- From: Henri Sivonen <hsivonen@iki.fi>
- Date: Sat, 28 Sep 2002 15:56:08 +0300
- To: www-html@w3.org
On Tuesday, Sep 3, 2002, at 10:45 Europe/Helsinki, Tim Bagot wrote: > At 2002-09-03T09:28+1000, Mark Stanton wrote:- > > [Henri Sivonen] >>> * Trying to do cache control via <meta> is delusional, because >>> proxies >>> don't see the <meta> tags. >> >> Sorry can you clarify this. Are they unable to see the <meta> tags or >> do >> they simply not bother looking? > > Most caching proxies do not attempt to parse the content of cached > documents, because it is too expensive. Right. And it isn't a defect, either. A HTTP proxy operates at the HTTP level in the protocol stack. It should not peek at the level below it (usually TCP but could be some WAP protocol as well) or at the payload (HTML, XHTML, PNG, etc.). > Such a way already exists, and can be read by the current generation of > proxies: HTTP headers. These also have the advantage that they can be > used > for all files, not just HTML documents. The usual response to this is > that > many free hosting providers do not allow their users to modify HTTP > headers. This is a problem not with HTTP or HTML, but with those > providers. Besides, the situations where an author needs to tune HTTP cache control tend to involve either server-side scripting or high-demand static content. In both cases the author has bigger problems if (s)he can't influence the HTTP headers. -- Henri Sivonen hsivonen@iki.fi http://www.iki.fi/hsivonen/
Received on Saturday, 28 September 2002 08:56:42 UTC