--- strk <[EMAIL PROTECTED]> skrev:
> On Mon, Nov 20, 2006 at 03:20:36PM +0100, Tomas Groth wrote:
>
> > > So, rationale is:
> > > - use 'whitelist' if you want to DENY by default
> > > - use 'blacklist' if you want to ACCEPT by default
> >
> > This seems logical to me, so fine by me :)
> >
> > I was thinking that maybe we should also support regular expressions (or
> > similar) when defining black/whitelist, so that it is easier to block
> > communication to standard server name such as
> > "violating.your.privacy.[a-z]*.com"...
>
> The manual page talks about "domains" rather then hostnames, meaning
> that we should consider the entries as regexp in the form .*.name.
> This would make presence of *bo&h* whitelist/blacklist meaningful.
> Example:
>
> blacklist spammers.com
> whitelist special.spammers.com
>
> Any use of regexp-like strings will require us to implement the concept
> of a *best matching* line, so that the 'stricter' mathcing is used.
> In the above example, www.special.spammers.com would be allowed, altogut
> matching both blacklist and whitelist.
>
> > Maybe an option to forbid connections to adresses not on the main sites
> should
> > be implemented as well.
>
> Could you expand this ?
>
The idea of this is based on normal web-ads, which uses cookies to identify
you. This can in many cases be blocked by disabeling cookies from "foreign"
sites. Example: You're browsing www.site1.com, which contains an ads which is
fetched from www.web-ads.com/ad1.html, and that code tries to make a cookie on
your pc, but because it's not on www.site1.com, it can be blocked. I think this
is a option in some browsers, but i'm not sure if this is actually needed in
gnash...
cheers,
Tomas
__________________________________________________________
Alt i én. Få Yahoo! Mail med adressekartotek, kalender og
notesblok. http://dk.mail.yahoo.com
_______________________________________________
Gnash-dev mailing list
[email protected]
http://lists.gnu.org/mailman/listinfo/gnash-dev
