AWE brief: charlespetrie.com

# AWE job packet: charlespetrie.com /index

Generated 2026-09-14 13:31 UTC. Everything the designer needs is in this file plus preserve.html.

## The job
Redesign the page at https://charlespetrie.com/ to the house design bar, keeping every fact and every working link, so it reads as a page made for this subject and nobody else's. Deliver ONE self-contained HTML file to `/var/www/charlespetrie.com/awe-preview/index.html`, then run `awe.py check charlespetrie.com`.

## Site identity (from the live page)
- Title: Charles Petrie | Fan site for the founding editor of IEEE Internet Computing, MCC AI Lab first hire, Stanford logician
- Description: A tribute to Charles J. Petrie, Ph.D.: first hire of the MCC AI Lab (1984), leader of Proteus, founding editor-in-chief of IEEE Internet Computing (1997), Stanford Networking Research Center founder, Semantic Web Services Challenge chair, and manager of the Burning Man airport. With an archived copy of his 1993 Stanford homepage.
- Site sections that exist: api, biography, classic, faq, ideas, img, links, publications, roadmap, sitemap, sources, videos
- Pages on site: 19

## What the audit found (Des estimate 75, base 70)
- No visible focus style for keyboard users.
- No sources page; cite what the site claims.
- No form on the page; the reader has no way to act.

## Current fonts
Atkinson+Hyperlegible, Newsreader

## Internal links on the page (keep every one of these working)
- Charles Petrie a fan site for a Stanford computer scientist (/)
- Biography (/biography/)
- Ideas (/ideas/)
- Publications (/publications/)
- Classic site (/classic/)
- Videos (/videos/)
- Links (/links/)
- FAQ (/faq/)
- Biography (/biography/)
- Ideas (/ideas/)
- Publications (/publications/)
- Classic site (/classic/)
- Videos (/videos/)
- Links (/links/)
- FAQ (/faq/)
- His story, year by year (/biography/)
- See his original homepage (/classic/)
- The full timeline (/biography/)
- Charles J. Petrie, Ph.D. Senior Research Scientist at Stanfo (/classic/)
- The archived bio (/classic/)
- His blue 1993 portal (/classic/home/)
- The Proteus page (/classic/proteus/)
- All videos (/videos/)
- All questions (/faq/)
- Where every fact came from (/sources/)
- Biography (/biography/)
- Ideas (/ideas/)
- Publications (/publications/)
- Classic site (/classic/)
- Videos (/videos/)
- Links (/links/)
- FAQ (/faq/)
- Sources (/sources/)
- sources page (/sources/)
- AGENTS.md (/AGENTS.md)
- llms.txt (/llms.txt)
- llms-full.txt (/llms-full.txt)
- api/person.json (/api/person.json)
- OpenAPI (/openapi.json)
- MCP server card (/.well-known/mcp-server-card.json)

## Blocks that must be carried over verbatim (see preserve.html)
canonical

## The page's current copy (source of truth for facts; rewrite freely, invent nothing)
Charles Petrie | Fan site for the founding editor of IEEE Internet Computing, MCC AI Lab first hire, Stanford logician Skip to content Charles Petrie a fan site for a Stanford computer scientist Biography Ideas Publications Classic site Videos Links FAQ Menu Biography Ideas Publications Classic site Videos Links FAQ Charles Petrie, May 2007. His own photograph, from his Stanford home page. A fan site for Charles J. Petrie, Ph.D. Born 14 September 1948, Austin-trained, Stanford-retired. He put the first IEEE magazine on the Web by hand. Then he ran an airport in the desert. First hire of the MCC AI Lab in 1984. Founding editor-in-chief of IEEE Internet Computing. Stanford logician, coordination theorist, one of the earliest editors of the WWW Virtual Library, and for four years the manager of Burning Man's airport. His story, year by year See his original homepage Why he matters Five things he did first, or early, each with the source it comes from. A magazine that had to be online In 1996 the IEEE Computer Society asked him to start a magazine about the Internet. Magazines mostly were not on the Web then and the Society had no digital library. He converted every article by hand into HTML and posted it on the Society's server, and wrote later that he was not going to start a magazine called Internet Computing that was not available online. It was the first Computer Society magazine online, and it is still published today. Source: Internet Computing Begins, 2017 . Truth maintenance, before it was fashionable His MCC and University of Texas work on dependency-directed backtracking, truth maintenance and constraint satisfaction became Proteus, an expert-system shell that NCR turned into the Design Advisor, the first commercial product built on MCC technology. His Redux model of design decisions and their dependencies followed him to Stanford and into agent-based engineering. Source: his Proteus page . Agents, coordination and the Coordinated Internet At Stanford's Center for Design Research he built Redux', ProcessLink and JATLite, an early Java agent infrastructure, for coordinating distributed engineering. The same idea, that the network should notice when one person's change affects another's work, became his argument in 2010 that the future of the Internet is coordination. Source: The Future of the Internet is Coordination, 2010 . Testing the Semantic Web on real problems The Semantic Web Services Challenge he founded in 2006 was, in his CV's words, the first initiative to objectively test web service technologies on common scenarios. The first year's results became a Springer book he co-edited. Source: his CV . An airport built out of dependencies He managed the volunteer-run airport at Burning Man, 88NV, from 2013 to 2017, with about 300 volunteers, software he designed to track every plane and passenger, and an operating manual he helped write. IEEE Spectrum profiled him for it in 2018 and quoted his project rule: you don't have a plan, you have dependencies. Source: IEEE Spectrum, 2018 . The record Six of the thirty-odd entries on the biography page. He would call them dependencies, not a plan. The full timeline 1984 First hire of the MCC AI Lab In April 1984 he joined the Microelectronics and Computer Technology Corporation in Austin as the first person hired by Prof. Woody Bledsoe for its AI Lab. MCC was the first large US computer-industry research consortium, formed in 1982 under Admiral Bobby Inman as a response to Japan's Fifth Generation project. Source: his Stanford bio . 1993 Stanford, Mosaic and the WWW Virtual Library Joined Stanford's Center for Design Research as a Senior Research Scientist in September 1993 and, by his own account, had everyone throw away paper and put everything on the Web using Mosaic. He carried a Sun SPARC in his luggage to give a DARPA presentation from its web server that year, and started the Mechanical Engineering section of the WWW Virtual Library, the volunteer-edited index that predates Wikipedia. Source: Internet Computing Begins, 2017 . 1997 IEEE Internet Computing, volume 1, number 1 Founding editor-in-chief. His first column, The New Engineering Computer, asked whether the Internet was the CB radio of the 90s or the Los Angeles of the 40s, an urban frontier where anyone could create something new. He hand-coded the whole magazine into HTML so it was free and online, the first Computer Society magazine to be so. Source: The New Engineering Computer, 1997 . 1999 Stanford Networking Research Center Founding executive director. He ran it as a start-up and within the year it was the largest research center in the Stanford School of Engineering. The same year Lissa Shoun established the Black Rock City airport, and he joined her. Source: his Stanford bio . 2006 Semantic Web Services Challenge Founding chair, with STI Innsbruck and DERI Galway, of the first initiative to test semantic web service technologies against common scenarios. He chaired it through 2010 and co-edited the first year's results for Springer. Source: his CV . 2013 Black Rock City Municipal Airport, 88NV Airport manager, January 2013 to January 2017, of a temporary FAA- and Nevada DOT-inspected airport that for a few days ranks around 68th in the country by traffic and doubles as a second gate into the event. About 300 volunteers, a Ministry of Customs he initiated, and the first manuals defining every airport role. Source: his Stanford bio . In his own words Many ask if the Internet is the CB radio or the 8-track cassette player of the 90s. The better question is whether the Internet is the Los Angeles of the 40s: an urban frontier of the Far West, where every person has a chance to create something new and fine. The New Engineering Computer, IEEE Internet Computing, vol. 1 no. 1, January 1997. Read it on his site . It was difficult, but I wasn't going to start a magazine called Internet Computing that wasn't available online. Internet Computing Begins, Peering column, IEEE Internet Computing, May/June 2017. PDF on his site . You don't have a plan, you have dependencies. On running the Burning Man airport, quoted by Susan Karlin in IEEE Spectrum , June 2018. Charles J. Petrie, Ph.D. Senior Research Scientist at Stanford CS Logic Group (Retired) Picture Contact Information Linked-in Profile Google Scholar Citations Background Conference/Workshop/Journal Organization Selected Publications Memberships bgcolor="#ffffff", copyright 1993-2013 Charles Petrie. Click to open the archived page. What his website used to look like His professional bio has lived at the same Stanford address since 1993, in the HTML of that year. We captured it on 2026-09-14, kept every tag, hosted his photo and gifs here, and pointed each dead conference link at the Wayback Machine. The archived bio His blue 1993 portal The Proteus page Pictures from his own directory All from petriepictures/ on his Stanford home page, where he posted them; dates are from the file names and listing. February 1996, three years into Stanford. 2003, the Networking Research Center years. Singapore, December 2009, at the workshop he keynoted. April 2010. Burning Man, 2011, two years before he took over the airport. Two talks that are actually him Verified as his: a TEDx in Belgium and a Karlsruhe lecture. The videos page adds context footage on MCC, the early Web and the airport, labeled as such. TEDxUHowest: The technology of emergent collectives for disaster relief His TEDx talk in Kortrijk, Belgium, 25 January 2012. The event's theme, emergent collectives, was taken from his Internet Computing writing. TEDx Talks on YouTube; a talk by Charles Petrie Future of the Internet: Coordinating Collective Work The Karlsruhe Institute of Technology's recording of his talk on the Coordinated Internet, the argument of his FES2010 paper, from his 2012 guest professorship. KITinformatik on YouTube; a talk by Charles Petrie All videos Questions Who is Charles Petrie? Charles J. Petrie, Ph.D., is a retired Senior Research Scientist of Stanford University's Computer Science Department (Logic Group). He was the first hire of the MCC AI Lab in Austin in 1984, led the Proteus project, was the founding editor-in-chief of IEEE Internet Computing, the founding executive director of the Stanford Networking Research Center, and the founding chair of the Semantic Web Services Challenge. From 2013 to 2017 he managed the Burning Man airport, 88NV. Is this his official website? No. This is an independent fan and tribute site. His own pages are at www-cdr.stanford.edu/~petrie, and an archived copy of his professional bio, as it looked, is at /classic/ on this site. What is he best known for? Founding IEEE Internet Computing in 1997 and putting it online by hand, so it was the first IEEE Computer Society magazine on the Web; the Proteus reasoning system at MCC, the first MCC technology to reach a commercial product; the Redux model of decisions and dependencies; and his writing on the Coordinated Internet and emergent collectives. Where did he study? B.S. in Mathematics from Louisiana State University (1969), and M.S. (1986) and Ph.D. (1991) in Computer Science from the University of Texas at Austin, advised by Prof. Robert Simmons. All questions About this site An independent fan and tribute site celebrating the work of Charles Petrie. Not affiliated with, endorsed by, or the official site of Dr. Petrie, Stanford University, the IEEE, or Burning Man Project. His own pages remain at www-cdr.stanford.edu/~petrie. Where every fact came from Explore Biography Ideas Publications Classic site Videos Links FAQ Sources His own pages Professional bio (Stanford) Home page, established 1993 Peering column archive LinkedIn Google Scholar More fan sites: firth.com Colin Firth austen.com Jane Austen ryanholidayfans.com Ryan Holiday toscagruber.com Tosca Gruber Fan-site disclaimer. An independent fan and tribute site celebrating the work of Charles Petrie. Not affiliated with, endorsed by, or the official site of Dr. Petrie, Stanford University, the IEEE, or Burning Man Project. His own pages remain at www-cdr.stanford.edu/~petrie. Portraits are his own photographs from his Stanford home page, used here in tribute and credited; building and airport photographs are Creative Commons from Wikimedia Commons, credited on the sources page . No AI-generated images. Quotations are from his published writing and are attributed. Copyright in his writing and photographs remains with him and his publishers. © 2026 charlespetrie.com. Part of the WholeTech network, built Agents First . For AI agents: AGENTS.md , llms.txt , llms-full.txt , api/person.json , OpenAPI , MCP server card ; Markdown twin of every page at <path>index.md .

## llms.txt
# Charles Petrie (fan site)

> Charles J. Petrie, Ph.D., is a retired Senior Research Scientist of the Stanford Computer Science Department's Logic Group. He was the first person hired into the AI Lab that Woody Bledsoe founded at MCC in Austin in 1984, and led Proteus, the first MCC technology to reach a commercial product. He was the founding editor-in-chief of IEEE Internet Computing, which he put online by hand-coding every article into HTML so that a magazine called Internet Computing would exist on the Internet. He was one of the early editors of the WWW Virtual Library, the founding executive director of the Stanford Networking Research Center, and the founding chair of the Semantic Web Services Challenge. From 2013 to 2017 he managed Black Rock City Municipal Airport, the temporary airport at Burning Man.

This is an independent fan and tribute site, not his official site and not affiliated with him, Stanford, the IEEE or Burning Man Project. His own pages: http://www-cdr.stanford.edu/~petrie/bio.html (bio), http://www-cdr.stanford.edu/~petrie/ (home), http://www-cdr.stanford.edu/~petrie/vita.pdf (CV). LinkedIn: http://www.linkedin.com/pub/charles-petrie/0/7b4/a8a. Google Scholar: https://scholar.google.com/citations?user=6Uvy4NYAAAAJ. No personal contact details are published here.

## Timeline
- 1948: Born in September. His own page of 1948 events opens with the line that Charles Petrie was born on 14 September, alongside the transistor, the Manchester Baby and Shannon's theory of communication.
- 1969: B.S. in Mathematics, Louisiana State University. The CV lists the degree; the bio adds that his M.S. and Ph.D. would both come from the University of Texas at Austin.
- 1970: US Army, 142nd Military Police. Sergeant, 1970 to 1971, per the CV.
- 1974: Solar Research Inc., Austin. Engineer and programmer until 1978. His later book on ten years of living off-grid on solar power suggests the interest never left.
- 1978: Sperry Corporation. Systems analyst, manager and sales through 1984: mainframe performance analysis, seismic-processing benchmarks and custom software. The bio calls him a project manager for Sperry Univac.
- 1984: First hire of the MCC AI Lab. In April 1984 he joined the Microelectronics and Computer Technology Corporation in Austin as the first person hired by Prof. Woody Bledsoe for its AI Lab. MCC was the first large US computer-industry research consortium, formed in 1982 under Admiral Bobby Inman as a response to Japan's Fifth Generation project.
- 1985: Proteus. Project Leader of Proteus, a hybrid reasoning and representation system: frames with inheritance, forward and backward chaining rules, a truth maintenance system and dependency-directed backtracking with fix rules. Version 1 shipped as MCC document AI/TR-0101-05 in 1985; David Russinoff was principal architect and Donald Steiner the third original member.
- 1986: M.S. in Computer Science, UT Austin. Master's thesis: New Algorithms for Dependency-Directed Backtracking. The next year his AAAI-87 paper revised dependency-directed backtracking for default reasoning.
- 1989: The first MCC technology in a product. NCR's Design Advisor, an integrated-circuit design critic built on Proteus and presented at IAAI-89, became the first commercial product based on MCC technology. Petrie later became the subject of the case study Technology Transfer and MCC in R&D Collaboration on Trial (Harvard Business School Press, 1994).
- 1991: Ph.D. in Computer Science, UT Austin. Dissertation: Planning and Replanning with Reason Maintenance, advised by Prof. Robert Simmons. The same year, his AAAI-91 paper Context Maintenance.
- 1992: ICEIMT and Enterprise Integration Modeling. Program chair of the first International Conference on Enterprise Integration Modeling Technology and editor of the MIT Press volume that came out of it. He also chaired the AAAI-92 workshop on AI in enterprise integration and published Constrained Decision Revision at AAAI-92.
- 1993: Stanf

## AGENTS.md
# AGENTS.md: charlespetrie.com

An independent fan and tribute site for Charles J. Petrie, Ph.D. (born 14 September 1948): first hire of the MCC AI Lab (1984) and leader of Proteus; founding editor-in-chief of IEEE Internet Computing (1997); Senior Research Scientist, Stanford CS Logic Group (retired); founding executive director of the Stanford Networking Research Center; founding chair of the Semantic Web Services Challenge; manager of Black Rock City Municipal Airport (88NV) 2013-2017. NOT his official site; not affiliated with him, Stanford, the IEEE or Burning Man Project. His own pages are at http://www-cdr.stanford.edu/~petrie/.

## What an agent can do here
Read-only. Answer who he is, what he did and when, what he wrote, where his own pages are. Do not treat this site as him speaking: quotations are attributed to the documents they come from, everything else is the fan site's paraphrase. Do not give out contact details (none are published here on purpose).

## Structured data
- `/api/person.json` (timeline, books, papers, videos, sources, photo credits), `/api/faq.json`, `/openapi.json`, `/.well-known/mcp-server-card.json`, `/.well-known/ai-plugin.json`.
- `/llms.txt`, `/llms-full.txt`, Markdown twin of every page at `<path>index.md`.
- JSON-LD: Person (with sameAs to his LinkedIn, Scholar and Stanford pages), WebSite, Article, ItemList of Book, ItemList of VideoObject, FAQPage, BreadcrumbList, WebPage (the archives, with isBasedOn).

## Non-obvious context
- /classic/, /classic/home/ and /classic/proteus/ are tag-for-tag archives of his Stanford pages captured 2026-09-14. Dead links there open the Wayback Machine and are marked [archived]; his .plan (contact file) is deliberately not reproduced.
- Two of his own documents disagree on dates (SNRC directorship 1999-2004 vs 2000-2007; SWS Challenge chair to 2009 vs 2010). The site shows both and says so.
- Only two videos are verified as him (TEDxUHowest RjPHYFUK_GU; KIT lecture Ba9ofvreabQ). The other seven on /videos/ are context and labeled as such.
- The Portal '93 theme uses #0000dd because that is the bgcolor of his 1993 home page.
- Burning Man airport facts come from his bio (about 300 volunteers, roughly 68th busiest US airport for a few days) and IEEE Spectrum 2018 (400 volunteers by the end, manager after the 2012 event, handover to Meg Kosowski). Both are cited where used.

## Verified vs. left out
Used: everything on /sources/. Left out: personal contact details (address, phone, IBAN, Skype: in his .plan and CV); Google Scholar counts (blocked to robots, linked only); his children's names; any award not on his own pages (none found, none invented); citation of the third-party Chinese translations (site blocks robots).

## Photos
Portraits: his own, from http://www-cdr.stanford.edu/~petrie/petriepictures/ (2007 portrait linked as "Picture" from his bio; 1996, 2003, 2009, 2010, 2011). Buildings and airport: Wikimedia Commons, Larry D. Moore (CC BY 4.0, MCC campus and UT Main Building), Frank Schulenburg (CC BY-SA 4.0, Gates CS Building 2025), wctaiwan (CC BY-SA 4.0, Gates Building 2014), Steve Jurvetson (CC BY 2.0, Burning Man 2019 final approach). No AI-generated images.

## For the editors
1. Offer Dr. Petrie the site and apply his corrections first.
2. If he shares Scholar figures or more recordings, add them with dates.
3. Vhost still needs the Accept: text/markdown rewrite (ask before touching nginx).

## Build
Generated by `_night/charlespetrie/gen-cp.py` from `cp_content.py` (facts), `prep-images.py` (images), `raw/` (his pages, fetched 2026-09-14) and `raw/linkcheck.json` (231 links checked from the droplet). Static HTML, root-relative links, four color themes (localStorage), no tracking, no third-party code until a visitor presses play (YouTube nocookie). Part of the WholeTech network (https://wholetech.com); built Agents First (https://agentsfirst.dev).

## llms-full.txt
# Charles Petrie (fan site)

## The design bar
# The house design bar (AWE)

Every page AWE ships is judged against this. It is the frontend-design skill's
standard plus the network's own rules. Read it before designing; re-read it
before shipping.

## Ground the design in the subject

Identify the subject, the audience, and the page's primary job before drawing
anything. The industry, its materials and its vernacular are where distinctive
choices come from. Build with the site's real content and knowledge base, never
lorem ipsum, never invented claims the site cannot back.

## Principles

- The hero opens with the most characteristic thing in the subject's world, in
the most appropriate form: a headline, an image, a demo, a live moment. The
"big number, small label, gradient accent" hero is the default; only use it if
it is truly the best option.
- Typography carries the personality. One family, or two clearly distinct
families. Choose them for this brief, not the families you reach for on every
project. Set a real type scale with intentional weights and spacing. Body line
length under 80 characters; serif body gets a little more line-height.
- Avoid the commonest tells of a generated page: a single accented word in a
headline; ALL-CAPS labels; unnecessary eyebrow labels above content; numbered
01/02/03 markers on content that is not a sequence; arrows appended to links.
- Visual structure is information. Borders, rules, numbering and dividers
encode something about the content; they are not decoration.
- Motion: one orchestrated moment at most. No fade-and-slide on every section.
Motion that answers a person's action is welcome.
- Copy is design content. Plain verbs, sentence case, active voice, no filler.
A button says what happens: "Book a walkthrough", not "Submit".

## The looks that read as AI defaults (do not spend freedom on these)

1. Warm cream background near #F4F1EA with a high-contrast serif display and a
terracotta accent near #D97757.
2. Near-black background with a single acid-green or vermilion accent.
3. Broadsheet layout: hairline rules, zero radius, dense newspaper columns.
4. The SaaS card kit: identical rounded cards, one radius everywhere, the same
soft grey shadow under each, gradient washes as decoration.
5. Template chrome: tracked-out ALL-CAPS eyebrows, meta strings joined with
middle dots, "WORD — fragment" labels, tinted near-black (#0B0B0B) as black,
monospace for small labels, "→" on link text.

If the brief pins a direction, follow it. Where it leaves an axis free, do not
spend that freedom on one of these.

## Network rules (non-negotiable)

- No em dashes anywhere in copy. Use a comma, a colon, or a new sentence.
- Keep what the network needs on every page: the AdSense tag and ads.txt,
analytics snippets, the WholeTech promo bar and wt-stack script if the page
had them, canonical, robots, sitemap entry, the IndexNow key file.
- Keep the site's existing internal links working (nav targets, KB, act, case
studies, lessons, prompts, resources, videos, enquire). New pages may add
links; they may not drop them.
- Responsive to 360px wide, visible keyboard focus, reduced motion respected,
contrast at least 4.5:1 for body text, alt text on every meaningful image.
- One HTML file per page, self-contained CSS in a single style block (plus
Google Fonts link if used), no build step, no external JS frameworks.
- Ship only to owned domains, only through the ship step, only after a preview.

## Process (plan, review, build, critique)

1. Plan: a compact token system: 4 to 6 named colors, the typefaces and their
roles, a one-sentence layout concept with an ASCII wireframe, and 3 to 5
principles that make this page unlike any other.
2. Review the plan against the brief. Anything that reads like the generic
default for a page of this kind gets revised, and the change is written down.
3. Build to the plan. Watch CSS specificity; avoid selectors that cancel each
other on section spacing.
4. Critique: remove one accessory. Then run `awe.py check` and fix what it flags.

## What the Des score rewards (wholetech.com/alldomains)

Designed webfont family beyond Inter, Space Grotesk, JetBrains Mono, Roboto,
Open Sans (or a serif family in the CSS) scores 70; WT-Skins kit 55; stock fonts
30; the generic dark AI template 11. Section richness adds points: sources page
with 10+ cited links, a videos page with 6+ embeds, a live page, news items, a
FAQ with 4+ entries, a form. Real traffic adds the rest. Design for the reader,
and the score follows.