I’m using Brave 1.76 (77) on iPad Air M2 and discovered in access.log, that Brave ask the server not only for visible elements of the static webpage, but also the JPG in meta propery=“og:image”. And that’s on every page of my website. Why?
Make page with og:image property and put this html on your server
Load page with Brave iOS browser
Check access.log of the webserver
What good is something like that? Do you get commissions from mobile operators for data transfers?
I know very well what og:image is for. But I’m not sure you understood my question. OK, once again, for you in more detail: Why does this Brave image load during normal page viewing? The image is nowhere to be seen and will not be used for anything. Other browsers don’t load it when reading the site. And the problem is all the more painful because Brave loads the image unnecessarily via mobile data.
@ArSt they responded by explaining the purpose behind that behavior. The metadata, like og:image, is meant for browsers to access so they can display things like previews or thumbnails. It’s not hidden code or anything sneaky.
What you’re seeing is actually normal. When a browser loads a page, it often requests not only the visible elements but also metadata like og:image. This is especially common on mobile browsers and apps that create link previews, and Brave does that too.
The og:image tag is used to generate rich previews for things like tabs, sharing, or history thumbnails. Brave is likely grabbing that image in case it needs to show it in any of those contexts. It doesn’t mean anything shady is going on, and it definitely doesn’t mean Brave is making money off your data usage.
There’s no commission from mobile data transfers. That’s just not how browser companies operate. Brave’s business model is built around privacy and optional user-controlled ads through the BAT system.
If you’re concerned about bandwidth or server activity, you could experiment with lazy-loading images or adjusting how metadata is handled. Just keep in mind that most modern browsers will still request og:image if it’s included in the HTML.
@ArSt part of what I’m struggling to understand is the issue. Metadata tags are meant to be seen and used by browsers. From my understanding you are inserting the metadata tags and then are upset that it’s being read and used. At least that’s how I’m understanding what you have shared. Am I missing something?
I see you mentioned it doesn’t happen on other browsers, but I’m not sure which ones you’re comparing to or the settings you have on each one. Also unclear is if you have tested on the same device(s).
What I am going to do though is tag @Mattches to see if he might be able to have some better answers. One of the things I also hope he can help determine is if the Web Discovery Project (WDP) is available on iOS. I don’t see the toggle for it like we have on other OS so I’m assuming it’s not. But if it is active at all and I’m just not seeing where the toggle is then that could also explain things.
It is good to use factual arguments in a debate. When you say:
What you’re seeing is actually normal. When a browser loads a page, it often requests not only the visible elements but also metadata like og:image. This is especially common on mobile browsers and apps that create link previews, and Brave does that too.
None of this is true. I’ve checked. Firefox, Safari and Vivaldi don’t load og:image until it is needed. Even desktop version of Brave doesn’t load the og:image without need it. So the conclusion is clear, it must be a bug. And a very serious bug, because if I go through 100 pages and Brave loads 100 images completely unnecessarily, then something is wrong.
I’ve come to accept it as part of the routine that I have to first wade through people who can’t see into the system, can’t verify the information, but are tasked with fending off people who complain just for fun. But this is a serious problem, there is simply a bug in some condition that unnecessarily increases the data transferred without that data being used.
User Agent: AppleWebKit/605.1.15 (KHTML, like Gecko) Version/18.0 Mobile/15E148 Safari/604.1
A Brave Search, with AI enabled, using criteria:
how does webkit load metadata og:image
Result:
WebKit, the open-source web browser engine, does not directly handle loading metadata like og tags.
Instead, when a webpage is loaded, WebKit parses the HTML and extracts the metadata tags, including og, which are then used by social media platforms when the link is shared.
The actual loading and rendering of the og are managed by the social media platform’s algorithms when they fetch the metadata from the webpage.
WebKit’s role is to ensure that the metadata is correctly parsed and available for use by other systems, such as social media platforms, which then use this information to display content appropriately when links are shared.
@ArSt Just to clarify, I wasn’t trying to start a debate. I simply shared what I understood based on my own knowledge and also checked with some other sources, including AI. That said, I know those sources can be limited or flawed, which is why I asked for more information and tagged someone from Brave who might be able to help.