Runs in your browser. We never see your photos.
Two numbers, counted since launch. A small nod to 1999.
Keeps the capture date, strips location, converts to JPG at 2048 px long edge, quality 85.
Rare plants get collected. A photo posted with GPS intact is a set of coordinates that anyone can read, and the same photo usually gets reposted to forums, groups and personal sites that do nothing to obscure it. Botanists have watched populations disappear after a well-meaning upload. So the default here is to remove coordinates from every file, in every mode, and you have to choose otherwise on purpose.
iNaturalist reads capture date out of the file and fills in the observation for you, which is why the iNaturalist preset keeps the date and drops everything else. Location is different: you can set it by hand on iNat after uploading, and iNat can obscure it for sensitive taxa. That is the safe path. Strip the coordinates here, then place the observation there.
Phones shoot HEIC by default and a lot of platforms still choke on it. Drop the files here and they come back as JPG, resized to a sensible long edge, with the camera and software fields gone. Nothing is uploaded to convert them; the decoder runs in the page.
ORF, NEF, CR2, CR3, ARW and DNG all decode here. This is the slow part, because a RAW file is large and the decoder loads on demand the first time you use it. Expect a few seconds per file on a laptop. The output is a plain JPG at the quality and size you set, with metadata handled the same way as everything else.
Coarsen rounds the coordinates to roughly 10 km before writing them back. That is county level, not trail level: enough to say which region the plant was in, not enough to walk to it. The before and after are shown as text so you can read exactly what will be written. There is no map pin, on purpose.
Yes. The page loads, and after that the work happens in your browser with no network calls for your files. The only thing sent anywhere is two integers per batch for the counters: how many photos, how many locations. You can watch it in your browser network tab.
After the first load, mostly. The HEIC and RAW decoders download the first time you use them, so the first RAW file of a session needs a connection. Plain JPG, PNG and WebP work with the network off.
Two integers per batch, no identifiers, no cookies. It exists because a running tally is satisfying and because it shows how many locations this has kept off the internet.
RAW files are large and decoding one is real work for your processor. A 25 MB ORF takes a few seconds on a laptop and longer on a phone. The queue tells you which file it is on.
No. Your originals on disk are never touched. This reads them, builds new files, and hands you a zip.