On 2006/04/13, at 10:46 PM, Ian Hickson wrote:
In any case this is largely academic. In practice there won't be
that many
resources that you'll be accessing cross-site, especially in the
course of
a single session.
That's a big assumption to make. I have use cases where a large
variety of resources would need to be accessed cross-site.
BTW, would you consider these URIs to have different policies?
http://www.example.com/search?a=b
http://www.example.com/search?c=d
Since in reality the problem is with the response, not the request,
I'm
starting to become of the opinion that there aren't any unsafe
methods.
POST, PUT and DELETE are unsafe; are you suggesting that we redefine
HTTP's concept of safety?
As stated before, I'm not sure the existence of one hole justifies
the
intentional opening of other holes.
It's not "one hole". Most of the Web works this way, always has.
I was referring to the ability to do a POST; obviously GET is
possible through a variety of methods, but that's OK, because it's safe.
Not following you; why should other media types be prohibited?
E.g., why
can't I POST or PUT some JSON or RDF to another site, if it wants
to let
me?
You can. My point is that the only thing that cross-site
XMLHttpRequest
lets you do (other than reading the data that is returned) which
existing
mechanisms don't let you do, is change the Content-Type header (and
other
HTTP headers). So the only vulnerability we need to worry about is
a site
that only accepts data with a particular type (or with particular
headers
set). Any other service is already "vulnerable". And that's the only
reason we're doing this GET-before-POST thing.
Sorry, I'm just an HTTP guy, so I'm a bit slow.
The attack I'm concerned about is an attacker write some XHR code in
a fashion that sends a request that has some side effect on another
server (say, your bank account). XHR introduces a new attack vector
here because it sends the request with your cookies; the user doesn't
have to initiate the interaction.
It's true that it's possible to muck around with script tags and HTML
forms to send an arbitrary POST without interaction (the "one hole"),
but the existence of one accidental attack vector isn't justification
for intentionally creating (and standardising) another bigger one
(not just POST, but other methods as well).
Certain users are concerned that referers will let other sites know
what
they are doing, and so disable Referer headers, sometimes at levels
that
the UA has no control over, for example in proxies.
Also, any request from an HTTPS page to an HTTP page has its Referer
header removed.
Thus we need a way to include the pertinent parts -- the domain and
the
protocol -- in the headers, so that the remote site can make an
educated
guess as to the intent of the first party and decide whether or not to
grant that page access to its data.
Fair enough.
I do wonder how long will it take for the browser preferences and
proxies to catch up; doubtless some people will want this blocked
too. I'm reminded of SOAPAction.
Cheers,
--
Mark Nottingham
[EMAIL PROTECTED]