What a Photo File Stores Besides the Picture
You open a photo and see a picture. That is the entire point of the format, and it is also why the rest of the file goes unnoticed.
If you want to look at your own photo instead of reading about it: drop one into the image metadata viewer. It lists every field it finds, right in your browser — no install, and the file never leaves your device.
I built pic2kb, a small set of image tools that run entirely in the browser. Earlier this year I added a metadata viewer to it, mostly because I wanted to answer a simple question for myself: what is actually inside the photos I take? I expected a dozen or so fields. The first photo I dropped in had well over forty, spanning three different standards, and one of them I had never heard of.
That gap between “a picture” and “a file” is worth closing, because the invisible half is not junk. Some of it is genuinely useful, and some of it is the reason a photo can say more about you than you meant it to.
Two things, one file
When a camera or a phone saves a picture, it writes two blocks into the same file:
- The pixels — the image itself, compressed.
- A description of those pixels — how they were captured, by what, when, and where.
The second block is what people mean by metadata. It is written automatically. You never turned a setting on to produce it, and most gallery apps never mention it, which is why it feels like it is not there.
What is actually in there
The fields fall into a handful of groups, and seeing them grouped is what makes the volume make sense.
The device. Camera make and model, the lens, and often the body’s serial number. This is the field that lets a photo identify exactly which camera took it — and if you take many photos with the same body, it lets someone else link them all to one device.
The exposure. Aperture, shutter speed, ISO sensitivity, focal length, whether the flash fired, exposure compensation, metering mode. This is the part photographers care about: it is the record of the decisions made at the moment of capture, and it is why you can look at a shot you like and learn how it was made.
Time. The exact moment, down to the second — and frequently down to fractions of a second, plus a timezone offset. The offset matters more than it looks: it narrows down where you were even when there is no coordinate in the file.
Location. If the camera app had permission to use location, latitude, longitude and altitude are written in, usually accurate to a few metres. Some files also record which direction the lens was pointing.
Editing and rights. The software that last saved the file, the author and copyright fields, the edit history, and in some formats a thumbnail of the image — which may be a copy of the frame from before you cropped it.
Three standards, not one
The reason a single photo can carry so much is that it is not using one container.
- EXIF is the photographic one — device, exposure, time, GPS. It is structured as numbered tags, which is why it is compact but also why it has hard limits on what it can hold.
- XMP is XML-based and far more flexible: edit history, author, copyright, and whatever a particular application decided to record.
- IPTC comes from news and publishing, and it is where human-readable text tends to live — captions, keywords, city and country names, creator and credit lines.
All three can sit in the same file at once. When you see a photo that has both coordinates and a plain-text city name, that is usually why: the coordinate came from EXIF, and the city was written as text in IPTC by whatever software touched the file last.
When there is nothing there
A photo with no metadata is normal, and the reasons are ordinary:
- It was uploaded to a platform that recompresses images. Most large social platforms re-encode on their servers and drop metadata in the process.
- It is a screenshot, a graphic, or an export. Nothing captured it, so there is nothing to record.
- It was cleaned on purpose. Removal deletes the data rather than hiding it.
That last point is worth sitting with. Cleaning a file is not reversible — once the data is gone, the cleaned copy is all you have. Which is a good argument for keeping one untouched original around.
Why it is worth one look
Two reasons pull in different directions, and both are real.
The first is that this data is the most practical way to learn from your own photos. If a shot came out well, the settings that produced it are sitting in the file. That is the reason photographers care.
The second is that the file knows things you did not tell anyone: where you were, what device you used, when exactly, and enough identifiers to connect this photo to others from the same camera. None of that is visible in the picture. That is the reason I ended up building the viewer in the first place — not because I wanted to inspect other people’s images, but because I wanted to know what my own were carrying before I sent them anywhere.
The gap is not a bug in the format. It is just that nothing in the normal flow of using a photo ever shows you the second half of the file.
If you have never looked, the metadata viewer is the fastest way to start — drop a photo in and read the list. If you then want any of it gone rather than read, the metadata remover strips it without re-encoding, so the picture itself comes out unchanged.
If the question is what all of it means once the photo leaves your device, that is the other post: What Your Photos Reveal Without Asking You.