Here is an interesting idea from someone on Get Satisfaction for one
way this feature could be implemented.

http://getsatisfaction.com/mozilla/topics/command_subscription_page_feature_to_warn_users_to_install_ubiquity


On May 8, 4:46 pm, Blair McBride <[email protected]> wrote:
> Great info and links - thanks!
>
> I think this is an important topic that needs to be considered when
> adding code that somehow touches content-space. Thankfully, there have
> been some really smart people looking into this already, so Ubiquity
> (and other addons) can learn from previous mistakes and research.
>
> It also shows that while these type of information leaks are getting
> fixed on the Firefox platform (eg, chrome:// URLs are no longer
> accessable unless whitelisted), its often up to extension authors to do
> things sensibly and with forethought.
>
> Whether or not this type of leak can be used malliciously or not is
> beside the point - if its not explicitly wanted (and expected) by the
> user, then its a security bug that needs fixing.
>
> - Blair
>
> On 8/5/09 4:30 PM, esquifit wrote:
>
> > There was some talk about this topic some time ago, see the links
> > below.  I don't know how vulnerable Firefox is these days.  The same
> > question arose repeatedly in the context of the Greasemonkey extension.
> > The GM developers eventually implemented some protection measures that
> > (I think) are now somehow handled by Firefox itself.   Anthony
> > Lieuallen's Karma Blocker extension [1] was also very helpful against
> > this vulnerability.
>
> > [1]https://addons.mozilla.org/en-US/firefox/addon/5230
>
> > There were essentially two approaches:
> > 1) When a message bar is displayed as a result of a script being
> > installed/discovered, the page can detect the vertical displacement
> > caused by the bar. For example a page could include the <link> tag
> > pointing to some (possible non existent) ubiuity script and check for
> > the vertical shift caused by the Ubiquity bar prompting for installation.
> > 2) Including a extension's chrome:// image resource from the web page.
> > Depending on whether the extension was installed or not, the included
> > image would have length 0 or greater than 0 and would thus affect the
> > width of some other element in the page.  This could be measured by
> > javascript code. I think newer versions of Firefox disallow inclusion of
> > chrome:// images from web content.
>
> > Here some interesting links:
>
> > Detecting FireFox Extentions ha.ckers.org <http://ha.ckers.org> web
> > application security lab
> >http://ha.ckers.org/blog/20060823/detecting-firefox-extentions/
>
> > Jeremiah Grossman: I know what you've got (Firefox Extensions)
> >http://jeremiahgrossman.blogspot.com/2006/08/i-know-what-youve-got-fi...
>
> > Greasemonkey Detect
> >http://wearehugh.com/public/2006/07/detect-greasemonkey.html
>
> > Greasemonkey Detectable? - greasemonkey-users | Grups de Google
> >http://groups.google.com/group/greasemonkey-users/browse_thread/threa...
> > <http://groups.google.com/group/greasemonkey-users/browse_thread/threa...>
>
> > On Sat, May 9, 2009 at 1:05 AM, Blair McBride <[email protected]
> > <mailto:[email protected]>> wrote:
>
> >     This would be a security bug if it were possible - it should never be
> >     possible for web content to detect which extensions a user is running.
>
> >     - Blair
>
> >     On 4/5/09 1:42 PM, Alphawolf wrote:
>
> >      > Hey there,
>
> >      > is it possible to check with javascript if Ubiquity is installed in a
> >      > user's Firefox? I'd like to display some install instructions to
> >     those
> >      > only who have it installed already.
>
> >      > Regards,
> >      > Oliver
--~--~---------~--~----~------------~-------~--~----~
You received this message because you are subscribed to the Google Groups 
"ubiquity-firefox" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to 
[email protected]
For more options, visit this group at 
http://groups.google.com/group/ubiquity-firefox?hl=en
-~----------~----~----~----~------~----~------~--~---

Reply via email to