Abyssal Claims Tools

Privacy

Your files stay on your computer

Files you drop on this site are read and checked by your own browser. They are never uploaded, whichever download mode you choose.

Browser mode (default)

In this mode your own browser downloads linked files directly from the data server. Some servers do not allow this; the tool then tells you so and asks you to download the file yourself and drop it here.

In this mode your browser connects to the data server itself, so that server receives your IP address and the address of this site, as with any website you open; its own privacy policy applies.

Server mode

If you choose the Abyssal Claims server, it downloads public files of public records for you, and only from these known data servers:

  • opendap.iopan.pl
  • geo.iopan.pl
  • geonetwork.iopan.pl
  • zenodo.org
  • doi.pangaea.de
  • api.datacite.org
  • data.crosscite.org
  • doi.org
  • repod.icm.edu.pl
  • danebadawcze.uw.edu.pl
  • api.gbif.org
  • api.obis.org

The server keeps no files, no copies and no cache. Its own application log records only the host name, the status code and the size of each download it makes for you, never the full address. That log is kept together with the hosting logs described below.

GBIF and OBIS datasets

If you paste a GBIF or OBIS dataset link or a GBIF DOI, the tool first asks the registry for the address of the dataset’s Darwin Core Archive: api.gbif.org for GBIF, api.obis.org for OBIS. In browser mode your own browser asks them directly, in server mode the Abyssal Claims server asks them for you. Either way the registry sees the dataset identifier you pasted.

The archive itself is then downloaded only from one of these known IPT servers, never from any other address a registry may name; this holds in both modes. A dataset whose archive lies elsewhere is reported by host name, and you can download the archive yourself and drop it here:

  • ipt.vliz.be
  • www.marine.csiro.au
  • hosted-datasets.gbif.org
  • ipt-obis.gbif.us
  • ipt.iobis.org
  • ipt.env.duke.edu
  • ipt.obis.org
  • members.oceantrack.org
  • ipt.inbo.be
  • nzobisipt.niwa.co.nz
  • ipt.biodiversity.aq
  • ipt.sibbr.gov.br
  • ipt.biodiversidad.co
  • ipt.medobis.eu
  • geo.abds.is
  • data.aad.gov.au
  • datasets.obis.org

In browser mode the IPT server receives your IP address and the address of this site, as with any website you open; in server mode it receives the server’s address instead. Their own privacy policies apply. Most of these servers do not allow a browser to download from another site: the tool then says so and offers server mode. Archives up to 100 MB are read; in server mode the server waits up to 90 seconds for an archive. It keeps no copy.

Collection review

A collection run reads metadata records only. DataCite (api.datacite.org) and Zenodo (zenodo.org) are asked by your own browser directly, so they receive your IP address and the address of this site, as with any website you open; their own privacy policies apply. Catalogues are read through the Abyssal Claims server, which passes the catalogue’s answer on to you and stores no catalogue content. Like every request to this site’s server, each catalogue request is logged (catalogue host, status, size), and a request counter per IP address is kept for at most 24 hours to enforce the rate limit and the daily budget. Only these known catalogues are used:

  • emodnet.ec.europa.eu
  • data.marine.ie
  • geonetwork.iopan.pl

The progress of a run and the parts of the fix list are stored only in your browser (IndexedDB), never on our server, until you delete them with “Delete all saved progress” on the collection page or discard the run. Apart from the catalogue requests, nothing about a collection run is sent to our server.

Hosting logs

This site runs on Google Cloud Run (region europe-west1). Like any web host, the platform keeps a standard log of every request to this site: the IP address, the time, the requested address, the browser (user agent) and the referring page. In server mode the requested address is the tool's own download address; the link you asked for is not in it. These logs are deleted automatically after 30 days.

They are used for security, to prevent abuse and for aggregated usage statistics, including which network (for example a university network) requests come from. No profiles of individuals are built from them, and they are not passed on to anyone else or sold.

Map preview

The map preview after a check loads its base map from CARTO (basemaps.cartocdn.com) directly in your browser, so CARTO receives your IP address and the address of this site.

CARTO's privacy notice: carto.com/privacy/

Analytics

Google Analytics 4 is used only if you accept it in the cookie notice. Until then nothing is sent to Google and no analytics cookies are set.

Events carry only counts (how many errors, warnings and notes), the type of input (file, Zenodo, GeoNetwork, DOI, PANGAEA or mixed) and the download mode. They never contain file names, DOIs, links or values from your data. You can change your choice at any time with “Cookie settings” in the footer.

Legal basis

Hosting logs and map tiles: legitimate interest (Art. 6(1)(f) GDPR), that is, the security and running of this site. Google Analytics 4: your consent (Art. 6(1)(a) GDPR).

Contact

The tool is run by Michal Mazurowski, an independent researcher. Privacy questions and requests: m.mazurowski@ai-wall.com

About the project: something-rare.com