(Internal) GlobaLeaks channels & specification

Eight questionnaires (GlobaLeaks calls the object you attach to a "Context" a Questionnaire, made of one or more Steps, each holding Fields). This document gives you, for each one: the context name and description to type into the GlobaLeaks admin panel, then every field with its exact label, the hint/description text shown under the field, the GlobaLeaks field type to pick from the dropdown, and whether to mark it mandatory.

Design principle used throughout: almost nothing is mandatory. A whistleblower who abandons the form because question 3 of 9 is a hard stop is a whistleblower we never hear from. Where a field is marked mandatory below, it's because the answer is needed to make the rest of the submission usable (e.g. "can you actually test this" gates the whole Ghostyle-testing flow) — everything else is opt-in.

GlobaLeaks field-type quick reference

In this document	Pick in GlobaLeaks admin
Short text	"Single-line text"
Long text	"Multi-line text"
Single choice	"Multiple choice question" with "Allow multiple answers" off
Multiple choice	"Multiple choice question" with "Allow multiple answers" on
Date	"Date"
File upload	"File upload"

For every "Single choice" / "Multiple choice" field below, add the listed options in the order given — order matters less than completeness, but keeping "Other" / "I don't know" last reads better.

For each context, also fill in the plain-language Context description field in the admin panel — that's the text GlobaLeaks itself shows above the form. Reuse the one-line description under each heading below for that.

1. System overview

"You work with facial-recognition technology." Context description: For anyone who understands how a specific system is built, deployed, or operated. Answer whatever you're comfortable with — most fields here are optional on purpose.

Field label	Description / hint shown to submitter	Type	Mandatory
What kind of place is this deployed in?	Pick the closest match.	Single choice: Retail · Transport & transit · Law enforcement · Airport & border control · Education · Stadium & events · Workplace access control · Other	No
If "other," what kind of place?	Only needed if you picked "Other" above.	Short text	No
Country or region of deployment	City-level is fine. Even a country alone is useful.	Short text	No
Vendor or product name, if known	The commercial name of the system, or the company that built or sold it. You can always add this later with your receipt code if you'd rather not name them yet.	Short text	No
What technology does it use?	Camera type (visible light, infrared, 3D depth), the matching algorithm if you know it, whether processing happens on-device or in the cloud, anything about the underlying model.	Long text	No
How many people or locations does it cover?	Rough numbers or scale are fine.	Long text	No
Who can access match results or footage?	Roles, departments, or third parties with access — job titles are enough, names aren't needed.	Long text	No
How long is data retained?	An exact policy if you know it, or your best estimate.	Short text	No
Is there a human review step before any action is taken on a match?		Single choice: Yes, always · Sometimes · No · I don't know	No
Anything else about how this system works?	Open field — use it for anything the questions above didn't cover.	Long text	No

2. Test our countermeasures

"You can run our test cases against a real system." Context description: For anyone with access to a live or staging facial-recognition system who's willing to try a Ghostyle or test image against it and tell us what happened.

Field label	Description / hint shown to submitter	Type	Mandatory
Do you have access to a system you could safely test against?	This is the one question we do need answered, since it decides whether the rest of this form applies.	Single choice: Yes, a live production system · Yes, a staging or test environment · Only a demo or sandbox · Not sure yet	Yes
What kind of system is it?	A sector or product name, whatever you're comfortable sharing.	Short text	No
Can you receive or upload a test image without it being logged in a way that could identify you?	Helps us understand what's safe to send you.	Single choice: Yes · No · Not sure	No
What could you test?	Select everything that applies.	Multiple choice: Detection (does it find a face at all) · Matching against a known identity · Liveness / anti-spoof detection · Something else	No
Describe the result	What happened when you tried it — detected as expected, failed, behaved unpredictably. Include confidence scores if you can see them.	Long text	No
Attach a redacted result, if you can share one	A screenshot or log works. Please blur or remove anything that could identify you first — your name, employee ID, an exact timestamp plus location together.	File upload	No
Willing to try more examples through your receipt code?		Single choice: Yes · Maybe · No	No

3. Weak spots & field recognition

"You know where a deployment falls apart." Context description: For anyone who's seen a specific facial-recognition deployment that's misconfigured, poorly maintained, or otherwise weaker than its spec sheet suggests — and for anyone who can describe what it looks like from the outside, so others can recognize the same setup.

Field label	Description / hint shown to submitter	Type	Mandatory
What kind of weakness is this?	Select everything that applies.	Multiple choice: Misconfiguration · Outdated software or firmware · Poor physical security of the hardware · Excessive or indefinite data retention · Weak or missing human oversight · Something else	No
Describe the weakness		Long text	No
How confident are you this still applies today?		Single choice: Confirmed recently · Confirmed a while ago, may have changed · Heard about it secondhand	No
What does this deployment look like from the outside?	Visible hardware, branding, signage, mounting position, LED or IR indicators — anything that would help someone recognize the same setup elsewhere.	Long text	No
Where else might the same setup exist?	Other branches, cities, or vendors known to use the same or a similar configuration.	Long text	No
Attach a photo of the hardware or signage, if it's safe to	Only if you weren't clearly recorded taking it yourself.	File upload	No

4. Contracts & market

"You've seen the paperwork." Context description: For anyone with knowledge of a contract, tender, budget, or vendor relationship behind a facial-recognition deployment — including legal, compliance, or procurement records.

Field label	Description / hint shown to submitter	Type	Mandatory
What kind of document or knowledge is this?	Select everything that applies.	Multiple choice: Signed contract · Draft or tender proposal · Internal budget · Vendor sales pitch · Impact or compliance assessment · Verbal knowledge, no document	No
Vendor or contractor name		Short text	No
Buyer / client organization		Short text	No
Approximate contract value or budget, if known		Short text	No
Contract duration or dates		Short text	No
What does the agreement actually cover?	Hardware, software licensing, maintenance, data hosting, analytics, staff training — anything bundled in.	Long text	No
Describe the market context	Competing bidders, exclusivity clauses, lobbying, prior relationships between vendor and buyer — anything about how the deal came together.	Long text	No
Attach a document, if you can share one safely		File upload	No

5. Operators & frontline staff

"You run the system, day to day." Context description: For security staff, control-room operators, or anyone whose job involves watching, acting on, or overriding what a facial-recognition system flags.

Field label	Description / hint shown to submitter	Type	Mandatory
What's your role in relation to the system?	E.g. security guard, control-room operator, help-desk staff, store manager.	Short text	No
How often does it flag someone incorrectly, in your experience?		Single choice: Often · Occasionally · Rarely · I don't know · We're not told	No
What actually happens when there's a match or alert?	Walk us through it — who's notified, what the written procedure says, whether it's actually followed.	Long text	No
Can staff override or dismiss a flagged match?		Single choice: Yes, easily · Yes, but discouraged · No · I don't know	No
Have you ever raised a concern internally about this system?	What you raised, and what happened as a result, if anything.	Long text	No
Anything about day-to-day use the manual doesn't mention?		Long text	No

6. Procurement & public officials

"You approved it, funded it, or signed off on it." Context description: For anyone on the decision-making or oversight side of a facial-recognition purchase — budget approval, tender evaluation, legal/DPO sign-off, or an oversight committee seat.

Field label	Description / hint shown to submitter	Type	Mandatory
What's your relationship to the decision?	E.g. budget approval, tender evaluation, oversight committee, legal/DPO sign-off.	Short text	No
Was there a public tender?		Single choice: Yes, competitive · Yes, single-bidder · No, direct award · I don't know	No
Was a data protection or human rights impact assessment carried out?		Single choice: Yes, and I can share it · Yes, but I can't share it · No · I don't know	No
Who reviewed or approved this beyond the immediate team?		Long text	No
What oversight exists after deployment?	Audits, public reporting, a complaints mechanism, a scheduled review or renewal date.	Long text	No
Attach a tender, assessment, or oversight document, if safe to share		File upload	No
Anything about how this decision was made that concerned you?		Long text	No

7. Public sightings

"You spotted something in public space." Context description: The lowest-barrier form here — you don't need inside access or technical knowledge. If you saw a camera, kiosk, or sign you didn't recognize, this is enough.

Field label	Description / hint shown to submitter	Type	Mandatory
Where did you see it?	City, neighborhood, or exact address if you're comfortable — even approximate is useful.	Short text	No
When did you see it?	Approximate date is fine.	Date	No
What did it look like?	Camera type, mounting, visible branding or model numbers, kiosks, signage.	Long text	No
Was there signage disclosing its use?		Single choice: Clear signage · Vague or hard-to-find signage · No signage · Didn't check	No
Attach a photo, if it's safe to take one	Avoid capturing yourself or bystanders' faces if you can — we can help redact if needed.	File upload	No
Anything else about the location or context?		Long text	No

8. Affected individuals

"You were stopped, flagged, or misidentified." Context description: For anyone who was personally stopped, denied service, questioned, or misidentified by a facial-recognition system. Your experience is evidence, whether or not you have technical details.

Field label	Description / hint shown to submitter	Type	Mandatory
What happened?	In your own words — stopped, denied entry or service, flagged, questioned, anything else.	Long text	No
Where and when did it happen?		Short text + Date (use two fields: "Location" as Short text, "Date" as Date)	No
Do you know which system or organization was involved?		Short text	No
Were you given an explanation at the time?		Long text	No
Did you try to challenge or appeal it?		Single choice: Yes, successfully · Yes, unsuccessfully · Yes, still ongoing · No	No
Would you be willing to be connected with a journalist or advocate about this?		Single choice: Yes · Maybe — contact me via my receipt code first · No	No
Attach any documentation you received	E.g. a denial letter or incident report.	File upload	No

After you set these up

Create each context in GlobaLeaks admin, paste in the description above, and build its questionnaire from the field table.
Copy the resulting https://raccontaci.nina.watch/#/submission?context=<uuid> URL for each one.
In report.html, replace the placeholder href="https://raccontaci.nina.watch" on each of the eight .audience-card links with its real context URL — they're marked with an HTML comment () right above the card grid.