W3C home > Mailing lists > Public > ietf-http-wg@w3.org > April to June 2010

Re: 'Age' for web pages

From: Nathan <nathan@webr3.org>
Date: Sat, 12 Jun 2010 14:39:12 +0100
Message-ID: <4C138E00.1060306@webr3.org>
To: Merin Tresa Willy <merintwilly@gmail.com>
CC: ietf-http-wg@w3.org
Merin Tresa Willy wrote:
> Doesn't the 'last-modified' detail actually work only for the static
> pages? For dynamic pages, it gives the then current time, right? So it
> might not prove useful when working with pages containing dynamic
> content?

AFAICT that issue is entirely orthogonal to HTTP, more of a design 
choice in how you choose to build applications, and indeed whether you 
create them in an HTTP friendly way.

For instance, you could create applications which produce 'static' 
files, and every time something changes on a page (a new comment is 
added, an edit is made, etc) the application regenerates the page and 
saves it back to a static file - to everybody looking at the system the 
pages are 'dynamic' - there are really no disadvantages to doing it this 
way, you can handle much more traffic, the application has less work to 
do, it's http friendly, your web server (typically) handles all the 
cache control and setting of headers and so forth. Even if the page 
updates every 10 seconds on a high traffic site, it's still a huge benefit.

I suppose a flip on the question may help, if you (or your system) knows 
that the content of a page hasn't changed since X, why would you want to 
keep generating the same content over and over for every request you 
receive?

Best,

Nathan
Received on Saturday, 12 June 2010 13:40:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Friday, 27 April 2012 06:51:20 GMT