[cap-talk] origin of the power box

[Sandro Magi]

Kenton Varda wrote:
> On a tangent:  Are power boxes necessary?  It seems to me that if the
> UI provides the ability to drag-and-drop capabilities, there's no need
> for the power box pattern.  Programs can simply present dialog boxes
> asking the user to drag such and such capability into them.

The window system is the powerbox in drag-drop scenarios.

Sandro

> It seems like with the power box approach, you would need some sort of
> namespace for your capabilities in order to allow programs to request
> them.  Then, you'd need (dun dun duuuuun) access control lists to
> decide what capabilities in this namespace the program is allowed to
> access.  I like the drag-and-drop approach since it totally bypasses
> these needs and is user-friendly to boot.
>
> As usual I am not well-versed on the literature, so if stuff has been
> written on this topic, please point me to it.
>
> -Kenton
>
> On 10/9/06, *Marc Stiegler * <marcs at skyhunter.com
> <mailto:marcs at skyhunter.com>> wrote:
>
>     It's been a long time since I read Ping's paper, but I do not
>     remember
>     this paper mentioning a powerbox. I just did a quick search for
>     "powerbox", "ping", and "interaction design", and the top hit is mark
>     seaborn, not ping's paper :-)
>
>     Ping may have described an object that would correctly be
>     described as a
>     powerbox, but I am quite confident that the first use of the term, and
>     the first detailed characterization of the powerbox pattern, appear in
>     the DarpaBrowser technical report, issued in June 2002:
>
>     http://www.combex.com/papers/darpa-report/html/index.html
>
>     So it is even more puzzling, what they did before then. Presumably
>     they
>     had powerboxes, they just didn't know it. This would make the
>     history of
>     the powerbox similar to the history of petnames: over and over
>     again we
>     can see very-nearly-complete petname systems (buddy lists for IM, the
>     newest cell phones that ask to add a contact to the address list when
>     you hang up from talking to a phone number the phone has not seen
>     before, the pgp web of trust), reinvented each time by folks who
>     do not
>     realize that what they are invented has been done before, in a context
>     just barely different enough to hide the relationship.
>
>     Also, powerboxes become important to characterize only when one gets
>     fierce about highly dynamic grants and revocations. The desktop is a
>     place where this dynamism is acute, and without a solution to the
>     dynamism there is no pola on the desktop. Server-side systems, while
>     they benefit from dynamic pola, can stumble along quite well
>     without it.
>     Most of the work on capabilities prior to ping's and our work was
>     server-side, so the powerbox was neither as important nor as visible
>     earlier.
>
>     --marcs
>
>
>
>     Neal H. Walfield wrote:
>     > What is the origin of the power box idea?  Mark Seaborn states on
>     > http://plash.beasts.org/ that
>     >
>     >   The powerbox concept appears to have first been proposed by
>     Ka-Ping
>     >   Yee and Miriam Walker in Interaction Design for End User Security
>     >   (December 2000).
>     >
>     > I find it amazing that this idea was suggested so recently.  How did
>     > earlier capability systems handle run-time access delegation?
>     >
>     > Thanks,
>     > Neal
>     > _______________________________________________
>     > cap-talk mailing list
>     > cap-talk at mail.eros-os.org <mailto:cap-talk at mail.eros-os.org>
>     > http://www.eros-os.org/mailman/listinfo/cap-talk
>     >
>     >
>
>     _______________________________________________
>     cap-talk mailing list
>     cap-talk at mail.eros-os.org <mailto:cap-talk at mail.eros-os.org>
>     http://www.eros-os.org/mailman/listinfo/cap-talk
>
>
> ------------------------------------------------------------------------
>
> _______________________________________________
> cap-talk mailing list
> cap-talk at mail.eros-os.org
> http://www.eros-os.org/mailman/listinfo/cap-talk
>