DEV Community

Nawal Shrestha
Nawal Shrestha

Posted on

Browser Based Image Compressor: Where Your Files Go

Browser Based Image Compressor: Where Your Files Go

A client sends you thirty product shots that haven't launched yet, and you need them smaller before they go into the deck. You drag them into the first free converter Google shows you. The progress bar zips across in two seconds, you download the results, and the job is done. Somewhere between the drop and the download, those unreleased photos sat on a server you can't name, in a country you don't know, under terms you didn't read. A browser based image compressor exists to close that gap, and this is how to tell whether the tool you're using is one.

What happens when you upload an image to a converter

Most free converters copy your file to their server, run the encoder there, and hand back the result. The copy can sit in a temp folder or a log, and the site's terms usually allow retention for debugging. You traded a file for a smaller file, and the price was a copy you no longer control.

The retention promise most of these sites make is "deleted after one hour." Maybe it is. The copy still passes through a server, a CDN, an error tracker, or a support ticket when something breaks, and each of those is another place the file lived. One photo is nothing. A photographer uploading a wedding shoot hands over a client's entire day, and a designer uploading unreleased product shots hands over work that isn't public yet.

Free bandwidth and free compute have to come from somewhere. Ads and pro upsells are the honest answers. If a tool has neither, ask what the product is, because the file you uploaded is a candidate.

What a browser based image compressor does differently

A browser based image compressor reads the file with the File API, runs the encoder in a Web Worker inside your tab, and writes the result to your downloads folder. The image never becomes an HTTP request, so no server copy exists to leak. If you cut the network after the page loads, the tool keeps working.

The browser reads the file into memory, decodes it with the built-in decoder or a WebAssembly codec, re-encodes it at your settings, and hands you a blob to download. Your CPU does the heavy lifting, which is why a big batch can warm up a laptop. SnapShrink's free in-browser compressor works this way: drop files in, set a target size, download the results, and nothing uploads. The first 10 images per session are free.

How to check a browser based image compressor in ten seconds

Open DevTools, go to the Network tab, and run a conversion while you watch. If you see a POST request carrying your file, the tool uploaded it. If the network stays quiet while the tool works, the file never left your machine. That one check beats every privacy policy, because policies describe intent and the Network tab shows behavior.

  1. Open the converter, press F12, and click the Network tab.
  2. Clear the log, drop in a photo, and run the conversion.
  3. Look for a POST or PUT request about the size of your file. Click it and the headers show the file going up; the response shows the converted file coming back.
  4. No upload request means no upload. A tool may fetch a small script or a font while it works, but that traffic is kilobytes, not your photo.

Run the check during the conversion, not the page load. The page load always downloads code, and that traffic says nothing about your files.

The metadata you did not mean to share

Phone photos carry EXIF data: GPS coordinates, the camera model, and the date and time the shot was taken. A server-side converter receives all of it, and whether it strips it before returning the file depends on the tool. A local compressor strips EXIF by default on your own machine, so the location data never leaves the room.

The GPS in a phone photo is precise enough to name the street the photo was taken on. Screenshots carry less metadata, but filenames can leak client and project names, so check those too before you upload a folder.

When uploading is fine, and when it is not

Uploading is fine for images you would post publicly anyway: memes, stock photos, your own blog screenshots. It is the wrong call for anything under NDA, client work, ID photos, medical images, or unreleased product shots. The test is simple: would you email this file to a stranger? If not, don't hand it to a server either.

FAQ

Does an in-browser compressor need the internet?

It needs the internet once, to load the page. After that the work happens in your tab, and you can prove it by going offline in DevTools and converting anyway. The page, the code, and your files are all already on your machine.

Is "deleted after one hour" good enough?

Better than kept forever, worse than never uploaded. The promise is only as good as the company's deletion jobs, logs, and backups. For files you'd rather not explain, "never left my machine" beats any retention policy.

Do all browser tools really process locally?

No. Some pages run the interface in your browser and still POST the file for the actual encode. The Network tab check above is the only claim that matters, and it takes ten seconds.

Those thirty product shots from the start of this article went through a free in-browser compressor: dropped in, shrunk, downloaded, and never uploaded. The deck shipped on time, and the photos stayed on the laptop until launch day. Next time a converter asks for your files, spend ten seconds in the Network tab first. If it stays quiet, compress away.

Top comments (0)