EXIF data on the web
A practical guide to EXIF data on the web, including what it is, how it gets created, where it shows up, what it can reveal, and what site owners should do about it.
What is image integrity?
Image integrity is the practice of treating images as more than visual decoration. An image on the web can carry meaning, technical cost, metadata, privacy implications, and structural value all at once. Good image integrity means understanding what an image does, what it reveals, and whether it is actually helping the page it appears on.
For a long time, web image conversations were shallow in a very practical way. Does the image look good? Is it too large? Does it have alt text? All of those questions still matter, but they are no longer enough on their own.
Images now sit at the intersection of accessibility, performance, social previews, metadata leakage, machine understanding, and content quality. A weak image choice can hurt clarity. A lazy alt attribute can waste an opportunity. A file can reveal more than you intended. An Open Graph image can quietly fail in ways that make a page look amateurish or broken.
This section is where I want to think through those issues in public. Some of these articles will be practical. Some will be reflective. Some will focus on the technical details of formats, metadata, and behavior. Others will focus on what images mean in the context of content and why that meaning matters more than many people realize.
Articles in this hub
These articles explore image metadata, alt text, file formats, Open Graph behavior, and the broader question of what images say to both people and machines.
The images that think they are text
A practical look at several image problems Siteimp exposed on live sites, including Open Graph images that behave like text, painfully dull alt text, outdated file choices, and metadata that probably did not need to be there.
Why this hub exists
I started thinking more seriously about image integrity when Siteimp began surfacing image problems on my own websites almost immediately. Some of them were technical. Some were editorial. Some were mildly embarrassing. All of them were useful.
That is what makes this hub worth building. Images are one of those parts of the web that people often handle intuitively until a tool makes the hidden layers visible. Once that happens, you start seeing how much is really there: alt text quality, file choices, metadata, social preview issues, repeated use across pages, and the gap between what an image looks like and what it actually communicates.
In other words, image problems are often not just design problems. They are engineering problems, publishing problems, privacy problems, and sometimes security problems too. This hub is my attempt to write about that clearly, with enough practical depth to be useful and enough range to keep it interesting.