Re: [Iotops] can we create protocols that securely transfer ownership?

'Transferring ownership' - like in selling property to somebody?  A
contract would be needed, and a notary with an electronic signature.

Le 30/10/2020 à 17:30, Michael Richardson a écrit :
> 
> Andrew G. Malis <agmalis@gmail.com> wrote:
>> Now that there seems to be at least a few people on the list, I
>> would like to suggest that this be put into the charter as
>> recommended reading:
> 
>> https://arstechnica.com/gaming/2020/01/unauthorized-bread-a-near-future-tale-of-refugees-and-sinister-iot-appliances/
>
>> 
> Tales of dystopian machine dominated futures abound.

There are also tales of futures dominated by a few oligarchies storing
the data.  It would be these private interests that dominate the world,
via computers, be them Things or information highways.  A sort of
facebooks extended along certain negative extension lines.

In these worlds indeed the person-to-person communication would be
needed to avoid negative situations.  But 'transferring ownership' would
not be desirable: one would not transfer ownership of information from
one's brain to Things and even less let these Things further
transfer ownership to other private interests.  There would be a need of
a protocol to make sure ownership is not transferred, but hardcoded in
silicium.

Or maybe 'securely transfer ownership' would be akin to a delegation of
identity, under the form of securely allowing somebody to act on one's
behalf, a sort of 'delegation of signature'.  Delegate signature but
maintain control, like in delegating to vote.

Alex

  There is an entire
> subarea of sociology that deals with the counter-cyclical nature of
> SF. (SF is happy when times are dark, and SF is dystopian when times
> are good) For those that like to keep up on the latest stupidity,
> given that you probably aren't reading comp.risks anymore, I
> suggest: http://www.firemountain.net/mailman/listinfo/dumpsterfire
> 
> Andrew, I presume that the thing we need to take home from this tale
> of abusive *DRM*  (nothing specific to do with IoT) is that we need
> real transfer of ownership.
> 
> That's why the charter proposes to work on issues that include: -
> factory provisioning of devices - onboarding of devices -
> administrative control of devices - software/firmware upgrades - end
> of life management of devices
> 
> There is a significant tussle between: 1) running the software *you*
> want on *your* devices 2) running the software that external entity X
> says is secure
> 
> Vernor Vinge, in _Rainbows End_, written in 2006, and set in 2025, 
> predicts a world where governments say, "Enough is Enough", and they 
> stop letting anyone other than them decide what software will run on
> personal computing devices, and core Internet routers. (X=government
> above) Sure, you can have as many layers of virtualization as we like
> (the protagonist's teenage granddaughter sets him up with
> win95^Wfvwm95), but they control the turtle at the bottom.
> 
> I coined a term Internet of Øwned Things => IøT. This is the set of
> things where the device can actually be owned by the person who
> bought it.  (vs pwned things, where a hostile owns things)
> 
> My take: if you can't control what software runs on your
> toaster^Wthing, then you don't really øwn it.
> 
> -- ]               Never tell me the odds!                 | ipv6
> mesh networks [ ]   Michael Richardson, Sandelman Software Works
> |    IoT architect   [ ]     mcr@sandelman.ca
> http://www.sandelman.ca/        |   ruby on rails    [
> 
> 
> 
> 
> -- Michael Richardson <mcr+IETF@sandelman.ca>   . o O ( IPv6 IøT
> consulting ) Sandelman Software Works Inc, Ottawa and Worldwide
> 
> 
> 
> 
> 