Graphics and design

Motion graphics, typography, illustration, layout and visual communication

• articleWhat exactly does WCAG 2.2 SC 1.4.3 require for text contrast, and how do I apply it as a check to AI-generated posters, title cards and thumbnails?
  Problem. Generated backgrounds are busy, and text laid over them often fails contrast. Conditions. WCAG 2.2 Understanding document for SC 1.4.3 (Level AA); it governs web content, so for posters and thumbnails it is a benchmark, not a legal rule. Method (documented). Text and images of text need a contrast ratio of at least 4.5:1, except large-scale text at 3:1; the Understanding text says 18 point text or 14 point bold text is judged to be large; computed values are not rounded (4.499:1 fails 4.5:1); text that is part of a logo or logotype is exempt; the document notes images of text do not scale well because they pixelate. Relative-luminance formula details are in C3-B066-G25-04 and C2-B22-E3-03. Checks on AI graphics [inference]. (1) After setting type (C3-B593-03), sample the background pixels under each glyph box, not the average of the image, and compute the ratio for the lightest and darkest background pixel against the text colour; the worst case must pass. (2) Gradients and...
• articleWhat style, preset and seed controls do Ideogram, Recraft and FLUX 3 document for repeatable image sets, and what is missing (e.g. FLUX 3 has no seed)?
  Problem. Repeating a look across a series: which knob does each tool give? Conditions. Docs of 23 Oct 2026. Method (documented). Ideogram 3.0 generate: seed (reproducible generation when set), style_reference_images (max 10, max 25 MB each; JPEG, PNG, WEBP), style_codes (8-character hex), style_type and style_preset (the preset cannot be combined with codes or references), color_palette, and magic_prompt (automatic prompt rewriting, default auto). Ideogram 4.5 edit: images to edit (max 5), the first is the one being edited and any others are extra references. Recraft: random_seed for reproducibility plus style references (C3-B593-07). FLUX 3 Image: the prompt carries everything; there is no negative_prompt, guidance or seed field, and bounding boxes go at the end of the prompt. Implications [inference]. (1) Seed reuse gives repeatability only for an unchanged prompt and settings on one model version; it does not carry a look to a different prompt, so for series work use style refere...
• articleWhich palette and exact-colour controls do FLUX, Recraft and Ideogram document, and how should I verify that a generated image kept the brand colours?
  Problem. Brand work needs specific colours; prompts alone drift. Conditions. FLUX.2 / FLUX 3 (BFL), Recraft API, Ideogram 3.0 API, docs of 23 Oct 2026. Method (documented). BFL: hex codes name exact colours in FLUX.2 prompts (docs claim brand-guideline matching 'with precision'); the FLUX 3 edit docs say to tie each hex code to one named object (for example only one bird in the middle) so the model knows where the colour goes. Recraft: controls.colors is an array of up to 10 {rgb:[r,g,b]} objects with an optional weight 0 to 1 (set for all or none, sum at most 1) and controls.background_color; the docs call them 'preferable colors'. A palette can also be stored with a style (Create style). Ideogram 3.0: a color_palette parameter exists (schema not expanded in the page read), plus style_codes (8-character hex codes) and style_preset; the preset cannot be combined with codes or references. Verify [inference]. Sample the output in sRGB, cluster to the dominant five to eight colours, an...
• articleWhen should I train a style LoRA instead of using reference images, and what do BFL's FLUX.2 klein guide say about dataset size, captions and trigger words?
  Problem. A brand look must hold across hundreds of images, offline or at low cost. Conditions. FLUX.2 [klein] LoRA training with an ai-toolkit-style config in BFL's guide; check the klein model licence before commercial use (not read here). Method (documented). The worked example trains a style LoRA on a fully synthetic dataset (images generated with FLUX.1 [dev] and illustrative LoRAs): 27 training images with caption files. Captions should describe only what is in the image and no style descriptors, so the model learns style from the pixels; a unique trigger word in every caption (example GFX_IMPR5N) activates the style at inference without polluting the base vocabulary. Training length guidance in the guide: style LoRAs (10 to 20 images) 1200 to 1800 steps, character LoRAs (10 to 15 images) 800 to 1200, medium datasets (20 to 40 images) 1200 to 2000, large datasets (50+ images) 2000 to 3000; the example config uses 3000 steps and inference with 8 steps for a soft painterly look. ...
• articleHow do Recraft's style references and style_id work for keeping a set of images on-brand (1-10 references, flexible vs precise match), and what are the rules?
  Problem. Producing many images that share one visual identity. Conditions. Recraft API V2/V3/V4/V4.1 (not V4.1 Flash), docs 23 Oct 2026. Method (documented). A custom style is built from one to ten reference images (PNG, JPG, WEBP; 64 MB total, each file under 20 MB) with no training; references can come from other generators, branding materials or existing assets. Use it by style_id (create once with Create style) or by attaching style_references / style_reference_urls to a generation, which creates a private style and returns its style_id for reuse; sending both is rejected. V4 and V4.1 styles have a match value: flexible (default) matches the general vibe, precise follows the style meticulously, including rendering technique, colour, composition and lighting; style_match in a request overrides the stored value. The dedicated V4 Styles models always require a style. Guidance: a single reference is enough to start; similar images sharpen the match and diverse ones widen the range. ...
• articleWhat do OpenAI and Google docs say about keeping characters and brand elements consistent across generations, and how many reference images can I give?
  Problem. Logos, mascots and products must repeat accurately across images. Conditions. OpenAI GPT Image docs and Gemini (Nano Banana) docs, 23 Oct 2026. Method (documented). OpenAI lists consistency as a limitation: the model may occasionally struggle to keep recurring characters or brand elements consistent across multiple generations; it also lists composition control (difficulty placing elements precisely in structured or layout-sensitive compositions). Google's model summary describes the premium Gemini 3 Pro Image model as offering accurate brand consistency and its Nano Banana 2.1 as improving multi-turn consistency (vendor claims); the Gemini reference table gives up to 4 or 5 character images for character consistency depending on the model. Use [inference]: treat logos as assets, not generations: place the real logo file (vector or high-res PNG) in the composite after generation; use the model for backgrounds and scenes; if a mascot must recur, use the vendor's reference-im...
• articleOverview: when I move a design from a layered source file to a web image or a print PDF, which things travel with the pixels and which stay behind?
  Synthesis supported by reviewed findings: layered files (PSD, ORA, Krita) are working formats whose merged composite is what other tools rely on; each export format carries colour identity differently (PNG chunks with priority, JPEG APP2 ICC segments, AVIF colr box), JPEG has no alpha so transparency must be flattened or carried elsewhere, browsers treat untagged images as sRGB, and print PDF/X needs a declared output condition.
• articleWhat spacing units do GOV.UK and Carbon define, and is there a baseline grid?
  The curator re-read all cited web sources (S2, S4); nothing was executed. GOV.UK spacing has ten fixed points (0, 5, 10, 15, 20, 25, 30, 40, 50, 60 px) and a responsive scale in which larger points shrink on mobile (e.g. point 4: 15 px on mobile, 20 px from tablet). No baseline-grid constant appears in the files read. Carbon and USWDS spacing were not read.
• articleWhat do Butterick's Practical Typography pages say about line length, point size and colour that I can turn into checks for AI-made posters, title cards and thumbnails?
  Problem. Generated layouts often have long lines, tiny type or loud coloured type on colour. Conditions. Butterick is body-text oriented (documents and web pages), so apply it to posters and title cards by analogy; thumbnails have very little text. Method (documented). Line length: aim for an average of 45 to 90 characters including spaces (or two to three alphabets per line). Point size: print body text 10 to 12 pt, web body 15 to 25 px; different fonts at the same point size do not necessarily look the same size, so judge by eye. Colour: contrast between light and dark draws the eye most; a thin or small font can carry a more intense colour than a heavy or large one; coloured type on a coloured background is usually regretted. Checks [inference]. (1) Count characters per line in any paragraph; credits and blurbs should sit within 45 to 90, and headlines of 1 to 6 words are exempt. (2) Compare cap or x-height of the type with its neighbours visually instead of trusting point-size n...
• articleWhat did Schloss and Palmer (2011) measure about colour-pair preference and harmony, and what do they say limits generalisation to palettes and to design practice?
  Schloss and Palmer (2011) show that people judge colour pairs on three distinct questions: how much they like the pair, how harmonious it is, and how much they like the figure colour against its ground. Across 992 two-colour concentric-square pairs, liking and harmony both rise with hue similarity (the best harmony model, with hue similarity, coolness and desaturation, explains 67.3% of variance), while liking of the figure colour rises with hue contrast. The authors caution about other layouts and about combinations of three or more colours.
• articleConversational creative agents inside tools (Artlist agent, Flova skills, Lovart brand kits, Combos.fun game agent): what they automate and where humans still steer
  Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. Six sponsored Mikkel Lassalle videos (Apr-Jun 2026) show four agent styles. Artlist AI agent: describe a goal in plain language, it picks a model (e.g. suggests 4K-capable models when asked), generates, and proposes next-step edits (moonlight, deeper shadow contrast, anamorphic flare) that you accept or override; same chat then animates a 5 s clip. Flova: "skills" (script-driven, story-driven, reference video, remix, music video, product commercial) wrap a pipeline; his manual path was Nano Banana Pro frame with "use the reference as the absolute identity source" -> GPT Image 2 character sheet -> Seedance 2.0 with both references, 15 s. Lovart: one prompt yields a brand...
• articleThe invented-brand product spot: packshot with rendered label text, then a 15 s multi-cut commercial (Apr-Jun 2026 creator recipe)
  Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. Recurring recipe in Artlist (2026-04-14) and Creative Fabrica (2026-06-02) videos: (1) generate a packshot of a product with an INVENTED brand name rendered on the label (Nano Banana Pro or GPT Image 2; e.g. a soda can or chocolate bar named after the creator), 16:9, one image; (2) image-to-video with Seedance 2.0: "Fast multiple cuts and cinematic camera movement. Premium beverage commercial. No music, only sound effects", 720p, 15 s, audio on; (3) iterate background by prompt ("place the product on a beach"). He reports good product and label consistency. Version-claim drift to note: in April he said Seedance 2.0 combines "up to 12 references"; later videos cite 50 re...
• articleBuilding a fictional brand campaign end to end in an editor (CapCut PC, Sep-Oct 2026): logo, products, model/product sheets, 30 s generation, scene split and chat-based edits
  Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. Mikkel Lassalle CapCut videos (2026-09-22, 2026-10-15) plus an Artlist Seedance 2.5 video (2026-08-11). Pipeline: invent the brand and make a logo with GPT Image 2 (his preference for logos and character sheets); develop products and style frames in a design workspace with Nano Banana Pro (his preference for cinematic frames); create one character sheet for the model and one for the product and tag both in the prompt; add camera/style cues (e.g. "ARRI Alexa 35 aesthetic"); generate up to 30 s with Seedance 2.5 (still capped at 720p in August per him; start/end frames also available); import into the editor, use "split scenes" to auto-cut the generation into shots, reord...
• articleYouTube rules (R1): chapters split a video into sections with individual previews and may appear in transcripts. Manual chapters: in the description list timestamps and titles, first must be 00:00, at least three in ascending order, minimum chapter length 10 seconds; manual chapters override...
  YouTube rules (R1): chapters split a video into sections with individual previews and may appear in transcripts. Manual chapters: in the description list timestamps and titles, first must be 00:00, at least three in ascending order, minimum chapter length 10 seconds; manual chapters override automatic ones. Automatic chapters are on by default for new uploads ("Allow automatic chapters", per video, in bulk, or via Upload defaults > Advanced settings) but not all videos are eligible; chapters are unavailable if the channel has active strikes or the content may be inappropriate for some viewers; channels without access must apply for Advanced features. Google Search (R2): key moments let searchers jump to segments; Google tries to detect segments automatically but "will prioritize key moments set by you, either through structured data or the YouTube description". For YouTube-hosted video, the description timestamps are the mechanism (all Search languages). For self-hosted or embedded ...
• articleWhat measured studies say about titles and thumbnails: rewriting titles for each community lifts engagement; outrage framing lifts engagement too - at a social cost
  Two studies (abstract level). (1) Hu et al.: on a large dataset of Reddit posts sharing YouTube videos, 21% of post titles barely change the YouTube title; statistically, rewritten titles get more engagement, and a controlled multi-phase experiment that neutralises video popularity, timing and community norms finds effective rewrites show emotional resonance, lexical richness and fit with community-specific norms; a fine-tuned BERT predicted the better title in pairs with 74% accuracy, beating near-random baselines including GPT-4o. Implication: when distributing a video to another platform/community, write a native title rather than reusing the YouTube one - and do not assume a general LLM can pick the winner without tuning. (2) Park et al.: a fine-tuned vision-language classifier of moral-emotional framing in thumbnails+titles across ~400,000 news-channel YouTube videos (Korea and US) finds "other-condemning" moral outrage consistently increases views, likes AND comments across bo...
• articlePer YouTube Help (R1): Standard features come with a new channel if there are no active Community Guidelines strikes (upload videos and Shorts, playlists, Posts - with limited daily limits). Intermediate requires phone-number verification and unlocks videos longer than 15 minutes, live streaming...
  Per YouTube Help (R1): Standard features come with a new channel if there are no active Community Guidelines strikes (upload videos and Shorts, playlists, Posts - with limited daily limits). Intermediate requires phone-number verification and unlocks videos longer than 15 minutes, live streaming from a computer, custom thumbnails and podcast creation, with higher daily limits. Advanced is granted automatically to creators with "sufficient channel history", or immediately via ID or video verification (not available to all creators), and must be kept by following Community Guidelines and maintaining positive history; it covers Content ID appeals, embedding live streams, applying for monetization (plus separate YPP requirements), clickable links in long-form descriptions/posts and channel details, adding a related video to a Short, pinning comments, RSS upload (podcasts), adding chapters, adding channels to the account, A/B testing (Test & Compare) and multi-language features. Daily li...
• articleYouTube multi-language audio (creator-uploaded dub tracks) vs automatic dubbing: eligibility via Advanced features, upload steps, localized thumbnails, copyright-mismatch removal, and YouTube's ">25% of watch time" non-primary-language claim
  Per YouTube Help (R2): multi-language audio lets creators upload their own dubbed tracks (also to Shorts) and, for long-form, language-specific thumbnails; it does not generate tracks. Available to channels with Advanced features. The default track follows the viewer's preferred language (from watch history); translated titles/descriptions are used by search. Upload: Studio (desktop) > Languages > video > Add language > next to "Dub" click Add > audio-only file roughly the same length as the video > Publish. If an automatic dub exists for that language it must be deleted first. Rights warning: if the copyrighted content in a secondary track differs from the original audio, the file may be removed - so music or effects beds must match the original mix. Edits to the original (trim, audio swap) propagate to secondary tracks. Localized thumbnails: upload the main thumbnail, then add per-language thumbnails under Languages. YouTube's own claim: "Creators who uploaded Multi-language audio...
• articleDo thumbnail, title and script features predict a YouTube video's views? What does measured evidence say vs creator advice?
  A Sept 2026 preprint (Tariq, arXiv 2609.32925v1; abstract read) links 3,728 thumbnails, 385 hours of speech (2.9M words), titles and tags to complete view histories of 56,419 vlog videos, following channels from their first upload including those that never succeeded. Findings as stated: 36 image/speech/text measures explain at most 1.1% of within-channel view differences; producing 40 more videos without a hit adds nothing; a channel's first hit raises views of everything published afterwards by 164% on average (robust to 17 checks); creators copy their hit's look and topic, but only copying the topic pays, copying the look adds nothing measurable. Implications for agents: creator packaging rules (C3-B344-01/02, C3-B336-02) are at best small within-channel effects in this vlog sample; topic choice and a breakout hit dominate. Caveats: single-author, unreviewed preprint; vlog genre only; observational design; "views" not watch time or revenue.
• articleDoes moral-outrage framing in titles/thumbnails raise engagement, and can misleading thumbnails be detected automatically?
  Park et al. (arXiv 2601.21815v2; abstract read) classify moral-emotion framing in thumbnail+title of ~400,000 videos from major news channels in Korea and the US and report that "other-condemning" moral outrage consistently increases views, likes and comments across both cultures - observational and limited to news outlets, with the authors warning about polarization; classifiers released. Naveed et al. (arXiv 2607.23739v1; abstract read) built a 2,843-video, eight-country dataset including 1,359 misleading-thumbnail videos (>7.6B views) and report LLM detection up to 93.8% accuracy (best: Claude 3.5 Sonnet) using thumbnail, video-to-text descriptions and subtitles. Together: outrage and misleading packaging can buy engagement, but platforms and researchers can detect misleading thumbnails, and YouTube's thumbnail policy and inauthentic-content rules apply. Agents should not recommend outrage or misleading packaging as a growth tactic.
• articleHow do USWDS colour tokens, contrast rules and Identifier logo guidance constrain graphics for US federal-style sites?
  USWDS asks sites to use its colour tokens rather than custom colours and expresses contrast as a grade difference: 40+ meets WCAG 2.0 AA for large text, 50+ AA (or AAA large text) and 70+ AAA. Its accessibility baseline is WCAG 2.0 AA, it aims at 2.1 AA, and it notes 2.2 is not yet legally required, so the magic-number rule is a 2.0-ratio shortcut. The Identifier must show the parent agency (not a product logo) and should not be used where it would mislead or on test or unofficial sites. Most USWDS code and guidance is CC0, but bundled icons (Apache-2.0) and fonts (SIL OFL 1.1) keep their own licences. Synthesis (producer, supported): an agent should not generate government Identifiers for non-agency work.
• articleWhat are sRGB, Display P3 and an embedded ICC profile, and what happens to an image that carries no colour tag?
  Browsers following CSS Color 4 (CRD, 20 October 2026) treat untagged images as sRGB and treat an image with a valid colour tag as being in that space. Display P3 has DCI-P3 primaries, a D65 white and the sRGB transfer curve. PNG 3 lists four ways to tag colour with precedence cICP, then iCCP, then sRGB, then cHRM+gAMA, and asks viewers to treat unknown-gamma PNGs like untagged images. ICC.1:2022 defines embedded profiles so images carry their colour description between systems; the ICC sRGB page advises not mixing v4 and v2 profiles. Inference (code_derived from the defaults, not a quoted statement): an untagged Display P3 image will be shown as sRGB and look less saturated rather than fail, so read the tag, keep it on export and tag output instead of guessing; a profile describes the numbers rather than changing them (curator agrees as reading of the spec's purpose, not a quoted sentence).
• articleWhich body-text size, line spacing and line length should I pick, and which of those numbers rest on measurement rather than one designer's opinion?
  Butterick's widely used numbers (10-12 pt print, 15-25 px web, 120-145% line spacing, 45-90 characters per line) are stated recommendations without measurements on those pages, while Legge and Bigelow's review of reading-speed data gives a fluent print-size range of roughly 0.2 to 2 degrees x-height (about 4-40 pt at 40 cm) and a critical print size around 0.15-0.3 degrees; the measured evidence constrains size relative to viewing distance but does not validate spacing or line-length rules.
• articleHow does YouTube's title/thumbnail A/B test (Test & compare) work: variants, winning metric, eligibility, limits?
  Per YouTube Help (captured 2026-10-23): test up to 3 titles, thumbnails, or title+thumbnail pairs; desktop Studio only, advanced features required; not available for Shorts, scheduled lives or premieres (premieres after conversion and live archives are allowed), nor made-for-kids, mature or private videos. Changing title/thumbnail mid-test stops it; tests take a few days to 2 weeks. Outcomes are Winner, Performed Same or Inconclusive, judged on watch time share (not CTR) with concurrent A/B/C delivery and a small held-out control group; with no clear winner the first-uploaded option is kept. Thumbnails below 1280x720 force all variants to 480p. YouTube warns that sequential third-party testers often optimise CTR and can pick different winners, and recommends starting with older videos and testing clearly different variants.
• articleWhat does YouTube's Inspiration tab offer (hooks, outlines, titles, thumbnails), when is it being deprecated, and what are its AI, provenance and data-use caveats?
  Problem: using YouTube's own AI ideation inside Studio. Conditions: Studio on desktop, English only, access varies; the page says that from August 2026 the Inspiration tab is gradually deprecated and creators should use Ask Studio for brainstorming. Method (documented): open Content > Get inspired; the tab shows nine suggested ideas from your channel data; refine by prompt or chips, base ideas on past videos, 'More like this', bookmark. Each idea card has a summary, image, audience-interest indicators, aggregated comment insights, prediction signals, channel alignment and related recent videos; selecting an idea gives an outline, title and thumbnail suggestion, with card types: Hooks (opening-segment ideas), Outlines, Titles, Thumbnails, Related videos. Provenance: generated thumbnails may carry a signed C2PA manifest which downstream platforms can read. Data: prompts connected with your channel are deleted automatically after 45 days; human reviewers may read conversations after di...
• articleWhat are YouTube's current custom thumbnail specs for videos, Shorts and podcasts (sizes, ratios, desktop vs mobile limits, the 4:5 crop)?
  Problem: preparing thumbnail files. Source: YouTube Help 'Add custom thumbnails on YouTube', read 2026-10-23. Source statements: recommended resolution 3840x2160 for videos and 2160x3840 for Shorts (minimum width 640 px for videos, minimum height 640 px for Shorts); JPG or PNG; file limits depend on device: mobile 2 MB for video thumbnails (10 MB for podcasts), desktop 50 MB for video, Shorts and podcast thumbnails; aim for 16:9 for videos, 9:16 for Shorts and 1:1 for podcast playlists. A vertical video with a 16:9 custom thumbnail is replaced by an auto-generated 4:5 thumbnail on the home, explore and subscription pages; the custom one still appears on watch feed, history and non-mobile platforms. Custom Short thumbnails can currently be added only in Studio on a computer and need a verified account; changes may take time to show. This matches the 50 MB desktop limit that the Oct 2025 TV blog said was coming (see P-L4-6-12). Method (inference): design a Short thumbnail at 9:16, che...
• articleWhat does YouTube's own TV guidance say creators should do for living-room viewing (device filter, 4K, thumbnails, channel art, episodic playlists, pacing, audio checks), and what is data versus advice?
  Problem: optimising for TV. Sources: YouTube Blog posts 'Maximize your reach: 5 tips for YouTube success on TV' (Nov 2024, product manager for connected TV) and 'Creators are taking over TV: 10 tips' (Apr 2025, based on a creator discussion). I found no dedicated YouTube Help page on TV for creators; the end-screens (6388789) and chapters (9884579) Help pages, fetched this round, contain no TV-specific text, and Shorts-on-TV appears only in news coverage, so these are blog posts, not documentation. Platform statements (self-reported, undated data): over a billion hours of YouTube are watched on TVs daily; viewers tend to watch longer sessions on TV; in Analytics Advanced mode filter by Device type to see TV watch time and views; one can also track new subscribers from TV; upload in 4K; thumbnails are supersized on TV, so use 1280x720 or higher, clean and readable; channel banner 2560x1440, 6 MB max, simple and legible from afar; episodic series in playlists with a consistent title f...
• articleWhat does YouTube itself advise for titles and thumbnails, how should their effect be measured, and what thumbnails are disallowed?
  YouTube Help (captured 2026-10-23): "90% of the best-performing videos have custom thumbnails" (YouTube statistic, no method given). Thumbnail advice: design for the target audience (familiar faces for subscribers; universal actions/emotions for casual viewers), rule of thirds, branding and readable text, avoid over-complex designs, make images large because display varies by device, keep current with community styles and experiment by updating older thumbnails. Titles: be accurate (inaccuracy causes viewers to stop watching and can hurt discoverability), succinct with key words first and branding/episode numbers last, limit ALL CAPS and emoji; choose searchable vs intriguing titles by intent. Measurement: for general audiences check impressions CTR on Home and Suggested in the first 24 hours on videos with above-average impressions; for subscribers check CTR in the Subscriptions feed in the first 24 hours; use research insights and the Audience tab for ideas. Policy: thumbnails vio...
• articleAre AI thumbnail "scores" (virality/clarity/curiosity/emotion) a substitute for YouTube's own title/thumbnail testing?
  Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. In a Pixels-sponsored demo (2026-08-23) an AI thumbnail tool generates thumbnails from short prompts, "recreates" proven thumbnail layouts with a new subject or a trained face persona, and scores thumbnails 0-100 on virality, clarity, idea, curiosity and emotion, reading title and thumbnail together; a "one-click fix" rebuilt two of his underperforming thumbnails from 44 and 33 to 99 and ~100. Useful design heuristics he states: get the message across at phone size within seconds; create a curiosity gap (show the verdict, hide the reason; opposite halves such as mask vs face); state rather than ask; an image that looks like it is happening now beats a concept; clarity w...
• articleWhat root font size and line-heights do Carbon and USWDS source define, so I can convert their rem tokens to pixels?
  By default both Carbon and USWDS treat 1rem as 16px (Carbon base-font-size 16px with a 100% root; USWDS respects the user's font size, a 16px equivalent), but USWDS can switch to a 10px root, which changes every rem conversion; USWDS line heights are unitless tokens 1-1.75, so pixel line height = font size x token.
• articleYouTube's Sept 2026 creator tooling: draft feedback, outlier research, channel-matched thumbnails, testing up to three video cuts, Gemini Omni editing in Shorts, voice+face likeness detection
  Made on YouTube 2026 (blog, 23 Sep 2026). Studio: new Insights and Research destinations to review past performance and find outlier videos; a draft feedback feature with personalised advice on pacing, structure and storytelling; channel-matched thumbnail generation; Ask Studio can run in the background to review older videos and suggest and TEST refreshed thumbnails; "coming soon" A/B testing of up to three VIDEO CUTS (beyond titles/thumbnails). Measured context from YouTube: creators have run more than 40 million title/thumbnail A/B experiments since the feature launched in 2024; Gemini Omni editing inside Shorts and the YouTube Create app was used by hundreds of thousands of channels per day in August 2026; a YouTube GenAI trends report says 72% of U.S. video posters aged 14-44 have used AI (statement as quoted; methodology not read). Editing assistant: conversational prompts to draft edits, reorder frames, trim, sync music to the beat. Community: opt-in AI comment moderation tha...
• articleHyperFrames: an open-source "HTML in, MP4 out" renderer built for coding agents - how creators use it for motion graphics and what it requires
  Primary README (heygen-com/hyperframes, Apache-2.0, Node >= 22; releases daily - v0.8.188 on 2026-10-23). It renders HTML, CSS, media and seekable animations into deterministic MP4 via a CLI, a Studio/playground, or as a rendering core, and is designed to be driven by coding agents: install as a Claude Code plugin (marketplace add heygen-com/hyperframes; /hyperframes) or via "npx skills add"/"npx hyperframes skills update" for Codex, Cursor, Gemini CLI, Copilot and others. Seven skills; the core set is /hyperframes (render contract and references for film, captions, overlays, motion, look), /hyperframes-cli (scaffold, check, inspect frames, edit timeline, preview, render locally or in the cloud, publish), /hyperframes-registry (~450 ready-made editable blocks: transitions, titles, kinetic type, captions, charts, maps, 3D, camera moves) and /media-use (music, SFX, images, voiceover, avatars, generated video, transcripts, cutouts, grades - via HeyGen services or free local tools); opt...
• articleDoes exact-match keyword titling make a YouTube video rank #1 in search? Creator "KVS" method vs YouTube's documented ranking factors
  YouTube's help page 'How YouTube search works' (undated, read 23 Oct 2026) says its search prioritizes three elements whose importance varies by search type: relevance (how well the title, tags, description and video content match the query), engagement (for example watch time of a video for that query) and quality (signals that a channel shows expertise, authoritativeness and trustworthiness on the topic); results may also be personalized from search and watch history, and YouTube says it does not accept payment for organic placement. A sponsored creator tutorial (13 Jan 2026) teaches picking a low-competition long-tail keyword, using it as the exact title and first in the description, mining competitors' comments for unmet needs, and treating tags as minor; according to discovery's reading of the auto-captions (not re-verified by the curator), the creator claims YouTube 'always ranks the most relevant result' and cites self-searches of exact titles as evidence. Reading: exact-matc...
• articleWorldbuilding from an author's design notes: a magic system from one constrained physics idea plus an era's intellectual mood, and a shared universe hidden behind standalone entry points
  Creator-reported: statements attributed to the creator(s) come from discovery's reading of auto-caption transcript excerpts; the curator checked the video metadata (title, date, length) and any cited documentation pages, not the spoken claims. Nothing was tested; promotional or sponsored framing applies where stated. Brandon Sanderson on two design origins. Allomancy (R1): seeded by a constraint noticed in a film (a telekinetic push in a Star Wars duel) - "If it could only push towards you, or pull towards you" along a centre of mass - then fused with the setting's era: late-Renaissance/early-industrial periods where alchemy and the periodic table were forming, so he wanted a system with "the feel of alchemy, but that actually works" -> ingesting metals, organised as a table. The first book using it was not publishable; on the rewrite he re-cut the powers to fit a heist crew's roles (each thief one power; the protagonist type has all). The Cosmere (R2): a connected universe planned ...
• articleIn JPEG XL and AVIF HDR stills, which of ICC, CICP/colr, Exif orientation and HDR light-level metadata take precedence, and what do tools document?
  Partially verified: the curator re-read S17, S18, S21, S24; S20 was not re-read by the curator and rests on the producer's pinned reading. JPEG XL keeps everything needed for correct display (including ICC colour and orientation) in the codestream: Exif/XMP boxes do not affect rendering and Exif orientation must be ignored because codestream orientation takes precedence; libjxl's ICC path cannot represent HDR transfer functions exactly. AVIF requires metadata OBUs to match mdcv/clli item properties and says alpha items should omit colr (readers ignore it); libavif also writes CICP into the AV1 sequence header for decoders that ignore colr, and lets a PNG cICP chunk override iCCP.
• articleWhat does the Netflix DUBCARD-Conformance QC error mean, how is it fixed, and where is that article now?
  DUBCARD-Conformance is an automated QC error for dub cards delivered as a multi-file ZIP whose images differ in dimensions or image format; the fix is to use one dimension and one image format matching the source asset. The article (Help Center id 115000610547) now redirects to the Knowledge Hub page dubcard-conformance, under IMF Fulfillment in Backlot > Automated QC Error Codes. Between the 2026-06-09 archive and 2026-10-22 only the pointer in the Solution line changed: from "page 27, section 5.7 of our current specification" to "our Branded Delivery Specifications or Non-Branded Delivery Specifications, depending on your workflow".
• articleHow does a language tag reach the font's 'locl' glyph substitutions and OpenType language-system tables, and where can that link silently break?
  OpenType 1.9.1 defines 'locl' as localized glyph forms that should always be applied and be driven by language metadata that selects a language-system (LangSys) table; language-system tags are typographic-convention tags, not language codes, and a run uses the specific LangSys or the script default, not both. CSS Fonts 4 requires renderers to derive the OpenType language system from the content language, keeps locl on by default, and provides font-language-override for cases where the font lacks the language; unknown override tags are silently ignored. HarfBuzz documents that shaping input is a (direction, script, language) triple per run with BCP 47 language tags, and activates locl by default. The three layers state their strength differently (OpenType 'may', CSS 'required', HarfBuzz API parameter).
• articleHas Adobe changed or clarified the Illustrator FAQ sentence about not adding Content Credentials, and do Illustrator release notes explain the 29.4 behaviour?
  The sentence is byte-for-byte the same in three captures: a Wayback capture dated 2025-10-06 (page date line 'Sep 11, 2025'), the cycle-2 fetch of 2026-10-21 and the live page of 2026-10-22 (date line 'Jun 1, 2026'). Between the 2025 and 2026 captures only the surrounding labels and small phrasing changed ('With version 29.4:' became 'With Illustrator versions 29.4 and later:'). The Illustrator desktop release-notes page, covering 29.8 (Aug 2025) through 30.8 (Aug 2026), contains no Content Credentials wording; 29.4 is outside its range. Two captures of that sentence do not show what happened between them or before September 2025.
• articleFrom Lutz's 1920 chapter XI, what check can an agent run to decide whether an explainer segment should be animated and how to sequence it (1920 practice, offered as a proposal)?
  Proposal (synthesis, not Lutz's own list): a six-step check that restates his scattered statements. 1) Can the camera show the thing? If yes and footage exists, photograph it. 2) If not, or if a static plan is the clearer picture, draw it as a moving diagram. 3) Is a textbook schematic available to copy, and a colour convention? 4) Can a draughtsman draw the successive states within the production means? 5) Write the scenario and the titles first; order parts whole-to-parts; add a pointer only where a fault or detail needs naming. 6) End on the complete or corrected state. Evidence level: 1920 practitioner statements, mostly conditional; no source supplies a comprehension test, so the check is structural. AI use is a hypothesis only.
• articleWhat limits or failure modes of animated educational film does Lutz (1920, ch. XI) name, and which obvious ones does he not name?
  Named limits are about means, not about viewers: the medium shows only in monochrome or simple outlines; colouring by hand is tedious and costly; existing colour processes lack some natural colours (from his own observation); refined stories need better drawing and figure skill; and some subject matter cannot be photographed, so diagrams must be substituted. He also states the premise 'all brains function in a similar way' as an assumption behind his case. He does not name comprehension failures (overload, speed, misread titles), cost figures for educational films, audience testing or teacher uptake: in Lutz's text of chapter XI the stems fail, mistake, confus*, overload, test, audience, attention, memory and retain have 0 hits (bounded scan).
• articleWhat does Lutz (1920, ch. XI) predict for animated educational film and the medium, and what does each prediction depend on?
  Lutz predicts: children will 'very soon' learn from film as well as be entertained; the possibilities of educational, travel and scenic film are not exhausted; colour cartoons will come, depending on a process that reproduces nature's full range of hues ('awaiting the inspired inventor'); with colour, animated paintings and gallery exhibitions of art works with movement; museums with projecting rooms and fireproof film libraries. He calls it 'fantastic dreaming' and argues that formerly visionary things became commonplace. He also states what the medium lacks now: it 'needs, for one thing, color'. This candidate only records what he predicted; comparison with later history was not done (zero-fetch batch).
• articleWhich statements in Lutz's chapter XI report existing practice or history, and which are conditional proposals, claims or predictions?
  Chapter XI is mainly a proposal in the conditional mood. In Lutz's text of the chapter (lines 4097-4371) 'would' occurs 32 times, 'could' 21 and 'will' 9, and there are 0 footnote anchors. The statements that report existing practice or history are few: Demeny's lip-reading apparatus (a note of 1892 appears only in the illustration credit), Desvignes' 1860 steam-engine pictures ('is recorded', no citation), animated maps 'frequently' used with news reels, educational/travel/scenic films shown in theatres (live photography), stencil tinting of photographic film, and a caption on drawings used in making a gasoline-engine film. None of these reports an animated educational film's audience, learning result or cost. An agent should therefore cite the proposals as Lutz's views, and the few reports as unreferenced assertions.
• articleHow does Lutz (1920, ch. XI) say an animated educational film should be planned and sequenced: scenario, order of parts, titles, schematic style, pointers?
  Lutz gives planning rules by example rather than as a rule list. The film "must be planned well" and a scenario would practically be written; titles and wording are thought out beforehand for educational value; for the muscle film the series of drawings is the first thing to prepare. His sequences go from the whole to the parts and from a fault to its correction: a sectional heart, then flow through the cavities, then a body-wide schematic with a colour code; a whole evolutionary tree, then separate drawings of branches; a poorly composed picture built on screen, faults pointed out with a cut-out pointer, then the components moved into the good arrangement. The chapter gives no durations, no number of titles and no simplification rule beyond 'schematic diagram like those in the books' (bounded scan).
• articleWhich subjects does Lutz (1920, ch. XI) say suit animated drawings rather than live photography, and what reason does he give for each?
  Lutz proposes animated drawings for subjects where the camera cannot show the thing: the heart and circulation, muscle action, the skeleton's levers, evolution as a tree; for machinery in operation, where a drawing replaces a static plan; and for change over time, such as changing maps. His reasons are (a) the camera cannot photograph some details, (b) the eye grasps a moving drawing in moments that paragraphs would take to explain, (c) a trained draughtsman can draw it, and (d) for maps and evolution, the subject is itself a change of a diagram. He keeps live photography for experimental apparatus and for physiological subjects where scientific moving pictures exist, and his art-method films film a demonstrator rather than animate. He does not name invisible, slow or microscopic phenomena as such (bounded scan of chapter XI).
• articleIn E. G. Lutz's 1920 Animated Cartoons, chapter XI, what does he count as an 'educational' animated film, and how does he separate instructional films from stories for children?
  Lutz gives two senses of 'educational film'. The strict sense covers only instructional films. A broader sense, which he credits to unnamed 'enlightened opinion', also admits any theme or story for children, even with a touch of humour; his example of that second type is Lewis Carroll's Alice tales as drawn by Tenniel, and he adds that such a story needs correspondingly better drawing than the ordinary comic cartoon. The examples after the opening paragraphs are instructional (art methods, physics, machinery, history, physiology, evolution). The classification is Lutz's own 1920 usage; he gives no test of learning for either kind.
• articleWhat should a design/text handoff record so a different renderer can reproduce the text correctly or at least detect that it cannot?
  SYNTHESIS (a proposal, not a source finding). From the other six candidates: per text run record language (BCP 47), script (ISO 15924), direction/writing mode, font identity (family, PostScript name, version string, file hash), font licence evidence, OpenType features and any language-system override, shaping engine and version with its default feature set, embedding/subset/outline state, fallback chain, and the languages the font claims to support, plus a render-check artefact. Each field is mapped to the source statement that makes it a real input of a renderer or a documented loss point. No inspected format or tool is shown to carry all of them; the Figma REST schema carries only some (candidate 06).
• articleDo the layout-exchange routes I can actually read (Figma REST schema, IDML as imported by Scribus, IDML as handled by a translation tool) carry language, direction, font version or licence, and what happens when the font is missing?
  Mostly a scoped negative result plus partial positive evidence. The Figma REST OpenAPI schema (v0.45.0, commit 7ec6254) exposes fontFamily, fontPostScriptName and an opentypeFlags map in text styles, but a whole-file search finds no lang, locale, bidi, rtl, writing or font-version field. The Scribus IDML importer (commit 52dd809) reads AppliedFont/FontStyle/AppliedCharacterStyle and shows no language handling in importidml.cpp; a missing font falls back to the application default font or a replacement dialog. A translation vendor's help page says IDML CharacterStyleRange elements can differ by an 'applied language' attribute and that an option merges such ranges, keeping the first. The official IDML specification could not be fetched, so IDML attribute semantics are not primary-verified here.
• articleWhen I subset or repackage a font for a handoff, which language-relevant OpenType data and which licence/version metadata do the defaults drop?
  The fontTools subsetter's own help (commit 16f88cb) says its default keeps locl and the features required for script shaping and all scripts, but anything else (e.g. onum, pnum, ss01, per the help's own example) must be added; the glyph set is closed over kept features unless --no-layout-closure is used. By default only name IDs 0-6 and English name records are kept, so the licence description (ID 13) and licence URL (ID 14) of OpenType 1.9.1 do not survive a default subset. WOFF 1.0 leaves font tables unchanged but its extended metadata (licence, licensee, credits) is informational and cannot change rendering. A subset handoff therefore needs the original font hash, the subsetter options, and the licence kept outside the font. Nothing was executed.
• articleIf I embed the fonts in a PDF, will the text look the same everywhere, and can someone edit it or reuse the font afterwards?
  ISO 32000-1 says results are most predictable when every font program is embedded; otherwise substitution and glyph selection are reader-dependent. PDF/A-1, PDF/A-2 and PDF/X (secondary) require embedding, with subsets allowed if every referenced glyph is present; PDF/A-1 requires a CIDSet for CIDFont subsets, and PDF/A-2 requires any CIDSet present to list every CID in the embedded program. But embedding is a view-and-print right by default, not an editing right, and the PDF text only says readers MAY use embedded OpenType layout data; it does not promise the original shaping features survive. For a handoff meant to be re-edited, the PDF alone is not enough: ship the licensed font file, its version and licence separately.
• articleWhat language, text-direction and Unicode-mapping information can a PDF carry, and what do PDF/A and PDF/UA validators actually check about it?
  ISO 32000-1 (PDF 1.7) lets a PDF set a default language in the document catalogue (Lang), override it on structure elements and on marked-content Span sequences, and even inside text strings via a language escape; the empty string means unknown, and tags are cited as RFC 3066 language tags (RFC 5646 says RFC-3066-valid tags remain valid). Direction is a separate Tagged PDF layout attribute (WritingMode). veraPDF's machine-readable rules show that PDF/A-2 Level A checks the Lang string syntactically (regex), PDF/UA-2 requires a non-empty catalogue Lang, and PDF/A-2 Level A/U requires a Unicode map for every font. None of the inspected rules checks that the declared language matches the text or the font's coverage.
• articleHow do I mark which human language each piece of text is in (HTML, SVG, CSS), and what in rendering depends on that mark?
  In HTML and SVG the language of a run is a BCP 47 tag in lang (and xml:lang in XML), inherited from the parent; the empty string means unknown. HTML and SVG both say the tag covers text-bearing attributes too. CSS cannot declare language, only read it (:lang) or override the OpenType language system; direction is a separate mechanism. In renderers that follow CSS Fonts 4, the content language is what picks language-specific glyph forms, so a handoff that drops the tag changes which glyph variants a conforming renderer may pick. This candidate states what the specifications require; no renderer was run.

More