I think it would also be a great feature if someone could enter the URL where a file can be downloaded from on another site, instead of only being able to upload the file to the webserver

What if you want to point users at a 700MB ISO image that's on an FTP site but you don't want to mirror the ISO image on your webserver

Comments

Anonymous’s picture

(not verified) CreditAttribution: commented

This would be really useful for a number of other reasons as well. First, for large files, a different non-threaded server (such as thttpd) could be used to take the load off Apache and free up some memory. There's no point in using a 4Meg Apache process to send a huge file to a user on a slow link for days on end!

In addition, this feature is important for sites hosting large media files. In many cases those files are served from dedicated media servers that should be referenced via a URL.