Two guides, and what each one argues

A page of thirty short articles about watermarking would be easy to produce and worth nothing to read. Almost every question people arrive with collapses into one of two: is a visible mark actually protecting anything, and how do I put the same mark on four hundred files without opening four hundred files. Each of those is genuinely long, so each got a piece of its own rather than a paragraph in a list.

Both are argued rather than summarised. The first one spends most of its length on the limits, because the limits are what nobody selling a watermarking tool wants to put in writing. The second is a method you can follow while the queue runs.

Three subjects come up often enough to deserve pages and do not have them, and saying so is cheaper than implying otherwise:

• Marking a scanned document rather than a photograph. The recipe is short enough that it sits as an answer on the front page instead of a guide: turn repeating on, set the lattice to about −30°, drop the opacity to around 20.
• Putting a logo file on properly — how large a mark file needs to be before it stops looking soft at full size, and why a flat JPG logo cannot be made see-through here.
• Whether a mark is worth anything when a picture ends up in a training set. Nobody can answer that honestly yet, which is exactly why it is not written up as though somebody could.

When those are written they will appear on this page. Nothing else will: this site marks pictures, and a guide about something it does not do would be a guide about somewhere else.

• The markerOne picture, the full inspector, and a file back in a few seconds.
• A folder at a timeSet the mark against file one, run it across the rest, get a ZIP and a list.
• AboutWhy this site applies marks and nothing else.