Frequently Asked Questions

HSCred is where high school students publish academic work, the way scholars publish in journals. A student masters the content of a unit, goes deeper into one part of it, and defends what they found in a recording of ten minutes or less. A licensed educator at their own school certifies the work is the student's own. Three blind experts, chosen by a university professor, score it against a public rubric. Work that clears the bar is published to a permanent, student-owned record of up to ten published works that any college can watch, weigh, and verify. Published. After the test.

A grade is a claim. A published work is a claim with the evidence attached.

A grade compresses a semester into one letter, set by one adult, unverifiable outside the building. A test score reports one morning of a student's life. A published entry is a dated record that links to the work itself: the student presenting and defending a real project, scored by three independent experts against criteria anyone can read. A reader who doubts the number can stop trusting and start watching.

No. We move the test to the front. Every channel opens with STAKE, a content check: a quiz, an exam, a teacher interview, whatever that channel requires. A student earns the right to project work by mastering the foundation first, and the record of that mastery travels with the published work.

What ends is test prep as the finish line. The cram comes early, and the thinking fills the rest of the unit. Not an end to testing. An end to test prep.

Three steps, in order.

• STAKE the content. The student proves they have mastered the unit's foundation. Their Guide checks the evidence and approves.
• PITCH the project. The student builds a project plan against the channel's rubric and works through at least two full rounds of feedback and revision with their Guide. When the work meets the criteria, it gets a title and an abstract.
• MINT the publication. The Guide certifies the work is the student's own. Three experts who do not know the student score the recording against the rubric. The median of their scores decides publication.

Notice what moved. A unit used to end with a cram for the final. Here the cram comes first, because on the other side of the STAKE waits the part students want: scripting, recording, producing, publishing.

A recording of ten minutes or less, audio or video, plus a title, an abstract, and the documentation built up through STAKE and PITCH: the sources, the feedback record, the revision trail. The recording can be as plain as reading a research paper to the camera or as produced as an edited engineer's journal from a robotics season. Students re-record as many times as they want before submitting. Substance is scored; polish is not.

Why a recording: it has a voice, so it creates a real audience. It travels anywhere digitally. Editing ten minutes down from hours of material consolidates the learning like nothing else in the unit. And it caps review time, which is what makes expert evaluation affordable at scale.

Three Expert Reviewers, appointed by the professor who runs the channel. They score independently and blind: they see the work and its documentation, never the student's identity.

Each rubric indicator gets a score from each expert: 2 (Above and Beyond), 1 (Publish), 0 (Not Ready for Publication), or -1 (Not Ready for Upload). The platform takes the median of the three per indicator, multiplies by that indicator's point weight, sums, and divides by the channel's cutoff. At 1.0 or above, the work is published. Below 1.0, nothing appears anywhere; only the student and their Guide are told.

So every score you see on a transcript sits between 1.0 and 2.0, and a 2.0 means the median reading was Above and Beyond on every criterion.

A channel is an academic journal for one field, run by a named university professor, the Channel Editor. The editor writes the rubric, sets the STAKE and PITCH requirements, and appoints the three-expert panel. The rubric is public: it is the channel's storefront, and students read it before choosing where to submit.

The name comes from media, but the resemblance ends there. No ads, no feed, no view counts, no algorithmic anything. Channels earn standing the way journals do: by what they publish. Institutions score the channels they trust, and students submit to the channels institutions trust. Work competes to be published; channels compete to be trusted.

• Guide. An educator at the student's own school: a teacher, counselor, or coach the student invites. Registered with a verified school email. Runs the feedback cycles, approves the STAKE, certifies authenticity. The only person in the process who knows the student.
• Channel Editor. A verified university professor. Defines what publish-worthy means in the field, recruits the expert panel, keeps grading on schedule. The named, accountable public face of every work the channel publishes.
• Expert Reviewer. A specialist the editor appoints: a postdoc, an advanced graduate student, a veteran educator. Scores blind, stays anonymous to the outside, is paid per evaluation.

The design in one line: the person who knows the student never judges, and the people who judge never know the student.

Students: nothing, ever. A student account is free for life, and so is sharing the record with any college.

Each expert evaluation costs $100, paid as one access code. Schools, districts, sponsors, and families buy the codes; students who have none choose the scholarship option at checkout and pay nothing.

Institutions: reading a transcript a student shares is free, always. Searching the full database is free for early institutional adopters now, and moves to a subscription as the record grows. Early adopters keep their access.

Four educators split the whole thing: $25 to each of the three Expert Reviewers, and $25 to the Channel Editor for keeping the panel on schedule. HSCred keeps $0.

Every evaluation dollar goes to the academics who judge the work. None of it goes to ads, to data brokers, or to us. The company earns its keep from institutional subscriptions, not from students publishing.

One card, one code, one expert evaluation: $100. Cards are sold singly online, or as a deck of fifty for $5,000, which is how a school or district funds a season of publishing.

A card pays for judgment, not results. The three experts are paid for their review whether or not the work publishes, which is what keeps the review honest. And a deck in a principal's hands is a strengths-based instrument: fifty chances to put the school's best work on the record, starting with the clubs and classrooms already doing it.

They publish anyway. At checkout the student picks the scholarship option instead of entering a code. The request goes to the Academic Capital Foundation and to any sponsor whose criteria it matches; one of them accepts it into a funding queue, and the queue funds in turn, with priority to students closest to their college applications. When it is funded, the work goes to the expert panel. A short delay at worst. Same rubric, same three experts, same publication, no fee.

No personal information travels with the request. Funded scholarships are waiting right now, so no student should hold back work for lack of $100.

Two organizations, one origin in Harlem. The Academic Capital Foundation is a nonprofit founded in 2015; it exists to make sure the upgrade to assessment reaches the students the old system overlooked, through scholarships, coalition work, and educator development. HSCred, Inc. is a for-profit company founded in 2025; it builds and runs the app.

They are separate organizations with separate boards, and the money moves in one direction only: HSCred gives to the Foundation and never takes from it.

Browse the channels and read the rubrics. The rubric is the whole game: it tells you exactly what that channel's experts call publish-worthy, line by line, before you commit to anything.

Then invite your Guide: a teacher, counselor, or coach at your school who knows you and your work. Pass the channel's STAKE. Build the project through PITCH with at least two rounds of real feedback. Record your ten minutes, re-record until you like it, and submit by the end of the month. No access code? Pick the scholarship option at checkout.

High school students 14 and older. A transcript holds up to ten published works, and you can publish at most one per calendar month, so the math sets the pace: roughly one every other month from sophomore year fills a record by graduation.

Starting as a junior or senior? You still have room for a strong record, since work you have already finished can count. See question 20.

One cycle, every month:

• Commit to your submission month by the 15th; before that, rescheduling is free.
• Submission and your Guide's certification are due the last day of the month.
• Experts are assigned on the 1st and finish scoring by the 20th.
• Results post on the 1st of the following month. Published work goes on the record that day.

About a month from upload to answer. The one-a-month rhythm is deliberate: real work, on a real schedule, with no application-season pile-up.

No. You bring HSCred to your school, not the other way around. Any educator at your school can serve as your Guide; you invite them and they register with their school email in minutes.

Money cannot gate you either. If your school has access cards, use one. If it does not, or nobody offers you one, the scholarship path is open to every student. No principal's signature stands between you and publishing.

Channels are our growing edge, and we will say it plainly: we want more of them. Three moves:

• Look one level up. Channels score intellectual moves, not topics. An experimental science channel can judge any investigation; a social studies channel can judge any source-based argument. Your subject probably fits a channel that exists.
• Tell us the field. Student demand is exactly what recruits the next editor.
• Know a professor? Introduce us. A channel takes one thirty-minute call to start.

Each project submits to exactly one channel, so choose the rubric that fits the work best.

Nobody sees it but you and your Guide. It appears nowhere, counts against nothing, and follows you nowhere.

The scale has no fail. Not Ready for Publication means not yet: the experts' rubric comes back to you, indicator by indicator, and it reads like a map of the revision. Fix the work and resubmit in a later month. The cost of a miss is time, never your record. Only your best work is visible, which is the point of a strengths-based transcript.

Yes. The retrospective path exists for exactly this: the robotics build, the science fair project, the documentary you made last spring. Work that should not disappear.

STAKE becomes documentation of the sources and influences behind the finished work. PITCH reconstructs how the work developed: at least two feedback checkpoints, with the Guide who watched it happen. The catch is that word: your Guide must have witnessed the original process to certify it is yours. Then MINT works exactly as it always does.

Two doors, both yours to open.

You share: your transcript is a link and a QR code. Paste it into any application, hand it to any recruiter, print it on a resume. Any admissions officer can open it free and watch you defend your work.

They search: colleges and scholarship programs that subscribe browse the whole record by subject, region, and the background fields you choose to share, and they can reach out first, to invite you to apply, to visit, to be seen. How much of your identity travels with the work is set by you, in stages. The work speaks either way.

You already vouch for students; a recommendation letter is your word against the world. Guiding puts evidence under your word.

Per project, the work is what you already do: approve the STAKE evidence (the quiz, the test, the notecard review, whatever the channel asks), run at least two rounds of feedback and revision during PITCH. Then, at MINT, score the rubric yourself and certify the work is the student's own. You are the only person in the chain who knows the student. That is by design, and it is why your certification carries the weight it does.

You score, but you do not gate. Your rubric score at MINT is calibration data; the publication decision belongs to three blind experts, full stop. A student who completes the process has the right to expert evaluation, whatever you or anyone at your school thinks of the work.

Your scores matter over time. The platform compares your ratings to the expert medians, and a track record of judging well is what qualifies you for paid review work. See question 27.

You testify to what you witnessed, and you witnessed everything: the STAKE mastery, the plan, the drafts, your own two rounds of feedback, the revisions that followed. Certification is testimony, not detection.

That distinction is the whole answer to the AI era. Detectors inspect a finished artifact and guess. You watched the thinking form in a specific teenager over weeks. Provenance gets built during the work; it cannot be bolted on afterward. Each channel also states its own AI disclosure policy in the rubric, so what must be declared is never a mystery.

The unit you already teach is the STAKE. The project you already assign is the PITCH. HSCred adds a destination, not a prep load.

Then differentiation starts working for you instead of against you. Students who pass the content gate move straight into production, self-directed, while you sit with the small group still mastering the material. The fast finishers stop waiting bored, and the cram lands at the start of the unit, where it motivates, instead of the end, where it exhausts. If your course already runs on projects, capstones, CTE builds, or Consortium-style performance tasks, the work maps one to one.

Because industrial schooling ran on literacy across the curriculum, and the digital era's version is journalism across the curriculum: literacy plus the tools of publication.

A newsroom is the most natural on-ramp to this whole model. It supplies an authentic audience, real deadlines, and a reason to verify, which are the three things a rubric can never fake. And once students can report their thinking, every subject has a pathway to a published artifact: a science investigation, a history documentary, a narrated proof. Start with a pop-up newsroom if you want proof in one class period: one story, two minutes, twenty-five minutes to make it.

Yes, as Expert Reviewers. Guiding your own students is unpaid; judging strangers' work is not. Guides who build a strong calibration record get invited to paid review: $25 per evaluation, at a pace of about six ten-minute reviews an hour, which is $150 an hour of subject-matter judgment, from anywhere, on your schedule.

One rule protects the whole system: you never review students from your own school or students you know. Blind stays blind. It is expert pay for expert judgment, and a way to stay in teaching without leaving for administration to earn more.

Because you already do this work. Editing a journal and refereeing submissions are ordinary scholarly service, done on your own standing, without a dean's approval. A channel is the same undertaking pointed at the field's youngest work: you define what publish-worthy means in your discipline, and your name stands behind every work the channel publishes.

Two returns. You see talent before anyone else does; the students publishing in your channel are the pipeline your department wishes it had. And the work pays, per evaluation, at rates set out in question 31.

One thirty-minute call, then editing, not authoring. You talk through the field; we draft the channel from that conversation: the overview, the STAKE content requirements, the PITCH plan template, and the full rubric. You mark it up until it is yours. Then you invite three Expert Reviewers by email, they onboard, and the channel is live.

The rubric bones, since you will ask: five to eight indicators totaling about sixteen points, the heaviest points on your discipline's core moves, presentation always present and always light. Four fixed levels platform-wide: Above and Beyond, Publish, Not Ready for Publication, Not Ready for Upload. The Publish descriptor gets written first; it is the benchmark everything else leans on. The scale itself never varies by channel, which is what keeps transcripts comparable.

The editor's recurring job is one thing: make sure three reviewers finish by the 20th each month. Reviewers set their own monthly capacity, and if your panel runs short, the platform supplies vetted graders. Grading is guaranteed; your channel never stalls on staffing.

The capacity math is friendlier than it looks. A three-person panel putting in one hour each weekday clears roughly 130 projects a month. You may also take one of the three seats yourself and do the reviewing you would otherwise assign.

• Expert Reviewer: $25 per evaluation. Recordings run ten minutes or less, so six an hour is a realistic pace: $150 an hour. One hour each weekday works out to about $39,000 a year.
• Channel Editor: $25 for every project the channel evaluates, for running the panel and keeping the calendar. An editor who also sits as one of the three experts earns $50 on that project.

Everyone is paid as an independent contractor through Stripe, on a W-9 or W-8BEN. The task itself will feel familiar: scoring focused work against a clear rubric is what you already do at a poster session or on a review board.

People the editor appoints, and no one else: postdocs, advanced graduate students, veteran educators, working specialists. Reviewers are verified, 18 or older, and never self-appointed.

They review blind and stay anonymous outside the platform, so no applicant, parent, or rival can lean on them; the editor is the named, accountable face of the channel. Reviewers never score students they know. Each decides how many projects to take in a month, from a handful to a steady practice.

Yes. Institutions endorse channels, put their name behind the rubrics they trust, and steer prospective students toward them; endorsed channels surface in the filters students browse. Admissions teams weight those same channels in their own sorting.

None of it is a prerequisite. An individual professor opens a channel on their own scholarly standing, the way they would join an editorial board. If your institution wants the formal version, write us and we will set up the conversation.

A link or a QR code that opens the record. At the top, a short overview in the student's own voice. Then up to ten entries, each with a date, a title, a score between 1.0 and 2.0, and a code identifying the university and professor whose channel published it, spelled out in a key on the record.

Every entry is one click from the work: watch the student present and defend the project, then go under the video to the sources and documentation. Student-shared records are free to read, with no vendor integration, no portal, and nothing to install. It reads like a transcript and audits like a portfolio.

Because the unit of comparison is the channel, not the school. Students from every school meet the same public rubric, the same fixed four-level scale, and the same blind three-expert panel; the median score shrugs off any one reader's mood. A 1.4 in a given channel means the same thing whether the student sits in the South Bronx or rural Mississippi, because the people who scored it never knew the school.

School reputation, the thing your reading process spends so much effort adjusting for, drops out of the signal. And when a number wants checking, the defense it summarizes is one click down.

No. We built the instrument; you set the values. Score any channel from 0 to 10 by how much you trust it. Weight the race-neutral background fields that matter to the class you are building. Set the size of the list you need, for offers, interviews, or an event, and tune the weights live while the list reorders in front of you.

Every officer on your team can run their own weights, so committee arguments get better: the disagreement resolves by opening the student's actual defense on the screen. The reduction of a student to one number still happens, but it happens last, on your desk, by your definition of merit.

The architecture was built for the post-SFFA era. Race and ethnicity are never collected from students, so they cannot enter any sort, weight, or filter, yours or ours. Context arrives through forty-five race-neutral fields, first-generation status, rural community, multilingual home, caregiving load, and their kin, each grounded in federal reporting standards.

What the ruling left standing is the individualized story, and that is the product: the student's own work, defended in the student's own voice. Your counsel can audit the whole chain, from fields to weights to the recording beneath every ranked row.

Opening a transcript a student shares with you: free, now and always.

Searching and sorting the full database is free for early institutional adopters, and moves to an annual subscription as the pool scales; early adopters are grandfathered in. Against the line items you already carry for travel season, list buys, and application reads that end in a guess, a searchable pool of pre-evaluated, watchable work is the cheap part of the funnel.

Start where you already are: applicants paste transcript links into their applications today, and any reader can open one in a browser mid-read. No integration required on day one.

From there it layers in rather than replacing anything. Holistic review keeps its shape; what changes is the quality of evidence inside it. Use search to source candidates your travel calendar would never reach, verification to check what an applicant claims, and committee playback to settle debates with primary evidence. The data is a menu, not a meal: it exists to point you at the student, the question, and the next conversation.

A Portrait is easy to declare and hard to see. Critical thinking and effective communication do not show up on an answer sheet, so a district can post the poster and still have no idea which classrooms are getting there.

A published project is those capacities, on the record: a student who investigates a question, builds something from it, and defends it on camera is doing the Portrait where anyone can watch. Channels can carry rubrics aligned to your Portrait's attributes, and a district can endorse the channels that match so students see them first. New York, which is phasing out its exit exam as the sole diploma gate while adopting a statewide Portrait, is the live case: the vision is adopted, and what districts need now is an instrument that makes it visible school by school.

Local portfolios do not scale; published ones do. The state portfolio programs of the 1990s died on scoring: no shared rubrics, no rater training, no moderation, so no two readers agreed. Every piece of this system exists because of that lesson. One fixed scale. Public rubrics. Three blind scorers per work, with the median as the score. Rater drift tracked, flagged, and retrained. Calibration paid for, not begged for.

Then the capacity question: evaluation here is paid work done by a distributed network, not a new duty for your teachers. A single three-person panel clears about 130 projects a month in an hour a day each, and panels multiply with demand. Your staff's role stays what it should be: teaching the students and certifying the work is theirs.

Start with one student. Find the one whose work should not disappear, and the scholarship queue publishes them at no cost to you. That one published work teaches your system more than a semester of vendor meetings.

Then scale on purpose: a deck of fifty access cards puts a strengths-based instrument in a principal's hands. Point it at the after-school robotics build, the sports analytics project, the community documentary, the work that never reached a transcript before. Pair one willing teacher with one professor in the same field and you have a full pipeline running.

Measurement comes built in, because the unit is countable and watchable: published works by school, by subject, by race-neutral background field, over time. Published portfolios are the scoreboard.

Three ways, each structural.

• External validation. A portfolio platform stores what students upload; a badge attests what an issuer asserts. Here, three blind experts under a named professor judge every piece before it exists publicly. The record is made of clearances, not uploads.
• Comparability. Badge taxonomies multiply and mean nothing across town lines. This is one fixed scale and public rubrics, so a score travels between schools, districts, and states without a crosswalk document.
• Ownership. The record belongs to the student, permanently, wherever they move and whatever your procurement does next year.

Underneath all three: this is performance assessment, not a competency checklist. We score what a student made, explained, and defended, instead of inferring skills from a list. Let the work speak, not the label.

A Sponsor funds evaluations for students in the scholarship queue. The mechanics match everything else on this page: each funded evaluation is $100, all of it flows to the four educators who grade, and HSCred keeps none of it.

You are not a passive check. Every scholarship request that meets your criteria comes to you for review, and you accept the ones you want in your queue. From there the platform funds them in turn. What a sponsor buys is more published students; what a sponsor gets back is a countable, watchable result.

You choose who is in your queue. Every scholarship request that meets your criteria comes to you for review, and you accept or decline each one. Nothing is assigned to you, and what you review is the request, never a name.

You do not choose the order. Once a request is in your queue, funding goes to the next student in line, with priority to those closest to their college applications, so nobody waits because a funder preferred someone else that month. You decide who is eligible for your money. The queue decides who is next.

No, and the design makes it impossible rather than merely against the rules. A sponsor sets criteria and reviews requests with no names attached, so there is no way to find a particular student in a queue. Funding is then released in queue order, not by anyone's pick. A donation routed through an organization cannot be pointed at a child.

Nor could it be allowed to work that way. A gift earmarked for a named individual is not a charitable gift, and the Academic Capital Foundation's standing as a nonprofit depends on that line holding. The queue is what makes it hold without anyone having to police it.

A family that wants to pay for their own student's evaluation buys an access card. It costs the same $100, it is the straightforward way to do it, and nobody needs a queue for it.

What no amount of money buys is the outcome. Three blind experts decide publication, they are paid the same whether the work publishes or not, and a student who cannot pay anything at all reaches that same panel through the scholarship queue. The fee buys judgment, never a result.

Yes. Sponsorship can be aimed where your mission points: a city, a subject, a population you serve. A youth media nonprofit can fund journalism publishing. A STEM foundation can fund investigations. A college access program can put a deck of cards behind its own cohort of first-generation students.

The forty-five race-neutral social history indicators are what you aim it with. Your criteria decide which requests reach you, you decide which of those to accept, and the queue funds them in turn.

The published work, in full. Every project your scholarships publish is yours to watch: the student's defense, the score, and the channel and professor who published it. You funded a real thing, and you can press play on it.

Work that does not publish stays private, exactly as it does for everyone else. It appears nowhere and reaches no one but the student and their Guide. On your side it shows up only in the overall metrics: how many requests you accepted, how many reached the expert panel, how many cleared the bar.

No. Not names, not contact information, not anything that identifies a student to you. That rule does not bend for the sponsor who paid for the review.

What travels to you is academic: the work, the score, the channel. Demographic information reaches you only in aggregate, only for published projects, and only inside the annual report document. A student decides for themselves when to attach their name to their record, and that decision belongs to the institutions they choose to approach.

Ready-to-insert outcomes reporting. We supply the content and the metadata; you assemble the report your board and your own funders expect, in your own format and your own voice.

The reporting is unusually easy to defend, because the unit of impact is a published work. It counts, and it plays. Aggregate demographic figures cover the published projects, and no figure is ever attached to a student. Few grant outcomes survive that much scrutiny. These do.

Yes, if you want it. Sponsors can take an embed code and put the work on their own site: the projects your scholarships published, or the queue of work still waiting on evaluation, whichever you choose to show.

It is an option, not an obligation, and the same privacy rules follow the work wherever it is embedded.

Yes, where it serves the learning, and every channel says exactly where in its rubric's disclosure policy. The house rule underneath is simple: AI to coach, humans to judge.

During STAKE, an AI tutor is a fine drillmaster: flashcards, practice problems, instant feedback on the content layer. That is the part of learning AI speeds up. Then the student sets it aside, because the project is theirs: the struggle is where the learning lives, and the defense is delivered in their own voice to a human audience. A machine can pass the test. That is why the test is the beginning and the defended work is the proof.

No. Three human experts score every submission, a human Guide certifies it, and a named professor stands behind the channel that publishes it. No model assigns a score anywhere in the pipeline.

The only algorithm in the system belongs to the institutions reading the record, each tuning its own sort with its own weights. HSCred is not a chatbot wrapped in a transcript. It is a publication system staffed by people whose names and reputations ride on their judgment.

By building authenticity in while the work is made instead of hunting for it afterward. Detection loses that game every time; witnesses do not.

The Guide watched the work form over weeks, through the STAKE and two rounds of their own feedback, and signs for authorship. Three blind experts then judge whether the thinking holds up under a ten-minute defense, which is a hard place for borrowed work to hide. The calendar does its quiet part: one project per month, due at month's end, so nobody can dump a pile of suspiciously fast work the season applications are due. And everything is permanent and attributable: the student's record, the Guide's certification, the editor's name. Every layer has skin in the game, and any doubter can press play.

Structure first: one fixed scale, a public rubric written in observable evidence, and the median of three independent scores, so one harsh or generous reader cannot move the result.

Method second: reviewers annotate while they watch, marking where the rubric's evidence appears, partially appears, or is missing, so a score hangs on highlighted moments in the recording rather than on impressions. Oversight third: the platform tracks each reviewer against their panels over time; drift gets flagged to the editor for retraining, and a reviewer who cannot recalibrate is removed. Because the work is paid, staying calibrated is in the reviewer's own interest. This is the machinery that scores AP essays and juried competitions, running in the open where anyone can check it.

Not by default. Published work lives inside the platform, visible to verified institutional users and to anyone the student hands the link. It is not indexed for the open internet.

Two exceptions, both opt-in: a student can choose to have work featured on open pages such as the Showcase, and a student can share their own link as widely as they like. Unpublished work is the strongest rule of all: it appears nowhere, to no one, ever, beyond the student and their Guide.

The strongest protection is refusal to collect. A student can earn a full record with nothing personal attached and be discoverable through the work alone.

From there, disclosure is staged and always the student's move: background context fields when they want to surface in institutional searches, name and contact last of all, when an institution they care about reaches out. Identifying information is stored separately from the academic record. There are no ads and no sale of student data to anyone; the only paying users are institutions with something to offer a student's future: colleges, districts, education agencies, and foundations. A record that can help a student and cannot be turned against them.

Every adult is verified and none is anonymous: Guides register with a verified school email, editors with a university address. The only adult with off-platform contact is the Guide, the educator the student already knows at their own school. Expert Reviewers see work, never contact information, and remain anonymous outside the platform.

Students can publish from behind the camera, documentary style, voice and evidence only, no face and no name. Participation starts at 14, with consent collected at signup. And the commercial rule doubles as a safety rule: nobody on this platform makes a dollar from a student's attention, only from an institution's search for talent.