Notice

Privacy

Block by Block does not know who you are.

Here is the short version. There is no account, no sign-up, no e-mail address, no password and no profile. A report you file carries no name, no user identifier and no device identifier, and no IP address is written to our database. There is no analytics, no advertising, no tracking of any kind. This page explains exactly what little we do hold, and why.

If you file a report and close the tab, nothing in our database links that report back to you. Not a hidden identifier, not an IP address, not a cookie. The honest answer to “what do you hold about me?” is: almost nothing.

Who runs this site

Block by Block is operated by Arcano Zero – AI Lab Srl, a company based in Rome, Italy, which is the data controller. It is a public demonstrator of a civic participation platform. It is not an official City of New York service, it is not endorsed by the City, and nothing you submit here reaches any city agency. Because the operator is established in Italy, the GDPR applies to everything described on this page.

What we store when you file a report

One row in a database. It contains:

• the text you wrote, capped at 280 characters;
• the category you chose (street & sidewalk, lighting, sanitation, housing, feels unsafe, parks & trees, something else);
• whether you marked it a problem or something working well;
• a latitude and longitude already snapped to a grid of roughly 150 meters;
• the derived block identifier, the district, the interface language you were using, a status field and a timestamp.

That is the whole row. There is no user identifier, no IP address, no device identifier and no cookie attached to a report.

Why your location is blurred to about 150 meters

A precise point on a map can identify a building, and sometimes an apartment. So the server rounds every coordinate to a grid of roughly 150 meters before writing it down. The exact point you tapped, or your exact GPS position, is never stored. It exists only in transit, for the moment it takes the server to snap it. Block-level precision is enough to see a pattern on a street; it is not enough to point at a door.

Corroborations and neighborhood alerts

When you tap “same here” on a report, we store the report identifier, a timestamp, and a SHA-256 hash of a random value your browser generated for itself. That value is random to begin with, is not derived from anything about you, and the hash cannot be turned back into a person. Its only job is to stop the same browser counting twice.

If, and only if, you switch on neighborhood alerts, we store the push endpoint URL your browser provides along with its cryptographic keys, a location snapped to the same 150 meter grid, and your interface language. Turning notifications off in your browser withdraws that consent.

One more table

A fourth table holds an administrative status for each priority cluster — new, under review, planned, in progress or done. It is set only from the protected administration area and contains no personal data.

What stays in your browser

Four values are kept in your browser’s local storage: “bbb-lang”, your chosen language; “bbb-session”, the random identifier described above; “bbb-same”, the list of reports this browser has already corroborated; and “bbb-cluster-status”, a local copy of administrative statuses. None of them is stored on our servers. “bbb-session” is the only one that ever leaves your browser: it is sent when you tap “same here” so the server can hash it, and only the hash is written down — the raw value never is, exactly as with your coordinates. These are functional, not tracking. Clearing your browser data erases all of them.

Cookies: there is exactly one

The only cookie on this site is “bbb_admin”, and it is set only when a member of district staff signs in to the protected administration area. It is httpOnly, SameSite=Lax, Secure in production, and expires after 12 hours. It holds a signed token proving a valid sign-in and contains no name and no identifier of a person. There are no analytics cookies, no advertising cookies, no tracking pixels and no third-party cookies. That is why you see no cookie banner: there is nothing to consent to.

Analytics and tracking

None. No Google Analytics, no Plausible, no Meta pixel, no advertising network, no session recording, no fingerprinting, no cross-site tracking. Web fonts are hosted on our own servers, so no font network sees you either.

Who else your browser talks to

• Vercel Inc. hosts the site and delivers it. Server functions run in the US East (iad1) region. Vercel processes standard server request logs.
• Supabase hosts the PostgreSQL database, in the AWS us-east-1 region in the United States. Your browser never contacts it directly; only our server does.
• OpenFreeMap / OpenMapTiles supplies the map tiles, and your browser contacts it directly every time a map is drawn. That means your IP address is visible to that service. We chose it over Google Maps precisely because it needs no API key and does no commercial tracking, but you should know it happens.
• NYC Open Data (data.cityofnewyork.us) is contacted only if you click a data-provenance link on the About page to check a figure for yourself.
• Browser push services run by Google, Apple, Mozilla or Microsoft, depending on your browser, are involved only if you enable neighborhood alerts. The push endpoint lives on your browser vendor’s infrastructure, so enabling alerts necessarily involves them.
• Anthropic (Claude API), in the limited way described next.

Screening, and where AI is and is not involved

Every report passes an automatic screen before it is published, looking for two things only: personal details about identifiable private individuals, and abuse. Opinions are never screened out. A flagged report is not deleted; it is held with a pending status until it has been reviewed.

The cluster briefings shown in the dashboard are generated offline by a maintenance script that reads only aggregated public 311 statistics. No visitor data is involved. Separately, the screening step is capable of sending the text of a report to the Claude API to double-check an automated flag, but only when an API key is configured for the deployment. No key is configured in production today, so that path is inactive. We disclose it in advance rather than switching it on quietly later.

Legal bases

We rely on legitimate interest to operate a public-interest civic demonstrator, and on your consent for neighborhood push notifications, which you can withdraw at any time by disabling notifications in your browser.

How long we keep it

Reports and aggregated data are kept for the duration of this demonstration project. Push subscriptions are deleted automatically as soon as the browser vendor reports them as expired or invalid.

Where the data is stored

The operator is in Italy; the database and the server functions are in the United States. So the data does cross the Atlantic, and we would rather say that plainly than hide it in a clause. What crosses is the content described above: 280-character texts, categories, block-level coordinates and timestamps, with no identifiers attached. Wherever the servers sit, this policy and the GDPR apply.

Your rights, and one honest limitation

You have the rights the GDPR gives you: access, rectification, erasure, restriction, objection and portability. There is a real limit on how we can exercise them here, and it comes from the same design that protects you. Because reports carry no identifier, we generally cannot tell which report is yours. Under Article 11 GDPR, where a controller cannot identify a data subject, the rights of access, rectification, erasure and portability do not apply unless you supply information that makes identification possible. Your right to object to processing, and your right to complain to a supervisory authority, are not affected.

We do not treat that as a way out. If you can point us to a specific report, for example by quoting its text and telling us roughly when you filed it, we will find it and act on your request. Write to us and we will help.

No sale, no sharing for advertising

We do not sell data. We do not share it with advertisers or data brokers. Not now, and not later. The third parties listed above are the only ones involved: Vercel and Supabase process data on our instructions, and the map, open-data and push services are reached by your own browser, as described above.

Children

This service is not directed at children, and it collects nothing that would identify anyone, of any age.

Contact

For anything on this page, including a rights request or a complaint, write to a.megacreator@gmail.com. This is the only contact address available today; a dedicated address will replace it. You also have the right to lodge a complaint with a supervisory authority, in Italy the Garante per la protezione dei dati personali.

Last updated: July 27, 2026