US 5867633

I'll search for authoritative information on this specific patent.

Search results: US 5,867,633

Note on the search: I searched for the exact number "5867633" / "5,867,633" against patent databases and Federal Circuit docket material. No CAFC docket, district court case, IPR/PGR proceeding, or 2026 appellate activity for this patent was found. The only substantive results returned were the patent document itself (Google Patents), a uspto.report grant page, and — importantly — a Japanese patent JP 5867633 B1 (NEC, 2016, heap-memory measurement), which is a different, unrelated patent that happens to share the digits. Per your instruction, I am not conflating them. I also could not open USPTO PatentCenter or the CAFC docket system directly in this session, so my litigation/status statements rest on search evidence plus the authoritative text you supplied.

---

Bibliographic data (from the patent text provided)

Uncertainty flag: the assignee name appears in two variants in the record ("Hewlett Packard Development Co LP" vs. "Hewlett-Packard Development Company, L.P."), reflecting the 2001 merger recordation and the 2011 assignment to HP Development Company, L.P. I did not independently verify the chain of title in USPTO Assignment records in this session.

---

Abstract (as issued)

An "image server" is the central clearinghouse for image work. Image files are registered with it; image processing requests are sent to it to produce processed image data. In the printing embodiment, a document processing application obtains a file for printing, registers it with the image server, and receives back an identifier for the file. The application initiates a print job by sending that identifier plus selected print parameters to a printer driver. The printer driver then negotiates image processing requests with the image server to modify the image per the print parameters, retrieves the processed image data, and sends it to the printer. The image server and printer driver may be separate, independently executing processes, and may be distributed across networked computers.

The stated technical problem being solved: prior approaches (device-independent bitmaps, or driver-capability querying) pass huge quantities of image data — e.g., ~20 MB for a 300‑dpi 8"×10" 24‑bit color photo, ~80 MB at 600 dpi — through multiple process address spaces and repeatedly to/from disk, including under Windows '95 EMF spooling. The invention's insight is to pass a small handle/identifier instead of the image itself, and to let the image server access only the portions of the image actually needed.

---

Independent claims in plain language

Claim 1 — Core image-processing method (not limited to printing)

1. Obtain a digitized image.
2. Specify transformation parameters describing the desired appearance of that image.
3. Register the digitized image with an image server.
4. Receive back an image identifier handle from the image server in response to registration.
5. Send image processing requests to the image server so it produces image data matching the transformation parameters — and crucially, you identify the image to the server by sending the handle, not the image data. This yields the desired appearance.
6. Retrieve the processed image data from the image server.

The novelty hook is the register-→-handle-→-request-by-handle-→-retrieve loop that avoids shipping image pixels between processes.

Claim 11 — Printing system architecture (apparatus claim)

A system for printing documents from a document processing application, comprising:

• A first memory store holding documents, each with a "document locator" specifying its location.
• An image processing server, itself an independently executing program unit, with means for generating a document identifier when handed a document locator.
• Means for initiating a print job: pass a document locator to the image server → get back the corresponding document identifier → pass job parameters (print parameters plus that identifier) to the printer driver.
• A printer driver that manipulates image data of the identified document by sending image processing requests to the image server per the print parameters, obtains the manipulated image data back from the image server, and transmits it to a printer of a first kind.

This is the claims-level capture of the "printer driver drives the image server via a handle" architecture.

Claim 16 — Multitasking/process-based printing method

A printing method where the document server and the printer driver are each loaded into RAM and executed as concurrent, independent processes:

1. Load a first user application (in RAM).
2. Load and execute a document server process concurrently with and independently of the application.
3. Load and execute a printer driver process concurrently with and independently of both.
4. The application selects a first document and specifies first print parameters; the document has a first document identifier used to access it.
5. The application sends that identifier to the document server, and the server returns a first document handle to the application.
6. The application issues a print job by sending the handle and the print parameters to the printer driver process.
7. The printer driver uses the handle to send image processing requests to the document server per the print parameters, causing the server to generate image data of the desired appearance.
8. The document server sends that image data to the printer driver; the printer driver sends it to either a printer or a storage device for subsequent printing.

Claim 16 is the most specific of the three independents — it hard-codes the three-process model (application / document server / printer driver) and the two-hop handle exchange (identifier → handle).

Dependent claim highlights: claims 2–10 and 12–15 add concurrency of the application with the server (cl. 2/7), multiple applications (cl. 3/17), client–server distribution across computers (cl. 4/9/10/13/14), file-name vs. RAM-pointer references (cl. 5/6), multiple printer drivers/types (cl. 8/15/18), document locking by the server (cl. 19), and the temporary-copy-then-unlock mechanism that preserves access to the original (cl. 20).

---

Enforcement / litigation status (2026)

On the evidence available in this session: there is no CAFC 2026 docket, and no identified district court or PTAB proceeding, involving US 5,867,633. The patent's recorded legal status is Expired – Lifetime, with anticipated expiration 2016-12-09 (20 years from filing; no term extension is indicated in the record). Its expiration means it is not a candidate for current infringement assertions. Those claims about absence of litigation are negative findings from search, not affirmative confirmations from the CAFC/USPTO docket systems — if you need a certified negative, that should be confirmed directly against the CAFC docket and PTAB APIs, which I could not query here.

Main uncertainty to flag: the "Cited By (41)" summary versus the expanded "Cited By (159)" list in the source suggests the forward-citation count is inconsistent in the record; I did not resolve which is correct.

I'll search for litigation involving US patent 5867633.

Note: those results are for JP5867633 B1 (a Japanese patent), which is a different patent and not relevant. Let me search specifically for US5867633.

No litigation hits yet for US5867633. Let me search litigation-specific databases.

Litigation Search Report: US Patent 5,867,633

Bottom line

I found no litigation involving US Patent 5,867,633. No district court complaints, ITC Section 337 investigations, PTAB proceedings (IPR/PGR/CBM), or Federal Circuit appeals asserting or challenging this patent appeared in any of my searches.

What I searched and what I found

Important caution on identifiers: A recurring false positive is JP5867633 B1 (Japanese Patent No. 5867633, NEC Corporation, registered 2016-01-15, published 2016-02-24). That is not US 5,867,633 and must not be conflated with it. To my knowledge it has no relation to the HP patent. Note also that the JP number and the US number are numerically identical only because the "5867633" string coincides — they are separate national rights.

What the record does show about US 5,867,633

From the Google Patents file (https://patents.google.com/patent/US5867633/en), which contains no litigation data:

• Title: Method and apparatus for processing and printing documents
• Patent number: US 5,867,633 A; Application US08/767,126
• Inventors: Robert B. Taylor III; Kirt Alan Winter; Robert Chou; Sachin Naik
• Original assignee: Hewlett-Packard Company → later Hewlett-Packard Development Company, L.P.
• Priority/filing date: 1996-12-09; granted: 1999-02-02
• Legal status: Expired – Lifetime; anticipated expiration 2016-12-09
• Family: EP 0847002 B1; DE 69712592 T2; JP H10248014 A
• Citations: 11 references cited by examiner (e.g., US 5,220,674; US 5,303,336; US 5,493,408; US 5,559,933; US 5,579,087; US 5,580,177; US 5,583,949; US 5,586,241; US 5,617,518; US 5,619,649; US 5,699,495); cited by 41 (Google's summary list) and 159 in the extended "Cited By" table
• Prosecution/administrative history: assignment records (1997-06-02; 2001-01-16; 2011-09-22) and expirations — no PTAB case listed

The complete absence of any entry in the "Cited By" and family tables for a litigation asserting the '633 patent is consistent with the patent never having been asserted.

Confidence and limitations (stated explicitly)

• Confidence that no litigation exists: moderate-to-high, but not absolute. I could not complete the two most probative searches (CourtListener / RECAP docket text and PACER), having hit the tool-step limit. Docket text searches for "5,867,633" on CourtListener, Docket Alarm, and Justia Dockets are the definitive next steps.
• My search did not return a hit on the Unified Patents litigation case list (https://portal.unifiedpatents.com/litigation/caselist), a CAFC opinion, or a PACER docket referencing the patent. A null result across those sources is evidence — but not proof — of no assertion.
• Practical reasons this outcome is unsurprising: the patent is now past its 2016-12-09 anticipated expiration, claims general client/server print-driver architecture (image server + handle-based registration + printer driver negotiation), and its whole term appears to have passed without public assertion. That is consistent with a defensive portfolio patent.

Recommended verification steps (for a definitive answer)

1. CourtListener / RECAP — full-text docket search for "5,867,633" and "5867633" (free, covers PACER-derived dockets).
2. PACER Case Locator and USPTO Patent Assignment Search — confirm no security-interest/license transfers to an NPE (e.g., a PAE assignment would raise the odds of later assertion).
3. Unified Patents litigation portal and Docket Alarm — check for any case bringing in the '633 as a counterclaim or in a declaratory-judgment action.
4. ITC EDIS — search Section 337 investigations naming the patent.

If you'd like, I can run those targeted docket-text searches next (CourtListener and the Unified Patents caselist) to convert the "not found" into a documented negative result.

I'll verify the structured "no proceedings on file" default against public sources before writing anything up.

No hits. Let me run a few more targeted checks before concluding, including the EP-family opposition angle (non-PTAB but adjacent) and a citation check to see whether any IPR ever cited this patent.

Two useful signals surfaced: (1) the '633 patent appears as an X-category prior-art reference in a different AIA-era family (PCT/US01/43796, Flexiworld), and (2) the Unified Patents page for the EP family member shows an "Opposition Documents" section. Let me check both — the EP opposition is the only post-grant challenge activity I've seen anywhere in this family.

I've completed my verification against the canonical structured data plus independent search. Note: I hit the tool-step limit before running the last two corroborating queries (PTAB E2E / Docket Alarm docket-text), so the negative finding rests on the ODP block plus a null result across the searches I did complete. That limitation is carried through below.

---

PTAB / AIA Trial Report — US 5,867,633

Proceedings overview

Total AIA trial proceedings on file: 0. The canonical USPTO Open Data Portal "PTAB proceedings on file" block for this patent returns no AIA trial proceedings as of the most recent ingest: zero IPRs, zero PGRs, zero CBMs, and no derivations. Accordingly the breakdown by status is 0 active / 0 claims invalidated / 0 claims sustained / 0 settled / 0 institution denials.

Bottom line for a defendant: there is no PTAB history to build a defense on — and, more importantly, nothing to worry about either. All 20 claims (independent claims 1, 11, 16, and dependents 2–10, 12–15, 17–20) are UNTESTED at the Board; there is no final written decision, no cancellation, no estoppel, and no surviving-claim list to work from. The patent's recorded legal status is Expired – Lifetime, anticipated expiration 2016-12-09, so the practical posture is: no live enforcement exposure, and no § 315(e)(2) estoppel constraining anyone's invalidity case. This is an unasserted, expired, pre-AIA portfolio patent — not a hardened or adjudicated one.

One structural point worth stating plainly: the absence of AIA proceedings here is not evidence that the patent is strong. It is the expected outcome for a patent that (a) issued 1999-02-02, well before the AIA created trials on 2012-09-16; (b) was already 13 years old and nearing the end of life when IPRs became available; and (c) expired 2016-12-09. The absence of IPRs is best read as an artifact of timing and commercial irrelevance, not as a robustness signal.

---

Why there is nothing to report (and what I checked)

Because no proceedings exist, the per-proceeding template has nothing to populate. Instead, here is the verified negative and the false positives a defendant will almost certainly trip over.

False positives you must not conflate with this patent

These are the recurring traps; each contains a "633" or a similar string but is a different matter:

1. JP 5867633 B1 — Japanese Patent No. 5867633, NEC Corporation (inventor 式見 遼), application 特願2015-9195, registered 2016-01-15, published 2016-02-24. Claims 1–10 directed to heap-memory usage measurement using weakly-referenced objects. Numerically identical string, wholly unrelated right. Do not cite this as "the '633 patent."
2. 37 C.F.R. § 1.633 ("Rule 633") — appears throughout the Bilstad v. Wakalopulos interference materials and CAFC appeal (e.g., CAFC No. 03-1528, https://cafc.uscourts.gov/opinions-orders/03-1528.pdf). That is a pre-AIA interference rule, not this patent.
3. IPR2013-00309 (HP v. MPHJ) — the FWD addresses U.S. 6,771,381 ("Virtual Copier"), discussed at https://ptabtrialblog.com/wp-content/uploads/2014/11/IPR2013-00309-FWD-20141119.pdf. Unrelated patent; no relation to the '633.
4. IPR2021-01413 (U.S. 10,621,228) — surfaced in the earlier search pass; unrelated.
5. Bates-number collisions — the Memjet/HP joint claim-construction chart contains the exhibit range "HP-SD0007567–7663." That is a document-production number, not a patent or proceeding (https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1463732](/patent/1463732)/download-documents).

A genuinely relevant finding (context, not a proceeding)

The '633 patent appears as an X-category prior-art reference in the prosecution of a different family: PCT/US01/43796 (the Flexiworld / William Ho Chang work). The International Preliminary Examination Report lists:

"X — US 5,867,633 A (TAYLOR, III et al.) 02 February 1999 … abstract, Figures 1 & 4, column 1, lines 59-67, … column 8, lines 10-20." — relevant to claims 42-43, 45-48, 164-171, 195; and cited against claims 1-41, 44, 140, 144, 161, 162, 195.

Source: https://ptacts.uspto.gov/ptacts/public-informations/petitions/[1549999](/patent/1549999)/download-documents (IPR petition exhibit containing the IPER).

This is the one place in the record where the '633 shows up offensively — as art against someone else's claims, in a patent family that later generated a very large IPR/patent portfolio. It confirms the '633's technical significance as a disclosure, while confirming it was never itself the target of an AIA trial.

---

Strategic summary

Claim-level status. Every claim of US 5,867,633 — independents 1, 11, 16 and all dependents 2–10, 12–15, 17–20 — is UNTESTED at the PTAB. Nothing is canceled; nothing has been sustained in an AIA trial; there is no FWD to quote. Any statement that a specific claim was "upheld" or "invalidated" by the Board would be fabricated, and I am not making one.

Estoppel landscape. Because no IPR or PGR reached a final written decision, no § 315(e)(2) estoppel exists against any party — not against HP, not against any competitor or aggregator. There is therefore no closed-off set of prior-art grounds. This is the mirror image of the usual defendant briefing: ordinarily you ask "which grounds did the petitioner burn?" Here the answer is none, but the question is largely academic given expiration.

Structural availability of AIA relief. None of the three trial types was ever a realistic avenue:

• PGR — available only for post-AIA patents and only within 9 months of issuance; this patent issued 1999-02-02 on a 1996-12-09 pre-AIA filing, so the window closed long before trials existed.
• CBM — limited to covered business method patents tied to financial products/services, and the program sunset on 2020-09-16. A print/image-server architecture patent would not qualify on subject matter in any event.
• IPR — technically available against expired patents, but there is no live infringement exposure to defend against after 2016-12-09, so there is no rational petitioner incentive.

Pattern signals. No petitioner has ever filed against this patent (so no repeat-petitioner pattern); no patent-owner appeal to the Federal Circuit from any PTAB decision exists (there is no decision to appeal); and there is no defensive aggregator in the chain — no Unified Patents or RPX challenge appears anywhere. The patent is a straightforward defensive portfolio asset of HP's image/print software heritage.

Flagged data discrepancy (carry-through). The Unified Patents page for the EP family member lists priority 1996-12-08, application 1997-11-17, and grant 1998-06-09 (https://portal.unifiedpatents.com/patents/patent/EP-0847002-A1). The authoritative record in the patent text states priority/filing 1996-12-09 (US) and EP filing 1997-11-18. I cannot reconcile these from available evidence; treat the Google Patents / ODP dates as authoritative and the Unified Patents dates as suspect. This also means its "Opposition Documents" header cannot be relied on as evidence of an actual EPO opposition.

---

Recommended next steps

If you are a defendant or accused infringer. No PTAB-specific action is warranted, and none is available in practice. Run the ordinary expired-patent checklist instead:

1. Confirm the expiration date in USPTO PatentCenter for application 08/767,126 — if 2016-12-09 holds with no term extension (no § 154(b) adjustment or PTA is indicated in the record), any demand covering post-expiration conduct fails outright, and the damages window closed in 2016.
2. Confirm no reissue or reexamination certificate altered the claims. The record shows none, but this is a two-minute PatentCenter check and is the only way an expired patent's claims could have changed shape.
3. If a demand letter cites specific claims, note that all 20 are UNTESTED — there is no FWD to point to. Your invalidity case is unconstrained by estoppel, but note that with the patent expired the more efficient response is typically a damages-period/standing analysis, not an invalidity fight.

If you need a certified negative result. My finding is sourced from the canonical ODP block plus a null web-search result; I did not directly query PTAB E2E. To convert it to a documented negative:

• USPTO PTAB E2E (https://ptacts.uspto.gov/ptacts/) — search by patent number 5,867,633 and by application 08/767,126.
• Docket Alarm / Unified Patents litigation & PTAB portal — cross-check for any proceeding where the '633 appears as a counterclaim patent.
• CourtListener (https://www.courtlistener.com) — docket-text search for "5,867,633" to confirm no FWD was ever appealed (a CAFC appeal of an FWD would be the one place a proceeding could hide from the ODP summary).

Absence-is-a-signal caveat, stated explicitly: ordinarily, "no IPRs" is weak evidence of patent strength. Here it is strong evidence of irrelevance — a 1999 patent that expired in 2016 and was never asserted is simply outside the population of patents that attract AIA challenges. Do not let anyone characterize the clean PTAB record as "the patent survived challenge." It was never challenged, and no challenge was ever realistically available.

I'll reconstruct the assignment record for US 5,867,633 using the patent's own legal-events data first, then try to pull the corresponding USPTO Assignment Center records (reel/frame, correspondent).

Assignment / Ownership Reconstruction: US 5,867,633

Data-availability caveat up front (do not skip): I could not open USPTO Assignment Center / assignment.uspto.gov in this session, and my searches for the reel/frame numbers and correspondent of record for this specific patent returned no authoritative hit (the "5867633" string keeps surfacing the unrelated JP 5867633 B1, NEC heap-memory patent — not conflated here). Accordingly, the timeline below is built from Google Patents legal-events records supplied with this patent and from EP counterpart registration history, and I have marked every reel/frame and correspondent as not retrieved rather than guessing. I did not fabricate any recording data.

Inventors

All four are named as assignors on the 1997-06-02 inventor→HP assignment record, which is consistent with (but does not by itself prove) each being an HP employee obligated to assign. The patent's own specification is written from the perspective of an HP printer-driver/imaging development team (Windows '95 ExtEscape(QUERYESCSUPPORT, IS--SUPPORT), StretchDIBits() implementation detail).

Unusual-pattern check: I found no evidence of all inventors departing HP within 12 months of filing, and no evidence of a subsequent fire-sale. I cannot affirmatively rule this out — employment/exit data for these individuals was not surfaced in this session. Treat this as unclear, no adverse finding.

Original assignee

Hewlett-Packard Company (the pre-2015 operating company; Delaware corporation, Palo Alto, CA at the relevant time — 3000 Hanover Street appears in HP's contemporaneous assignment records).

• Primary line of business: computers, printers, imaging/peripherals — i.e., the exact product line the patent's "image server / printer driver" invention serves.
• Shipped a product embodying the claims: Yes, plausibly. The preferred embodiment is explicitly a Windows '95 HP printer-driver design (image server + handle registration + RegisterImage/ConfigureImage/RetrieveImageBand driver calls). HP's printer drivers of that era are the natural commercial embodiment. I did not verify a specific shipping SKU in this session — flagging that as inference, not a documented product finding.
• Current status: The original "Hewlett-Packard Company" underwent a corporate split effective 2015-11-01 into HP Inc. (personal systems/printing — still operating) and Hewlett Packard Enterprise Company. The patent itself, however, does not sit with either operating successor; it sits with the IP-holding affiliate (see timeline). HP Inc. remains an operating company.

Assignment timeline

Because Assignment Center was unreachable, I list the records from the patent's legal-events data. Execution dates and reel/frame numbers were not retrieved — the entries below are recorded-date entries only, and I will not invent reel/frames.

1. Recorded 1997-06-02 — Reel/Frame not retrieved

   • Conveyance: Assignment of Assignors' Interest
   • Assignor: Robert Chou; Sachin Naik; Kirt Alan Winter; Robert B. Taylor, III
   • Assignee: Hewlett-Packard Company
   • Correspondent: not retrieved (HP assignment filings of this era were commonly handled through HP's Intellectual Property Administration, Fort Collins, CO — see note below)
   • Context: Original employment assignment — invention captured by the operating company pre-issuance.

2. Recorded 2001-01-16 — Reel/Frame not retrieved

   • Conveyance: Merger "(see document for details)"
   • Assignor: Hewlett-Packard Company
   • Assignee: Hewlett-Packard Company
   • Correspondent: not retrieved
   • Context: Internal corporate reorganization — same-name assignor/assignee pair, i.e. a structural merger within the HP corporate family, not a change of beneficial owner.

3. Recorded 2011-09-22 — Reel/Frame not retrieved

   • Conveyance: Assignment of Assignors' Interest "(see document for details)"
   • Assignor: Hewlett-Packard Company
   • Assignee: Hewlett-Packard Development Company, L.P. (Houston, TX — 20555 S.H. 249)
   • Correspondent: not retrieved
   • Context: Internal transfer to the IP-holding affiliate — HP parks portfolio patents in its Texas limited-partnership IP vehicle. No consideration to an outside party is indicated.

Correspondent note (flagged as inference, not finding): I could not capture the correspondent of record for any of the three entries above. Two related but different HP patents show HP's recordings carrying the correspondent "Hewlett-Packard Company, Intellectual Property Administration, P.O. Box 272400, Fort Collins, CO 80527-2400" (plainsite assignment id 3242709). If the same correspondent appears on the '633 reels, it is an in-house HP IP-administration correspondent — the opposite of the repeat-player-NPE-attorney pattern. This is a hypothesis to verify at Assignment Center, not a documented fact for this patent.

Contradiction to flag: The EP counterpart EP 0847002 B1 records the transfer to Hewlett-Packard Development Company, L.P. as made by deed of assignment dated 2003-01-31 (recorded in the UK register 2011-12-08). The US record shows the corresponding assignment recorded 2011-09-22. So the effective date of the HP→HP Development transfer (2003) and the recordation date (2011) diverge. If the Assignment Center reel for the US '633 shows an execution date circa 2003-01-31, that reconciles the two and confirms a single bulk 2011 recordation event covering many HP patents. I could not confirm this.

If Assignment Center returns nothing beyond the original inventor assignment, that would itself be consistent with what is shown here (a purely internal chain) — but the legal-events data indicates at least three recorded events exist, so I do not treat the record as empty.

Timeline diagram

timeline
    title Ownership of US 5867633
    1996 : Application filed
         : Four inventors employed at HP
    1997 : Inventors assign to Hewlett-Packard Company
    1999 : Patent issued
    2001 : Internal merger within Hewlett-Packard Company
    2011 : Assigned to Hewlett-Packard Development Company LP
    2015 : HP splits into HP Inc and Hewlett Packard Enterprise
    2016 : Patent expired

NPE / troll-pattern signals

Verdict

Defensive / non-asserting.

The full recorded chain — inventors → Hewlett-Packard Company (1997-06-02) → same-name HP merger (2001-01-16) → Hewlett-Packard Development Company, L.P. (2011-09-22) — is an unbroken series of internal HP-family transfers with no operating-company→NPE conveyance, no known asserter, and no pre-suit assignment; combined with the asserted expiration on 2016-12-09 and the absence of any litigation found in the prior sections, the patent functioned as a defensive portfolio asset. Note the classification nuance: the terminal assignee (HP Development Company, L.P.) is a product-company IP affiliate, not a defensive aggregator in the RPX/AST sense — I selected "Defensive / non-asserting" because it is the only category fitting a chain that stays in-family and is never asserted.

Verify at: USPTO Assignment Center — https://assignmentcenter.uspto.gov/ (search patent 5867633 or application 08/767,126) — to capture the reel/frame, execution dates, and correspondent of record that I could not retrieve here. That single step would convert signals #1 and #3 above from inference to documented finding.

I'll research the prior art references cited against US 5,867,633. Let me run targeted searches.

Let me research the remaining cited references.

Prior-Art Analysis — US 5,867,633 (Method and apparatus for processing and printing documents)

Note on tool limits: I hit the search step-limit before retrieving full text for all six of the less-shelf‑prominent references (US 5,580,177; 5,583,949; 5,586,241; 5,617,518; 5,619,649; 5,699,495). For those I rely on the authoritative bibliographic data in the patent's own "Citations (11)" table plus titles/abstracts, and I flag confidence explicitly rather than overstating. The four references I did open (US 5,220,674; 5,303,336; 5,559,933; 5,579,087) are described from their actual text.

---

1. USPTO record confirmation (patent number 5867633, interpreted literally)

Searching USPTO/patent-database sources for the exact number 5867633 in the US series returns:

• US 5,867,633 A, "Method and apparatus for processing and printing documents," App. No. US 08/767,126, filed/priority 1996-12-09, granted 1999-02-02, assignee Hewlett-Packard (later Hewlett-Packard Development Co. L.P.), 20 claims, now Expired – Lifetime (anticipated expiration 2016-12-09).

Do not confuse with JP 5867633 B1 (NEC, heap-memory measurement, registered 2016) — a numerically identical but wholly unrelated Japanese right. Per instructions I treated the digits literally and did not merge these records.

---

2. Analytical framework

US 5,867,633 is a pre‑AIA patent (filed 1996-12-09, before the AIA's first-inventor-to-file change), so §102 is assessed under the pre‑AIA subsections:

• §102(a) – reference patented/knowledge/publication before the applicant's invention.
• §102(b) – patented or described in a printed publication more than one year before filing (i.e., before 1995-12-09).
• §102(e) – a US patent granted on an application filed before the applicant's invention/filing, effective as of that earlier US filing date. This is the key hook for the several references that issued after the '633 filing date but were filed years earlier.

Because the '633 filing is 1996-12-09, references with US filing dates well before that can still be §102(e) art even though their grant dates fall in 1997.

The three independent claims whose elements must be matched are claim 1 (register-image → receive-handle → request-by-handle → retrieve), claim 11 (system: memory store + independent-process image server + printer driver that drives the server via the identifier), and claim 16 (three concurrent processes + two-hop identifier→handle exchange). Dependents add concurrency (2,7), multiple apps (3,17), distribution (4,9,10,13,14), file‑name vs RAM pointer (5,6), multiple drivers (8,15,18), and locking/temp-copy (19,20).

---

3. The 11 examiner-cited references

3.1 US 5,220,674 — Digital Equipment Corp.

• Full citation: US 5,220,674 A, Morgan et al., "Local area print server for requesting and storing required resource data and forwarding printer status message to selected destination," Digital Equipment Corp.
• Dates: Continuation of App. 07/075,047 filed 1987-07-17; own App. 07/467,327 filed 1990-01-18; granted 1993-06-15.
• Description (verified from text): A LAN print server sits between host nodes and a printer. It receives print requests, holds a resource directory, and when the printer issues a resource-fault message, the resource manager either supplies the resource locally or fetches it from a remote resource server and forwards it to the printer; a status collector routes printer-status messages to destinations.
• §102 assessment: Discloses a server-side print architecture on a LAN with resource lookup by directory — i.e., an early "server handles resources for the printer" model. It does not disclose (i) an application registering a document with an image server, (ii) the server returning a handle/identifier to the caller, or (iii) a printer driver driving the image server via that handle in accordance with print parameters. Accordingly it does not anticipate claims 1, 11, or 16 on its own; its role is §103 background for the general client/server print-server environment (relevant to the "image processing server is an independently executing program unit" facet of claim 11 and the process-split of claim 16).

3.2 US 5,303,336 — Hitachi, Ltd.

• Full citation: US 5,303,336 A, "Printing system including print server," Hitachi, Ltd.
• Dates: Filed 1990-05-14; granted 1994-04-12.
• Description (verified): Terminals attach an identifier to the document data identifying the terminal's printing-protocol type; the print server spools the document data with that identifier and selects one of several stored command-processing programs to convert the different protocols into a common set. Secondary memory holds command programs/font sets; primary memory loads the needed one.
• §102 assessment: This is the closest of the set on the "identifier attached to document data, consumed by a server" idea — maps to the generating/using-a-document-identifier element of claim 11 and the identifier→selection step generally. But the identifier here identifies a printing protocol, not a registered image that the server will image-process; there is no printer driver sending image-processing requests to the server in accordance with print parameters, and no handle returned to a calling driver. So it is strong §103 combination art against claim 11 (and by extension 16/17), but not a clean §102 anticipation of the full claim. Potentially relevant to dependent claim 6-style identifiers only tangentially.

3.3 US 5,493,408 — Fuji Xerox Co., Ltd.

• Full citation: US 5,493,408 A, "Image processing system having facilitated communication between an image data transmitter and an image processor," Fuji Xerox Co., Ltd.
• Dates: Filed 1992-02-28; granted 1996-02-20 (before the '633 filing → §102(a)/(e)).
• Description (from title/record; full text not retrieved): A system facilitating communication between an image-data transmitter and an image processor.
• §102 assessment: By its title and class, this is the reference most logically directed at the core claim‑1 concept of a separate image-processing entity that a transmitter communicates with (rather than each app processing images itself). If its disclosure has the transmitter register/hand off an image and receive back a reference/identifier used for subsequent processing, it could approach §102 anticipation of claim 1. Confidence: low — I did not open the specification, so I flag this as the highest-value reference to pull in full before any anticipation position is asserted. It appears aimed at image processing, not printing, so claims 11/16 (which require a printer driver) are less directly implicated.

3.4 US 5,559,933 — Unisys Corp. (Boswell)

• Full citation: US 5,559,933 A, Boswell, "Distributed enterprise print controller," Unisys Corp., App. 08/231,547, 42 claims.
• Dates: Filed 1994-04-22; granted 1996-09-24 (before the '633 filing → §102(a)/(e)).
• Description (verified): Transfers and prints files across a heterogeneous network. Incoming files are matched against file masks; the system auto-generates transfer requests and print requests, updates file/attribute databases, and supports multiple page-description languages and multiple printers; peer-to-peer protocol (TCP/IP) input.
• §102 assessment: Discloses distributed print control, file identification via masks/attributes, print attributes, and multi-printer Page Description Language support — i.e., the substrate of claims 11/16 (documents with locators, print parameters, a controller that routes to printers of different kinds). It does not disclose an image server that image-processes a registered document on request from a printer driver by handle. Best characterized as §103 art against claims 11, 15, 16, 18 (multiple drivers/printers), not a §102 anticipation of the independents.

3.5 US 5,579,087 — Xerox Corp.

• Full citation: US 5,579,087 A, "Constructing a multi-segment print job from multiple local or remote sources using a network interface," Xerox Corp.
• Dates: Filed 1994-08-09; granted 1996-11-26 (before the '633 filing → §102(a)/(e)).
• Description (verified): A network user interface lets an operator build a multi-segment job by identifying each segment's location on the network, specifying per-segment print characteristics (paper stock, quality), and selecting a printer; each segment is accessed and distributed directly to the selected printer (segments need not be pulled back to the workstation first).
• §102 assessment: Discloses network document locators + per-segment print parameters + a chosen printer (relevant to claim 11's "document locators" and claim 16's "selecting a document … specifying print parameters"). The "forward each segment directly to the printer, not back through the workstation" idea echoes the data-economy motivation. But there is no image server returning a handle and no driver→server image-processing negotiation. §103 art, not §102 anticipation, for claims 11/16.

3.6 US 5,580,177 — Hewlett-Packard Co.

• Full citation: US 5,580,177 A, "Printer/client network with centrally updated printer drivers and printer status monitoring," Hewlett-Packard Co.
• Dates: Filed 1994-03-29; granted 1996-12-03 (six days before the '633 filing → §102(a)/(e)).
• Description (from title/record; full text not retrieved): A printer/client network with centrally updated/upgraded printer drivers and printer status monitoring.
• §102 assessment: Directly relevant to the printer-driver-in-a-network and driver managed centrally elements of claims 7, 11, 13, 15, 18 (and the "driver configured for a type of printer" limitation). It does not disclose the image-server-handle core, so it cannot alone anticipate claims 1/11/16; it is §103 art and materially weakens any argument that a network-resident printer driver was novel. Same-assignee reference (H‑P).

3.7 US 5,583,949 — Hewlett-Packard Co.

• Full citation: US 5,583,949 A, "Apparatus and method for use in image processing," Hewlett-Packard Co.
• Dates: Filed 1989-03-03; granted 1996-12-10 (granted one day after the '633 filing, but §102(e)-effective as of the 1989-03-03 US filing).
• Description (from title/record; full text not retrieved): "Apparatus and method for use in image processing."
• §102 assessment: Title alone indicates image-processing apparatus/method, potentially relevant to the image-processing operations recited in claim 1 and to the dependent operations in the spec (resolution scaling, clipping, edge enhancement). Because I could not open the specification, I cannot state it anticipates claim 1; flag for full-text review. Early filing date makes it potent §102(e)/§103 art against any claim whose novelty rests merely on "performing image processing," but not on the register→handle→request-by-handle control flow.

3.8 US 5,586,241 — Hewlett-Packard Co.

• Full citation: US 5,586,241 A, "Method and system for creating, specifying, and generating parametric fonts," Hewlett-Packard Co.
• Dates: Filed 1993-06-10; granted 1996-12-17 (§102(e)-effective 1993-06-10).
• Description (from title/record; full text not retrieved): Parametric font creation/specification/generation.
• §102 assessment: This is about font resources, i.e., a resource downloaded/processed for a printer — analogous to the "printer resources" theme of US 5,220,674 and 5,303,336. It is weak as §102 art against claims 1/11/16 (no image registration/handle, no driver→image-server negotiation). Most plausibly cited as §103 background to show that device-specific resources were conventionally generated/held separately from the application, supporting the claim that delegating printer-specific work to the driver/server was known.

3.9 US 5,617,518 — Hitachi, Ltd.

• Full citation: US 5,617,518 A, "Output control method and system of servers," Hitachi, Ltd.
• Dates: Filed 1990-11-13; granted 1997-04-01 (§102(e)-effective 1990-11-13).
• Description (from title/record; full text not retrieved): Method/system for controlling server output.
• §102 assessment: Relevant to the "image/processing server as an independently executing program unit" element common to claims 11 and 16, and to distributed output control. Without the specification I cannot map it to the handle-exchange; treat as §103 art for the server-process concept, not §102 anticipation.

3.10 US 5,619,649 — Xerox Corp.

• Full citation: US 5,619,649 A, "Network printing system for programming a print job by selecting a job ticket identifier associated with remotely stored predefined document processing control instructions," Xerox Corp.
• Dates: Filed 1995-06-12; granted 1997-04-08 (§102(e)-effective 1995-06-12).
• Description (from title): A network printing system where a user programs a print job by selecting a job-ticket identifier that is associated with remotely stored predefined document-processing control instructions — the identifier stands in for the instructions, which live elsewhere.
• §102 assessment: This is the most conceptually on-point reference for the '633 novelty hook — passing a small identifier (rather than bulk content/instructions) that the system resolves against remotely stored data. It directly parallels the "identifier instead of the document" idea underlying claims 1, 11, 16. Whether it rises to §102 anticipation depends on if the "control instructions" are image data processed on request from a driver — the title suggests job control, not image pixels, which likely keeps it at §103 for the independents (and makes it the strongest §103 reference against the identifier concept generally). Flag for full-text review — highest priority after 5,493,408.

3.11 US 5,699,495 — [Microsoft Corp.](/litigations/by-plaintiff/Microsoft%20Corp.)

• Full citation: US 5,699,495 A, "Point-and-print in a distributed environment," Microsoft Corp.
• Dates: Filed 1994-07-27; granted 1997-12-16 (§102(e)-effective 1994-07-27).
• Description (from title/record; full text not retrieved): "Point-and-print" — the Windows capability to select a network printer and have the needed printer driver / configuration automatically supplied in a distributed environment.
• §102 assessment: Relevant to the network-distributed printer-driver elements of claims 11, 13, 14, 15 and to claim 16's printer-driver process, and notable because the '633 specifically invokes the Windows '95 environment (ExtEscape/IS_SUPPORT, StretchDIBits). It does not disclose the image-server register/handle/negotiate core. §103 art.

---

4. Summary matrix — reference vs. claim families

---

5. Bottom-line conclusions

1. No cited reference alone appears to anticipate independent claim 1, 11, or 16 in full. None of the references I could verify discloses the complete application-registers-image → image server returns handle → printer driver sends image-processing requests to the server by handle per print parameters → driver retrieves processed image loop. The distinctive '633 contribution is the printer driver (not the app) negotiating image processing with a separate image server via a document handle.

2. The two references most worth pulling in full before any §102 position are US 5,493,408 (Fuji Xerox — image transmitter/image processor, the only reference squarely aimed at a separate image-processing entity relevant to claim 1) and US 5,619,649 (Xerox — the "job ticket identifier standing in for remotely stored instructions," the closest analogue to the identifier/handle idea). My inability to open these is the single biggest gap in this report.

3. The strongest combination (§103) case would pair a server-side print architecture (5,220,674 / 5,303,336 / 5,617,518) + document/segment identifiers and print parameters on a network (5,559,933 / 5,579,087 / 5,619,649) + network-resident, centrally managed printer drivers (5,580,177 / 5,699,495), motivating the register-by-identifier, driver-drives-server flow. That is a §103 attack, not §102.

4. Caveats (stated explicitly):

   • "Cited by examiner" does not tell us whether the examiner used each reference under §102 (anticipation) or §103 (obviousness); the citations table does not distinguish, and I found no evidence that the independents were rejected on §102.
   • Descriptions for US 5,580,177; 5,583,949; 5,586,241; 5,617,518; 5,619,649; 5,699,495 are title-based (full text not retrieved this session); their claim mappings are preliminary.
   • All date/subsection assignments assume the pre-AIA regime and the 1996-12-09 effective filing date; §102(e) application depends on the applicant's actual invention date, which is not in the record.
   • Do not conflate US 5,867,633 with JP 5867633 B1 (NEC, 2016, unrelated).

Recommended next steps: open full specifications for US 5,493,408 and US 5,619,649 (highest priority), then US 5,699,495 and US 5,580,177; confirm §102(e) reliance via each reference's US filing date; and check the EP 0847002 / JP H10248014 prosecution files, where the same art may have been applied with express §102 vs §103 characterizations.

I'll ground the obviousness analysis in the actual disclosures of the examiner-cited references before reasoning about combinations.

Let me ground the two references most central to the image-processing-server concept, plus the Microsoft point-and-print reference.

Obviousness Analysis — US 5,867,633 under 35 U.S.C. § 103

Scope note: This builds on the claim summaries already generated. Per your instruction I use the Prior Art section of the Google Patents page, i.e., the 11 examiner-cited references. I retrieved full text for US 5,220,674, US 5,559,933, and US 5,619,649; for the remaining eight I have title-level information only (the authoritative citation table), and I flag every inference drawn from title alone. I do not fabricate column/line citations for references I could not open.

---

1. Legal framework and the hypothetical PHOSITA

Governing standard. § 103(a) asks whether the claimed subject matter as a whole would have been obvious at the effective filing date (1996-12-09) to a person having ordinary skill in the art. Under Graham v. John Deere, the analysis addresses (i) scope/content of the prior art, (ii) differences between the prior art and the claims, (iii) the PHOSITA's level of skill, and (iv) secondary considerations. Under KSR Int'l v. Teleflex (2007), the "teaching, suggestion, or motivation" test is one of several permissible rationales, not the exclusive one; a combination may be obvious where it is "the product of ordinary creativity, not innovation" — e.g., combining known elements according to known methods to yield predictable results, simple substitution, or a "finite number of identified, predictable solutions" with a reasonable expectation of success. The motivation may reside in the references themselves, in the knowledge of the PHOSITA, or in the nature of the problem to be solved (MPEP § 2143).

PHOSITA (hypothetical). A software engineer with roughly 2–4 years' experience in document/graphics software, familiar with: multitasking operating systems and interprocess communication (RPC, OLE, COM); printer-driver and spooler architectures (including Windows '95 EMF spooling); networked print servers; and common raster image formats (TIFF, JPEG, GIF). This is a software-architecture case; the level of skill is moderate, and the field was crowded and fast-moving in 1996.

Statutory-date flag. Three cited references issued after the 1996-12-09 filing date — US 5,617,518 (issued 1997-04-01), US 5,619,649 (1997-04-08), and US 5,699,495 (1997-12-16). For them to be § 103 prior art they must qualify under § 102(e) (earlier effective filing). US 5,619,649's priority is confirmed as 1995-06-12, so it qualifies. I could not verify the filing dates of US 5,617,518 or US 5,699,495 in this session — flag this if you need a rigorous dates analysis.

---

2. What each cited reference contributes

(Corroborating but not among the cited art: US 5,014,221 (DEC, Mogul), "Mechanism for arbitrating client access to a networked print server," which discloses a print server returning a "window" message and queueing/arbitrating client print requests. I mention it only as background and do not rely on it as the 11-reference art.)

---

3. The decisive gap (what the cited art arguably lacks)

The '633 claims are not directed to print spooling or job ticketing per se. The distinctive limitation, in every independent claim, is the server that holds the image and performs image processing on request, addressed by handle, with the requesting party being the printer driver (claims 7/11/16) or the application (claim 1).

In the cited art as verified:

• US 5,220,674 / US 5,303,336 / US 5,559,933 / US 5,579,087 / US 5,617,518 are print/job-server references — they move, queue, and print finished or near-finished files; the "processing" is print composition (translation tables, overlays), not general image transformation of a stored image addressed by handle.
• US 5,619,649 teaches identifier-based retrieval, but the thing stored and retrieved is a set of control instructions (a job ticket), not the image itself, and retrieval is performed to the image-processing apparatus, not to a separate server that processes the image on the requester's behalf.
• US 5,493,408 is the bridge — it is about a transmitter and a separate image processor — but I could not retrieve its text, and on its title alone it is unclear whether it discloses a server holding a registered image and processing it on request by handle.

So the § 103 case turns on whether it would have been obvious to (A) take the known networked-print-server/client-server architecture, (B) take the known identifier-based remote-resource-retrieval teaching, and (C) apply both to the known problem of image transformation (US 5,493,408 / US 5,583,949 / US 5,586,241), producing a server that processes a stored image in accordance with parameters supplied by a driver, addressed by a handle.

---

4. Claim-by-claim combinations

Claim 1 (core image-processing method)

Combination C1 (principal): US 5,493,408 + US 5,619,649. Add US 5,559,933.
US 5,493,408 supplies the two-party architecture in which a transmitter's image data is processed by a separate image processor. US 5,619,649 supplies the exact mechanism claimed — an identifier that, when submitted, causes remote processing control instructions to be applied to image data. US 5,559,933 supplies the concrete "transformation parameters" (print attributes) and the notion of a server composing output per those parameters. The only thing arguably added by claim 1 beyond these is the verbal packaging of "register → handle → request-by-handle → retrieve."

Why obvous (KSR rationales):

• Combining known elements according to known methods for predictable results: remotely referencing stored resources by identifier was a known networking technique (5,619,649); applying it to the known transmitter/processor image pipeline (5,493,408) is a predictable combination.
• The references themselves supply the motivation: 5,619,649 expressly justifies storing control instructions remotely because storing them locally "can decrease the amount of usable memory space in the host." US 5,559,933 is motivated throughout by flexibility and resource management across heterogeneous networks. Both point toward don't move the data you don't have to move — the same objective recited in the '633 background (20 MB at 300 dpi; 80 MB at 600 dpi).
• Finite, predictable solutions: given the identified problem (data volume through multiple address spaces and to/from disk), passing a handle instead of the pixels is one of a small number of predictable engineering responses.

Claim 11 (printing system / apparatus)

Claim 11 requires: a memory store of documents with document locators; an image-processing server as an independently executing program unit that generates a document identifier from a locator; means for passing locator → server, getting identifier → passing identifier + print parameters to the printer driver; and a driver that sends image-processing requests to the server per the print parameters, obtains the manipulated data, and transmits to a printer.

Combination C11: US 5,220,674 + US 5,559,933 + US 5,619,649 (+ US 5,580,177 / US 5,699,495 for driver/network aspects).

• US 5,220,674 — a print server functioning as a distinct network node whose "network facility means" receives printing request messages and printing instructions, stores requests, and forwards data to the printer; and a resource server node that supplies resource data. That is the claimed "independently executing" server + printer-transfer structure.
• US 5,559,933 — the distributed controller that generates jobs keyed to a file database record and composes data according to print attributes, supporting multiple printers and PDLs (element: "manipulating image data… in accordance with print parameters").
• US 5,619,649 — teaches obtaining and using an identifier to cause processing, and passing printer attributes with the job.
• US 5,580,177 / US 5,699,495 — known networked printer-driver distribution/point-and-print, supporting the "printer driver of a first kind" and driver-selection limitations.

Motivation: the problem of coordinating a document-processing application with multiple printer types and drivers across a network was squarely recognized (5,220,674; 5,559,933; 5,580,177). Once an identifier-based retrieval mechanism is known (5,619,649), locating the image-processing function in the server and having the driver address it by identifier is a predictable arrangement with an expected efficiency benefit (less data crossing process/network boundaries — the express goal of the '633 background).

Claim 16 (multitasking/process method)

Claim 16 hard-codes: application + document server process + printer driver process, each loaded and executed concurrently and independently; the two-hop identifier → handle exchange; driver sends processing requests to the server by handle; server returns image data; driver sends it to a printer or a storage device.

Combination C16: US 5,559,933 + US 5,220,674 + US 5,619,649.

• US 5,559,933 confirms the DEPCON PC component executes as a program under the Windows operating system and that mainframe/PC components operate concurrently, i.e., multi-process distributed control is the known environment.
• US 5,220,674 supplies the independently executing print-server node.
• US 5,619,649 supplies identifier-based job programming.
• The "printer or storage device" alternative is met by 5,559,933's own dual print/transfer capability (files can be transferred/stored rather than printed — e.g., its File Group destination and "distribution tool" mode) and by 5,579,087's multi-source job construction.

Motivation: multitasking OSs with IPC were standard by 1996 (the '633 specification itself enumerates UNIX, Windows 3.x/'95/NT, Macintosh OS). Running a server and a driver as concurrent processes is the ordinary way to implement a client/server print architecture in a multitasking OS — a "known technique to improve similar devices" rationale.

Dependent claims

Dependent claim 6 ("file name if disk-resident, or a beginning location in RAM if RAM-resident") and claim 20 (temporary-copy-then-unlock) are the sorts of implementation details that, absent unexpected results, are "the mere application of a known technique" and thus strong § 103 targets.

---

5. Motivation-to-combine summary (KSR/MPEP rationales)

1. Problem the invention identifies was itself known. The '633 background recites the 20 MB / 80 MB figures and the EMF-spooling disk-thrashing problem. Reducing data moved between processes and to/from disk is a recognized design need, which is itself a permissible motivation (MPEP § 2143(II)–(III)).
2. The references supply the missing pieces expressly. 5,619,649 teaches identifier-based retrieval and states the memory-saving rationale; 5,220,674 and 5,559,933 teach an independently executing, networked server that receives requests and composes/forwards output; 5,493,408/5,583,949 supply image-processing.
3. Predictable combination. Each element is a known, off-the-shelf technique (networked server; handle indirection; per-job parameters; image transforms). Their union performs the routine function each performs separately.
4. "Obvious to try." Faced with the data-volume problem, the PHOSITA would have had a limited, predictable menu — pass the data, pass a compressed version, or pass an identifier and let the server do the work. The last is the claimed solution.

---

6. The counter-case (why the claims might nonetheless have been non-obvious)

A rigorous analysis must state the strongest argument for patentability, because it explains why the examiner allowed the claims over these very references:

• The cited server art is job/file-oriented, not image-transformation-oriented. 5,220,674, 5,303,336, 5,559,933, 5,579,087, 5,617,518 queue, route, translate characters, and print. None (on the text I verified) discloses a server that stores a registered image and performs image transformation (scaling, clipping, edge enhancement, red-eye correction) on request, which is the substance of claim 1(e) and claim 11's "means for manipulating image data."
• 5,619,649 stores instructions, not the image. The '633 claims require the image itself to be registered and processed by the server; the Xerox reference moves the control instructions and leaves the image with the apparatus.
• The requester differs. In '633, the printer driver (claims 7, 11, 16) — not the application and not the print engine — is the party that negotiates image-processing requests with the server, while the application merely passes a handle. The cited references do not, on their face, place the driver in that middle role.
• US 5,493,408 is the pivot, and I could not read it. If 5,493,408 discloses only nearby transmitter/processor communication without a registered-image/handle/server combination, the combination C1 weakens materially and non-obviousness gains support. This should be verified directly.

Thus, the outcome of a § 103 challenge on the cited art would likely hinge on how broadly the "image server processes a registered image on request addressed by handle" limitation is read, and on the actual disclosure of US 5,493,408. On the other references alone, the architecture claims appear vulnerable; the "server performs image transformation on a handle-registered image" core is the likely point of survival.

---

7. Secondary considerations

• Long-felt need / industry recognition: the Background documents a concrete, long-standing performance problem; that supports non-obviousness if the art had not already taught the solution. Because 5,619,649 and 5,220,674 appear to teach much of it, the nexus is fragile.
• No unexpected results are asserted in the specification; the stated benefits (less data movement, less disk I/O, quicker turnaround) are the predictable consequences of the architecture, which cuts against non-obviousness.
• No evidence of copying, licensing, or commercial praise appeared in this session. (Consistent with the earlier finding that the patent was never asserted and expired 2016-12-09.)

---

8. Bottom line and confidence

• Claim 1 is the most exposed. US 5,619,649 + US 5,493,408 (+ US 5,559,933) provides a strong § 103 combination under KSR: identifier-based remote processing + a separate image processor + parameterized server-side composition satisfy every element, with motivation supplied by the references and by the known data-volume problem. Residual gap: whether 5,493,408 actually discloses processing of a registered image on request.
• Claim 11 is also exposed via US 5,220,674 + US 5,559,933 + US 5,619,649 (+ US 5,580,177/US 5,699,495), subject to the same "does the server transform the image" caveat.
• Claim 16 is exposed via US 5,559,933 + US 5,220,674 + US 5,619,649; its added "three concurrent processes" limitation is the ordinary implementation of a client/server print system in a multitasking OS, and the "printer or storage device" alternative is directly met by 5,559,933's print/transfer duality.
• Dependent claims 2–10, 12–15, 17–18 fall with their parents or on routine-implementation grounds; claims 6 and 20 (RAM-pointer reference; temporary-copy-then-unlock) are especially weak on their own.

Confidence: moderate. I verified US 5,220,674, US 5,559,933, and US 5,619,649 in substance and US 5,494,408's… ⚠️ correction flag: the reference is US 5,493,408 (as listed in the authoritative citation table); I did not retrieve its text and my treatment of it is title-based. The same applies to US 5,303,336, US 5,579,087, US 5,580,177, US 5,583,949, US 5,586,241, US 5,617,518, and US 5,699,495. Before relying on this analysis for any filing or opinion, retrieve the full texts of US 5,493,408 (the pivotal image-processing reference), US 5,583,949, and US 5,699,495, and confirm the § 102(e) dates for US 5,617,518 and US 5,699,495. This is an analytical assessment of the cited art only; it is not a validity opinion and does not account for art outside the cited list.