Re: Fwd: Question re: Bad ideas from Microsoft et al...

From: Anders Sandberg (asa@nada.kth.se)
Date: Wed Apr 09 2003 - 16:49:54 MDT

  • Next message: Lee Daniel Crocker: "Re: Trusted Computing (was Bad ideas...)"

    Good post, Hal.

    I guess the real fear should rather be about who owns the operating
    system rather than the TC system. The one who owns the OS owns the
    contents ("Yes, Microsoft is allowed to delete you - didn't you read the
    EULA before clicking?").

    The problem with the TC infrastructure is lock-in and lock-out
    situations. If software relies on TC, it might be unable to interoperate
    or transfer to systems with other TC systems or newer TCs - imagine what
    happens when TC V1.0 (4096 bit keys) is shown to be insecure and it is
    time to migrate to TC V2.0 (16384 bit keys), and people can't transfer
    their files because the software won't let them move to the safer (but
    from the software's point of view, untrustworthy) system. This suggests
    that backwards compatibility will be extremely important, and that may
    lead to lock in effects too ("Yes children, once upon a time there were
    two standards of a little chip. And that is why the galaxy now is
    divided into the Solarian and the Macrostar empires and we can't
    communicate with each other").

    The second problem is that untrusted software will have a disadvantage
    within formal institutions such as companies, governments etc against
    trusted software. It makes a lot of sense to only trust TC systems and
    handle outside systems much more carefully or not at all. But it is
    within the untrusted environment the creative commons lie, it is here
    where much of the bottom up ideas emerge. One can of course take the
    best of this and turn them into trusted systems (I guess a TC linux is
    quite doable, it is just that any changes in the code will not be
    trusted, even if they are entirely allowable), but that will require a
    costly (in time, effort, trust building) code review that will act both
    as a limiting threshold and cause it to leave the creative common.

    To me, these problems doesn't seem insurmountable. They are messy
    legal-engineering-social-philosophical problems, but they can likely be
    solved. Or rather, there are probably sufficiently good solutions that
    we can live with them. The great risk right now is that we get trapped
    into a premature brittle solution that is suboptimal or has a strong
    lock in factor. We need limited prototypes of this to play with and find
    what really can work, not wholesale standards enforced by law or
    monopolies.

    -- 
    -----------------------------------------------------------------------
    Anders Sandberg                                      Towards Ascension!
    asa@nada.kth.se                            http://www.nada.kth.se/~asa/
    GCS/M/S/O d++ -p+ c++++ !l u+ e++ m++ s+/+ n--- h+/* f+ g+ w++ t+ r+ !y
    


    This archive was generated by hypermail 2.1.5 : Wed Apr 09 2003 - 16:55:50 MDT