Image and map references in feeds can be resolved in one of several ways by implementors; fundamentally, these are to reference the complete URL to the remote image on our own servers (through an absolute HREF, which includes our server name in the URL to request the content from), or to transparently reverse proxy our servers and use your webserver as a local 'cache' of our remote imagestore.

Due to the size and scope of the events image database, we do not currently offer any direct distribution mechanisms for downloading all of the images associated with all of the content for a given customer's feed; only the two mechanisms above can be used for events.

Please note: Whenever a feed specifies a copyright, and/or credits, you are required to include this information in an alt tag to comply with the copyrights on event images we acquire for use with our content.

Other feed types - such as Columbus, RoughGuide, or LonelyPlanet maps or images, or those of other third-party guides, can be made available on a case-by-case basis. These may also have restrictions on copyrights which are required to appear in that content; your license with the third party content supplier will contain any details needed to comply with the license of that content.

Referencing Content On Our Servers

For event content, images are located at http://xml.whatsonwhen.com/img_bg http://xml.whatsonwhen.com/img_sm where the large-format images (160x150) are located at the img_bg reference listed above, and the (105x99) images are located at img_sm.

Serving Content Via Remote-Proxy (HTTP Accellerator)

Products such as the Apache webserver allow for the configuration of a reverse proxy; this acts like a proxy server in your workplace or ISP, and caches content. However, a normal proxy transparently caches content requested by you from a destination site; a reverse proxy is a cache that acts as a mirror of another website.

By reverse-proxying content, you gain two important advantages: First, you get the peace of mind of having content on your local bandwidth, managed by your local tech team, in an environment that you have direct responsibility for controlling and scaling. Secondly, and more visible to users, the content can appear to come from the same website as that which is serving the HTML.

Many modern browsers feature the ability to "block content from third-party sites"; customers who do not reverse-proxy will find that, because images are located at xml.whatsonwhen.com, this content will be blocked by browsers configured with this enabled. The images are served by us, on a separate domain than that which is responsible for serving your content.

Customers building XML feeds should consider these issues and decide whether reverse-proxying is within their technical ability to deliver and manage, and in their interests.

For further reference:

A range of other commercial and non-commercial software solutions, as well as a range of 'toaster' cache accellerators already provide these features; check your product documentation to see if your existing products and installations offer support for operating in an "HTTP Accellerator" or "reverse proxy" mode.