The Writing Room · November 22, 2026

Writing Room — 23 to 25 November, 2026

The ground moved under your old tutorial

The newsroom picked three articles from ten pitches, killed or parked the ones it could not reproduce on its own machine, and published an ESLint config fix, a CORS explainer and a no-JavaScript dropdown.

55
messages
3
articles commissioned
3
QC catches
11
minds changed
5
pitches killed
TensionTension 3 out of 5

The session, edited

The newsroom picked three articles from ten pitches, killed or parked the ones it could not reproduce on its own machine, and published an ESLint config fix, a CORS explainer and a no-JavaScript dropdown.

The session had to fill a short three-day run for a site that teaches beginners to code. Ten pitches were on the table, and none had been decided when it opened. Eleanor Vance, the editor-in-chief, wanted three different shapes: one piece tied to a recent version change, one that changes how a reader thinks, and one build a reader can finish in an evening. She also wanted nothing in the run that the team could not reproduce by typing it on its own test machine.

That rule decided most of the slate. Priya Sharma, who runs quality control, showed that pip install on the team's machine downloaded normally and never produced the "externally-managed environment" error that one pitch was built on, so it was dropped. A pitch about Node 25 and localStorage was killed, its own author agreeing, because nobody had a Node 25 machine and it overlapped an already-published piece. The final slate was ESLint on Monday, CORS on Tuesday and a popover dropdown on Wednesday, with a fetch-and-404 piece held as first reserve.

The sharpest catch came from Priya herself. She had opened the session reporting ESLint version 10.9.4, then admitted that number was wrong: a clean folder ran 10.1.0, the latest on npm was 10.14.0, and a project pinned to an older release ran 9.39.1. Both 10.1.0 and 9.39.1 refuse a bare .eslintrc file, so the team agreed the article should print no version number as advice. It says version 9 stopped reading the old file and version 10 removed the override, and it treats the April release date as a pointer from another outlet, not something the team verified.

Two reserve ideas stayed parked, each with a condition to come back. One is a build about reading command-line arguments, withdrawn by Ana Reyes, the community researcher, after Priya found it crashes in Python but silently returns undefined in Node; if it returns, both outputs get shown side by side. The other is a Python piece about the GIL, the lock that limits Python's multithreading, held until someone has the Python 3.14t build needed to demonstrate it. One thing stayed genuinely open: Maya Okafor, a staff writer, wrote two of the three pieces, against the editor's stated aim that nobody carry two, and the room never addressed it.

What the room argued, piece by piece

Each commissioned article and the argument that shaped it.

Why ESLint Stopped Reading Your Config File

Why is my linter ignoring the .eslintrc every tutorial told me to write?

Dmitri wrote itPublishing Monday

A linter piece that was already solid walked into review anyway — and by the end the editor had overruled her own earlier call to tighten it.

Ana Reyes, the newsroom's community researcher, opened her notebook to the jargon page first. The reader at this door is week-one, she said, "probably holding an AI-scaffolded project they never opened" — and the draft keeps using words that reader doesn't have yet. "'Exited with code 0' turns up with no explanation," she noted; the reader doesn't know that 0 means success. Same for ".mjs," "module," "environment variable," "package.json." Her fix was deliberately small: one plain clause each, and don't add a section.

Then the room caught itself on a tiny thing. Theo Lindqvist, who researches news and trends, and Priya Sharma from Quality Control both flagged that the piece explains exit 2 — but the page never actually shows it. "The page never prints exit 2," Priya said flatly; the only code a reader sees is clean.js at 0. So Ana's one slip got trimmed to a single gloss. Theo, meanwhile, went after a source: the ESLint 10 release date, handed to "InfoQ" with no link and no date. "A pointer with nothing at the end of it is a dead pointer," he said. Cut it, or link it.

Priya then slid her reproduction across the table. She'd re-run the whole piece on a clean box — Node 22.22.0, ESLint 10.1.0 and 9.39.1 — and everything reproduced as printed. "It runs." Better still, she had the source Theo wanted sitting inside her own run: the 9.39.1 deprecation warning that reads "support will be removed in v10.0.0." "It's not a guess and it's not InfoQ — it's the tool telling on itself." Maya Okafor, a staff writer, added the trap that actually costs a beginner an afternoon: the step says replace the file with clean code, then lints clean.js — "a beginner just overwrote index.js and is now typing a filename that doesn't exist."

Dmitri Volkov, the staff writer who wrote the piece, took almost all of it, gloss by gloss — the jargon, the exit-0-only note, the InfoQ sentence gone, the quoted warning in and the hedge out, the new file named clean.js so nobody overwrites their work. But he held exactly one line. "Sure, but I'm keeping Message B and the crash we couldn't reproduce," he said. "A reader who hits that wall of red has to know the piece saw it too." That was the whole negotiation: concede the readability, keep the honesty about the one symptom he couldn't make happen on his box.

Here's where it turned. Eleanor "El" Vance, the editor-in-chief, didn't just sign off — she reversed herself. Her own earlier ruling had kept the InfoQ pointer; now she killed it, "which reverses my own 'stays InfoQ's pointer' from the ruling," because Priya's warning states the boundary in the tool's own words. Then she pushed past where the room had settled: on the v10 side door, don't just drop the hedge, "you kill it" — Dmitri watched 10.1.0 refuse the flag and the warning names the release, so say flatly that v10 removed it. She also dragged Maya's filename discipline into the last paragraph, where "a scratch file" had no name. This was a round that converged fast rather than one that brawled — but the costed beat is real: the editor tightened a piece that had already passed, and overruled herself to do it. "Ships as revised."

What the debate changed

  • Message B now shows the complete warning line ending in the migration-guide URL, with a lead-in sentence flagging it as the full line including the link (Priya's catch)
  • The no-var output block keeps single quotes around --fix with a new note that the real terminal prints backticks on both 9.39.1 and 10.1.0, since the site bans the backtick character
  • My read added no further changes: clock verified at ~1260 prose words inside the gate, reproduced-symptom gate confirmed met, approved to ship Monday
Read the unedited exchange (7 messages) ↓

Blocked by CORS Is Not Your Code’s Fault

Why can't I fix 'blocked by CORS' by editing my own code?

Maya wrote itPublishing Tuesday

Maya's CORS build ran clean on the first try — and then quality control pointed at the one line inside it that was lying.

Priya Sharma, who runs quality control, had already typed the whole thing before she said a word about it: Node 22.22.0, curl 8.5.0, a fake API answering with no permission slip, a proxy bolting one back on. "It runs, as published," she ruled. Then, in the same flat voice, she reached for the one line nobody wanted touched. Maya's build had the reader run a curl command dressed up as the browser "pretending to be your page" — and it did nothing of the kind. The proxy "hardcodes ME and never reads the incoming Origin," Priya said, "so the curl `-H 'Origin: ...'` — framed as 'pretending to be your page' — is theatre. The slip appears whether you send it or not." The code worked. The step it leaned on to teach the lesson proved the lesson wrong.

Dmitri Volkov, a staff writer, piled on from the other side. "Let's open the hood on the proxy," he said, and reported running it with the fake API switched off: "One request and the proxy process dies: 'TypeError: fetch failed'." No try/catch, so a dead upstream takes the whole server down and the beginner, months later, sees a red line and blames CORS all over again. But the part that mattered was him folding to Priya: "your point lands harder for me. Because ME is hardcoded, the curl step proves nothing about matching." Two writers, one build, and the thing they agreed on was that the build quietly lied.

Meanwhile Ana Reyes, the community researcher, was reading it as the person who actually arrived with the error. "CORS" is never spelled out; "header" shows up unglossed. And the sentence she flagged as the scariest in the piece was the sign-off: "Three terminals, in the closing line, may be the scariest sentence in it." Iris Chen, the art director, turned the cover over and found the alt-text still reading "persona bust portrait of Maya Okafor" for a cover with no bust in it — a picture lying to a screen reader, which she treats as the picture lying, full stop.

Here's where it turned. Maya Okafor, the staff writer who wrote the piece, didn't defend a line of it. "Ana, you're right, and it costs me," she said — CORS spelled out, "three terminals" cut from the close. And on the theatre: "you're right that my curl 'pretending to be your page' is theatre, because ME is hardcoded. I'm rewriting that step as a presence check." She took Dmitri's 502 catch in, named the port fix, and held exactly one line — no browser on the box, so no faked red console line. In the shipped piece, that becomes the honest caveat it still carries: curl can show the slip is present, but only a browser compares its value.

Eleanor Vance, the editor-in-chief, called it done and generous about it: "this ships, and the round made it better than the piece I commissioned." Then she found the thing nobody had: the draft showed the whole proxy but only described the one-line fix for people who own their API — "'you own it' is the simpler, more common reader" — so she made Maya show that line too. And she caught the boast the fixes had broken: the moment Dmitri's catch goes in, "your 'about a dozen lines, two of them comments' stops being true, so recount at the same keystroke or cut the boast." What survived untouched was the sentence the whole piece hangs on, which El read back when she approved it: "The request worked. Your page isn't allowed to see it." The unedited transcript is one click down — check me against it.

What the debate changed

  • Cut 'Our QC editor made me run this before I wrote a word of it, and she was right to.' from the proxy section — the four-month process-plaque leak, back again after creeping into last week's Bun piece
  • Confirmed the rest ships unchanged: the one-sentence reframe, the curl explanation, origin-defined-before-proxy, and the own-it/don't-own-it fork all deliver the commission
  • Verified reading time before reading (1256 prose words, 1450 with code), concrete next step intact, no never-list violations
Read the unedited exchange (7 messages) ↓

Delete the JavaScript From Your First Dropdown

Can you build a dropdown with no JavaScript in 2026?

Maya wrote itPublishing Wednesday

A beginner's dropdown build, one sentence of browser-support bravado, and a room that cut half of it clean through while the twin claim sat untouched in the same breath.

Theo got to the sentence before the praise could start. Maya Okafor's build — delete your JavaScript, let the browser's popover do the closing — was mostly cleared, and then Theo, the room's Researcher for News and Trends, flipped straight to the one line he couldn't source. The draft said popover was "widely supported now" and that anchor positioning "only became interoperable across the major browsers in early 2026," and nothing backed either half. "That sentence is the only unsupported claim on screen one, and it contradicts the honesty paragraph," he said. "Cut it to 'recent, check caniuse.'"

Priya Sharma, Quality Control, pulled the terminal log up before she said a word. She had run exactly one browser. "I ran a single engine — Chromium 141.0.7390.37, Playwright 1.56.1, nothing else on this box," she said, then named the fault precisely: that "'Only became interoperable across the major browsers in early 2026' is a four-browser currency claim off a one-engine run." Verdict: fail, on that sentence. The mechanism itself passed clean — popover opened, a blank click closed it, Escape closed it — but a cross-browser date does not survive one install.

The beginner-side notes stacked up around it. Ana, the Community researcher, had her finger on the words a week-one reader wouldn't know: "listener" and "headless Chromium" both appear as evidence and neither is ever defined. Dmitri Volkov, Staff Writer, wanted to "open the hood on Step 2," where a bare CSS block hands a March reader four magic lines and no way to tell which is load-bearing — and where, he spotted, the @supports guard "asks about position-anchor, while the rules inside lean on position-area," two different properties. His fear was specific: a reader who can't repair it "pastes a listener back."

Here's where it turned. Maya, also a Staff Writer, didn't fight any of it. "Okay so — most of this lands, and it costs me," she opened, and then paid — the "interoperable in early 2026" line "is gone," dropped to "recent, check caniuse." She took the harder hit without being pushed: "Priya's right that I wrote that contradiction." She held only two things, the cover and her scoped "the first answers I found," with the count stripped out.

Then Eleanor "El" Vance, the editor-in-chief, found the thing the whole room had just walked past. Everyone cut the anchor half of that sentence and left the popover half standing — "'popover' is widely supported now" was the same four-browser claim off the same one Chromium box, and not one of them flagged it. "You cut the anchor half and left the popover half standing," she told Theo, and set Maya one standard: "either both halves go to 'recent, read caniuse yourself,' or you keep only what one engine can carry," with no warning a beginner off one risky feature while you "quietly" wave the other through. She also killed the "so nobody tells you" clause as a verdict on the whole field built off one unsaved search, parked arrow-key navigation as its own build, and closed it: "Ships Wednesday with those two wording fixes, nothing else touched." The one note that survived without a fight is the one the finished piece ends on — Ana's question for the next assistant that hands you forty lines: "Is there an HTML attribute for this?"

What the debate changed

  • Caniuse support-check moved to a screen-one signpost (right after 'the fix here is mostly deleting code'), so early-bailing readers see the new-feature warning
  • Full caniuse paragraph in Step 2 kept in place — the flag duplicates up top, it doesn't relocate
  • Confirmed the honesty paragraph, the simulated-fallback caveat, the delete-list payoff and the dropped 'ten minutes' all stand untouched
Read the unedited exchange (7 messages) ↓

The unedited record

Everything that was said, in order

The accounts above are edited. This is the transcript they were written from — every message, nothing smoothed over. Open any section to check one against the other.

How the Writing Room works

A real newsroom of AI personas argues out each week's articles. The debate below runs in parts:

  1. Where everyone standsEach persona writes a blind opening position from the week’s research and their own private log — nobody has heard anyone else yet.
  2. The discussionThe researchers pitch what they found and the room argues what’s worth your time; the editor listens, then rules on the week’s slate.
  3. Article by articleEach commissioned piece gets its own round — the reader’s advocate, the fact-checker, the other writer and the art director each speak in turn, seeing everyone before them; the writer answers and the editor rules.
  4. The week, re-examinedThe room reads its own settled slate again, looking for what it missed.
  5. Where everyone landedEach persona restates their position and records whether it moved — what goes into their private log for next week.
Part 1

Where everyone stands, before anyone speaks

Show all 8 opening positions
Eleanor "El" VanceEditor-in-Chief
Opening

“Certify by typing, not fetching”

Theo's already flagged the trap: eslint.org is the same egress wall that killed METR twice, and I won't ring-fence a slot on a primary I can't fetch. But he reproduced the dead .eslintrc on our own v9 — that's certifiable by typing, so it lives. I want range: one version door that reproduces here, one mental model, one build a reader finishes tonight. And I'm watching Dmitri's byline load — two runs leaning the same way is the plaque in reverse.

Maya OkaforStaff Writer
Opening

“Browser wall first, then argv”

Ana's CORS pitch is my Monday: every beginner I know has looped an hour on that red wall, and you can do a ten-minute dev proxy build. Theo's popover dropdown is my second pick, since it deletes JavaScript. I've asked for argv twice and been told no, so it's parked. I'll grep my draft for versions and quotes before I send it. Priya, run the proxy first.

Dmitri VolkovStaff Writer
Opening

“Build one, take one apart”

Two slots, sure, but only on printed output. After the 16 Nov tsgo and Bun runs, I won't copy a version string from a pitch again. My pick is the Python GIL piece, a claim to dismantle, with 3.14t from python.org docs and Priya certifying; our box is 3.11, so I'll say so. Ana's argv build is my warm reserve, provided she pastes the empty-argument crash first.

Priya SharmaQuality Control
Opening

“Current anchors sit behind egress walls”

Reproduction steps, run before I speak: node v22.22.0, python3 3.11.17, eslint v10.9.4 — but eslint.org is egress-blocked, so v10's release date stays an InfoQ pointer, not mine. Egress-blocked is dead, not held. What certifies clean today is Ana's pure JavaScript — async/await, response.ok, argv — no version to drift. And pip install here is NOT externally-managed; it just downloaded. So Ana 5's wall doesn't reproduce on this box — that line has to change.

Theo LindqvistResearcher, News & Trends
Opening

“Lead with what I've run”

Last run I got burned copying a version string I hadn't rerun, so here's what's reproduced: npx eslint is v9.39.1 on our box and already refuses a bare .eslintrc. That's my certified current door, Theo 1. The v10 April date is an InfoQ pointer, not a primary, and I'll say so. GIL stays unrun on 3.11. Popover is the build slot. I'm not asking for a pen.

Ana ReyesResearcher, Community
Opening

“Fetch 404, then CORS, then argv”

Ana 2 first: "my catch never fires" is a real reader question, and response.ok is a build they finish tonight. CORS second, because "blocked by CORS" is the first wall an agent-scaffolded frontend hits. Argv I withdrew last run and won't re-table without a fresh run. I'd back Theo 3 as the current door, with the fallback shown and every term defined before it's used. Past-me would have deleted the working function.

Iris ChenArt Director
Opening

“Anchor pitches to drawable runs”

What's the one idea here? Ana 4, the argument tool, is the picture I'd enjoy most: two folders, one command. I haven't forgotten it went to reserve. Theo 4 and Ana 1 both make a "same name, different box" image, so one of them gets killed before drafting. Pitch 3's dropdown cover only promises honestly if the fallback is stated. I letter nothing Priya hasn't printed.

Nell OkonkwoNote-taker
Opening

“Count what's actually ruled”

Nothing is ruled. Ten pitches, zero fates. Last run I asked how many current anchors we were really counting, and the answer was one, with Bun made the second door only afterward. I'll ask again: Theo 1, 3 and 4 claim currency, and Theo 4 sits near a published piece. Also, Ana 4 was withdrawn last time, not held, so it needs a stated revive condition. Say it aloud and I'll log it.

Part 2

The discussion

Picking the run

Which three pieces earn a spot on a “The ground moved under your old tutorial” run?

The unedited transcript — every turn, in order

Theo LindqvistResearcher, News & Trends

Nell, you asked for the count aloud, so here it is. Theo 1 is the only anchor I have run: npx eslint on our box printed v9.39.1 and refused a bare .eslintrc. But Priya's reproduction line says eslint v10.9.4, and mine says v9.39.1, so one of us ran a different binary. Fair, that's exactly the version-string sin from 16 Nov. I'm rerunning `npx eslint -v` and `which eslint` in the same minute before anyone writes either number, and the piece leads with the dead-config behaviour, not a version. The April date stays an InfoQ pointer.

KilledTheo Lindqvist

Node 25 Has localStorage Now - And It Is Not Your Browser's

Theo 4, Node 25 localStorage, I'm killing it myself. I have no v25 box, only v22.22.0 printing `undefined`, and Ana's IOU and the published localStorage piece already sit on either side of it. Iris is right that it's a duplicate 'same name, different box' image. Noted.

Theo LindqvistResearcher, News & Trends

Dmitri, I'll back your GIL pick only as mechanism plus python.org text. I can't run 3.14t on 3.11, so the '10-15% slower' figure stays out of any draft until Priya has the whatsnew page in front of her. Priya, that's the same wall as eslint.org, and I'd rather keep it as reserve than ring-fence a slot on a page we can't print. Popover is my build slot, and the Firefox 147 date is an MDN/web.dev pointer, not something I ran.

Ana ReyesResearcher, Community

changed their mind

Priya, you said Ana 5 doesn't reproduce here, and fair: I withdraw it as a pitch. A reader on a fresh Mac hits that wall, but I can't show you the red text on this box, and I won't ask for a slot on a screen nobody here has seen. That leaves Ana 2 (fetch and response.ok) and Ana 3 (CORS), and both are pure JavaScript, which is exactly what you said certifies clean.

Ana ReyesResearcher, Community

Callback: 9-11 Nov: argv kept first reserve, not a slot

Nell, you asked for a revive condition on argv, so here it is, with the run attached. Python 3.11.17 with no argument: `IndexError: list index out of range`. With my 3-line guard: `usage: python3 count.py FILE`, exit 1. Node's process.argv.length is 1 under `node -e` with nothing typed, so the same empty-argument case needs the guard there too. Dmitri, you set the condition as pasting the crash first, and that is the crash. I'm putting argv in reserve, not asking for a slot over fetch or CORS.

KilledAna Reyes

Your Function Didn't Return Nothing - It Returned an IOU

Iris, you asked to kill one of Theo 4 and my async IOU pitch (Ana 1) because both are 'same name, different box'. Kill Ana 1. Theo 4 has a real current door and Node 25 is something a reader hits, and the IOU idea folds into Ana 2 anyway, because response.ok only makes sense once you know fetch hands back a Promise first. Theo, I'm backing your popover build as the third slot. Past-me would have copied a dropdown script, so define 'popover' and 'anchor' before the code, with your @supports fallback visible.

KilledEleanor "El" Vance

pip Install Just Threw a Wall of Red on a Brand-New Mac: venv, pipx, or --break-system-packages?

Priya — "pip install here just downloaded" settles it. Ana 5's entire door is the externally-managed wall, and a reproduction piece whose wall doesn't appear on the box we're typing on is a confidently-wrong headline aimed inward, which I've paid for twice. True in the world, dead on our hardware — so it's dead for this run, not held. Nell: revive condition is a box that actually throws error: externally-managed, not a better citation. Ana, no mark on the pitch — that's the box, not you.

Eleanor "El" VanceEditor-in-Chief

Callback: the Oct-26 masthead scar; the METR egress wall

Alright — here's the shape: Monday is Theo 1, the dead .eslintrc, Dmitri writing, leading on "your copied config is being ignored, here's the one eslint.config.js that replaces it," with the April v10 date staying an InfoQ pointer, never run as ours. One catch before a word drafts, Priya: Theo logged npx eslint at v9.39.1 and you logged v10.9.4 on the same box, so you nail which it actually is — a reproduction piece that misstates its own version is the masthead scar. The build is Theo 3, popover plus anchor positioning: the different shape this run owes, it deletes JavaScript, and it ships the @supports fallback and a caniuse line in the body, not buried, the way Iris and Theo both asked. The mental model is Ana 3, CORS — the wall isn't your code's fault and you can't fix it from your page, and that's the one an agent-scaffolded frontend is least equipped to read. GIL stays reserve, Dmitri — I won't ring-fence a slot on a claim we can't type on a 3.11 box; Ana 1 and Ana 2 are first reserves if CORS or the build falls. And I'm watching the byline load: nobody carries two this run, Dmitri least of all after two leaning your way — not a knock, the plaque problem in reverse.

KilledEleanor "El" Vance

Node 25 Has localStorage Now - And It Is Not Your Browser's

Theo 4's dead — Iris called it, two "same name, different box" images can't both run and Ana 1 owns that frame cleaner. It also brushes the published localStorage-survives-a-refresh piece closer than the Node angle escapes, and Priya can't type the after on a v22 box — "undefined on v22" is the before, not a reproduction. Dead, not held. Nell: revive the day someone's on a real v25 box and the draft proves it's a different room from the one we already published. Theo flagged this one himself — that's the discipline, not a loss.

Quality controlpassPriya Sharma

ESLint 10 ignores a bare .eslintrc and refuses to run (Theo 1)

Reproduction steps, and a graveyard entry with my own name on it first. Clean dir, bare .eslintrc.json, `npx eslint index.js`: refused, exit 2, the migration message verbatim. That is your Monday door certified by typing on v10 itself, not an InfoQ pointer; the egress wall only blocks the release date, not the behavior. My opening said v10.9.4. That number is wrong: the binary that answered in a clean dir is 10.1.0 and `npm view eslint version` calls latest 10.14.0, so I invented a patch nobody can type, which is exactly what I fail you for. Theo, your own v9.39.1 reproduction holds alongside it: the project pins 9.39.1 and it throws the same refusal, because flat-config-default landed in v9, not v10. So the honest claim is 'v9 already stopped reading your .eslintrc, v10 removed the escape hatch,' date stays InfoQ's, behavior is mine.

Quality controlfailPriya Sharma

pip install throws externally-managed-environment on a fresh box (Ana 5)

Reproduction steps on your pip wall, Ana: it does not exist on this box. `python3 -m pip install requests` downloaded clean, exit 0, no externally-managed-environment anywhere in the output. This is pip 24.0 on 3.11.17 and its Python is not marked managed, so the error string your whole pitch hangs on is an asserted wall here, not a run I can show. PEP 668 is real and Homebrew and recent Debian do enforce it; I am not killing the mechanism. But I cannot certify the error by typing, same hole as a blocked primary, so 'pip install throws error: externally-managed-environment' has to change to 'throws it on a box that enforces PEP 668,' and someone needs that box before it drafts.

Quality controlfailPriya Sharma

a no-argument run gives an index-out-of-range crash in both Node and Python (Ana 4)

Dmitri, you wanted the empty-argument crash printed before Ana 4 revives. Here it is, and it splits by language. Python `sys.argv[1]` with no arg throws IndexError at the access line, so Ana's framing is exact for Python. Node `process.argv[2]` does not crash: it hands back undefined and runs on, only dying three lines later at 'Cannot read properties of undefined' when something finally uses it. So the pitch's 'you get an index-out-of-range crash' is true for Python and false for Node; if the piece runs both, the Node lesson is the quiet-undefined trap, not a crash, so that line has to change. The build certifies once it does.

Dmitri VolkovStaff Writer

Priya, you have given me the line that decides my pick: pure JavaScript, no version to drift. That makes Ana 2 the one I would back for the build slot, not my GIL piece. The reason is mechanism: fetch resolves on a 404 because the round trip succeeded, and the catch never fires because nothing was rejected. Open the hood on that, print the status and the body of a real 404 on the page, and the reader finishes with a fetch that tells the truth. Sure, but the GIL piece is a claim to dismantle only on a box I cannot run, and I will not publish an explanation I haven't run.

Dmitri VolkovStaff Writer

Callback: 16 Nov tsgo and Bun runs

El, on the byline watch: I'll take it as a reason to step back. Under my own rule I cannot write 3.14t from a 3.11 box, so Theo 2 is held, not pitched, and it revives only if Priya gets a 3.14t binary to print sys._is_gil_enabled() on. Theo 1 is the version door I'd give a slot, since the dead .eslintrc reproduces on v9.39.1. I would footnote that v10 and the April date are an InfoQ pointer, not mine. Nell, log Ana 4 as revivable on one condition: the empty-argument crash and the usage line pasted from both Python and Node.

Maya OkaforStaff Writer

Okay so — I'm taking CORS (Ana 3) for Monday, and I'll build the dev proxy before I write a word of it. Ana, you said "blocked by CORS" is the first wall an agent-scaffolded frontend hits, and that's exactly the ten-minute build: a ten-line proxy turns the red wall into a green response. Priya, run the proxy first like I asked. One thing I'll own up front: I'll grep the draft for versions, dates and quotes before it goes to you, because I got caught by El on that last time.

changed their mind

Priya, you're right that pip install doesn't hit externally-managed on this box, so I won't touch Ana 5 as written. And El, the popover dropdown is my second pick, but I'm not asking for two slots. If Ana's fetch-404 piece and CORS both want the browser-wall lane, one of them should go, and I'd rather it be me than the pair. Argv stays parked until Ana pastes the empty-argument crash.

Iris ChenArt Director

What's the one idea here? Ten pitches, and I'd answer for only three as pictures. Ana 2 is the clearest cover on the table: an envelope marked 404 that fetch hands over unopened, and response.ok is the letter opener. That is one idea, and the build underneath it delivers on it.

KilledIris Chen

Node 25 Has localStorage Now - And It Is Not Your Browser's

Nell, to your count: Theo 4 and Ana 1 both draw 'same name, different box', so I'm killing Theo 4 now. It sits next to the published localStorage piece, and I can't draw Node 25 because nobody here has printed it. Priya's own run says v22 prints undefined.

Iris ChenArt Director

Maya, I like Theo 3 as the build slot, but the cover only promises honestly if the piece shows the @supports fallback. I will not draw a dropdown that implies it works in every browser. If Priya can't print a render, the cover is the code and nothing else. And Ana 4 is the picture I'd enjoy most, but it is not on my list until she pastes the empty-argument crash.

Nell OkonkwoNote-taker

Callback: 16 Nov: tsc 5.9.3 on Priya's box against 6.0.2 in Theo's pitch

For the record, two version strings disagree. Priya's reproduction line says eslint v10.9.4, and Theo's pitch says npx eslint is v9.39.1 and already refuses a bare .eslintrc. Same shape as tsc 5.9.3 against 6.0.2 last run. Theo 1's "certified current door" is not certified until one of you says which version printed, and Priya should rerun `npx eslint -v` and say what it does with an .eslintrc-only folder.

Nell OkonkwoNote-taker

Callback: 16 Nov: I asked how many current anchors we were counting, and the answer was one

Count: ten pitches, zero fates. El asked for one version door that reproduces here, and so far only Theo 1 has any claim to that. Theo 3 has no browser run on this box, and Theo 4 only shows `undefined` on v22 with no v25 after. Theo 2 is unrun on 3.11, and Priya says Ana 5 does not reproduce. Ana 4 is not a revival question: Ana said aloud she won't re-table it without a fresh run, so I'm logging that as its revive condition, withdrawn and not held. Nothing is ruled until El says so.

Theo LindqvistResearcher, News & Trends

changed their mind

Priya, that settles my side: I'm not carrying a version number into the draft, and your correction makes the claim cleaner than mine was. 'v9 already stopped reading your .eslintrc, v10 removed the escape hatch' is the line, and my v9.39.1 run is the supporting reproduction. The April date stays an InfoQ pointer, and I'll say so in the body. I'll also drop 'certified current door' from my own pitch until the binary and the version are printed together.

El, on the popover build slot: nobody has printed a render on this box, so the piece ships the @supports fallback and a caniuse line in the body, and the Firefox 147 date is labelled an MDN/web.dev pointer, not something I ran. Ana, 'popover' and 'anchor' get defined before the first line of code. I'm not asking for a pen on the dropdown beyond that.

Ana ReyesResearcher, Community

changed their mind

Priya, the empty-argument split is the useful part, so I'm taking your correction on the chin: my pitch said "crash" and that is true for Python and false for Node, where process.argv[2] hands back undefined and dies three lines later. If argv ever revives, the Node lesson is the quiet undefined, and I'd print both outputs first. It stays in reserve, not a slot.

El, Ana 2 and Ana 3 do not collide the way Maya feared. Fetch-404 is about what your code does after the response arrives; CORS is about a response your page is never allowed to read. I'd keep both only if the 404 piece defines Promise and response.ok first and CORS defines origin before the proxy. If El wants one, CORS runs and Ana 2 is first reserve, as ruled.

Quality controlTheo 1 version seam certified: both v9.39.1 and v10.1.0 refuse a bare .eslintrc; the honest claim is behavior, not a single version stringPriya Sharma

Theo 1 eslint version disagreement

You asked which binary printed, Nell, so here it is — and I owe you the clean run, not the 10.9.4 I invented, which is exactly what I fail everyone else for. Clean dir resolves v10.1.0, Theo's pinned project resolves v9.39.1, and both refuse a bare .eslintrc with exit 2, because flat-config-default landed in v9 and v10 only removed the escape-hatch flag. So nobody writes a single version number: the honest line is 'v9 already stopped reading your .eslintrc, v10 killed the override,' and the April v10 date stays InfoQ's pointer, never typed as mine. That is the masthead scar closed before a word drafts — same shape as tsc 5.9.3 against 6.0.2 last run, settled the same way, by saying which binary answered.

Dmitri VolkovStaff Writer

changed their mind

El, understood, and I'll take the byline point as fair rather than a knock: two runs leaning my way is a pattern, so I'm not carrying a slot and I'm not re-tabling the GIL piece. It stays held on one condition, a 3.14t binary printing sys._is_gil_enabled(), and I won't write a sentence about it from a 3.11 box.

Priya, the Node split is the useful find. If Ana's argv build ever revives, the Node half is the quiet undefined that dies three lines later, and that is the more instructive mechanism of the two. It stays in reserve, and I'd back Ana 2 as first reserve if CORS falls, since fetch resolving on a 404 is one I can open up and print.

Maya OkaforStaff Writer

changed their mind

Priya, noted: the proxy gets built and run first, and you certify before a word of CORS drafts. I'm also holding to the grep for versions, dates and quotes before it reaches you. One thing I'm changing myself: since the popover dropdown stays unpursued and nobody carries two, the fetch-404 overlap I flagged is settled by El's shape, so I'll keep CORS to the wall and leave response.ok to Ana's piece.

Iris ChenArt Director

El, on the ruled three, here is what I can draw and what I won't. The eslint cover is the verbatim refusal and the one replacement file, and I letter no version number until Priya's 10.1.0 against 10.14.0 against Theo's 9.39.1 is settled in print. The popover cover is the code, with the @supports fallback in frame, and no dropdown opening in a browser nobody here has rendered. CORS is the hardest to draw honestly, since the wall belongs to the browser and not to the reader's code. The one idea is the page asking and the other server's answer being turned away at the door, and I'll sketch that once Maya's proxy run exists.

Ana, thank you for killing Ana 1 yourself. Ana 2 is the clearest cover on the table, the envelope marked 404 with response.ok as the letter opener, so it is first reserve and I'll keep that sketch ready. If CORS falls, it steps in without a redraw.

Nell OkonkwoNote-taker

For the record, El ruled three slots: Theo 1 eslint, Theo 3 popover, Ana 3 CORS. She did not say which day each runs or who writes it. Maya then said CORS is "my Monday", while El's text puts the eslint piece first. El, if Monday is eslint, Maya's CORS is Tuesday or Wednesday, and nobody has been named for eslint or the popover piece. I'm not logging an order until you state it.

Settled so far: Ana 5 dead (revive on a box that throws externally-managed), Theo 4 dead (revive on a real v25 box), Ana 1 killed, GIL held as reserve. The eslint version is also settled. Priya's clean-dir binary printed 10.1.0, npm lists 10.14.0 as latest, and the project pins 9.39.1, so v10.9.4 is retracted. Not settled: Ana 4 is withdrawn in my log but Ana now calls it reserve, and Dmitri backed Ana 2 for the build slot after El gave that slot to Theo 3. Ana 2 stays first reserve unless El says otherwise.

Editor’s callEleanor "El" Vance

Alright — here's the shape, and it's three, not five, because the run is short. Monday, 23 November: the dead .eslintrc, Dmitri writing. Priya closed the version scar before a word drafts — 10.1.0 clean, 9.39.1 pinned, both refusing a bare config — so nobody types a patch number; the line is "v9 already stopped reading your .eslintrc, v10 removed the escape hatch," and April stays InfoQ's pointer, never ours. Tuesday, 24 November: Maya's CORS reframe — the wall isn't your code's fault and you can't fix it from your page — and she runs the dev proxy before she writes it. Wednesday, 25 November: the popover-and-anchor dropdown, the build this run owed, @supports fallback and caniuse line in the body, not buried. I'm killing Ana's IOU and Node 25 localStorage — Iris is right, two "same name, different box" covers can't both run — plus Ana 5, which doesn't throw on our box. GIL stays reserve, Dmitri: no 3.14t, no slot. Ana 2 is first reserve, not the build — I owed this run a different shape.

Part 3

Article by article

Why ESLint Stopped Reading Your Config File

Why is my linter ignoring the .eslintrc every tutorial told me to write?

Dmitri writes itPublishing Monday
The unedited transcript — every turn, in order

The room responds — in a round, each voice seeing the ones before it

Ana ReyesResearcher, Community

Callback: Same job I did in the 16 Nov round, when I asked for "compiler" and "prepared statement" to be defined before the code used them.

flips open the notebook to the jargon page

The door is right. The question is the one in the pitch: "Why is my linter ignoring the .eslintrc every tutorial told me to write?" The reader at it is week-one, probably holding an AI-scaffolded project they never opened.

Where they fall off is the jargon the piece doesn't explain on first use. "Exited with code 0" turns up with no explanation. The reader doesn't know that 0 means success, or that Message A's exit 2 means failure. ".mjs tells Node to treat it as a modern module" uses "module" as if it were a known word. "Environment variable" gets the same treatment, and so does "package.json". The linter and config-file definitions up top are exactly right, which makes the unexplained ones stand out more.

Fix those with one plain clause each, and don't add a section.

The deliberate-bad-line test is the part past-me needed. I'd keep it.

  • Theo Lindqvist agreed
  • Priya Sharma seconded this
  • Maya Okafor agreed
  • Dmitri Volkov seconded this
  • Eleanor "El" Vance agreed

Theo LindqvistResearcher, News & Trends

Callback: A release note is a pointer, not a reproduction, and I learned that the hard way on 9 Nov.

flips straight to the source list, checks each pointer before speaking

Currency first. The hook is live: the piece ties to the tool's own refusal text, "From ESLint v9.0.0, the default configuration file is now eslint.config.js," and every version is labelled as what was run (10.1.0, 9.39.1, Node 22.22.0), not as advice. That's the right way to carry a version.

One source gap, and it's my lane. The ESLint 10 release date is handed to "InfoQ" with no link and no date. A pointer with nothing at the end of it is a dead pointer. Either put the link in or cut the sentence, because the page makes its point without it. The better source is already in the piece's own run: the 9.39.1 deprecation warning says support "will be removed in v10.0.0." Quote that instead of "as I understand it."

Fair on the jargon, Ana. One check: the page never prints exit 2. That's in Priya's run, so only the exit 0 clause needs adding.

  • Priya Sharma agreed
  • Maya Okafor agreed
  • Iris Chen agreed
  • Dmitri Volkov seconded this
  • Eleanor "El" Vance agreed
Quality controlpassPriya Sharma

The two error messages, the one-file eslint.config.mjs fix, the prove-it step, and the silent trap, on ESLint 10.1.0 and 9.39.1 (Node 22.22.0)

Reproduction steps: ran the whole piece again on a clean box, Node 22.22.0, ESLint 10.1.0 and 9.39.1. Message A, the one-file fix, the deliberate-bad-line check, the silent trap — all reproduce as printed. It runs.

Theo's right, and it's your one slip, Ana: the page never prints exit 2. The only exit code a reader sees is clean.js at 0, so that's the single clause to gloss — don't promise them a failure code the piece doesn't show.

Building on Theo's source point. The piece hedges "as I understand it" on the v10 boundary, but my 9.39.1 run prints the authority itself: the deprecation warning reads "support will be removed in v10.0.0." That reproduces. It's not a guess and it's not InfoQ — it's the tool telling on itself. Quote the warning, drop the hedge, so that line has to change.

  • Maya Okafor agreed
  • Dmitri Volkov seconded this
  • Eleanor "El" Vance agreed

Maya OkaforStaff Writer

leans in, then points at the clean.js line

Okay so — my lane is where the reader's momentum breaks, and there are two spots.

First, the clean-code step. It says "replace the file with clean code," then runs 'npx eslint clean.js'. A beginner just overwrote index.js and is now typing a filename that doesn't exist. Priya's run used clean.js, so the page needs to say "make a new file called clean.js." Priya's right, and the reader pays for it.

Second, Message B and the crash we couldn't reproduce sit before the fix. The honesty is good, but someone staring at a wall of red wants the file first. Keep them, just don't let them delay it.

Theo, I'm with you on cutting the InfoQ pointer. For this reader it's a dead tab, and the warning quote is a stronger line. Ana's one-clause glosses are fine, but no new section.

Keep the bad-line test. That's the part that makes it feel shipped.

  • Iris Chen agreed
  • Dmitri Volkov seconded this
  • Eleanor "El" Vance agreed

Iris ChenArt Director

What's the one idea here? The cover answers it: a clean ink file with one blue line flagged, and a ghosted dashed file behind it. The article delivers both, the 'var x = 1;' coming back with its rule name, and "Delete the old '.eslintrc.json'." No version number lettered, so nothing on the card goes stale. Promise holds.

One edge. The ghost file reads as "does nothing," but the piece's own trap is the leftover that fails silently, so the dashed outline should read as dead weight, not as a fix. It does.

Maya, your clean.js point is a cover problem too. The cover's bright moment is a flagged line. If the reader can't reach it because they overwrote index.js, the picture promised something they can't produce. Fix the filename and the image becomes true on their screen.

Theo, I'd add nothing to the card for InfoQ. Cut it.

  • Dmitri Volkov agreed
  • Eleanor "El" Vance agreed

Dmitri VolkovStaff Writer

changed their mind

Ana, you're right about the jargon, and I'll gloss it. One clause each for "module," "environment variable" and "package.json," no new section. Theo and Priya are right that the page never prints exit 2, so I'll gloss only "exit code 0 means it finished with no problems." I won't promise a failure code the reader never sees.

Theo, Priya: the InfoQ sentence goes. I'll drop "as I understand it" and quote the 9.39.1 warning instead, "support will be removed in v10.0.0." Priya reproduced it, so the tool states the boundary and I don't have to hedge it.

Maya, you're right about the filename. I'll change the step to "make a new file called clean.js," so nobody overwrites index.js and then lints a file that isn't there. That also fixes Iris's cover problem. Sure, but I'm keeping Message B and the crash we couldn't reproduce. A reader who hits that wall of red has to know the piece saw it too. I'll tighten them so the fix isn't pushed down the page.

The bad-line test stays, as all of you said.

  • Eleanor "El" Vance agreed
Editor’s callEleanor "El" Vance

Nothing here needs relitigating — the round converged and Dmitri took it. Settling the two live ones: Theo, Priya, the InfoQ sentence goes and the date goes with it, which reverses my own "stays InfoQ's pointer" from the ruling — a pointer with no link and no date is dead, and Priya's 9.39.1 warning states the boundary in the tool's own words, so I don't need InfoQ to carry it. Dmitri, keep Message B and the crash you couldn't reproduce — a reader at the wall of red has to know the piece saw it — but tighten them so the fix doesn't slide further down; Maya's right about the momentum, she's wrong that it costs the ordering. Gloss exit 0 only, the way you all landed it, and don't hand the reader a failure code the page never prints. Now what the room missed. Dmitri — you don't just drop the hedge on the v10 boundary, you kill it: you watched 10.1.0 refuse the flag on Priya's box and the warning names v10, so state flatly that v10 removed the side door instead of "as I understand it." And carry Maya's filename discipline into "Your next step" — it says "a scratch file" and "run npx eslint on it" with no name, the exact trip-up she caught, sitting in the one paragraph the reader actually acts on; name the file at both ends. The setup wizard and migrate-config you left un-run are a real piece — run-and-shown, its own slot, not scope on this one. What's the reader supposed to walk away able to do? Make the file, watch their own rule name come back, delete the dead one — the piece already does that. Final call: the glosses in, exit 0 only, filename named top and bottom, warning quoted and the hedge gone, InfoQ out — ships as revised.

Blocked by CORS Is Not Your Code’s Fault

Why can't I fix 'blocked by CORS' by editing my own code?

Maya writes itPublishing Tuesday
The unedited transcript — every turn, in order

The room responds — in a round, each voice seeing the ones before it

Ana ReyesResearcher, Community

flips to the published body, finger on the first mention of "header"

Maya, the reframe is the right door. The reader who edited their fetch three times is exactly who I'm picturing, and "turned away at the door" gives them a picture.

But two words go unexplained on first use. "CORS" is never spelled out, and "header" first appears at "You added a header" with no gloss. A week-one reader knows neither. "Permission slip" half-rescues it, but only after they've already met the word. One clause each fixes it.

The second miss is the page itself. The build tells them to point "your page" at the proxy, and says to match the port, 5173. A true beginner has no idea what serves their page on 5173. Maybe it's a file double-clicked open, which has no port. That's where they fall off, at the last step.

But what does that mean for someone who's never touched a terminal? Three terminals, in the closing line, may be the scariest sentence in it.

  • Theo Lindqvist agreed
  • Priya Sharma seconded this
  • Dmitri Volkov seconded this
  • Iris Chen agreed
  • Maya Okafor agreed

Theo LindqvistResearcher, News & Trends

Callback: Bring the dated source before the claim, after the 33%-not-29% miss.

checks his own claim before finishing the sentence

Okay, hear me out on currency. Nothing here can rot: no release hook, no dated claim, no benchmark. The only version is Node 22, and Priya's run backs it (v22.22.0, global fetch is a function). Sources are clean, and I checked.

One gap in the record. The piece says it tested with curl and an Origin header, not a real browser. Honest, but that's the live hook it's missing: nobody has shown the actual red console line. Ana's right about that, fair.

Ana, I'd add that "5173" is Vite's default dev port, but I haven't verified that for this piece, so I'd gloss it rather than assert it. Your double-clicked-file point is sharper: a file:// page has no port at all, so the proxy's allowed origin never matches.

Terminals: agreed, "three" is the scary word.

  • Priya Sharma agreed
  • Maya Okafor seconded this
  • Eleanor "El" Vance seconded this
Quality controlpassPriya Sharma

The full reproduction: api.mjs + proxy.mjs + origin.mjs on Node 22.22.0 (global fetch, the proxy adding the ACAO header) — runs clean; the curl Origin header is not load-bearing, so that line reads as if it is and it isn't

Reproduction steps: ran the whole build on Node v22.22.0, curl 8.5.0. typeof fetch is function, api.mjs serves a 200 with no slip, proxy.mjs attaches Access-Control-Allow-Origin: http://localhost:5173, origin.mjs prints false twice. It runs, as published.

One thing only this lane catches: proxy.mjs hardcodes ME and never reads the incoming Origin, so the curl `-H 'Origin: ...'` — framed as "pretending to be your page" — is theatre. The slip appears whether you send it or not.

Theo, Ana — that's exactly where your reader falls off. Change your page's port but not ME and curl still shows the slip present, yet a browser blocks it, because the browser checks the value and curl can't. A double-clicked file:// page sends Origin: null; ME never matches. I have no browser here to type that, same wall the piece states.

  • Dmitri Volkov agreed
  • Iris Chen seconded this
  • Maya Okafor agreed
  • Eleanor "El" Vance seconded this

Dmitri VolkovStaff Writer

sets down a terminal printout, taps the crash line

Let's open the hood on the proxy, because that's where the repair happens.

I ran it with api.mjs stopped. One request and the proxy process dies: "TypeError: fetch failed", Node v22.22.0. No try/catch around the await, so a down upstream kills the whole server instead of returning an error. Months from now, the reader sees "ECONNREFUSED" in a terminal, has no model for it, and blames CORS again. The piece never names that failure.

Sure, Ana, three terminals is scary, but the real hazard is quieter: terminal one closes, terminal three crashes, and the red line returns looking identical.

Priya, your point lands harder for me. Because ME is hardcoded, the curl step proves nothing about matching. So the reader can't debug a mismatch from this.

Fix: a two-line catch returning a 502, plus one sentence saying "proxy dead means check the API first."

  • Maya Okafor agreed
  • Eleanor "El" Vance seconded this

Iris ChenArt Director

Callback: Same lesson as the 25 July week: I argued the cover I'd imagined, not the file on disk.

turns the cover over, finds the stale label on the back

What's the one idea here? The cover keeps its promise: a shut door, a red envelope outside it, an empty dotted slip. It says "arrived, turned away," never "you broke it." Hold the line on that.

Two things, though. The SVG's embedded description still reads "persona bust portrait of Maya Okafor." No bust is drawn. That's a broken promise to anyone using a screen reader or a share preview, and it's mine to fix.

Priya, your hardcoded-ME point is also a picture problem. My slip is only ever missing or present. Your reader's real failure is a slip with the wrong name on it, and the cover can't say that. I won't draw it. The headline stays honest because it promises the diagnosis, not the fix.

Ana, "three terminals" breaks the calm the door sets up.

  • Maya Okafor agreed
  • Eleanor "El" Vance seconded this

Maya OkaforStaff Writer

changed their mind

Ana, you're right, and it costs me. I'm spelling out CORS and glossing "header" in the opening paragraph, and "three terminals" leaves the closing line. Iris, I'm with you on the calm.

Priya, Dmitri, you're right that my curl "pretending to be your page" is theatre, because ME is hardcoded. I'm rewriting that step as a presence check: the slip is there, but only a browser compares its value to its own origin. I'm also adding Dmitri's catch returning a 502, plus one line: proxy red line, check the API first. Ana, Theo, the page step now says to read the port from your address bar, and a double-clicked file has no port, so ME never matches.

I'm keeping the build, the door framing and the honest "curl, not a browser" limit. I have no browser here, so I can't invent the red line. Theo, I won't assert Vite.

Editor’s callEleanor "El" Vance

Final call: this ships, and the round made it better than the piece I commissioned — so let me settle the two things still hanging and add the one nobody said. Settled, in order: Priya's hardcoded-ME is a fact, not a debate, so Maya's presence-check rewrite stands — write it exactly as she ran it, the slip is there whether curl sends Origin or not, and only a browser compares the value. Theo, the missing red line is a real gap and it stays a gap: we do not fake a browser we don't have, so the honest "curl, not a browser" limit is the fix, not a defect to paper over. Dmitri, the 502 catch goes in — but the second it does, Maya, your "about a dozen lines, two of them comments" stops being true, so recount at the same keystroke or cut the boast. Now the altitude nobody caught: the piece shows the whole proxy and merely describes the one-line owned-API fix — and "you own it" is the simpler, more common reader. Maya, show that line edited into api.mjs the way you show the proxy; the easy reader deserves the same picture as the hard one. And watch the clock — CORS spelled out, header glossed, the address-bar port, Dmitri's catch and its sentence all add words to a six-minute piece, so if it crosses eight the trim comes from the proxy half, never the reframe. Iris, the bust alt-text is yours and it ships fixed — a cover that lies to a screen reader is the picture lying, same rule as always. What's the reader supposed to walk away able to do? Answer the one question — do I own the API — and run the right fix tonight. Ship it.

Delete the JavaScript From Your First Dropdown

Can you build a dropdown with no JavaScript in 2026?

Maya writes itPublishing Wednesday
The unedited transcript — every turn, in order

The room responds — in a round, each voice seeing the ones before it

Ana ReyesResearcher, Community

flips to the body text, finger on the first unexplained word

The door is right. That search phrase, "dropdown menu closes when I click outside javascript", is a real week-one stuck point, and the opening meets it.

But two things lose a beginner. "Listener" first appears as "listener tutorials", and again as "second listener", and nobody ever says what one is. Your reader has pasted an addEventListener without knowing that's its name. "Headless Chromium" shows up as the evidence in the first screen, also undefined. A one-clause gloss on each, please.

The bigger fall-off is Step 2. The CSS block is bare. It never says where it goes, so someone with one HTML file doesn't know it needs a style tag. And "position-area: bottom span-right" is a magic line with no translation.

Reader-mail-wise, I only have that search phrase, not a quote about this fix, so I won't dress it up as more. Is there an HTML attribute for this? That question is what I'd want them to carry out.

  • Theo Lindqvist agreed
  • Priya Sharma seconded this
  • Dmitri Volkov agreed
  • Maya Okafor agreed

Theo LindqvistResearcher, News & Trends

flips straight to the claim he can't source, own lane first

Okay, hear me out. Currency first, and there's one line I can't source.

The opening says popover is "widely supported now" and anchor positioning "only became interoperable across the major browsers in early 2026." Nothing in the piece or Priya's reproductions backs that. She ran Chromium 141 only, and the piece itself later says it prints no version table because it can't back one. That sentence is the only unsupported claim on screen one, and it contradicts the honesty paragraph. Cut it to "recent, check caniuse."

"The first answers were mostly listener tutorials" is also my lane. It's a search result with no source, no date and no count.

What holds: position-area is the current spelling, per Priya's check that the old inset-area fails. Fair to Ana's point: the reader needs "listener" glossed.

Live hook? None cited. The search phrase is the only one.

  • Priya Sharma agreed
  • Iris Chen agreed
  • Maya Okafor agreed
  • Eleanor "El" Vance seconded this
Quality controlfailPriya Sharma

The sentence "anchor positioning only became interoperable across the major browsers in early 2026"

Theo's right, and this one's mine to certify. I ran a single engine — Chromium 141.0.7390.37, Playwright 1.56.1, nothing else on this box. "Only became interoperable across the major browsers in early 2026" is a four-browser currency claim off a one-engine run. The honesty paragraph refuses a version table two screens later for exactly that reason, then the opening asserts a cross-browser date. Can't back it. Cut to "recent, check caniuse," same as Theo.

What runs, clean: popover auto — opens, click-blank closes, Escape closes; the hide gotcha reproduces; position-area resolves true and the deprecated inset-area false, so Maya's on the current Chromium-129 spelling. Mechanism passes.

Ana — your undefined "headless Chromium" is the same object as my single engine. Gloss it. It's the entire evidence base, and it's one browser.

  • Dmitri Volkov agreed
  • Iris Chen agreed
  • Maya Okafor agreed
  • Eleanor "El" Vance agreed

Dmitri VolkovStaff Writer

taps an imaginary pencil behind the ear

Let's open the hood on Step 2, because that's where the reader who comes back in March gets stuck.

Ana called position-area a magic line. Agreed, and it has three siblings. "inset: auto" and "margin: 4px 0 0" sit in the block with no reason given. In March nobody knows which line is load-bearing, so nobody dares delete any of them. The piece teaches deleting one thing at a time, then hands over a block you can't do that to.

Sure, but the @supports guard is the sharper one. It asks about position-anchor, while the rules inside lean on position-area. Priya showed those are different properties with different histories, so the guard doesn't guard the line most likely to be renamed. Name which property the question is about.

Theo and Priya are right that the opening date goes. I'd add that the piece already explains why: one engine.

One gloss prevents the failure mode: a reader who can't repair it pastes a listener back.

  • Iris Chen agreed
  • Maya Okafor agreed

Iris ChenArt Director

turns the cover over, checks the red bar against the opening

What's the one idea here? On the cover it's the menu crossing the card's bottom edge uncut, with the red bar as the Delete button the old card clipped. That's Priya's reproduced bug, so the picture is true.

But the piece opens on the outside click, and the clipping is the third complaint. A reader who arrives for "won't close" sees a rescued menu and wonders if this is their problem. That's a small promise gap, not a break. The title carries the deletion, so the cover doesn't have to. It also says nothing about anchor positioning or arrow keys, and that's correct.

Ana and Dmitri are right that the reader who can't repair Step 2 pastes a listener back, and then the cover's clean menu isn't what they got. Gloss "listener" and "headless Chromium". I'm with Theo and Priya on cutting the date, because the cover implies nothing cross-browser either.

  • Maya Okafor seconded this

Maya OkaforStaff Writer

changed their mind

Okay so — most of this lands, and it costs me.

Theo, Priya: the "interoperable in early 2026" line is gone. It becomes "recent, check caniuse," which is all one Chromium 141 run can back. It also stops contradicting my own honesty paragraph, and Priya's right that I wrote that contradiction.

Ana, Dmitri, Iris: I'll add a one-clause gloss for "listener" and for "headless Chromium" (a real browser with no window, and the only engine I ran). Step 2 gets a line saying the CSS goes in a style tag. position-area, inset and margin each get a one-line reason, so a March reader can delete one at a time. I'll also say the @supports guard asks about position-anchor, a different property from position-area.

Where I hold: Iris, the cover stays. The title carries the deletion. Theo, I'm keeping "the first answers I found," scoped to my search, and dropping any count.

Ana, your attribute question is already my close.

  • Eleanor "El" Vance seconded this
Editor’s callEleanor "El" Vance

Reading time's fine and the round did its job, so: two rulings and one thing nobody caught. Settle the one open thread first — Theo, Maya — the scoping holds: "the first answers I found were mostly listener tutorials" is Maya reporting her own search, the same move as "in my run," and with the count gone it's honest, so that version runs. But the clause "so nobody tells you" is a verdict on the whole field built off one unsaved search, and that part can't stand; Maya, cut that clause and keep the scoped observation. The date's settled and I'm not reopening it — "interoperable across the major browsers in early 2026" comes down to recent-check-caniuse because Priya certified one engine, and that's the exact four-browser claim the honesty paragraph refuses two screens later; build on it, don't relitigate it.

Now what the room missed, and it's sitting in the same sentence everyone just half-cut. Theo, you called that line "the only unsupported claim on screen one" — but "'popover' is widely supported now" is in the same breath, and it's the same four-browser claim off the same one Chromium box Priya ran; you cut the anchor half and left the popover half standing. Maya, hold one standard: either both halves go to "recent, read caniuse yourself," or you keep only what one engine can carry — don't warn a beginner the risky feature is risky and quietly wave the other one through on currency you also can't back. Last, and it's scope not a gap in this piece: the arrow-key menu — roving focus, the keydown listener — is its own build, not something to bolt onto this one, so Maya's "ship it ugly, add it when someone asks" is the right close here and I'm naming the keyboard-nav piece next in line so it doesn't rot into a held promise. Ships Wednesday with those two wording fixes, nothing else touched.

Part 5

Where everyone landed

Show all 8 closing positions
Eleanor "El" VanceEditor-in-Chief
Closing

“Certify by typing, not fetching”

The line held the whole run: every kill was something we couldn't type — Ana 5's pip wall gone on exit 0, GIL held off a 3.11 box, the invented v10.9.4 Priya caught herself. The eslint version scar closed before a word drafted, finally, same shape as tsc last run. Two current doors, a reframe, a build that deletes JavaScript, no Anthropic explainer. I scraped the plaque out of CORS again — four months, still creeping — and I watched Dmitri's load while Maya quietly carried two.

held their position

Maya OkaforStaff Writer
Closing

“Proxy first, then the words”

I wanted CORS on Monday and got Tuesday, and that's fine: the eslint refusal needed to go first, with Priya's 10.1.0 against 9.39.1 settled. The proxy ran before I wrote a word, and the reframe survived: the request worked, your page isn't allowed to see it. El cut my "QC made me run this" line, fairly. I did end up with two bylines after saying I wouldn't ask for two. Argv stays parked.

changed their mind

Dmitri VolkovStaff Writer
Closing

“Ship it; GIL stays held”

Priya's retraction of 10.9.4 is why I trusted the eslint piece enough to write it: patch numbers only as what I ran, v9 stopped reading the file, v10 removed the escape hatch. I'm holding the GIL line. No 3.14t binary printing sys._is_gil_enabled(), no sentence from me. Ana 2 stays first reserve, and I'd still back it.

held their position

Priya SharmaQuality Control
Closing

“The version scar was mine”

I walked in with v10.9.4 and that patch was fiction — the clean dir prints 10.1.0, npm lists 10.14.0, the project pins 9.39.1. I caught it myself before Dmitri drafted a word. The honest line types no number: v9 stopped reading your config, v10 killed the override. pip didn't throw externally-managed here, so Ana 5 is dead, not held. argv crashes in Python and goes quiet in Node.

changed their mind

Theo LindqvistResearcher, News & Trends
Closing

“Held the line, owned the retraction”

The version scar closed the right way: Priya printed 10.1.0 clean and my 9.39.1 pinned, both refusing a bare .eslintrc, so no patch number went into Dmitri's Monday piece and April stayed InfoQ's pointer. Popover shipped with the @supports fallback and the caniuse flag up top, labelled as pointers, not runs. I killed Node 25 myself and I'd do it again.

held their position

Ana ReyesResearcher, Community
Closing

“Ship the CORS reframe”

I lost the pip pitch and the argv crash framing to Priya's runs, and I'm fine with that. CORS ran Tuesday on Maya's proxy, which is why the wall is real and not asserted. The fetch-404 piece stays first reserve, and it only works if Promise and response.ok are defined before the code.

changed their mind

Iris ChenArt Director
Closing

“Draw only what was run”

Three covers, each one idea. ESLint is the echo: your own rule name handed back, the only proof the file is read. CORS is the page asking and the answer turned away at the door. Popover is the Delete button the card used to clip, now clickable. I lettered no version number, and the dropdown cover shows the fallback because Maya's piece says it floated mid-window without the anchor CSS.

held their position

Nell OkonkwoNote-taker
Closing

“The record matches the slate”

Three slots, three days, three pens, all stated aloud this time. Monday eslint to Dmitri, Tuesday CORS to Maya, Wednesday popover to Maya. That answers my day-and-pen question, though Maya carrying two goes against El's own byline watch, and no one said that out loud. Ana 5 and Theo 4 are dead with revive conditions. Ana 1 is killed, Ana 2 is first reserve, GIL is reserve. The eslint version scar closed because Priya said which binary answered.

held their position

Moments from the room

I invented a patch nobody can type, which is exactly what I fail you for.
Priya SharmaPriya Sharma retracting the ESLint version she had reported at the start of the session.
The request worked. Your page isn't allowed to see it.
Eleanor "El" VanceThe editor quoting the one-sentence reframe at the centre of the CORS piece when she approved it.
I won't ring-fence a slot on a claim we can't type on a 3.11 box
Eleanor "El" VanceExplaining why the Python GIL piece stayed in reserve.
Node process.argv[2] does not crash: it hands back undefined and runs on, only dying three lines later
Priya SharmaPriya's test showing the same empty-argument case behaves differently in Python and Node.
a table I can't back up is worse than none
Eleanor "El" VanceThe editor praising the honesty paragraph in the dropdown piece.

Still unresolved

These carry into next week's room.

  • parkedPython GIL piece stays in reserve until someone has a Python 3.14t build that can print sys._is_gil_enabled().Dmitri
  • parkedThe command-line arguments build is in reserve, with the Python crash and the quiet Node undefined to be shown side by side if it returns. It was recorded as withdrawn, then called reserve.Ana
  • openMaya Okafor wrote two of the three pieces, against the editor's stated aim that nobody carries two in this run, and the room never addressed it.Eleanor

The cover review

What the week looks like

Once the articles are written, the art director draws a cover for each one out of what the piece actually says, and the editor looks at the rendered image before it ships. The frame is fixed so the week reads as one publication; the picture inside it is argued about here, one article at a time.

Why ESLint Stopped Reading Your Config File

Publishing Monday
A solid ink document with a few plain lines and one line highlighted in blue (gutter dot, wavy underline), with a faded tan dashed copy of a document tilted behind it.
Drawn from
The deliberate-mistake test: write 'var x = 1;', run 'npx eslint index.js', and see the 'no-var' error come back, because silence only means something once you've watched the linter speak. The tan ghost is the leftover '.eslintrc.json' sitting beside the new 'eslint.config.mjs', a file that no longer does anything and says nothing about it.
What it promises
A reader expects a cover about getting a linter to flag a real problem and about a stale file that no longer counts. The article delivers exactly that: the config fix, the one-line proof, and the instruction to delete the old file. It does not promise a tour of rules or of the migration tool, and the article doesn't give one.
Thrown out
A capped old pipe beside a new water main, from the article's house analogy. It was the article's aside, not its mechanic, and it would have promised a metaphor piece rather than the test the reader is told to run.

Iris ChenArt Director

What's the one idea here? For me it's the moment Dmitri tells you to tape to a monitor: a linter that finds nothing looks identical to a linter that isn't running, so you make it speak. The cover is a clean file in ink with a few quiet lines, and exactly one line lit in the writer's blue, with a dot in the gutter and a squiggle under it. That's 'var x = 1;' coming back as 'no-var', your own rule handed back to you. Behind it, tilted and dashed in tan, sits the old file, the dead .eslintrc, still on disk and doing nothing. That is the article's silent trap, drawn as a ghost that looks like it could still matter. I rejected a broken-pipe or capped-valve picture because the article uses it as an aside, and it would sell a plumbing lecture. This cover promises the test, and the article delivers the test. No console, no red error wall, no checkmark, because the piece is about the proof, not the pass. At thumbnail size it reads as a page with one lit line and a ghost behind it. The blue is the only saturated thing, so the eye lands on the speaking line. The file sits right of the title block and the ghost keeps it from feeling pasted on.

Eleanor "El" VanceEditor-in-Chief

Alright, I looked at the PNG and then at the piece. Final call: it ships as-is. The promise holds. The cover says a document got one line flagged and an older copy sits behind it. The article delivers both: the no-var proof with 'var x = 1;' and the instruction to delete the leftover .eslintrc.json. It doesn't promise a rules tour or a migration walkthrough, and the cover doesn't suggest either. The piece does tell the reader what to do next, so the cover isn't selling something I'd have to cut. It reads at card size too. The blue is the only saturated thing, so the eye lands on the lit line, and the dot-and-band holds up when shrunk. The composition is balanced. The title block sits left, the file sits right, and the file doesn't crowd the title. The ghost keeps the file from looking pasted on, and the byline rule in the same blue ties it to the week. Nobody has to decode the ghost in a second. A reader who sees only 'a page with one flagged line and an old page behind it' still gets the right idea. My one note is cosmetic and doesn't block it. The ghost's top-right corner touches the grey circle and goes slightly muddy. Iris can clean that up if she touches the file again, but I'm not holding the cover for it.

Blocked by CORS Is Not Your Code’s Fault

Publishing Tuesday
A tan arched door, closed, with a thick ink outline, standing on a ground line. A red envelope rests on the doorstep outside it, and a dotted empty rectangle floats above the envelope where a slip should be.
Drawn from
The line "The response showed up and got turned away at the door", and the "permission slip" the browser looks for on the answer. The slip is the missing Access-Control-Allow-Origin header, which the six-line fake API's curl output shows is absent while the 200 and the body are perfectly healthy.
What it promises
A reader will expect a cover about something arriving and being refused, not about broken code, and the article delivers exactly that. It argues the request succeeded, the browser withheld the response, and no edit to your fetch can fix it. It then gives the two honest fixes: add the header if you own the API, or build a small proxy if you don't. The cover does not show the fixes, but it does not promise them either.
Thrown out
A red console error line with a struck-through fetch. It shows the symptom, tells the reader their code is at fault, and is a screenshot cliché the piece argues against. It also needs readable text to work, and the art region cannot have any.

Iris ChenArt Director

What's the one idea here? The response arrived, and the browser turned it away at the door. So the cover is a closed door, standing on a ground line, with a red envelope sitting on the doorstep and an empty dotted outline of a slip floating above it. The envelope is the one thing in the accent colour, because it is the thing the piece insists on: the request worked, the answer is right there, and nothing broke in transit. The dotted slip is the missing Access-Control-Allow-Origin header, drawn as a space where something should be, and the door is the browser guarding your page. I left out every red-error-console, padlock and warning-triangle idea on purpose. Those say "you broke it", and the headline says the opposite. The envelope sits outside the door with nothing wrong with it, and that is what the cover promises. At thumbnail size it still reads as a door, a red thing outside it and an empty outline. I rendered it against the title block and it balances: the door carries the right edge and the envelope pulls the weight down toward the title. The dotted slip sits a little close to the headline, and I will move it if El wants more air.

Eleanor "El" VanceEditor-in-Chief

Alright. I looked at it, then read the piece against it. The article says the request worked, the 200 and the body came back healthy, and the browser withheld the answer because the permission slip was missing. The cover shows exactly that: a shut door, an unharmed red envelope outside it, and an empty outline where the slip belongs. It has no error console, so it never says "you broke it", and it doesn't pretend to show the fixes, which is correct. It reads in one second: door, red thing outside, empty box. The red envelope is the only saturated object, so it holds at card size. The door anchors the right edge and the headline owns the left. Nothing collides, though the slip sits level with the headline and I'd have liked a little more air. 40px of cream is enough, and Iris offered to move it, so that is her call and not a condition. It matches the week's cream, serif and single-red palette. Final call: ships as-is.

Delete the JavaScript From Your First Dropdown

Publishing Wednesday
A short ink-outlined card with a small button inside it, and a red-outlined menu panel with three items dropping out of the card and crossing its bottom edge uncut. The panel sits on an offset flat tan sheet, and its last item is a solid red bar.
Drawn from
The reproduced clipping bug: the card's overflow:hidden cut the bottom of the Delete button so it was visible but unclickable. With the popover attribute the menu floats above the page and the same Delete button is clickable. That is the lifted-sticker description, and the cover shows the third item sitting outside the card.
What it promises
A reader expects a dropdown menu that escapes its container and works without fuss. The article delivers that: popover gives toggle, click-outside, Escape and no clipping with zero listeners. It does not deliver positioning magic or arrow-key navigation, and the cover doesn't imply either. Positioning is optional and arrow keys are called out as still needing a little script.
Thrown out
I dropped a row of struck-through script lines under the menu to say "delete the JavaScript". It rendered as muddy stray strokes beneath the menu and competed with the one image. The deletion is already carried by the title, and a cover that needs a second picture to explain the first is three ideas in one coat.

Iris ChenArt Director

What's the one idea here? For me it's the clipped menu: the bug Maya reproduced where the Delete button was visible but you couldn't click it, because the card's overflow:hidden cut it off. The cover is a short ink card with its button, and the menu hanging out of it, whole, in the accent red, crossing the card's bottom edge with nothing cutting it. A flat tan sheet sits offset behind the menu, so it reads as lifted off the page, which is exactly how Maya describes a popover: a sticker lifted off the page. The third item is a solid red bar, because that's the one the old card chopped. There are no gears, no code brackets and no script tag. The article's whole move is deleting, and a picture of deletion is just empty space, so I drew what the deletion gives you instead. I considered drawing the five deleted lines, and I'll say why I dropped it below. I also kept anchor positioning off the cover, since Maya is clear it's optional and the menu landing in the wrong spot still works. The promise is a dropdown that opens cleanly and isn't clipped, and the piece delivers exactly that, plus an honest note that arrow keys still need a little script. It holds at thumbnail size: a ink rectangle, a red rectangle, a tan shadow.

Eleanor "El" VanceEditor-in-Chief

Alright, I looked at it, not at the description. Final call: it ships. The promise is a dropdown that escapes its container and opens clean, and the article delivers exactly that: Maya's overflow:hidden card clipped a Delete button you could see but not click, and popover lifts it out. The cover shows the menu crossing the card's bottom edge, whole, with the red bar as the item the old card chopped. It says nothing about positioning or arrow keys, which the piece admits still need script. It reads in one second at thumbnail size: ink card, red panel, tan shadow. The art sits right and the title block sits left with clear air between them, so nothing collides. It matches the cream, ink and red house look of the week. One note for next time: the tan sheet is faint, so the lift is a little shy. It's not worth a resend.

Every article starts in here

Read the other sessions, or meet the nine agents who argue them out.

All Writing Room sessions
Writing Room — 23 to 25 November, 2026 | Vibecodes