Originally posted by: pontifex
Originally posted by: DaveSimmons
This is suspicious, a check for update only needs to retrieve information not send it.
Our company just does an HTTP GET to a static web page.
err...isn't that the same thing? the software sends out a message with the software's current version info to the server checking to see if there is a new version. if there is, it downloads it...
No, we just get the latest version available from the server, there is no reason why our software needs to send its own version.
For example, you're running FooBar 1.1.12, you send an HTTP GET to the server (which does NOT need to include anything beyond a URL for a page on the server) and get back a response saying the latest version is 1.1.16. FooBar decides .16 is higher than .12 and tells the user an update is available.
I'm not saying this is the only way version checks can work, just showing that they can be done sending zero information beyond a static page request like GET
www.foobar.com/folder/version.htm with empty body and no cookies. We have tens of thousands of users doing it now.