From Bugzilla Helper: User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; de-DE; rv:0.9.2) Gecko/20010726 Netscape6/6.1 BuildID: all version The browser says in all http-requests: User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; de-DE; rv:0.9.2) Gecko/20010726 Netscape6/6.1 Accept-Encoding: gzip,deflate,compress,identity - so it wants to get a file as gzip - the webserver gzip's the file: Content-Encoding: gzip But if the page or file is downloaded, the file is not decompressed. Our customers are not able to download any word file. Reproducible: Always Steps to Reproduce: - Open www.newtron.net/mp/componet - In the navigation menu right klick and choose Save... - Save as a file - Look into it with notepad: not readable - rename it to myfile.html.gz - inflate it (not gunzip) - Look into it --> perfect source code Expected Results: Instead of evaluation the Content-Encoding only in the render engine, it should be evaluated in the connection part (the part who write the accept-encoding) From Netscape 4.x present. Should we ignore the accept-encoding from all mozilla and netscape browsers?

*** This bug has been marked as a duplicate of 51852 ***