Re: lynx-dev Client side pull, revisited

[Henry Nelson]

> I know that W3C says that one should not RELY on the HTTP-EQUIV="refresh"
> to work, and that one should also always include content in the body of
> a page so that a user does not get stranded, and I even agree that the
> use of this feature to redirect is completely silly.
> 
> However, I am writing an application in which I have user sessions (which
> expire after a certain amount of time). Now it would be very nice for the
> users if their session were not to expire while their browser points to
> the page in question. The only way I can think of achieving this is through
> client side pull. If the page is refreshed (say once every 20 mins), then

I have seen a few worthwhile client-side-pull pages to update graphs, but
in general they're bandwidth wasters if you ask me.  It's hard to believe
there is no way to keep a session alive besides using client-side-pull.

I'm no programmer, but wouldn't your application be more robust if you
wrote it so that it would handle the expiration and dynamic updating.  If
some user can continually, indefinitely and automatically "refresh" their
session, what's the point in having it expire in the first place?  If you
must have the session expire in 20 minutes, seems to me it would be most
polite to inform your users of that fact, and tell them what to do in case
they want to extend their time.  That way you could control how much time,
and how many times, they can extend.

Just a suggestion.

__Henry

; To UNSUBSCRIBE: Send "unsubscribe lynx-dev" to address@hidden