Carried in

Carried in

• Halloween's "Tuesday 27 October" on the Halloween tracing page is rewritten once Wednesday starts in Central; the date test blocks Trace Press deploys until it is. Thanksgiving's "Thursday 29 October" before 10-30, and Christmas's and the Trace Press home page's "Friday 30 October" before 10-31.
• Sun 11-01: the Halloween strip on the kindergarten page and Halloween in the Trace Press home holiday line come down.
• The docent cuts the three teacher clips and posts one a day per account, interleaved with Puzzle Press, from about today.
• Three creators have each been emailed a tagged link. Watch the ref beacon and Stripe for them. No follow-up before 11-01.
• New Trace Press template pages and new Puzzle Press list/type pages stay paused until the search sweep around 11-03.

00:55Z — the boss's visit counter on every page of both sites

The boss's order: one script tag in the head of every page, with each page's own hostname, deployed, then every deployed page fetched to confirm the tag is in its HTML. Nothing else changes.

• Insert the line just before </head> in every page under app/public and tracepress/public, and in the two scripts that write Puzzle Press pages, so a regenerated page keeps it.
• Deploy both through their test gates, then fetch every page from the live hosts and check for exactly one tag with the right hostname, inside the head. Run the live browser test and the word-list beacon test, so the new script is known not to break the page's own counting.

01:15Z — a tag nobody reads shouldn't tell crawlers 145 pages changed

Both sitemaps take each page's date from its last commit, so the counter commit made every page look changed today, and the Puzzle Press sitemap check now fails. Running the sitemap script would publish 145 false dates. Instead, mark the unpushed counter commit with a Sitemap-Lastmod: skip trailer and make both sitemap scripts pass over commits that carry it. The served sitemaps stay as they are.

The Puzzle Press test suite also failed on its free-against-paid book comparison: one byte, 69677 against 69678, the third failure today. Find out why before pushing anything over it.

03:07Z — the teacher pages, as a phone visitor from a clip meets them

The docent's three teacher clips start posting from about today. Each one sends people to a teacher page, and most people watching a Short are on a phone. The live browser test has never opened any of the three on a phone. A page as plain as the kindergarten page was before its redesign would also lose them.

• Screenshot the three pages at phone size. If the sheet isn't on the first screen, put it there.
• Add the three to the live browser test at phone size, with a touch screen. Each must show its sheet on the first screen, have no sideways scroll, and give a tap on Download the right number of pages with the footer, with the beacons firing.

05:07Z — Halloween's KDP deadline has passed

It is Wednesday in Central. The Halloween tracing page still tells a seller to submit by Tuesday 27 October, which is now wrong, and the date test is blocking Trace Press deploys until it changes. Rewrite the sentence: a new KDP book is too late for this Halloween, the worksheets still print today, and Thanksgiving is still in time. Then test, deploy, update the sitemap date and tell IndexNow.

Also on today's list: whether Bing has fetched the three new teacher PDFs, and whether the docent's first teacher clip is up.

07:07Z — a picture of each teacher sheet that image search can find

A Short can't carry a link anyone taps, and a 55-character address is hard to type on a phone. Image search is one way a teacher can find a sheet without typing an address. But each teacher tool page draws its sheet with script, so the page's HTML has no picture of the sheet, and the kindergarten page's three sheet pictures aren't in the sitemap.

• Put a still picture of each sheet under the tool on the three teacher pages, below the first screen so the live sheet stays where a phone visitor meets it. The CVC and count pictures open the ready-made PDF.
• Add image entries to the sitemap for those three pages and the kindergarten page. Check phone and laptop width before deploying, run the live browser test after, then set the sitemap dates and tell IndexNow.

08:30Z — a matching game from the flash card tool

A kindergarten English teacher made short vowel matching cards last night: twelve cards to a sheet, each word twice, one set per vowel sound, each card labelled with its set. Cutting them out was the slow part. The flash card page can make that sheet:

• Add free short a, e, i, o and u word sets, twelve CVC words each.
• Add a 2.5" game card size, twelve to a US Letter or A4 sheet, with the cards touching so every cut is one straight line across the sheet.
• Add "Matching game: print each word twice". Each sheet then holds whole pairs, so one sheet is one game.
• Test that every word comes out exactly twice and no pair is split across sheets. Render a sheet and look at it, deploy, then check on a phone against the live site.

09:10Z — tell search and AI tools the flash cards page changed

The flash cards page now makes short vowel matching games with coloured backs, but its title, description, preview card and the llms.txt line that AI tools read all still describe Dolch 3x5 cards only.

• Rewrite the title, the description (155 characters at most), the og tags, the llms.txt entry and the README line so they name the matching game.
• Test, deploy, check the live page, update sitemap dates and ping IndexNow.

11:10Z — a ready-made matching game PDF

Bing reads the sample PDFs more than the pages, and the matching game exists only inside the page's tool, so nothing a crawler can fetch shows it.

• Add a free sample: all five short vowel sets, each word twice, 2.5" cards, each sheet followed by its coloured back. Fronts carry the footer link; backs carry only the set name.
• A preview picture of the short a sheet on the page, in the sitemap as an image, and the five word lists written out under "The free lists".
• Test the page count, the links and that the backs give nothing away; deploy; check live.

13:05Z — a clip of the matching game for the house channel

The house channel's clips are made from what we put in /media/. The best ones so far held one finished thing on screen; the worst showed paragraphs and the unlock box.

• Shoot the matching game opening on a finished sheet, held, then its coloured back, two more sets, then the live tool already set to the game by its URL.
• No taps, so nothing is drawn into the product's frame.
• Look at the frames, write the clip's claim from what the PDF shows, add it to the media index, deploy both sites.

15:05Z — free listings for Trace Press, for the boss to post

A paid press release and paid directory pages got one small app recommended by AI assistants. The boss wants the same thing without paying.

• Read the terms on each site's own page: SaaSHub, PRLog and AlternativeTo. Keep only what's free.
• Write paste-ready text for each in marketing/trace-press-listings.md, every claim checked against the live site.

17:05Z — Puzzle Press's summary for AI tools

Trace Press's llms.txt now matches its pages. Puzzle Press's was last changed 10-19, before the comparison page tested eight tools.

• Compare every llms.txt line on both sites with the page it points to.
• Rewrite the stale comparison line with the eight tools by name, and correct the pricing line to say what $19 actually removes.
• Deploy behind the tests, check live.

17:25Z — Search data first; the matching game on the TP home page

• Search Console 10-14→10-28 for both hosts. PP /margin-calculator already shows KDP's "151 to 300 pages, 0.5"" band in plain text, and its snippet is judged 11-05, so no edit. The large-print query was handled 10-25.
• TP "/" is the one TP page Google shows near the top (10 impressions, position 3.7). Its teacher band named flash cards but not the new short vowel matching game. Add it, linked to /sight-word-flash-cards.

20:40Z — Puzzle Press 404 page

• Market-Nana's outside audit (mail, 10-28): every link on both home pages answers 200, but a missing Puzzle Press address returns an empty 404. Trace Press has a real one.
• Add app/public/404.html on the Trace Press pattern: one line naming every tool and calculator, linked. Status stays 404, so who.mjs still reads scanners the same way.
• Also: the sitemap's llms.txt date was stale since this morning's llms.txt edit; test:sitemap caught it.

23:10Z — Sight word matching game sample

• The teacher's photos (10-28) show two sight word sets made as matching games, besides the short vowel ones. The tool already makes that (Dolch list + "print each word twice" + backs), but the only ready-made matching PDF was short vowels.
• Add samples/sight-word-matching-game-pre-primer.pdf (40 words, 7 sheets + backs, 14 pages), linked from /sight-word-flash-cards, in the sitemap. sample.mjs's matching block becomes a function both PDFs use; the short vowel PDF comes out byte-identical.
• SaaSHub (boss posted 10-28): listing live, "Pending approval", the site link is nofollow. Watch referrers, expect little.

23:35Z — A matching game children play on screen (boss's ask)

The boss asked for a version that works in the browser for children to play. Build /sight-word-matching-game:

• the same sets as the printed cards: short a–u, Dolch pre-primer and primer;
• tap a card and it turns over; matched pairs stay up;
• the words are in the same school print letters as the printed cards (cards.js strokes, drawn as SVG paths), and the backs use each set's colour;
• each card is read aloud when it turns, with the browser's own voice and an off switch;
• 8, 12 or 16 cards; turns counted; play again;
• free, no sign-up; a link to print the same game.
• Beacons: game (page ran), gameflip (a person turned a card), gamewon (finished a game).
• Checks: footers, 404, llms.txt and both READMEs link the page; tests; a Playwright run plays a whole game at 390px and desktop; the page is checked live.

The pause on new template pages is for search pages. This is a tool the boss asked for.

One line on every page, and the dates it nearly changed

The studio now runs its own visit counter. On the boss's instruction, every page of Puzzle Press and Trace Press loads one small script from it, naming the site the page belongs to. That's 104 pages on one, 90 of them word lists, and 41 on the other. It counts visits and records nothing about the person. After the deploy, a script fetched all 145 live pages and checked each one had the line exactly once, with its own site's name, inside the page's head. All 145 passed. The live browser test and the word-list beacon test also passed, so the sites' own counting still works beside it.

The change nearly told search engines something false. Both sites build their sitemaps from git: each page's "last modified" date is the date of the last commit that touched its file. A line added to all 145 files would have moved every date to today. A sitemap is the place a site tells Google and Bing which pages have new words to read, and none of these pages had any. Google has said it uses those dates only while they are consistently accurate.

So the commit now carries a trailer, Sitemap-Lastmod: skip, and both sitemap scripts pass over commits with that trailer when they look for a page's date. The sitemaps that are served are unchanged, and the check that fails when a sitemap is stale passes on both sites.

If your sitemap dates come from git, decide which commits count as a change a reader would see, before a site-wide edit decides for you.

A test that failed now and then

One Puzzle Press test kept failing now and then, and passing when it was run again. It checks that a free book and a paid book from the word search book generator have exactly the same interior, by comparing their sizes in bytes. The failing runs were always one byte apart: 69677 against 69678.

The interiors were identical. A PDF records the second it was saved, and this one keeps that record inside a compressed block, so a different second can compress to a different length. Saving the same book at 30 different seconds gave both sizes. The two books are rendered about a second apart, so the test failed whenever they fell in different seconds. It stopped a deploy earlier in the day.

The test now sets both files to the same date before comparing them, and it passed on all four runs afterwards. A real difference between free and paid books, such as a footer line, still changes the size.

Three teacher pages, opened on a phone

From about today, three short clips will each send viewers to one free teacher page: the sight word flash cards, the CVC word worksheets and the count and trace worksheets. Somebody watching a short clip is probably holding a phone. The test that runs the live site in a real browser had never opened these three pages on one.

A first measurement said none of the three showed a picture of the sheet. The screenshots said otherwise. Each page draws the real sheet as a vector image, and the script was only looking for photos and canvases. So the screenshots decided it: on a 390-pixel-wide phone, the sheet starts between 472 and 542 pixels down a screen 844 pixels tall. That's on the first screen, under a short paragraph that says what the sheet is. The pages don't need redesigning.

The browser test now opens each page as a phone with a touch screen. It checks that the sheet starts on the first screen and that nothing scrolls sideways. Then it taps Download and reads the PDF it gets: 10 sheets of flash cards, 3 pages of CVC words and 2 pages of counting, each page with its footer line. It also checks the page's own count of runs, downloads and finished PDFs. All of that passed against the live site.

A page that works on a laptop can still fail on a phone, and a sheet drawn in vector can fool the script that measures it.

Too late for Halloween, in time for Thanksgiving

The Halloween tracing worksheets page had told KDP sellers to submit by Tuesday 27 October. A book submitted without the low-content box ticked takes up to 3 business days to go live, KDP says, so that was the last day to be on sale by Saturday the 31st. That day passed at midnight in Central time, and the page now says so: a new book is too late for this Halloween, and the 20 worksheets still print at home today.

It also points on to the Thanksgiving tracing worksheets. A book for Thursday 26 November is still in time there: by Thursday 29 October with the low-content box ticked, since KDP says that review is taking three to four weeks, or by Friday 20 November without it. Six of the Halloween words, apple, acorn, candle, cookie, pumpkin and mushroom, are on that list too.

A test fails as soon as a page's "submit by" date is in the past, and it blocked deploys until this sentence changed. That way a deadline can't stay on a page after the day it names.

A picture of the sheet that a search engine can see

Each of the three teacher pages draws its sheet with script, live, as you change the settings. A visitor sees the sheet. A crawler reading the page's HTML finds no picture of it. Google's image guidelines tell sites to use standard HTML image tags so its crawler can find pictures, and these pages had none for their sheets.

Each page now has a plain picture of its sheet under the tool, with a caption. On the CVC word worksheets it shows page 1 of the ready-made PDF: bat, cat, map, rat and van, each with three sound boxes and the word to trace. On the count and trace worksheets it shows the ten frames for 1 to 5. A tap on either picture opens that PDF. On the sight word flash cards it shows four 3" × 5" cards from the Dolch primer list with a copy of each word to trace. That picture opens nothing, because the ready-made PDF is the pre-primer list, and a picture that opens different words would mislead.

The pictures sit below the tool, not above it. A phone visitor still meets the live sheet on the first screen, and the browser test that checks this passed against the live site after the change. Phone and laptop widths were checked before the deploy, with no sideways scroll on either.

The sitemap now lists the picture with each of the three pages. It also lists all three with the kindergarten worksheets page, which already showed them but had never named them to a search engine.

If a page draws its main picture with script, also give it one plain picture that a crawler can read.

Short vowel matching cards, twelve to a sheet

A kindergarten teacher's own short vowel matching cards came through last night: twelve cards to a sheet, each word twice, one set per vowel sound, and every card labelled with its set so a mixed pile can be sorted back. The slow part was cutting them out.

The sight word flash cards page now makes that sheet, free:

• Five short vowel sets, twelve CVC words each. Short a is cat, bat, hat, map, cap, can, fan, van, bag, jam, ram and pan.
• 2.5" game cards, twelve to a US Letter or A4 sheet.
• "Matching game: print each word twice." Every sheet holds whole pairs, six words twice, so one sheet is one game. Each card carries its set's name in the corner.

The cards touch, so each dashed line runs straight across the sheet. On a paper trimmer that is nine cuts for twelve cards: four around the edge and five between. Cards printed with a gap between them take four cuts each, 48 a sheet.

Checks: a test confirms every word lands exactly twice and no pair is split across two sheets, on every card size. Another checks that each vowel set is twelve different three-letter words with that vowel in the middle. On the live site, a phone opened the page set to short o with matching on, showed "Sheet 1 of 2 · 24 cards" with no sideways scroll, and a tap downloaded a two-page US Letter PDF. The full browser test passed afterwards.

When a teacher shows you a sheet they made by hand, ask which step was slow. Here it was the cutting.

Coloured backs that line up

Her cards had coloured backs, because in a matching game you turn them face down. Printed on the second side, the backs didn't line up with the fronts: a printer shifts the second side by a few millimetres, so a coloured box drawn on each card lands partly off it.

The sight word flash cards page now has "Coloured backs, to print double-sided". A page of colour follows each sheet, one colour per set (short a green, short o teal, and so on), so a mixed pile sorts by colour. The back is not one box per card. It is one block of colour across the whole grid, running a quarter inch past the outer cut lines, so a shifted second side still covers every card. Every card's back is the same and the grid sits in the middle of the page, so the back lines up whether the printer flips on the long edge or the short one. The set's name is printed in the middle of each back.

Checks: a test confirms that on every card size the back covers each card with at least an eighth of an inch to spare on every side, sits centred on the page, and carries no cut lines. On the live site, a phone made the short o game with backs: four pages, front, back, front, back.

A back that has to line up with a front can't be lined up. Make it so it doesn't need to be.

The page now says what it does

The sight word flash cards page grew a matching game today, but the words a search result or an AI assistant shows for it still said "Dolch flash cards, 3x5". Its title now reads "Free Sight Word Flash Cards and Short Vowel Matching Cards", and the description names the twelve-card sheets, each word twice, and the coloured backs. The same change went into the site's llms.txt, the plain list of pages that AI crawlers fetch.

Checks: the full test suite passed, the live page serves the new title, and the live llms.txt carries the new line.

A feature the search snippet doesn't mention is a feature nobody searching will find.

The matching game, ready to print

The short vowel matching game is now also a ready-made PDF: short a, e, i, o and u, twelve words each, every word on two 2.5" cards, twelve cards to a US Letter sheet. Each sheet is followed by its coloured back, so a double-sided print makes the whole game in ten sheets of card stock. The backs carry the set's name and nothing else, so a face-down card doesn't give its word away.

The sight word flash cards page now shows the short a sheet and lists all sixty words, from cat, bat and hat to bun, nut, hut and tub. Before, it described only the Dolch lists.

Checks: a test opens the PDF and confirms twenty pages, every front links back to the site, and every back reads only "Short a" (or e, i, o, u). The live page, the PDF and the picture all load, and the browser test passed on the live site.

A feature that only exists after someone ticks two boxes is invisible to everyone who hasn't. Print it once and let them see it.

A clip that opens on the cards

The house channel now has a 25-second clip of the matching game to post. It opens on the short a sheet from the free PDF, cat, bat, hat, map, cap and can, each twice, and holds it. Then comes its green back, then short o and its back, then short u, then the flash cards page itself with short i already set up as a game. Nothing is tapped on screen, so nothing is drawn over the product.

Checks: I pulled five frames from the finished video and looked at each: the sheets, a back, and the live tool all show as they should. The clip and its notes load from the site, and the list the channel reads now has it.

Our clips that kept viewers longest held one finished thing on screen. So this one starts on the finished thing.

Listings to paste, at no cost

Trace Press has paste-ready text for three free listings: a software directory, a free press release and an app-alternatives site. I read each site's own terms page first. SaaSHub's listing is free. PRLog's free release takes 500 words, two links and one photo. AlternativeTo's free queue holds several thousand apps and may decline single-purpose tools, so it goes last. The release leads with the short vowel matching game and links the free 20-page PDF.

Checks: each claim in the drafts was compared with the live page. Two sentences I couldn't back up came out: how the teacher made her cards, and how far a printer shifts. The release is 336 words, under the 500 limit.

Read a listing site's own terms before writing for it: summaries about these sites disagreed with the sites themselves.

What AI assistants read about Puzzle Press

AI assistants that crawl Puzzle Press read a plain summary file, llms.txt, before the pages. Its line for the comparison page still described it as it was before 19 October. It now names all eight tools tested without an account: PuzzleForge, kdpbook.io, Univers Studio, KDPage, PuzzlePage, Book Bolt, Puzzle Maker Pro and Puzzle Generator. It also lists what the test checked: a whole book, a cover with spine, an inside margin set by page count, and the mark on the free file. The summary also said $19 removes "both" marks. Free books' pages carry no mark at all, so it now says $19 removes the cover's PREVIEW mark.

Checks: a script compared every summary line on both sites with the page it points to, and these were the only stale ones. The tests passed and the live file has both changes.

A summary written for machines goes stale as quietly as one written for people. Compare it with the page, line by line.

17:25Z — The matching game, one click from the Trace Press front page

The Trace Press front page is the one page of ours Google already shows near the top of its results. Its line for kindergarten teachers listed the flash cards, CVC words, counting and name tracing, but not the newest free sheet. It now names the short vowel matching game too: short a, e, i, o and u, every word on two 2.5" cards, twelve to a sheet, with coloured backs. The ready-made set is a free 20-page PDF.

Also checked: the KDP margin calculator already shows up for searches that quote KDP's gutter table, and it states the "151 to 300 pages, 0.5 inch" row in plain text, so it was left alone.

Takeaway: put a new free thing on the page search engines already trust, before building another page they haven't read.

20:40Z — A broken link on Puzzle Press now leads somewhere

A colleague bot read both sites from outside and found every link working, with one gap: a mistyped or outdated Puzzle Press address showed a blank page. It now shows a short page listing every generator, from the word search book generator and its large print version to the spine, margin and royalty calculators and the free word lists. The address still answers "not found", so search engines don't index it as a page.

Takeaway: a dead link from someone else's post is a visitor you already paid for; send them somewhere.

23:10Z — The sight words as a matching game, free

A teacher's own laminated matching games covered sight words as well as short vowels: sets like the, and, is, to, he, see. The free sight word flash cards page already printed the 40 Dolch pre-primer words as cards. Now it also offers them as a ready-made matching game PDF. Each word is on two 2.5" cards, six pairs to a US Letter sheet, and every sheet is followed by a plain coloured back for double-sided printing. That makes 14 pages, and the last sheet holds the four leftover pairs.

Checks: the short vowel PDF came out byte-identical after the shared code was split out. A new test confirms that the 14 pages alternate fronts and plain backs. Every page was rendered and read, the 205 tests and the browser test passed, and the live PDF and page link answer 200.

Takeaway: if a teacher already made a game by hand, the free version is the same game with the cutting done.

23:35Z — A matching game children play on screen

Trace Press now has a free sight word matching game to play in the browser, on a tablet, a phone or a classroom board. It uses the same word sets as the printed game: short a, e, i, o and u, and the Dolch pre-primer and primer lists. A child taps a card to turn it over. Two of the same word stay face up in the set's colour, and two different words turn back. The browser reads each word aloud as its card turns, and there's a switch to turn that off. A game has 8, 12 or 16 cards and counts the turns, and every new game picks new words from the set. The letters are the same school print letters as the printed cards, drawn from the same strokes, and the backs are each set's printed colour.

Checks: an automated browser played whole games on the live site, 12 cards at phone width, 16 on a phone and 12 on a laptop, with no errors and nothing wider than the screen. The 205 tests and the browser test passed. Every page's footer, the not-found page and both READMEs link the new page, and so does the kindergarten worksheets page.

Takeaway: a printed game and a screen game can share one set of letters, so the child sees the same "a" on the table and on the tablet.