> Thanks for the answer, but cannot agree that the situation is > "unlikely on the real internet". A simple example: a page containing
It is very unusual on the internet to have the result of a POST be Content-Type: image/gif. What normally happens is that the response is text/html, which then contains an img element that causes the browser to GET the image/gif. > quotes and an image of the real-time dynamics of a stock. The data > change continuously, if to reload image, it will be async with text > data. > > And also take into account that "post" is becoming more and more popular > in the net, as well as that sort real-time pages. POST is irrelevant, because the images are accessed with GET, not POST. If the images are tied to the page, either the URL for the image should contain some function of the time, so that the correct version can be regenerated, or the site should generate a temporary image file at the same time as it generates the HTML, and include the URL to that, temporary, file in the HTML. Temporary files are a common tactic for security by obscurity on software download and map sites, to stop people redistributing the link to the underlying resource and bypassing the HTML page and its advertising payload. ; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to [EMAIL PROTECTED]
