- From: Shel Kaphan <sjk@amazon.com>
- Date: Fri, 29 Nov 1996 11:27:58 -0800 (PST)
- To: Jeffrey Mogul <mogul@pa.dec.com>
- Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Jeff, There's another category of cache-busting that you did not mention in the statistics you reported. This is the use of unique URL components, which may be "once-only" URLs, or are at least unique for a single user. These are typically dynamically generated and inserted as components of hyperlinks in HTML documents, so that following these links makes a uniquely identifiable request to the server. This is done not so much to make sure the response is not cached, but to make sure that the response is not shared. (It is of course also done sometimes to identify the client to the server -- much as cookies are sometimes used). This type of response is even less cache-friendly than "ordinary" cache-busting, because the responses may well be cachable, and it is pretty well guaranteed that nobody else will ever request the same URL. Not to beat a dead horse or anything, but the reason people use these techniques is because they are the only way to guarantee some degree of control over the user experience. To really beat a _thoroughly_ dead horse, this is the case because caches and history mechanisms are improperly conflated in most browsers. The "correct" methods of controlling cachability, with HTTP headers, are interpreted by browsers in ways that screw up the user experience as well as controlling the cache. Until/unless this ever gets properly addressed, people will have to use cache-unfriendly workarounds. --Shel
Received on Friday, 29 November 1996 11:32:40 UTC