Tested, dated, and checkable
Guides
Longer pieces that answer a question the tools themselves cannot. Each one is a test I actually ran, on a stated date, with the method written out so you can run it yourself and disagree with me if the result differs.
The Norway problem, and other ways YAML changes your config
Eleven ordinary config snippets through two real parsers. Eight came out different, three of those with no warning from either, and one renamed a key.
What PDF compression actually does to your file
This site’s own compressor run across its three presets: 79% smaller on a scan, larger on every setting for a text PDF, and selectable text gone from all of them.
What your photos reveal, and what removing it actually costs
A phone photo carries your position to about a tenth of a metre, the time, the device and often its serial number. Three ways to remove it, measured: one leaves the picture byte-for-byte untouched.
PNG vs JPG vs WebP: which should you use?
Four test images encoded to PNG, JPEG and WebP in the same browser encoder the tools here use, with the byte sizes: why a photographic PNG is enormous, and what a JPEG-to-PNG round trip really costs.
Are free PDF tools safe?
One test PDF sent to 13 popular online PDF tools, with the network panel open: which ones uploaded the file, and what each privacy policy actually promises about deleting it.