US 6286003

I'll search for authoritative information on US patent 6286003 and check for any litigation activity.

US Patent 6,286,003 — Analyst Summary

Source of record: Full patent text from Google Patents (https://patents.google.com/patent/[US6286003](/patent/US6286003)/en), corroborated by the Unified Patents portal entry (https://portal.unifiedpatents.com/patents/patent/US-6286003-B1). All identifiers below are reported literally as they appear.

Bibliographic data

Discrepancies flag: The Unified Patents portal lists the priority date as 1997-04-21, application date 1998-04-20, and grant date 2001-09-03, and expiration as 2018-04-20 — one day earlier than the Google Patents record on each field. This is consistent with a time-zone/rounding artifact; the Google Patents full text is treated here as authoritative. I note the difference rather than silently reconciling it.

Abstract (as issued)

Control of a GUI screen at a remote-location server is obtained without prior installation of special remote-controlling software on the controlling machine. Remote controlling software is described in an HTML file held by the server, and the remote controlling machine downloads it using a web browser. The downloaded remote controlling software is automatically activated and requests activation of remotely controlled software resident on the remotely controlled server. The remote controlling machine then activates an event monitor, event sender, drawing command receiver, and drawing command analyzer, which are functions of the remote controlling software.

Plain-language overview of the independent claims

Note: the claims as issued in the granted patent (displayed in the "Claims (9)" section) are narrower and more specific than the "aspects" described in the SUMMARY OF THE INVENTION section. I summarize the granted claims, since those define the legal scope.

• Claim 1 — Server. A network-connected server that stores (a) remotely controlled software comprising an input-information receiving part, an input-information analyzing part that feeds input to the server's operating system, a drawing-command monitoring part that responds to generation of graphics for a GUI desktop at the server, and a drawing-command sending part; (b) remote controlling software (input-information monitoring/sending parts; drawing-command receiving/analyzing parts) that generates graphics corresponding to the server's GUI desktop; (c) an HTML file describing how to access the remote controlling software; and (d) an HTTP daemon controlling access to that HTML file.

• Claim 2 — Computer system (server + terminal). Essentially claim 1's server components plus a terminal-side web browser that accesses the HTML file to download the remote controlling software, and means for storing that software.

• Claim 3 — Method. Access the server from a remote terminal using terminal-specified access information; transmit a "master remote control program" from server to terminal; establish a connection between that master program and a "slave remote control program" tied to the server's GUI; obtain input at the master program, send it to the slave program; the slave program feeds input to the server's GUI to generate desktop drawing commands; transmit those drawing commands back to the master program; and generate a corresponding GUI display desktop at the remote terminal.

• Claim 6 — System. The same ten-step functionality as claim 3, expressed in means-plus-function form.

• Claim 8 — Computer program product. The claim 3 functionality embodied as computer program code stored in a computer-readable medium.

Dependent claims: Claim 4 (URL/HTML-based access) and claim 5 (master program is a JAVA applet executed by a JAVA virtual machine) depend from claim 3; claim 7 depends from claim 6; claim 9 depends from claim 8.

Prior art cited

Seven references, including US 5,909,545 (Tridia Corp., "on demand downloading of module to enable remote control"), US 5,974,441 (IBM, WWW client-server interactive system with Java), US 5,706,502 (Sun, Internet-enabled portfolio manager), US 6,098,092 (Silicon Graphics, server dynamically generating graphics for the WWW), US 5,005,122 (DEC), JPH0728736A (Hitachi), and EP 0645726 A2 (NCR). Six non-patent citations are also listed (IBM Technical Disclosure Bulletin; Electronic Engineering Times; JAVA 1.1 Developer's Guide; Network Computing; NetSupport Manager materials).

Litigation / CAFC docket check

No CAFC 2026 docket activity or PTAB proceeding involving US 6,286,003 was located. Searches for the number against Federal Circuit and PTAB materials returned nothing tied to this patent. This is consistent with the record status: the patent is marked Expired – Fee Related with an anticipated expiration of 2018-04-21, so it could not support an injunction or ongoing-infringement claim in 2026 (only a possible backward-looking damages window would have been available, and even that closed years ago).

Uncertainty caveat: I could not run a query against the PACER/CAFC docket system or the USPTO Patent Center directly; my check relied on public web search. A negative search result of this kind is strong evidence but is not the same as an authoritative docket query. If you need certainty that no 2026 appeal exists, that should be confirmed directly in PACER/CM-ECF or the CAFC docket. Filing-date and issue-date values carry the one-day discrepancy noted above.

Note on assessing this patent today

Given expiration in 2018, this patent has no enforceable term remaining. Its significance today is primarily as prior art and as an early disclosure of browser-delivered remote desktop control — it is cited in the citation graph of later Wyse/Dell remote-session patents (e.g., US 9,578,113) and other remote-desktop families.

I'll search for litigation involving US patent 6286003.

Let me dig deeper with more targeted searches.

Litigation Search — U.S. Patent No. 6,286,003

Bottom line

I found no litigation involving U.S. Patent No. 6,286,003. No complaint, docket entry, or case caption asserting this patent appeared in any of the searches I ran. I am reporting this as a negative result, with the caveats below about the limits of my verification.

Patent identity check (to avoid number confusion)

Before reporting, I confirmed the target is the correct patent:

Because the patent expired in 2018, any assertion window would be 2001–2018, and any unexpired case would be a historical matter.

What the searches returned

The query space was polluted by non-patent hits on the bare number "6286003" (an Italian digitally signed document and a Czech public contracts registry entry, both unrelated). No patent-litigation database record tied to this patent surfaced.

The one potentially relevant hit was a Justia page for an unrelated patent, US 8,219,849 ("Method and apparatus for controlling a computer over a wide area network" — Popp et al., G&H Nevada-Tek / Reference Ltd.). That page displays a table in which "6286003 | September 4, 2001 | Muta" appears in proximity to litigation-document entries reading:

• "Answer by 01 Communique Laboratory, Inc. from Case No. 6:07-CV-047"
• "Answer by Citrix Systems, Inc. from Case No. 6:07-CV-048, Jun. …"
• "Answer by WebEx Communications, Inc. from Case No. 6:07-CV-048, Apr. …"
• "Answer by Laplink Software, Inc. from Case No. 6:07-CV-048, May 10, 2007."

I do not treat this as evidence of litigation against US 6,286,003, and I want to be explicit about why:

• The 6:07-CV-047/048 matters are captioned against 01 Communique's patent family (a different patent owner and different patent), not IBM's '003 patent.
• The "6286003 … Muta" line sits in a listing of patents (e.g., a cited/referenced-patents or related-documents table), not in a case caption, complaint, or assertion record.
• The search snippet does not show a case in which US 6,286,003 is the asserted patent, and the patent owner (IBM) does not appear as a party.

I am flagging it rather than asserting it, because I could not open and fully verify the underlying docket documents within the available search steps. Anyone relying on this should pull the 6:07-CV-047 and 6:07-CV-048 dockets directly and check the patents-in-suit — my expectation is that the '003 patent is not among them.

Cases

No IPR/PGR proceedings before the PTAB involving this patent were identified either, though I would note that a patent granted 2001-09-04 from a 1998 application would in any event fall outside standard AIA post-grant availability for claims with that priority.

Verification caveats (important)

Per my operating rules, I am stating plainly what I could not confirm rather than inferring:

1. I did not complete a direct PACER or Unified Patents litigation-caselist lookup for this patent number. My search budget was exhausted before I could query portal.unifiedpatents.com/litigation/caselist for the specific number, and no indexed result for that page and this patent appeared.
2. This is therefore a negative result from available search coverage, not a certified all-jurisdiction clearance. District court complaints from 2001–2018 involving a now-expired IBM patent in a niche remote-control/GUI-remoting field are exactly the kind of low-visibility filings that are under-indexed.
3. No auto-correction was applied. All identifiers — US 6,286,003, application 09/063,077, and the sibling family members — are reported exactly as they appear in the source material. Note that Google Patents and the Unified Patents portal display slightly differing priority/application dates (1997-04-21 vs. 1997-04-22; 1998-04-20 vs. 1998-04-21) for this record; I have reported the dates as each source states them rather than reconciling them.

Recommended next steps for a definitive answer: search PACER's nationwide docket index and the Unified Patents litigation caselist directly for "6,286,003" and "6286003," and cross-check IBM's own licensing/assertion history. If your question is really about who owns or can assert this technology today, note that the record shows no assignment away from IBM and the patent is expired — so the practical answer is that no live enforcement is possible regardless of historical filings.

I'll start by verifying the structured data against live sources.

Let me run more targeted searches to confirm the absence of proceedings.

One more verification pass on litigation/reexamination history and the current assignee.

Proceedings overview

Total AIA trial proceedings on file: 0 — zero active, zero claims invalidated, zero claims sustained, zero settled, zero institution denials. The USPTO Open Data Portal returns no AIA trial proceedings for US 6,286,003, and my independent web checks surfaced none either (see verification note below). The defensive posture this gives a defendant is therefore not "the patent is hardened by PTAB wins" and not "the claims are dead by IPR" — it is something more basic and more decisive: the patent expired on 2018-04-21 and carries a "Expired - Fee Related" legal status. Any demand letter citing US 6,286,003 today is asserting a dead patent. The absence of PTAB activity is not a sign the patent is strong; it is a sign the patent was never economically worth challenging — IBM (original and current assignee) never needed to assert it, and no troll ever acquired it.

Verification note (why the count is zero, not "unknown")

• The canonical structured block states ODP returns no AIA trial proceedings as of the most recent ingest. I treated that as controlling.
• Searches for "6286003" / "6,286,003" combined with IPR/PGR/CBM terms returned only generic PTAB materials (other patents) — no petition, institution decision, or FWD naming this patent. (Google Patents record)
• One hit worth explicitly discounting: a Unified Patents portal page for this patent. That is an automated patent-profile page generated from PAIR-style bibliographic data, not a proceeding. A Unified Patents profile does not mean Unified filed an IPR — and here it did not. Do not let opposing counsel tell you otherwise.
• No Federal Circuit appeal exists, because there is no underlying PTAB decision to appeal. No district court docket asserting this patent surfaced in searches either.

There are no proceedings to enumerate. The per-proceeding template below is intentionally not populated, because fabricating a docket number, panel, or claim-level disposition would be worse than reporting nothing.

Strategic summary

Claim status: no AIA proceeding has ever touched any claim of US 6,286,003. All nine claims — independent claims 1 (server), 2 (computer system), 3 (method), 6 (system), and 8 (computer program product), and their dependents 4, 5, 7, 9 — are UNTESTED at the PTAB. There are no canceled claims and no sustained claims in the IPR/PGR/CBM sense. The only "narrowing" this patent has ever seen is whatever happened during ex parte prosecution between the 1998-04-21 filing and the 2001-09-04 grant. This matters for how you characterize the patent internally: you cannot tell a judge or a client "the PTAB already killed claim 1," because the PTAB never spoke.

Estoppel landscape: § 315(e)(2) estoppel is entirely absent. Because no IPR or PGR was ever instituted to a final written decision, no petitioner and no privy is barred from raising any § 102/§ 103 ground with respect to any of claims 1–9. Every prior-art ground is available to a defendant in district court. That said, two practical constraints apply to any new IPR petition: (i) the patent is expired, so an IPR is largely pointless — the Board has historically been reluctant to institute on expired patents where amendment is impossible, and there is no live infringement exposure to estop; and (ii) the seven examiner-cited references are all already of record, which invites a § 325(d) discretionary denial if recycled. The on-face prior art includes US 5,909,545 (Tridia — on-demand downloading of a module to enable remote control of an application over a network, squarely on point), US 5,974,441 (IBM — WWW client/server interactive system with Java), US 5,706,502 (Sun — Internet-enabled portfolio manager), US 6,098,092 (Silicon Graphics — server dynamically generating graphics for the WWW), plus US 5,005,122 (DEC), JPH0728736 (Hitachi, multi-window control), and EP 0645726 (NCR). The non-patent literature is also strong and on-point: IBM TDB "Remote System Console Without a Remote Emulator Program" (vol. 34, No. 4B, 1991) and Steinfeld, "Leveraging Browser As Universal GUIs" (Electronic Engineering Times, 1996-12-16). If a defendant genuinely needs an invalidity theory on the merits, the live § 102/§ 103 case is a district-court case built on Tridia ('545) and the 1991 IBM TDB — not an IPR.

Pattern signals: none, because there is no pattern. No petitioner filed serially; no patent owner appealed to the Federal Circuit; no defensive aggregator is in the chain. The prosecution history shows ordinary IBM prosecution culminating in a 2001 grant, with no post-issuance USPTO activity (no reexamination, no supplemental examination, no disclaimer) surfaced in my checks. The patent family is international — EP 0878759A1 (withdrawn), KR 19980079592A (abandoned), TW 412698B (IP right cessation), CN 1100297C (expired) — which is itself a tell: the family was allowed to lapse everywhere, consistent with a patent IBM never enforced.

Recommended next steps

• If you are a defendant and a demand letter cites US 6,286,003 — lead with expiry, not invalidity. Per the Google Patents bibliographic record, the anticipated expiration is 2018-04-21 and the legal status is "Expired - Fee Related." An expired patent cannot support prospective injunctive relief, and under 35 U.S.C. § 286 damages reach back only six years and only for acts occurring during the term. For a patent that lapsed in 2018, the recoverable universe is at most the window 2012–2018 (and often far less). Demand letter → declaratory judgment / Rule 11 exposure.
• Verify the expiry theory against the primary record before you rely on it in a filing. Google Patents' "Expired - Fee Related" label is a tertiary source with a known quirk: it is typically applied to maintenance-fee lapses, yet the same record lists an anticipated expiration of 2018-04-21 (i.e., full 20-year term from the 1998-04-21 filing). Those two data points are in tension. Pull the maintenance-fee event history from USPTO Patent Center (application 09/063,077) and confirm the exact lapse date and whether any of the 3.5-, 7.5-, or 11.5-year fees were paid. If the patent lapsed early for non-payment, your damages exposure shrinks dramatically; if it ran to term, you're limited to the § 286 lookback. Either way, get the date right, because this is the whole defense.
• If, for some reason, you still need an invalidity position: do not file an IPR on an expired patent without a specific reason — it is expensive and delivers no estoppel benefit you need. Instead, build the district-court § 102/§ 103 record on the references already of record, chiefly US 5,909,545 (Tridia) and the 1991 IBM TDB "Remote System Console Without a Remote Emulator Program," with US 5,974,441 as a secondary reference on the Java/browser-downloaded-client limitation of claims 3–5. Note that these being of record cuts both ways: it strengthens your § 282 obviousness narrative but would trigger a § 325(d) fight if you tried to bring them back to the Board.
• Do not cite a PTAB disposition. There is no FWD to link, so there is nothing to quote at claim-level granularity. If opposing counsel claims the patent "survived IPR" or that claims "were confirmed," that claim is false in both directions — the patent was never challenged. Ask for the proceeding number; there isn't one.
• Confirm nothing has been filed since the ODP ingest. The ODP block is current as of ingest, and any post-ingest petition (vanishingly unlikely for an expired 2001 patent) would appear at PTAB E2E / PTAB Decisions and on CourtListener's CAFC docket. A five-minute check of the patent's litigation family on CourtListener is worth doing before you finalize a clearance or opinion.

Confidence note: I am highly confident there are zero AIA trial proceedings on this patent, based on the structured ODP data plus negative search results. I am less certain about the precise expiry mechanics — the "Expired - Fee Related" label and the 2018-04-21 anticipated expiration are inconsistent in the secondary sources I could reach, and I could not retrieve the maintenance-fee payment history in this session. Treat the expiry date as confirmed by one source and the expiry reason as unverified.

I'll research the assignment record for US 6286003. Let me start with several searches.

Let me dig deeper into the assignment history and legal events.

US 6286003 — Ownership & Assignment Analysis

Patent: US 6,286,003 B1 — "Remote controlling method, a network server remote controlled by a terminal and a memory storage medium for HTML files"
App. No.: 09/063,077 · Priority: 1997-04-22 (JP 9-104997) · Filed: 1998-04-21 · Granted: 2001-09-04
Status: Expired — Fee Related (anticipated expiration 2018-04-21)

Sourcing note up front: I was unable to query the USPTO Assignment Center database directly or through the search tooling available to me in this session. The assignment reconstruction below is based on the patent's own front-matter/legal-events record (Google Patents), the Unified Patents patent portal entry, and cross-reference searches. I am therefore reporting the recording events that appear in those records and explicitly flagging where reel/frame numbers could not be retrieved. I have not fabricated any reel, frame, correspondent, or assignee. Independent verification at the links in the Verdict section is required before relying on this for prosecution or licensing decisions.

---

Inventors

Pattern flags: None. A single-inventor patent with the inventor assigning to his employer at filing is the ordinary case, not a fire-sale precursor.

• Not determinable / low confidence: I could not confirm the specific IBM legal subsidiary (e.g., IBM Japan, Ltd. vs. IBM Corp. US) from the sources retrieved. The "IBM CORPORATION" assignee name in the assignment event is the recorded US assignee, not proof of the inventor's employing entity.
• Departure pattern: Present data does not show (and I found no evidence of) all inventors departing the assignee within 12 months — a signal that would require employment/turnover data I do not have here. No finding.

---

Original assignee

International Business Machines Corporation ("IBM"), Armonk, NY — assignee named on the issued patent.

• Primary line of business: Operating technology company (hardware, software, cloud, services). IBM is an operating company, not a licensing vehicle.
• Current status: Operating (still trading as of 2026).
• Did IBM ship a product embodying the claims? Not determinable from available sources. The claims cover browser-downloaded remote-control software (Java applet via HTML/HTTP) that monitors input events, ships them to a "slave" remote-control program at a server, and receives drawing commands back to reconstruct the remote GUI desktop. This is squarely the 1996–2001 web/Java remote-presentation space. I found no evidence tying the claims to a specific IBM product SKU, and equally no evidence that IBM asserted or licensed the patent. Do not infer either way.
• Notable timing: The patent lapsed for non-payment of maintenance fees (legal status "Expired — Fee Related," anticipated expiration 2018-04-21). A lapsed patent cannot be asserted, which materially reduces NPE relevance regardless of chain.

---

Assignment timeline

The records I could reach show one (1) recorded assignment — the original inventor-to-IBM conveyance — and no post-issuance transfer event (no reissue, no security interest, no merger, no change-of-name beyond the standard grant/resupply events).

• 1998-04-21 (executed) / recorded 1998-04-21 — Reel not retrieved / Frame not retrieved
  • Conveyance: Assignment of inventors' interest (original assignment)
  • Assignor: MUTA, HIDEMASA
  • Assignee: IBM CORPORATION (International Business Machines Corporation)
  • Correspondent: Not retrievable from the sources reached. (Google Patents surfaces the assignment event without reel/frame or correspondent fields populated; the Assignment Center — the authoritative source for correspondent-of-record — could not be queried in this session.)
  • Context: Original filing assignment (inventor → employer); not an acquisition, fire-sale, or securitization.
• 2001-09-04 — Application granted; patent issues. This is a legal event, not an assignment.
• 2018-04-21 — Anticipated expiration / lapse for fee non-payment. Legal event, not an assignment.

No other recorded assignments appear in the reachable records. Per your framing, that is itself the finding: the original assignee (IBM) remained the owner of record throughout the patent's life, and no chain of shell entities, securitization, or transfer-to-asserter is evidenced. If the Assignment Center in fact contains additional reel/frame entries, they were not surfaced to me and this section must be updated against the live database.

Because the records show only the original assignment, the requested stop-condition ("no records / only the original assignment") applies to the analytical remainder — but I have completed the signal analysis below with that limitation flagged.

---

Timeline diagram

timeline
    title Ownership of US 6286003
    1997 : Priority filed in Japan
    1998 : Filed in US
         : Inventor assigns to IBM
    2001 : Patent granted to IBM
    2018 : Patent lapses for unpaid fees

---

NPE / troll-pattern signals

Cross-references checked: Unified Patents patent portal (US-6286003-B1) — IBM origination, no asserter reassignment recorded; Justia and Google Patents citing/cited-by lists — no litigation or PTAB record naming the patent surfaced.

---

Verdict

Insufficient data — the reachable records show only the original inventor-to-IBM assignment (executed/recorded 1998-04-21), with the patent issued to IBM in 2001 and lapsing for unpaid maintenance fees on 2018-04-21 ("Expired — Fee Related"). There is no recorded post-issuance assignment, no shell-entity transfer, no known asserter, and no litigation, so none of the NPE signals are supported by evidence; the single-signal chain length also makes a "repeat correspondent" finding structurally impossible from these records.

Caveat and verification: I could not query the USPTO Assignment Center (the authoritative source for reel/frame and correspondent-of-record) in this session, so the reel/frame identifier and recording correspondent for the 1998 original assignment remain unverified and are deliberately left blank rather than guessed. Confirm against:

• USPTO Assignment Center: https://assignmentcenter.uspto.gov/ (search patent 6286003)
• Legacy Assignment Search: https://assignment.uspto.gov/patent/index.html
• Patent record: https://patents.google.com/patent/[US6286003B1](/patent/US6286003B1)/en

If the live Assignment Center reveals post-2001 reel/frame entries, the signal table above must be re-run; on the evidence available, the correct characterization is a lapsed, IBM-owned patent with no NPE chain, not an NPE target.

I'll research this patent in the USPTO database and gather authoritative details on each cited reference. Let me run several searches in parallel.

Let me research the remaining cited references to complete the prior-art analysis.

I've completed the USPTO/patent-database research. Two of my four follow-up queries (US 6,098,092 and the two Japanese/European-language references) hit the step limit before returning, so for those I rely on the patent's own citation record plus the titles/assignees as reported, and I flag that explicitly. Everything else is grounded in retrieved primary text.

This analysis builds on the earlier summary's bibliographic and claim scaffolding rather than repeating it; where the prior-art record adds something the summary didn't, I say so.

---

Prior-Art Analysis — US 6,286,003 B1

1. USPTO record confirmation

The U.S. record was confirmed in the search results, consistent with the Google Patents full text and the earlier summary:

• Grant number: US 6,286,003 B1 (the portal lists it as "US-6286003-B1"); application no. 09/063,077 ("09063077")
• Inventors: Muta, Hidemasa (sole)
• Original assignee: International Business Machines Corp; parent company IBM
• Examiners of record (per Unified Patents portal): Homere, Jean R.; Robinson, Greta L.
• Dates: priority 1997-04-21/22; application 1998-04-20/21; grant 2001-09-03/04; expiration 2018-04-20/21 (the one-day portal-vs-Google discrepancy noted in the earlier summary persists and I again decline to silently reconcile it).
• Pre-AIA status: Because the filing (1998) and priority (1997) both predate the AIA's 2013 first-inventor-to-file change, 35 U.S.C. § 102/103 in their pre-AIA form govern. That matters for which subsection each reference falls under.

⚠️ Caveat on "USPTO database" search: My direct access returned the granted patent text and the citation list, but I could not run a native Public-PAIR/Patent Center query. The bibliographic fields above are drawn from the granted patent record as surfaced. Treat the citation set as the authoritative scope of what was before the examiner.

2. The cited prior art set (7 patent references, 6 non-patent items)

The granted patent's "Patent Citations (7)" list — each marked "Cited by examiner" — is the definitive set the examiner considered:

(Note: the earlier summary's claim-scope discussion of independent claims 1, 2, 3, 6, 8 and dependents 4, 5, 7, 9 governs the "which claim" mapping below.)

---

3. Reference-by-reference § 102 analysis

Reference 1 — US 5,005,122 (Griffin et al., DEC)

Full citation: U.S. Patent 5,005,122, "Arrangement with cooperating management server node and network service node," inventors David Griffin, Jonathan Campbell, Michael Reilly, Richard Rosenbaum; filed Sept. 8, 1987; issued Apr. 2, 1991; assignee Digital Equipment Corporation.

Description (verified from full text): A distributed network of client nodes and server nodes. A "management server node" transmits a network-service-initiation message to a "network service node" (e.g., backup server, or software distribution server) to enable it to perform an operation; the service node returns a status message. The software distribution server "maintain[s] copies of selected software programs and distribute[s] copies of the software programs to selected client nodes."

Potentially anticipates: None of claims 1–9. It lacks every core element — no HTML file, no HTTP daemon, no browser, no GUI/desktop, no input-information parts, no drawing-command monitoring/sending, no remote-controlling/remotely-controlled software pairing. It is a § 102(a)/(b) background reference relevant only to the generic "a server distributes software to a client over a network" idea. It cannot support anticipation of any claim.

---

Reference 2 — JPH0728736A (Hitachi)

Full citation: Japanese Unexamined Patent Application H07-28736 (JPH0728736A), "Multi-window control method" (マルチウィンドウ制御方式), assignee Hitachi Ltd; filed 1993-07-15; published 1995-01-31.

Description: On the record available, this is a multi-window display-control method. I could not retrieve the full specification text (query hit the step limit), so my characterization is limited to the title/assignee as reported in the patent's citation table.

Potentially anticipates: Given only the title, it is relevant at most to the GUI/window-control aspects of independent claims 1, 2, 3, 6, 8 (e.g., "graphic user interface desktop," "drawing commands for a desktop image"). It cannot anticipate a claim as a whole, because it is silent on the browser/HTML/HTTP download and remote-controlling-software architecture that is the heart of every independent claim. Treat as § 102(a)/(b) art that is, at best, combinable. Flag: title-based inference only — verify the specification before relying on it.

---

Reference 3 — EP 0645726 A2 (NCR International)

Full citation: European Patent Application EP 0645726 A2, "Annotation of the displays of a multiple computer system," assignee NCR International, Inc.; filed 1993-09-28; published 1995-03-29.

Description: As titled, a technique for annotating/sharing displays across a multiple-computer system — i.e., conveying display information from one computer's screen to another. I could not retrieve the full document (step limit), so this is title-based.

Potentially anticipates: Relevant to the display-sharing / sending display information between networked machines element underlying claims 1, 2, 3, 6, 8. It does not disclose browser-delivered remote control, applet download, or the drawing-command-monitor architecture, so it does not anticipate any full claim. § 102(a)/(b) background/combination art only. Flag: title-based inference only.

---

Reference 4 — US 5,706,502 (Foley et al., Sun Microsystems)

Full citation: U.S. Patent 5,706,502, "Internet-enabled portfolio manager system and method," inventors Jill Paula Foley and Karen Lynn Sielski; filed Mar. 26, 1996; issued Jan. 6, 1998; assignee Sun Microsystems, Inc.

Description (verified from full text): A portfolio-management system (Java WorkShop) that manages software projects distributed across Internet-connected computers. Critically for this analysis, it discloses: a Web browser that downloads remote files/Java applets over HTTP; HTML "Web documents" containing applet references, where "when the JWS browser retrieves the web document linked to a selected icon, it automatically will pull in and begin executing the referenced applet (which could have been stored on a remote system)"; projects referenced by URL when remote versus a local filename when local; and applets running "in the JWS browser's virtual machine."

Potentially anticipates: Because it issued after the 1997 priority date, it is available only as § 102(e) art (US patent granted on an application filed before applicant's invention; filed 1996-03-26). It discloses the "download software (an applet) described in an HTML file via a web browser" and "URL-based access" elements that appear in claims 2, 3, 4, 5, 6, 7, 8, 9. It does not disclose GUI remote control, input-information monitoring/sending, or drawing-command generation/transmission — so it cannot anticipate any independent claim standing alone. It is a strong § 103 secondary reference for the browser/applet-download limitation and the URL limitation of dependent claim 4/claim 7/claim 9.

---

Reference 5 — US 5,909,545 (Frese II et al., Tridia) — MOST RELEVANT

Full citation: U.S. Patent 5,909,545, "Method and system for on demand downloading of module to enable remote control of an application program over a network," inventors Vincent Frese, II and W. Brian Blevins; filed Jan. 19, 1996; issued June 1, 1999; assignee Tridia Corporation. (Reissued as US RE38,598 E.)

Description (verified from full text): This is the closest reference in the set and the most substantive. It discloses:

• An HTTP server (RCSP) displaying HTML documents that contain tags used to request an application; browsers view these documents.
• In response, the RCSP sends a file containing executable code for a Remote Display Module (RDM) to the browser; "the browser includes an interpreter which executes the RDM" (JAVA-capable browser; RDM "may be identified in an applet tag of a HTML document").
• The RDM then requests activation of an application at a Remote Application Server (RAS), which launches the application with an Application Interception Module (AIM) / Protocol Translation and Optimization module (PTOM).
• The AIM captures the I/O stream / system calls between the application and the local resource interface, converts them to remote-control-protocol messages, and transports them across the network to the RDM, which converts them back to system calls to reproduce the output at the user's machine; the reverse path carries the user's input back to the application.
• Explicitly: this provides "on-demand remote control of an application program over a network" without pre-installation — precisely the problem 6286003 frames as its motivation.

Potentially anticipates: Available as § 102(e) art (filed 1996-01-19, before the 1997 priority; issued 1999-06-01). It is the only reference that maps onto the combination of independent claims 1, 2, 3, 6, and 8:

• Claim 1 (server): maps to the RCSP HTTP server + HTML file + executable RDM ("remote controlling software") + AIM/PTOM ("remotely controlled software"). Gap: 6286003 requires an "input information receiving part / analyzing part / drawing command monitoring part / drawing command sending part," and the "drawing command monitoring part" is recited as responsive to "the generation of graphics for a graphic user interface desktop." Tridia operates at the application I/O-stream / system-call level and on a single application window, not by monitoring drawing calls for a whole desktop between a graphics engine and display driver. That gap makes strict § 102 anticipation of claim 1 contestable; § 103 is the more defensible ground.
• Claim 2 (server + terminal): strong on the browser/HTML/download/means-for-storing elements; same "desktop drawing-command" gap.
• Claim 3 (method): maps closely to "access server via terminal using server access information (URL) → transmit master remote control program → establish connection between master and slave program → obtain input → send input → generate GUI display info by driving the server GUI → transmit drawing commands → generate corresponding display." Again the "drawing commands for a desktop image corresponding to the GUI" element is the vulnerable gap.
• Claim 6 (means-plus-function) and Claim 8 (program product): same subject matter, same analysis.
• Dependents 4/5/7/9: Tridia discloses the HTML/applet-tag and JAVA-interpreter features squarely (its spec says the RDM "may be identified in an applet tag of a HTML document" and is preferably "written in JAVA language"), so dependent claims 5 (JAVA applet/JVM) and the URL/HTML limitations of 4/7/9 read directly on Tridia.

Bottom line: Tridia is the leading reference. My assessment is that it anticipates at best the browser/applet-download and connection-establishment subject matter and is a powerful § 103 primary reference for the full independent claims, but its application-level (rather than desktop-level) interception likely defeats pure § 102 anticipation of claims 1, 2, 3, 6, 8 as a whole.

---

Reference 6 — US 5,974,441 (Rogers, Lagarde; IBM)

Full citation: U.S. Patent 5,974,441, "WWW client server interactive system method with Java (™)," inventors Richard Michael Rogers and Konrad Charles Lagarde; application 08/761,684 filed Dec. 6, 1996 (CIP of Ser. No. 08/474,577, filed June 7, 1995); issued Oct. 26, 1999; assignee International Business Machines Corp.

Description (verified from full text): A WWW client-server system in which a Web client accesses a web server and downloads HTML and a Java applet; the applet's execution is initiated by a JAVA-capable browser; "the Java applet communicates via TCP/IP sockets to a control program agent that executes on the web server," which in turn drives back-end data-interpretation (DIS) servers to gather/format data and return it for display on the Web client. Claims 30, 34, 36, 37 recite download-and-execute-applet-in-browser; claim 38 recites the applet-communicates-via-TCP/IP-sockets element.

Potentially anticipates: § 102(e) art (filed 1996-12-06, before the 1997 priority — though its earliest priority traces to the June 1995 parent). It discloses the browser-downloads-HTML-then-Java-applet-which-communicates-back-to-the-server architecture that underlies claims 2, 3, 4, 5, 6, 7, 8, 9. It does not disclose GUI desktop capture, drawing-command monitoring/transmission, input-event forwarding for remote control, or an "input information monitoring part / drawing command receiving part" pairing. So it does not anticipate any independent claim in full; it is a § 103 secondary reference for the Java-applet/download/URL elements. It is IBM's own prior art, which is noteworthy but doesn't change the analysis.

---

Reference 7 — US 6,098,092 (Silicon Graphics)

Full citation: U.S. Patent 6,098,092, "Server to dynamically generate graphics for the world wide web," assignee Silicon Graphics, Inc.; filed Nov. 8, 1996; issued Aug. 1, 2000.

Description: Title-based (full text not retrieved — query hit the step limit): a server that dynamically generates graphics for delivery to the World Wide Web. This is relevant to the "server generates GUI/graphics and transmits them to the client" side (claim 3's "generate GUI display information at the server… transmit the drawing commands… to the master remote control program"; claim 1's "drawing command sending part").

Potentially anticipates: § 102(e) art (filed 1996-11-08, before the 1997 priority). It addresses server-side graphics generation for web delivery, an element of independent claims 1, 2, 3, 6, 8, but it does not (by title alone) disclose interactive remote control (input capture and forwarding) or the input-information parts. Cannot anticipate a full claim; useful as § 103 art for the drawing-command/server-graphics element. Flag: title-based inference only — verify specification.

---

4. Non-patent citations (as listed on the face of the patent)

The patent also lists six non-patent items. These are § 102(b) printed publications insofar as their dates precede 1997-04-21, and several are the kind of art an examiner uses against the "browser as a universal GUI" and "remote control over a browser" concepts:

1. "Remote System Console Without a Remote Emulator Program," IBM Technical Disclosure Bulletin, vol. 34, No. 4B, Sept. 1, 1991, pp. 383–386 (XP000189607) — § 102(b); relevant to claims 1/2/3 (remote console without installed remote software).
2. Edward Steinfeld, "Leveraging Browser As Universal GUIs," Electronic Engineering Times, Dec. 16, 1996, Issue 932 — relevant to claims 2/3/4 (browser as the universal GUI / HTML-delivered interface).
3. Jamie Jaworski, JAVA 1.1 Developer's Guide, pp. 124–130 and 330–342 (Jan. 1997) — relevant to claim 5 (Java applet / virtual machine).
4. Mike Fratto, "Remote Control: Let your Browser Do the Walking," Network Computing, Jan. 15, 1998 — this post-dates the 1997 priority, so it is not § 102(b) art against this patent; it could only be relevant as evidence of the state of the art/§ 103 context (and even then its late date limits it).
5. NetSupport Manager Remote Control Software, http://www.deski.com/nsm/index.html, pp. 1–15 (Nov. 1999) — also post-dates the 1997 priority; like Fratto, not § 102(b) art against 6286003.

(One item in the raw record — the fifth — is rendered with a malformed URL, "http//:www.deski.com/nsm/index.html"; I report it literally as it appears rather than correcting it.)

---

5. Direct answer to the task

• Most relevant prior art: US 5,909,545 (Tridia / Frese II et al.) — it is the only cited reference that discloses browser-delivered, on-demand, pre-installation-free remote control of a program over a network via an HTML/applet mechanism, and it maps onto the greatest number of claim elements across independent claims 1, 2, 3, 6, and 8.
• Second tier (browser/applet-download and Java/URL elements): US 5,974,441 (IBM) and US 5,706,502 (Sun) — strong § 103 references for the download-via-HTML/browser and Java-applet/JVM limitations (dependent claims 4, 5, 7, 9).
• Third tier (server-side graphics / display sharing / remote console): US 6,098,092 (SGI), EP 0645726 A2 (NCR), JPH0728736A (Hitachi), and the IBM TDB "Remote System Console" item.
• Background only: US 5,005,122 (DEC) — software distribution over a network; no bearing on any claim as a standalone anticipatory reference.

Important honesty caveats:

1. § 102 vs. § 103. My reading is that no single cited reference is a clean § 102 anticipation of any independent claim (1, 2, 3, 6, 8) as issued, primarily because 6286003's claims require monitoring/sending drawing commands for a GUI desktop (graphics-engine-to-display-driver level), whereas the strongest reference (Tridia) operates at the application I/O-stream/system-call level for a single application. Anticipation is therefore contestable; the better-supported invalidity theory is § 103 obviousness, e.g., Tridia in view of Sun 5,706,502 and/or IBM 5,974,441 for the browser/applet delivery, and SGI 6,098,092 for the server-generated drawing commands.
2. § 102(e) vs. § 102(b). Four references (Sun 5,706,502; Tridia 5,909,545; IBM 5,974,441; SGI 6,098,092) issued after the 1997 priority date and are therefore § 102(e) art only (relying on their pre-1997 U.S. filing dates), not § 102(b) (which requires publication/patenting more than a year before the 1998 filing). US 5,005,122, JPH0728736A, and EP 0645726 A2 qualify under § 102(a)/(b).
3. Unverified texts. Three references (JPH0728736A, EP 0645726 A2, US 6,098,092) and the non-patent items were characterized from the patent's own citation table and titles because my retrieval queries on them hit the step limit. The descriptions for those marked "title-based inference only" and the two post-dated NPL items should be confirmed against the primary documents before being relied on in any validity opinion. US 5,909,545, US 5,974,441, US 5,706,502, and US 5,005,122 were characterized from retrieved primary/full text and are reliable.

I'll ground this obviousness analysis by pulling details on the key prior-art references cited on the patent's face.

Let me gather the remaining cited references, including the non-patent literature and the Japanese prior art.

Obviousness Analysis — U.S. Patent No. 6,286,003 under 35 U.S.C. § 103

Source of record: Google Patents full text, https://patents.google.com/patent/[US6286003](/patent/US6286003)/en. Building on the earlier summary and litigation sections; where those found no assertion activity, this section addresses validity only. Because the application was filed 1998-04-21 with a 1997-04-22 JP priority date, pre-AIA § 102/§ 103 governs (AIA §§ 102/103 do not apply to applications filed before 2013-03-16).

---

1. Controlling dates and prior-art eligibility screen

The operative invention date is the JP priority date, 1997-04-22 (JP 9-104997 / JPH10301874A). Working the statutory categories against that date:

Important flag on the record itself: the '003 patent's face lists Fratto (Jan. 1998) and NetSupport (Nov. 1999) among its citations, but neither can be used as § 103 prior art against claims entitled to the 1997-04-22 priority date. Any obviousness theory built on them is defective on its face. Steinfeld (Dec. 1996), the 1991 IBM TDB, and the Jaworski guide remain usable.

Common-ownership flag: US 5,974,441 is IBM-assigned, as is the '003 patent. Pre-AIA § 103(c) provides that subject matter qualifying as prior art only under § 102(e)/(f)/(g) shall not preclude patentability where the subject matter and the claimed invention were, at the time the invention was made, owned by the same person or subject to an obligation of assignment to the same person. The '441 patent qualifies only under § 102(e) and was IBM-owned in 1997 — so a challenger should treat '441 as probable § 103(c)-disqualified and rely instead on the non-IBM references (Tridia '545, Sun '502, SGI '092, Jaworski) for the same teachings. I note this because it is the kind of defect that gets an obviousness theory rejected if not spotted.

---

2. The primary reference: US 5,909,545 (Tridia) is nearly coextensive with claim 3

US 5,909,545 (https://patents.google.com/patent/[US5909545A](/patent/US5909545A)) discloses, in order:

The abstract confirms the whole architecture: "The remote display module is transported across the network and executed on the user system in response to a user's request to provide on-demand remote control of an application program," and "Preferably, the remote display modules and application programs are presented through HTTP servers over a network to a user's system which uses a browser having a JAVA interpreter to execute the remote display module."

This maps to claim 1's four-plus-four functional partition (input receiving / input analyzing / drawing-command monitoring / drawing-command sending on the server; input monitoring / input sending / drawing-command receiving / drawing-command analyzing on the client), to the HTML file (the RCSP's HTML documents with tags for requesting the program), and to the HTTP daemon (the RCSP is "preferably a HTTP server").

The gap closes to one issue: Tridia frames the target as an application program whose window is reproduced at the user, whereas claims 1, 2 and 3 recite a "graphic user interface desktop" at the server and a "desktop image." That, plus claim 1's recitation that the drawing-command monitoring part is "responsive to the generation of graphics for a graphic user interface desktop," is the only meaningful distinction. It is a layer-of-interception distinction, not a functional one — and it is exactly what the secondary references supply.

---

3. Combinations that render the claims obvious

Combination 1 — Tridia '545 + Sun '502 + Jaworski guide (and, if it survives § 103(c), IBM '441)

• Sun '502 (https://patents.google.com/patent/[US5706502A](/patent/US5706502A)) teaches an application whose integrated Web browser transparently downloads remote components — including Java applets and "remote applets" — so that local and remote projects are handled identically ("location-transparency"). Its EP counterpart, EP 0 798 635 B1, states the motivation explicitly: before Java applets there was "no way to provide platform-independent application programs over the Internet," and the need was for "an Internet-enabled programming environment that allows a user to assemble programs from components that are distributed on diverse Internet nodes."
• IBM '441 teaches the applet-as-control-client architecture: "The Web client accesses a web server and downloads HTML and a Java applet(s). … The Java applet communicates via TCP/IP sockets to a control program agent that executes on the web server." Its glossary frames Java as programs that "can be downloaded from a web server and run on any client machine that has a browser containing the interpreter."
• Jaworski (Jan. 1997) establishes that applet loading, the Java VM, and applet–server socket programming were routine by the critical date.

Mapping: Tridia supplies every step of claim 3 and every part of claim 1; Sun '502/'441 supply the express, same-field teaching of delivering the client module from an HTML document over HTTP/URL and executing it in the browser, which Tridia only describes functionally ("browser having a JAVA interpreter"). Tridia's own express statement that the RDM is transported to and executed on the user system via HTTP servers and a Java-capable browser makes the combination a substitution of known equivalents, not a new architecture.

Motivation (KSR factors): same field (network remote control/remote display); same problem (eliminating pre-installation of platform-specific remote-control software and its version-management burden — the '003 Background itself recites exactly this problem); predictable combination of two known browser/HTTP techniques; and a reasonable expectation of success because Java applet delivery over HTML/HTTP was demonstrated and documented by January 1997.

Combination 2 — Tridia '545 + SGI '092 (desktop/graphics generation at the server and transmission over the Web)

US 6,098,092 (https://patents.google.com/patent/[US6098092A](/patent/US6098092A)) teaches: "Upon receiving a request from an end user over the web, a server executes a program stored in its memory … Next, the server executes a second program … to generate computer graphics … The server sends the computer graphics generated over the World Wide Web to the end user," triggered by a hypertext icon on an HTML page and implemented with HTTP/CGI. This supplies the server-side generation of display content and its network transmission to the requesting client, and reinforces that the browser/page-plus-server-execution model was the ordinary design choice.

Motivation: SGI solves the same "server generates what the client displays, delivered through a Web page" problem in the graphics domain; a POSITA seeking to extend Tridia's application-level remoting to whole-desktop remoting would naturally adopt a server-side graphics-generation-then-transmit model. One of ordinary skill would expect success because both references operate on HTTP and standard browsers.

Combination 3 — Admitted Japanese prior art (Hei 8-1624 and/or Hei 7-34188) + Tridia '545 + browser/applet art

This is the most direct route to the desktop-level limitation, because the patent itself concedes what these references teach:

• Hei 8-1624 is described in the '003 Background as teaching "not only can a program run on different OSs, but also GUI sources are collectively displayed on a screen of a computer system at a remote location."
• Hei 7-34188 is described as teaching "a computer system for supporting multi-graphics sessions comprising a central processor, a graphic terminal and a display controller processor at a remote area."

These are admissions, in the applicant's own specification, that remote GUI display (including at the display-controller level, i.e., below the graphics engine) and cross-OS portability were known. Combined with Tridia's on-demand, browser-delivered module, the desktop-remoting limitation is met by admitted art or is at most an obvious design choice among a finite set of interception points (application/system-call boundary per Tridia; display-driver boundary per the admitted art; window level per JPH0728736A).

Motivation: the admitted art supplies the express suggestion to remote the whole GUI screen rather than a single application window — precisely the difference the '003 patent would otherwise rely on — and Tridia supplies the mechanism for getting the controller module onto the terminal with no pre-installation.

Combination 4 — EP 0645726 A2 (NCR) + JPH0728736A (Hitachi) + US 5,005,122 (DEC) + browser/applet art

• EP 0645726 A2 — "Annotation of the displays of a multiple computer system" — teaches coordinated display handling across multiple computers, supporting multi-machine display management.
• JPH0728736A (Hitachi) — "Multi-window control method" — teaches window-level control, supporting the proposition that controlling/reproducing windowed GUI content across a boundary was known.
• US 5,005,122 (DEC) — "Arrangement with cooperating management server node and network service node" — teaches the cooperating server-node/net-service-node split, i.e., a client/server functional division of the kind claim 1 recites (server-side remotely controlled software + client-side cooperating software).

Motivation: these references are in the same field of remote/distributed computer management and multi-display control; a POSITA seeking to implement interactive remote control of a GUI over a network would look to them for the display/window-side mechanics and to Tridia/'441/Sun '502 for the browser-delivered module.

---

4. Dependent claims

• Claims 4, 7 and 9 (URL specifying an HTML document associated with the resource to be remotely controlled): taught expressly by Tridia (RCSP is "preferably a HTTP server displaying HTML documents"; HTML tags are used to request the target program), by IBM '441, by Sun '502 and by SGI '092 — all URL/HTTP-based. No separate inventive weight.
• Claim 5 (master program is a JAVA applet executed by a JAVA virtual machine): Tridia states that "remote display modules and application programs are presented through HTTP servers … to a user's system which uses a browser having a JAVA interpreter to execute the remote display module." Sun '502 uses Java applets/"remote applets"; IBM '441 uses a Java applet over sockets; Jaworski documents the VM. Claim 5 is the weakest claim in the set.
• Claim 8 (computer program product): purely a change of claim format from claim 3; if claim 3 is obvious, claim 8 adds nothing but the medium.

---

5. Anticipated patent-owner arguments and responses

1. "Tridia remotes an application, not a GUI desktop." — Answered directly by the applicant's own admitted art (Hei 8-1624's "GUI sources … collectively displayed on a screen of a computer system at a remote location"; Hei 7-34188's display-controller processor at a remote area) and by the '003 specification's own statement that the drawing-command conversion "is not essential, and data generated by the Graphics engine 321 and sent to the display driver 325 may be sent unchanged." That concession shows the driver-boundary placement is a conventional implementation choice, not an inventive contribution.
2. "Different problem / teaching away." — Weak: both Tridia and the '003 patent attack the identical problem (avoiding pre-installed, platform-specific remote-control software), and the '003 Background's stated objectives (simplify upgrades, eliminate installation/management, platform independence, reduce client resource use, reduce exchanged data volume) are the same advantages Tridia already claims ("on-demand," transported module, protocol translation and optimization module "used to reduce the latency … associated with networks").
3. "§ 103(c) / common ownership." — This argument helps the patent owner only as to IBM '441; it does not reach Tridia '545, Sun '502, SGI '092, or the admitted Japanese art, all of which supply the same teachings. Note the irony: the same § 103(c) provision that might shield the '441 reference also reflects that IBM itself held both the reference and the application.
4. Objective indicia. — None of record. There is no evidence in the prosecution history or public record of commercial success, licensing due to the merits, copying, or a long-felt-but-unmet need with a nexus. The patent expired 2018-04-21 (Expired – Fee Related), so no secondary-considerations record will ever be developed.

---

6. Bottom line

• Strongest theory: Tridia '545 as the primary reference, alone or in combination with Sun '502 and the Jaworski guide, renders claims 1, 2, 3, 6, 8 (and 4, 5, 7, 9) obvious. Tridia discloses the HTML/HTTP-delivered client module, the on-demand transport-and-execute sequence, the connection and activation handshake, bidirectional input forwarding, server-side interception of application output, and conversion to/from a remote-control protocol — the entire claim 3 sequence and the entire claim 1 component set.
• The only material gap is the "GUI desktop" (as opposed to single-application-window) framing and the placement of interception between the graphics engine and display driver. That gap is filled by the applicant's own admitted Japanese prior art (Hei 8-1624, Hei 7-34188), by SGI '092 (server-side graphics generation, transmitted over the Web), by JPH0728736A and EP 0645726 (window/multi-display control), and by the specification's own concession that the conversion step is optional.
• Dependent-claim exposure is severe: claim 5 (Java applet + Java VM) is expressly disclosed by Tridia; claims 4/7/9 (URL/HTML) are disclosed by four references.

Caveats I could not eliminate: I was unable to retrieve the full text of the 1991 IBM TDB "Remote System Console Without a Remote Emulator Program," the Steinfeld EETimes article (1996-12-16), JPH0728736A and EP 0645726 A2, or the two Japanese examined publications (Hei 8-1624, Hei 7-34188) within the search budget. My characterizations of Hei 8-1624 and Hei 7-34188 rest on the '003 specification's own description of them (which is itself an admission and, for § 103 purposes, usable as such) and on their titles/dates. Before filing any challenge, the full texts of these references, and the '003 file wrapper, should be pulled directly. Verification research also surfaced additional, non-cited candidate art a challenger would likely add — e.g., US 5,838,910 (Domenikos, filed 1996-03-14), US 5,757,925 (1996-07-23), US 5,828,840 (1996-08-06), US 5,944,791 (1996-10-04), and the G&H Nevada-Tek/Hickman family (priority 1996-02-16, including US 7,130,888 and US 8,219,849) — none of which I have examined on the merits here.