceased, but "ceasing operations" ≠ "bankruptcy sale," and I do not have the filing to cite.

7. Privateering — NOT PRESENT.
No operating-company sponsor is documented on whose behalf Aloft/Oso IP asserted. The asserter is an independent monetization entity, not a proxy for a product company.

8. Defensive aggregator — NOT PRESENT.
The chain does not terminate at RPX, AST, LOT Network, Unified Patents, or OIN. The opposite: Unified Patents is adverse to this patent family, running a $2,000 PATROLL prior-art contest on sibling US 10,372,793 (Mar 27 – May 3, 2026), which is the inverse of neutralization.

Verdict

NPE — high confidence.

At least three independent strong signals converge. (i) Shell-entity transfer: an operating startup, Instantbull, LLC, divested the family to a chain of non-practicing licensing vehicles — Stragent Technologies, LLC (2008‑11‑11) → Stragent, LLC (2008‑12‑31) → Aloft Media, LLC (2009‑04‑21 / 2009‑08‑21) → Boadin Technology, LLC (2011‑11‑12) → Stragent, LLC (2014‑11‑25) → Jenam Tech, LLC (2016‑12‑02) — with Unified Patents independently calling Aloft Media "an NPE and entity of Oso IP, LLC." (ii) Known asserter in the chain: Aloft Media LLC is indexed in the Stanford NPE Litigation Database as an Individual-inventor-started asserter, having filed 2:08‑cv‑00292 (E.D. Tex., 2008‑07‑29) and companion suits against Google, Yahoo!, AT&T, AOL, and Oracle. (iii) Cascading transfers with a repeat cast: six transfers in eight years, two pairs inside 24 months, all with Kevin J. Zilka and Patrick E. Caldwell recurring as counsel/principals across the differently named LLCs.

Two practical caveats for a licensing/deal file: (a) US 10,078,623 expired 2026‑03‑20 (20-year term from the 2006‑03‑20 priority date), so the asset has no forward assertion life — only historic damages exposure survives; and (b) the family is under active invalidity attack on a sibling (PATROLL on US 10,372,793) and under inequitable-conduct/unenforceability attack in the Brotherhood DJ action, which is a material risk to any residual value.

Verify: USPTO Assignment Center (search patent 10078623) · Google Patents US10078623B2 · Stanford NPE Litigation Database — Aloft Media LLC · Unified Patents PATROLL notice

I'll research the prior-art citations for US 10,078,623 B2. Let me pull from multiple sources.

Let me pull the actual "References Cited" front page and citation lists.

Let me get the front-page "References Cited" list for US 10,078,623 B2 specifically.

US 10,078,623 B2 — Prior-Art Citations and § 102 Analysis

Patent confirmed as the correct one (not a look-alike number): US 10,078,623 B2, "Hyperlink with graphical cue," Gal Arav, app. 15/721,939, filed 2017‑10‑01, granted 2018‑09‑18, priority 2006‑03‑20, Aloft Media LLC. I excluded the unrelated "10078623" hits (a Novartis requisition ID and a UK school URN).

⚠️ Sourcing caveat (read first)

I could not cleanly isolate the face-of-patent "References Cited" block of US 10,078,623 B2 itself. The reference lists I retrieved are from sibling members of the same Arav "Hyperlink with graphical cue" family — chiefly the granted US 10,372,793 B2, US 10,691,874 B2, US 10,482,164 B2, US 11,308,260, and the earlier US 8,683,003 B2 — all of which share the same specification and were examined against substantially the same art. Because these are the same family/technology and the citations overlap heavily, they are a strong proxy, but they are not a substitute for the exact 10,078,623 front page. Any per-claim mapping below should be re-verified against the USPTO PatentCenter "References Cited" tab for 15/721,939.

I am also flagging that I could not verify every reference's title; where I give one it is drawn from the search snippet, and where I only give the field/subject I am inferring from the patentee, classification, or context.

A. U.S. patent documents cited (family list)

Older family list (from US 8,683,003 B2 front page), cited for the aggregation/message-board aspects: US 5,754,176 (Crawford, 5/1998); US 5,854,630 (Nielsen, 12/1998); US 6,065,051 (Steele, 5/2000); US 6,199,077 (Inala, 3/2001); US 6,199,157 (Dow, 3/2001); US 6,208,339 (Atlas, 3/2001); US 6,216,141 (Straub, 4/2001); US 6,253,325 (Steele, 6/2001); US 6,282,548 (Burner, 8/2001); US 6,314,458 (Steele, 11/2001); US 6,336,133 (Morris, 1/2002); US 6,366,947 (Kavner, 4/2002); US 6,393,468 (McGee, 5/2002); US 6,405,238 (Votipka, 6/2002, "Quick Navigation upon Demand to Main Areas of Web Site"); US 6,421,651 / US 6,430,537 (Tedesco, 7/2002, 8/2002); US 6,456,303 (Walden, 9/2002); US 6,484,149 (Jammes, 11/2002); US 6,498,835 (Skladman, 12/2002); US 6,515,681 (Knight, 2/2003); US 6,633,311 (Douvikas, 10/2003, "E-service to Manage and Export Contact Information"); US 6,661,877 (Lee, 12/2003); US 6,691,158 (Douvikas, 2/2004); US 6,748,449 (Dutta, 6/2004); US 6,874,126 (Lapidous, 3/2005); US 6,879,691 (Koretz, 4/2005); US 6,889,213 (Douvikas, 5/2005); US 6,952,730 (Najork, 10/2005); US 7,017,109 (Douvikas, 3/2006); US 7,024,451 (Jorgenson, 4/2006); US 7,069,308 (Abrams, 6/2006); US 7,117,254 / US 7,188,153 / US 7,478,078 (Lunt); US 7,188,080 / US 7,340,419 (Walker); US 7,194,552 (Schneider); US 7,231,428 (Teague); US 7,233,997 (Leveridge); US 7,373,338 (Thompson); US 7,451,161 (Zhu); and others. (Names/dates taken from the Justia/Google patents snippets; treat individual dates as approximate to the month shown.)

B. Foreign patent documents cited

WO 02/03243 A1 (Jan 2002); WO 02/061610 A1 (Aug 2002); WO 2004/029780 A2 (Apr 2004); WO 2007/109264 A3 (Sep 2007).

C. Non-patent literature (NPL) cited

This is the most important NPL for the two-tier-reveal claims, because it discloses hover/preloaded dropdown behavior:

• "Hybrid CSS Dropdowns," A List Apart (Mar 30, 2005).
• "Suckerfish Dropdowns" and "Son of Suckerfish Dropdowns," htmldog.com (Nov 25, 2005).
• "CSS / DHTML Hybrid Navigation Bar," brothercake.com navmeister (Apr 10, 2003).
• "Bloglines," archived Jan 26, 2005.
• Archived commercial sites: Lucent, Intel.com, CNET, Macromedia Flash MX 2004, "pure CSS menus" (Meyerweb), WSJ Online, TV Guide, (mt) Media Temple.

D. § 102 anticipation analysis (by claim concept)

Because the granted claim numbering of 10,078,623 could not be verified (consistent with the caveat in the earlier section), I map to the independent-claim elements recited in the abstract/claims: (i) display first set of hyperlink representations; (ii) first input → display hidden second set of graphical hyperlink representations; (iii) second input → navigate to destination; with dependents adding "initially hidden," "textual/hypertext," and "without an additional server access."

• US 5,983,245 (Newman) — strongest § 102 candidate. Starred by the examiner and classified in hyperlinking (G06F 17/2235). A single-reference § 102 rejection of the broad two-tier hyperlink claims is the most plausible challenge here; I could not verify the exact disclosure, so treat this as the lead reference to read in full.
• US 2004/0205514 (Sommerer) — starred; graphical interface. Potential § 102 against the "graphical representations of hyperlinks" limitation if it discloses user-revealed graphical selectable objects.
• US 2001/0038395 (Holtzblatt), US 2006/0020615 (Keohane), US 2006/0075347 (Rehm), US 2010/0205547 (Boegelund), US 2014/0315179 (DeGross) — all starred (*). These are the references to scrutinize for a § 102 hit on the two-input reveal-then-navigate step sequence; DeGross (2014) post-dates the 2006 priority date and so can only be prior art if the priority claim is broken.
• The NPL dropdown references (A List Apart / Suckerfish / navmeister / Bloglines) — strongest § 102/§ 103 art against the "second set initially hidden" + "revealed without an additional server access (preloaded/AJAX)" dependent limitations, because CSS/DHTML dropdowns inherently preload content with the page and reveal it on hover.
• Autocomplete references — US 6,564,213 (Ortega), US 6,704,727 (Kravets), US 7,032,183 (Durham), US 7,035,864 (Ferrari), US 7,647,312 (Dai), US 8,438,487 (Lin-Hendel), US 8,539,024 (Smit), US 9,348,911 (Egendorf) — these are directed at the text-input/symbol-lookup (FIGS. 4A–4C) subject matter in the same specification, not at the hyperlink-cue claims. As such they generally cannot anticipate the hyperlink claims standing alone; they matter only if the granted claims include lookup/autocomplete limitations (as some family claims do).
• Message-board aggregation references (the Steele, Douvikas, Lunt, Walker, Votipka, Tedesco, etc. group) — cited against the "Message Aggregator" (US 7,529,795) lineage; they bear on the aggregation claims, not the hyperlink-cue claims.
• Self-family items (US 10,042,823, US 10,078,623, US 2019/0034391) — same-family members; not prior art to 10,078,623 (they share the priority chain). Ignore for § 102.

Bottom line: For the hyperlink-cue claims, the references that can plausibly anticipate under § 102 are US 5,983,245 (Newman) and, secondarily, the starred UI references (Holtzblatt, Sommerer, Keohane, Rehm, Boegelund) plus the preloaded-dropdown NPL. The large autocomplete and message-board-aggregator citation groups are aimed at other aspects of this omnibus specification and are unlikely to anticipate the hyperlink claims alone.

Uncertainty notes

• The exact 10,078,623 front-page citation block was not directly retrievable; the list above is the same-family proxy (10,372,793 / 10,691,874 / 10,482,164 / 8,683,003) and should be confirmed at USPTO PatentCenter.
• I did not verify each reference's title; where titles are given they come from the retrieved snippet, and several subject descriptions are inferred.
• Exact granted claim numbers for 10,078,623 remain unverified, so the § 102 mapping is stated in terms of claim elements, not claim numbers.

If you want, I can next attempt to pull the verbatim granted claims of US 10,078,623 B2 and the exact face-of-patent citation list so each reference can be tied to a specific claim number.

I'll start by checking the prior-art landscape that the task refers to. I don't see a "Prior Art" section in the material supplied to me, so let me try to locate the cited/similar-art references for this patent and the key structural prior art in this field. I'll flag explicitly what I can and cannot verify.

I now have the cited-reference lists from the sibling/continuation patents in this family (US 8,683,003 and US 10,372,793), plus some highly relevant external art. Let me pin down the granted claim text and a couple of reference identities before analyzing.

Obviousness Analysis — U.S. Patent 10,078,623 B2 ("Hyperlink with graphical cue") under 35 U.S.C. § 103

0. Important scope and evidence caveats (read first)

Before the analysis, three flags — these materially affect confidence:

(a) I do not have a "Prior Art section" from the page you referenced. No prior-art citation list was included in the material supplied to me. To avoid fabricating one, I reconstructed the prior-art universe from two verifiable sources:

• The "References Cited" lists that appear on the face of sibling members of this family (US 8,683,003 B2 and US 10,372,793 B2, both "Hyperlink with graphical cue," same inventor Gal Arav), which I retrieved via search of patentimages/Justia; and
• External art I located and verified by snippet (patent text, CAFC/Board summaries, and a district-court claim-chart brief).

If the page you intended me to use lists a different set of references, the specific mappings below must be re-run against that list. I say so explicitly rather than assume.

(b) The granted claim text is still not fully verified. I could confirm via Justia that pre-grant publication US 2018/0024973 A1 is the publication of application 15/721,939 and that Justia lists "Patent Grant number: 10078623" for that publication — i.e., the publication claims are a strong proxy for the granted claims, but they are not the granted claims. Claims are commonly amended between publication (2018-01-25) and grant (2018-09-18). I was able to retrieve verbatim claim 23 and fragments of claims 1, 15–18, 22 (quoted below). All element-by-element mapping is done against that language, with the caveat that final wording must be confirmed against the granted B2 PDF.

(c) Consistency with the earlier-generated sections. The previously generated summary correctly flagged (i) that the granted claims were unverified and (ii) that no 2026 CAFC appeal was found. My new search corroborates both and adds nothing that contradicts them. The previously generated summary also reported "apparatus claim 1 with dependents 16, 17, 22; apparatus claim 23." My results are consistent with that. No contradiction to report.

1. Governing framework and the operative date

• Priority / effective filing date: March 20, 2006 (claimed). Unless a claim loses priority under § 112 (e.g., lack of written description for a later-added limitation), the prior art must predate March 20, 2006 (pre-AIA § 102(a)/(b)/(e)). I note that application 15/721,939 was filed October 1, 2017, post-AIA — but its claims, if supported, get the 2006 date. Any limitation not supported by the 2006 specification would only get a 2017 date, which would open a much larger art space. This is a live issue worth reserving.
• Anticipated expiration: March 20, 2026 (per Google Patents). As of the current date in this task (April 26, 2026) the patent would presumptively have lapsed. This does not affect the § 103 merits but affects enforcement/remedy.
• Applicable law: Graham v. John Deere, 383 U.S. 1 (1966) (scope/content, differences, PHOSITA level, secondary considerations); KSR Int'l v. Teleflex, 550 U.S. 398 (2007) (predictable combinations; "obvious to try"); In re Gartside, 203 F.3d 1305 (Fed. Cir. 2000) (legal conclusion reviewed de novo, factual findings for substantial evidence).

Level of ordinary skill in the art (POSITA)

A person of ordinary skill as of March 2006 would have a bachelor's degree in computer science (or equivalent) plus about one to two years of experience building web pages/applications using HTML, CSS, and client-side scripting (JavaScript/DHTML), with working familiarity with browser DOM events such as onMouseOver/onMouseOut, HTTP request-response dynamics, and preloading techniques. This is a modest skill level; the invention is a GUI/HTTP-layer interaction, not an algorithmic advance.

The claim limitations that must be met

From the verbatim published text (proxy for granted claims), the independent-claim core is:

Dependent claims add: the second-tier hyperlinks "include additional information that is preloaded and hidden" (cl. 15); "static object" (cls. 16–17); display is "immediate" (cl. 18); second set "displayed below and to a side" / "immediately below" the first-set representation, and pre-downloaded "prior to the first user input" (cl. 22).

(Note the tension: the patent is titled "Hyperlink with graphical cue" and the specification's FIG. 5A embodiment uses logo graphics, but the retrieved independent apparatus claim 23 recites the second set as textual, non-AJAX representations. This is exactly why the granted wording matters. I map both permutations below.)

2. The prior art

Excluded from art (and why): US 8,683,003; US 10,037,793; US 10,078,623 itself; US 10,372,793; US 10,691,874; US 10,482,164; and the other Arav "Hyperlink with graphical cue" continuations. These share the same priority chain and inventive entity, so they are not prior art against a continuation claiming that priority (pre-AIA § 102(e)/(f) "by another" requirement; § 103(c)). They are useful as evidence of claim scope, not as art.

3. Combination A (primary): Gennaro + Ingram + Hill

Targets claims 1, 15–18, 22, 23 and the method/CRI counterparts.

(1) Gennaro → E1, E3, E4, E5, E6.
Gennaro's hot spots are "representations … of one or more menu items" (E1) displayed in connection with a web page. Positioning a pointer over a hot spot "designates" it without selecting it, and "a corresponding menu is displayed to provide links to multiple additional web pages" — i.e., the second tier is a set of representations of a set of hyperlinks, revealed in response to a first input by hovering (E3 → E4). The user then "select[s] a link by positioning the pointer over one of the links and initiating an action such as clicking a mouse button" — the second input and resultant navigation (E5, E6). Critically, mousing over a menu item "highlights the selected link … and displays the destination URL" — the "graphical cue" concept of telling the user where a link goes before committing. The menu is displayed in the same page/window, and it is an applet-based menu, i.e., non-AJAX (directly meeting claim 23's negative limitation).

(2) Ingram → E3/E4 structure and the motivation.
Ingram supplies the express teaching that a pointer-adjacent, hover-triggered menu/toolbar of user-selectable items should be presented on a web page ("a visual clue and/or menu"), optionally after a brief dwell delay, and explicitly contemplates that it can overlay content. Ingram also frames the problem the '623 patent purports to solve: reducing user disorientation about where a link leads. Ingram therefore both discloses the reveal-on-hover mechanism and supplies the reason to adopt it.

(3) Hill → E2 and E7 (the crux limitations).
Hill discloses the exact architecture recited in claim 23:

• second-tier content "embedded therein but not displayed unless the hypertext link is activated" → "initially hidden";
• content "stored locally with the rest of the current WWW page" → "pre-downloaded with the web page" / "preloaded and hidden";
• "immediately called forth" upon activation → "immediate display";
• and because the content is already local, no round-trip is needed → "without requiring an additional access to the server."

Motivation to combine (KSR rationales).

• Same field, same problem, same solution class. All three are web-browser hyperlink-UIs and all aim to inform the user about a link's destination before navigation — the precise problem recited in the '623 Background ("the textual representation … may not make the destination … clear").
• Known technique improving a similar device in the same way. Gennaro already couples hover-reveal to a multi-link menu; adding Hill's local-prefetch/latent-content mechanism is a predictable, mechanical substitution of "fetch on activation" with "fetch with the page, reveal on activation" — the classic latency-hiding tradeoff Hill itself describes.
• Predictable results. Each element retains its function: hover still reveals a menu (Gennaro/Ingram); prefetch still eliminates latency (Hill). Nothing in any reference disparages the combination; Hill expressly teaches the combination of hidden embedded content + immediate display concurrently with the destination page load, which is a stronger disclosure than required.
• No teaching away. None of the references criticizes combining text-level representations, hover triggers, and prefetch. Meiresonne v. Google is directly on point: the Federal Circuit affirmed that a rollover viewing area (Finseth) plus adjacent descriptive link text (Hill-type) was unobvious-rejecting — i.e., the Board and CAFC held that such combinations of textual link descriptions with graphical rollover previews are obvious, and that neither reference taught away.

Claim-by-claim scorecard (Combination A):

Graphical-logo permutation. If the operative claim (likely granted claim 1) instead requires the second set to be graphical representations/logos, add Finseth (graphical thumbnails in a rollover area) and/or WO 02/39207 (concurrently displaying a visual image of the linked page). Company logos/favicons as link identifiers were notoriously well known in the art by 1999–2003, so the substitution of a logo for a thumbnail is a predictable design choice. This makes the graphical variant at least as vulnerable as the textual variant.

4. Combination B (alternative): Corel CA 2279145 + Finseth + Hill

This combination independently reaches the same limitations and is useful as a second, independent ground.

• Corel discloses every information-flow element: a first tier of link representations on the page; a cursor-position-triggered selection of a link (E3); retrieval and caching of preview data for each link in advance (E2 pre-fetch); display in an overlay window adjacent to the identified link (E4); cessation when the cursor leaves; and, in the multi-link embodiment, a set of previewable links (the second tier).
• Finseth supplies the graphical nature of the preview via a rollover viewing area.
• Hill supplies the "hidden until activated" + "stored locally with the page" + "immediate, no server access" limitations (E2/E7).

Motivation: Corel's stated purpose is to avoid wasted user time and disorientation from committing to a link — the same objective as Ingram/Gennaro and the same objective recited in the '623 Background. Combining Corel's caching with Hill's locally-embedded latent content is the routine engineering step of client-side preloading to hide latency, a technique acknowledged as conventional in the '623 specification itself (it describes pre-downloading "with the web page" via AJAX as a known implementation detail). Finseth's rollover graphics supply the "graphical cue" with a reasonable expectation of success, and the CAFC's Meiresonne decision confirms such combinations are non-teaching-away and yield predictable results.

5. Combination C (paper trail): IBM 2002/0129051 + Ingram + Hill

IBM '051 provides "means … for enabling a user to select to view only a portion of a hypertext document linked to a hyperlink" and "accessing and storing at said one receiving display station said portion," displayed "responsive to said user selection." Combined with Ingram's hover-triggered selectable menu and Hill's local latent content, this yields E1–E7 with the same motivations. This ground is weaker on the "second set of hyperlinks" element (IBM previews a document portion, not a set of links) but is a useful secondary/back-up ground, and the EP 2270682 framework reference (WO 02/39207) can be layered in to supply the "concurrent display of linked-page visual context."

6. Rebuttal anticipation and counter-arguments

The patent owner would likely argue, and my responses:

1. "The references preview a document, not a set of hyperlinks." This fails against Gennaro, whose menu 46 expressly "provide[s] links to multiple additional web pages" — i.e., the second tier is literally a set of representations of a set of hyperlinks, and hovering a menu item "displays the destination URL."
2. "Hover is not 'selection.'" The '623 specification itself says the first input may be "clicking on or hovering" (repeatedly). A patent owner cannot read "selection" narrowly to exclude hover when the specification defines it that way. Both Gennaro and Ingram disclose hover/dwell-triggered reveals.
3. "The art fetches content on demand; the claims require pre-download with the page and no additional server access." Hill squarely discloses hidden content "stored locally with the rest of the current WWW page," "immediately called forth" on activation — pre-download + no round-trip. Corel's claim 7 discloses advance retrieval and caching of preview data.
4. "Non-AJAX elements" is a negative limitation the art can't meet. This is a claim-drafting artifact (AJAX post-dates the 2006 priority and was a known technique by 2017). Gennaro's applet menu and Hill's locally-stored embedded content are, by construction, non-AJAX. Negative limitations of this type do not confer patentability where the art simply predates the disclaimed technology.
5. "Graphical logos are a specific improvement." Finseth, WO 02/39207, and the ubiquity of logo/favicon link identifiers defeat novelty here; and logo-tile navigation predates 2006 generally.
6. Secondary considerations. I found no evidence in the available material of nexus-qualified secondary indicia (unexpected results, licensing success attributable to the claimed invention, copying, long-felt need). Absent such evidence, the KSR "predictable results" analysis controls.

7. Bottom line

• Under the claim language best available to me (publication claim 23 / fragments of claims 1, 15–18, 22, all as published for application 15/721,939), claims 1, 15–18, 22, and 23 — and their method/CRI counterparts — appear vulnerable to § 103 over Gennaro + Ingram + Hill, with Corel CA 2279145 + Finseth + Hill as an independent alternative ground. The strongest, cleanest mappings are on the "pre-downloaded / initially hidden / immediate display / no additional server access" cluster (Hill), the "hover reveals a multi-link menu and then navigates" cluster (Gennaro), and the "pointer-adjacent hover toolbar of selectable items" cluster (Ingram). The Meiresonne v. Google affirmance is strong supporting authority that this class of rollover/link-context UI combinations is obvious.
• Confidence is high on the technical mapping for the limitations I could verify from the art; confidence is medium on the mapping against the granted claims, because I have not seen the issued B2 claim text.

Open items to resolve before relying on this analysis

1. Pull the granted claims of US 10,078,623 B2 (USPTO PatentCenter "Claims" tab or the B2 PDF) and diff against the 2018/0024973 publication claims. My analysis is only as good as that text.
2. Verify Gennaro's patent number and its publication/filing date (currently cited here from a district-court brief only).
3. Confirm which of the face-cited references (from US 8,683,003 / US 10,372,793) the page you referenced actually lists, and supply that "Prior Art" list — if it differs from my reconstruction, re-run § 3–§ 5 against it.