1. Data Controller
FjordSpot is operated by Cato Tangen (private individual) from the Bergen area, Norway. Contact: support@fjordspot.no
2. What Data We Collect
- Account: Email address and password, your language choice and the account's moderation status (for example whether it is paused). You do not need to provide a username when registering.
- Profile: Display name (the app gives you a random name, which you can change), home region, country, how long you have been fishing and a short text about yourself. Everything except the display name is optional.
- Location data: GPS coordinates and map selections you register or choose in the app for analysis and display. Precise location is stored with the catches and spots you register, linked to your account, until you delete them. We do not track your location in the background.
- Catch data: Species, weight, length, photo, notes, and position you voluntarily register, and what the app adds itself: weather, moon and tide at the time of the catch and the habitat score (HSI) for the spot.
- Saved locations: Coordinates and names you save as fishing spots, with the tags, folders, spot type and colour you choose. Your saved locations are private and not visible to other users. Access may occur in limited cases for operations, security, backup, or troubleshooting, but is not used for sharing or publication. For operations and troubleshooting we use AI assistants from Anthropic (Claude). They only retrieve aggregated figures from the database and the operations logs, not information about individuals. This is a rule we have given them, not a technical barrier. If information about one user has to be viewed to fix an error, the controller does it personally, without the assistants.
- Community: If you share a catch, other signed-in users see the photo, the species, the weight, the time of the catch, the trophy mark, the region and an approximate position (rounded to about half a kilometre, or no position if you have turned that off), together with your display name. Comments and reactions you give are shown with your display name, and the owner of the catch gets a notification. Your profile is private by default. Private means that others see your display name, both on what you share and if they look up your profile, but nothing more. If you choose a public profile, they also see your home region, fishing years, the text about you, when the profile was created, and the counts of shared catches and reactions. The “friends” option counts as private until a friends feature exists. The catches you share are controlled by a separate setting (“Show catches in the community”), not by the profile setting, and they always show your display name. Profiles that existed before this update were also set to private; if you want to show more, choose public yourself. If you report content, we store the report with its category and anything you wrote. If you block someone, we store who, and the two of you no longer see each other's content. You control sharing, position and visibility under Settings → Privacy. Photos you share in the community are seen by an admin or a moderator before they become public. Moderators also see reports about content, but not who made them. Moderators sign a confidentiality agreement before they are given the role. The person who is reported never learns who made the report. What cannot be hidden is that you can see which of your content is hidden, and roughly when. In a thread with only two people, you may therefore be able to guess who reported it.
- Error reporting: Pseudonymized error reports are sent to Sentry (EU region) for troubleshooting. For signed-in users, reports are linked to an internal user ID (not your name or email) so we can follow up on errors affecting your account. A report can contain an approximate position (about 1 km), for example the place you asked for weather or waves for. Performance measurements (how long pages and requests take) are also sent to Sentry, but only for roughly one in ten page loads or navigations in the browser. If you are signed in, they are linked to the same internal user ID. The iPhone app sends no performance measurements, only error reports. The server also sends measurements for roughly one in ten requests, without a user ID, but with a trace ID that also appears in the server's operational log. Our own error log is stored without a user ID. The count of how much AI you have used (the quota) is stored with your user ID and deleted after 90 days.
- Security logs: We store technical logs related to login, security, and abuse prevention for a limited period (see “Retention Period”). The server also has an operations log. It may contain your internal user ID and approximate positions from map lookups, but it is not meant to contain your name or email.
- Usage statistics: When a page is opened, we store its address (without parameters), the page you came from and your browser's technical identifier (user agent). We do not store your IP address or your account with this. Page views are recorded once you have accepted the terms, and on the information pages (including terms, privacy, help and changelog) also without that. In the iPhone app, page views are only recorded if you have said yes to it. The question is on the consent screen, it is off by default, and you can change the choice at any time under More → Privacy, without signing in. The choice is stored on your device, not on your account. The Android app is the website in Chrome, and there what is said about the browser applies.
- Push notifications: If you turn on push notifications, we store the subscription your browser creates (an address at your browser's push service and encryption keys) and your browser's user agent, linked to your account. When you turn notifications off in the app, we try to delete the subscription on our side. If that deletion fails, it may remain until you delete your account. Subscriptions that the push service reports as expired during a delivery are removed, and everything is deleted when you delete your account. Beyond this, there is no fixed deletion deadline. If you have allowed notifications in the iPhone app, the next two Fish Alone check-in reminders (at the deadline and five minutes after) are stored as notifications on your phone, not with us. They are removed when you check in or end the trip in the app. If you end the trip from another device, a notification may still appear until the app has confirmed that the trip is over. The notifications do not replace the email reminder when it is turned on.
- IP address: We use your IP address to limit the number of requests and prevent abuse. The counters for this only apply within a time window of at most one hour. When you sign in, we count attempts per email address and IP address in time windows of at most 15 minutes, to stop password guessing.
- Photos for AI species identification: Photos you take for species identification are first assessed by FjordSpot's own image model, which runs on our own machine in Norway. If the model is uncertain or unavailable, the photo may be sent to Anthropic (Claude) as a data processor for interpretation, but only if you have consented to AI and usage limits allow it. If you have not said yes, the photo is not sent there. If our own model can make suggestions, you will see them marked as uncertain. You choose the species yourself. On a successful identification, FjordSpot stores a downscaled copy of the image on our own server in Norway to improve species recognition; the legal basis is legitimate interest (Art. 6(1)(f)), the copy is never shared publicly and is deleted when you delete your account. Photos you add to a catch are stored by FjordSpot on our own server in Norway, privately and access-controlled, until you delete the catch or your account. If someone has reported the photo and the case is open, we keep it hidden as evidence for at most 30 days after you have deleted the catch. It is not shown to others. If you delete your whole account, that photo is deleted at once as well. Anthropic normally retains the data for 30 days, and longer only in the exceptions Anthropic describes in its terms (for example violations of its usage policies or legal requirements).
- Voice-controlled catch logging: Your speech is turned into text in one of two ways, and only one of them sends audio to FjordSpot. In English and German in the browser, the browser's own speech recognition is used (Google in Chrome, Apple in Safari), which may send the audio to the vendor's speech service for transcription; the same applies to the Android app, which is the website in Chrome. The iPhone app does not use the browser's speech recognition, and there voice logging is only available in Norwegian. With the browser's speech recognition FjordSpot receives only the text and up to four alternative interpretations, never the audio, and stores no audio recording. If you use voice logging in Norwegian, the app records the audio and sends the recording to FjordSpot, where our own speech model (NB-Whisper) on our own machine in Norway turns it into text. Then the audio recording is stored by FjordSpot on our own server in Norway. Audio recordings on FjordSpot's server are normally deleted during nightly cleanup after 90 days. Recordings linked to cancelled drafts may be deleted earlier. If the service has just failed, the app uses the browser's speech recognition for the next recording instead and says so on screen. If the device cannot record or store the recording locally, for example in private mode, with blocked storage or in an older browser, the app uses the browser's speech recognition without a message. The browser's speech recognition is also used, without a message, while an older version of the app is still running on the device, or when the app does not respond quickly enough. In all of these cases FjordSpot receives no audio. The browser's speech recognition may still send the audio to Google or Apple, as described above. The iPhone app uses no such fallback. Whether the text is interpreted by AI (Anthropic) is your choice the first time you use the feature, and you can change it under Settings → Privacy. If you say no before recording starts, the regular catch form opens. The text from your speech is not forwarded to Anthropic without a current yes. If the app cannot read your AI choice, it may still try to send to FjordSpot, which rejects new processing when your choice is no. Recordings waiting on your device can be deleted, or processed after you have said yes to AI. Your answer (yes or no, with version, time and language) is stored on your account. The text is sent to FjordSpot and on to Anthropic (Claude) for interpretation, together with up to four alternative interpretations from the speech recognition. To get place and species names right, we also send the names of up to 30 of your saved spots and up to 10 species you have logged recently — the audio recording itself is never sent to Anthropic. The text from your speech is kept as part of the catch, also after the audio recording has been deleted. We also store when the confirmation card for a voice catch was shown on your screen. We use it to see how often the card is seen before the catch is saved automatically. The time is deleted together with the draft. Drafts are normally deleted during nightly cleanup when they are more than 90 days old. Cancelled drafts may be deleted once they are more than 24 hours old. Age is counted from when the draft was created.
- Fish Alone contact: If you use the Fish Alone feature, we collect the name and email address of the contact person you provide so we can notify them of missed check-ins or expired sessions. When an alarm is raised, the contact person receives an email with your name, your last known position, your battery level and a map link. Through the link in the start email, the contact person can follow the trip while it is running: name, status, times, the latest check-in (position, accuracy and battery) and up to the 120 most recent positions — never your email address or phone number. The link stops working when the trip ends. This information is used solely for this feature and is deleted when the session ends, with the exceptions below.
- After an alarm in Fish Alone: If an alarm was sent during the trip, or the trip ended after the check-in deadline, we keep your name and email address as they were when the trip ended, the contact persons' names and email addresses, the check-ins with position, the position track and the times for 7 days after the trip. The purpose is to be able to give them to the police or the rescue services (the Joint Rescue Coordination Centre or a search team). This only happens when they ask for it in connection with a search for you. The legal basis is the protection of vital interests (GDPR Art. 6(1)(d)). The data is used for nothing else and is deleted automatically after 7 days. If you ask to delete your account during this period, the deletion is postponed until the 7 days have passed, and then happens automatically.
- When a trip ends by itself: If a Fish Alone trip ends automatically without an alarm having been sent and without the check-in deadline having passed, we keep the following for 7 days: the latest check-in and the latest check-in with a position (coordinates, accuracy, battery level and time), and the most recently received position point (coordinates, accuracy, time and, if your phone provided them, speed and direction). The contact persons, the other check-ins, the rest of the position track and details of a catch used as a sign of life are deleted then, and the sharing link stops working. The reason is your safety: if you are reported missing after the trip, we can give the coordinates and the time of your latest check-in to the police or the rescue services (the Joint Rescue Coordination Centre or a search team). This only happens when they ask for it in connection with a search for you. The legal basis is the protection of life and health (vital interests, GDPR Art. 6(1)(d)). The data is used for nothing else and is deleted automatically after 7 days. If you ask to delete your account during this period, the deletion is postponed until the 7 days have passed, and then happens automatically.
3. Purpose and Legal Basis
Contract (GDPR Art. 6(1)(b))
We process personal data necessary to provide the service, including for:
- Creating and operating user accounts
- Login and authentication
- Storing catches and fishing spots
- Map analysis and habitat assessment (HSI)
- The community: sharing catches, profile, comments and reactions, when you choose to use it
- Fish Alone while the trip is running: position, battery level and email to the contact persons you entered
- Species identification from photos and interpretation of voice logging
Legitimate interest (GDPR Art. 6(1)(f))
- Security and abuse prevention
- Troubleshooting and technical stability
- Improving the service and habitat model
- Usage statistics (page views without IP address or account)
- Training copies of species photos on a match, to improve species recognition (you can object, see “Your Rights”)
- Keeping audio recordings so we can check the interpretation if something went wrong. The retention period is described below.
- The names and email addresses of the contact persons in Fish Alone: your and their interest in someone being told if you do not report in
- Moderation in the community: reports and blocking
Vital interests (GDPR Art. 6(1)(d))
- Data from a Fish Alone trip for 7 days after an alarm or an expiry, for use in a search (see section 2)
Consent (GDPR Art. 6(1)(a))
- Push notifications for fishing conditions (turned off under Settings → Notifications)
- Use of AI (Anthropic) for species photos and voice catches (changed under Settings → Privacy)
- Usage data (page views) in the iPhone app (changed under More → Privacy, no sign-in needed)
The consent to AI covers only the sending to Anthropic. The training copies of species photos and the use of location data in the habitat model (section 4) do not rest on consent but on legitimate interest, and you can object to them (see “Your Rights”). You can also turn off the use in the habitat model yourself under Settings → Privacy (section 4).
4. Use of Location Data in the Habitat Model
The position, species, time and conditions of your catches are also used to check and improve the habitat model. We calculate the habitat score for the catch location and store it with the catch. The calculation runs on our own server. It does not use your name or your account, and the result is stored only as numbers per species. No fishing spot is shown or shared with others. The legal basis is legitimate interest. You can turn this off under Settings → Privacy (“Use my catches to improve the map”). The habitat score is then no longer stored for your catches, the stored habitat scores are deleted, and none of your catches are used to improve the model. If you turn it back on, your catches are used again, including older ones, and the habitat scores are calculated again. The FjordSpot score shown on your catches is a different number and is kept. Catches you delete are no longer used.
5. Data Sharing
We do not sell personal data. Service providers that process data on our behalf:
- Sentry (EU): Error reporting and performance monitoring.
- Cloudflare (USA): The service sits behind Cloudflare, which protects against attacks and delivers the traffic. All traffic to the app passes through Cloudflare, which therefore sees your IP address and the content of the requests. Cloudflare is a US company.
- Google (map link in Fish Alone): The emails to the contact person and the contact page contain a link to Google Maps with your position. If the contact person opens the link, Google receives the position and the contact person's IP address. We send nothing to Google ourselves.
- Resend (USA): Sending emails (password reset, verification, Fish Alone notifications). The Fish Alone alarm email contains your name, your last known position and your battery level, and it passes through Resend. Emails are sent from Resend's EU region (Ireland), but Resend states that its primary processing takes place in the USA. The transfer is based on the EU-U.S. Data Privacy Framework and the EU Standard Contractual Clauses (SCC).
- Anthropic (USA): AI interpretation of photos for species identification and text from voice logging (transcript, alternative interpretations and the names of saved spots and recently logged species) — the audio recording itself is never sent to Anthropic. Transfer to the USA occurs under EU Standard Contractual Clauses (SCC). Anthropic normally retains the data for 30 days, longer only in the exceptions Anthropic describes in its terms. For operations and troubleshooting Anthropic receives only aggregated figures, not personal data.
Other external services:
- Google / Apple (browser speech recognition): Speech-to-text in voice-controlled catch logging (only when you use the feature). Your browser may send the audio to the browser vendor's speech service for transcription (Google in Chrome, Apple in Safari, and in the Android app Chrome). The iPhone app does not use the browser's speech recognition. For voice logging in Norwegian they are used only as a fallback: when our service has just failed, when the device cannot record or store the recording locally, for example in private mode, and while an older version of the app is still running on the device or the app does not respond quickly enough.
- Your browser's push service (for example Google in Chrome and Android): Delivers push notifications if you have turned them on. The push service sees the subscription address and technical delivery information (such as time and message size). The notification content itself is end-to-end encrypted and cannot be read by the push service.
- Kartverket / GeoNorge: Map and depth data. Map tiles load directly in your browser, also in the position picker when you log a catch and on the contact page in Fish Alone, so Kartverket receives your IP address and which part of the map you are viewing. We send no account information.
- Miljødirektoratet (Norwegian Environment Agency): Marine habitat layers in the map. These tiles load directly in your browser when you enable the layer, so Miljødirektoratet receives your IP address and which part of the map you are viewing. We send no account information.
- Norwegian Meteorological Institute: Weather and wave forecasts for the analysis point. Fetched by our own server, not by your browser — only the coordinates of the analysis point are sent, no identity.
- Kartverket (tides): Tide data for the analysis point. Fetched by our own server — only the coordinates are sent, no identity.
- NGU (Geological Survey of Norway): Seabed type for the analysis point. Fetched by our own server — only the coordinates are sent, no identity.
- BarentsWatch: Ship traffic (AIS) and lice counts for aquaculture sites. Fetched by our own server — only the map extent or the site number is sent, no identity.
- FjordSpot's own models: The image model for species recognition and the speech model (NB-Whisper) run on our own machine in Norway. Nobody else receives the photos or the audio from there.
- Apple (App Store) and Google (Google Play): If you download the app from the App Store or Google Play, Apple or Google processes information about the download under their own privacy policies. From them we receive aggregated statistics, and crash reports from devices whose user has agreed to share them. We do not learn who you are.
Where third parties process personal data on our behalf, this is done as data processors under data processing agreements where required.
6. Storage and Security
- Primary service data is stored on servers in Norway. Certain technical service providers may process limited data as part of operations, security, email, or error reporting.
- Passwords are stored as secure hash values (bcrypt) and never in plaintext.
- All traffic is encrypted with HTTPS/TLS.
- CSP, CSRF protection, and rate limiting are implemented.
- Backups are taken daily. Locally we keep the latest 14 daily, 8 weekly and 12 monthly copies, which normally reaches about one year back. As protection against losing the whole server, two external copies are also kept: one on another machine in the same house (the server's host machine) and one encrypted copy with a third party at another address. The copy with the third party is pruned manually today, not on a fixed schedule. Deleted data may therefore remain in backups for about a year after deletion, and in the external copies until they are pruned.
7. Retention Period
Account data is retained as long as you have an active account. Inactive accounts are deleted after 24 months — you are notified by e-mail about one month in advance, and a single sign-in is enough to keep the account. Catch data and saved locations are deleted when you delete your account. If someone has reported a photo and the case is open, we keep it hidden as evidence for at most 30 days after you have deleted the catch; if you delete your account, it is deleted at once. Audio recordings on FjordSpot's server are normally deleted during nightly cleanup after 90 days. Recordings linked to cancelled drafts may be deleted earlier. With the browser's speech recognition no audio is stored, only the text from your speech, which is kept in the catch. Training copies of species photos and the results of species identification have no separate deletion deadline and are kept until you delete your account. Reports you have submitted are deleted when you delete your account, including those that have not been handled. If you keep your account, a handled report is deleted 12 months after it was handled; a report that has not been handled stays until it is handled. Moderation notifications are deleted 12 months after they were created. Moderation decisions about you are anonymised by a nightly cleanup once they are older than 12 months, counted from when the decision was made. If a decision concerns something still in effect, such as a pause in the community or hidden content, it is anonymised by a nightly cleanup once more than 12 months have passed since that was lifted. Other notifications you receive have no separate deletion deadline and are kept until you delete your account. Our own error log is deleted no later than 90 days after the error last occurred. Pseudonymized error reports at Sentry are deleted no later than 90 days after they were received. Performance measurements at Sentry are deleted no later than 90 days after receipt, although a downsampled selection may be kept for up to 13 months after the first 30 days. Security logs are retained for up to 90 days. Page views in the usage statistics are deleted by a nightly cleanup once they are older than 13 months.
8. Your Rights
Under the GDPR, you have the right to:
- Access: Request a copy of your data.
- Rectification: Correct inaccurate information.
- Erasure: You can delete your account and all associated data directly in the app under Settings → Data → Delete account. Alternatively, contact support@fjordspot.no. The following is not deleted at the same moment:
- backups: normally gone after about a year; the copy with the third party is currently pruned manually, not on a fixed schedule (see “Storage and Security”);
- our own error log: no later than 90 days after the error last occurred; error reports at Sentry, with a pseudonymous user ID: no later than 90 days after the report was received;
- performance measurements at Sentry: at most 90 days after receipt, although a downsampled selection may be kept for up to 13 months after the first 30 days;
- the operations logs on the server: up to 90 days;
- the emails and the delivery log at the email provider: normally 30 days (applies to all emails we send, including the Fish Alone notifications);
- what Anthropic has received of photos and voice text: normally 30 days (see section 5);
- reports others have sent about your content: they remain without a link to your account and are deleted 12 months after they were handled; the snapshot of the reported content (the evidence) is deleted no later than 30 days after the report was sent; the moderation decisions are anonymised at the moment the account is deleted;
- if Fish Alone sent an alarm, or a trip ran out of time, the whole deletion is postponed until 7 days after the trip, and then happens automatically (see section 2).
- Data portability: In the app you can export your catches (JSON and GPX) and your saved spots (GPX) under Settings → Data. If you want a copy of the rest, such as your profile, photos, comments or audio recordings, write to support@fjordspot.no and you will receive it within 30 days. The copy also includes the reports you have sent, the moderation decisions about you and your photos, the periods the community has been paused for you, when content you have posted has been hidden, your comments, your answers to the AI consent, your choice about using your catches to improve the map and, if you are or have been a moderator, when you were given and lost the role. For decisions, pauses and hidden content, the copy gives only the day, not the time. It does not include who has reported your content.
- Objection: Object to processing based on legitimate interest.
- Restriction: Ask us to restrict the processing of your data, for example while we check whether it is accurate or whether an objection should be upheld.
- Withdraw consent: At any time: push notifications under Settings → Notifications, the AI choice under Settings → Privacy, and usage data in the iPhone app under More → Privacy.
Contact us at support@fjordspot.no to exercise your rights. You may also file a complaint with the Norwegian Data Protection Authority (Datatilsynet).
9. Cookies and Storage on Your Device
FjordSpot stores the following on your device. None of it is used for tracking or advertising, and we have no third-party tracking or advertising cookies.
- Login: Two cookies that keep you signed in, valid for 7 days. They are renewed when you use the app during the last day before they expire. A third cookie is set if a sign-out did not manage to delete the session on our side. It lasts 7 days, may be set again at sign-in until the revocation is settled, and only applies to the sign-in service.
- Language: One cookie with your language choice, valid for one year.
- Settings: About 25 choices you make in the app (theme, map layers, notifications, Fish Alone, voice and the like), the fact that you have accepted the terms, and in the iPhone app your choice about usage data (yes or no, with version and time). They live in the browser's local storage until you clear site data.
- Queues for voice catches: Audio recordings, position and your user ID sit in the browser's database until they have been sent to us. The new audio queue holds up to 50 recordings or 15 MB. The older queue for the browser's speech recognition also stores the text from your speech, and has no such limit. Analyses you request without a connection are queued with their coordinates.
- Map cache: Map tiles you have viewed (up to 2000 tiles, about 200 MB) and the latest weather and tide responses may be cached on your device for faster display. This is not an offline map, and the app does not promise one.
You remove all of this by clearing site data for FjordSpot in your browser, or by uninstalling the app.
10. Children
You must be 13 or older to create an account in FjordSpot. Children under 13 may use the app together with a parent or guardian, who creates and owns the account and is responsible for its use. We do not check age at registration. If we become aware that a child under 13 has created an account on their own, we delete the account or restrict the processing.
11. Changes
We may update this policy. In the event of significant changes, we will notify you in the app or on the website before the changes take effect.
