On Thursday 11 December 2008 23:00, Daniel Cheng wrote:
> On Fri, Dec 12, 2008 at 4:09 AM, Matthew Toseland
> <[email protected]> wrote:
> > If we're sure that what is connecting to us is a web browser, and the data 
is
> 
> "If we're sure" ...
> But how?
> Cookie? We have cookie disabled.
> HTTP-Agent? All browsers is MSIE and Mozilla...

User-Agent should work fine. If we're not using javascript, we can send a 
progress page with a 1 second refresh on seeing a User-Agent that looks like 
a web browser. If wget etc is configured to spoof a User-Agent, that's not 
our problem.
> 
> HTTP Push with  multipart/x-mixed-replace is designed for this,
> but almost no program support it.
> 
> > not in cache, we should show a loading screen when a user clicks on a 
link.
> > With javascript this could be updated in real time, without it it could 
still
> > be shown with a series of redirects.
> >
> > ADVANTAGES:
> > - Reduce the level of user aggravation by showing them that something is
> > happening.
> > - Possibly reduce the number of browser connections needed and thereby 
reduce
> > the impact of running a non-optimised profile. (We WILL still need a 
custom
> > browser profile however for security reasons).
> > - Look better.
> > - We could probably get rid of the 2MB stream-to-browser file size limit 
if a
> > loading screen is displayed: We could just show the loading screen until 
the
> > data is available and then send the whole lot to the user.
> >
> > https://bugs.freenetproject.org/view.php?id=2781
> >
> > Thoughts? Should this be a priority for 0.8? Does anyone want to implement 
it
> > (I'm pretty busy atm!). Does anyone know Javascript well?

Attachment: pgpAZu0UeXO7T.pgp
Description: PGP signature

_______________________________________________
Devl mailing list
[email protected]
http://emu.freenetproject.org/cgi-bin/mailman/listinfo/devl

Reply via email to