g'day,
I'm surprised nobody shouted "get out of here, this is a
technical forum!", but hey there's still time... :-)
Caitlin Bestler wrote:
On 7/30/02, Peter Deutsch wrote:
. . .
A clock, whether embedded or external, is only useful if it
is used by both the sender and all the recepients. Given the
goal of having a single clock, there are indeed many
different ways to achieve the same result.
Actually, I think one of the subtext threads here is that we
need to stop thinking in terms of a single clock for the
system as being one of the goals...
You cannot, however, have two of the devices on an I2C bus
decide that the clock the rest of the bus is using is
defective, and just decide to use a better clock on their
own.
Actually, in the cited example, you most certainly *can*.
I2C permits multimastering on a bus (that is, more than one
device can act as a master, although only one can do so at a
time). To continue to thrash this poor analogy a bit
further, I think it's time to consider that in addition to
permitting multiple clocks on a bus, it time to consider the
possibility that not every device needs to be on the same
bus. All that you need is that the set of devices that need
to communicate together have a mechanism to negotiate and
agree upon a common protocol. Anyone not affected by this
newly defined relationship are then free to ignore it
without harm...
Additionally, I think it is also important to recognize that
domain names are now an integral part of trademarks and have
meaning far beyond translating a name to an IP address.
Actually, this observation is relevant for the operators of
commercial resource location services based upon deployed
technology but is irrelevant for the purposes of a
*technical* discussion about how to best design next
generation directory service technologies. This is clearly
one to take over to the ICANN side of the house. We don't
want to upset out new-found enthusiasm for technical
issues... ;-)
Suppose the IETF were to somehow get the crazy idea to
radically change the entire domain registration system, and
as a result Disney no longer owned "disney.com"?
Does anyone really think the courts would back the IETF?
Errrr, to be a little more precise Disney might own the
trademark rights to "disney.com" in the context of Internet
resource naming in certain jurisdictions. This might be
somehow relevant to operators of global resource discovery
services but what exactly does that have to do with this
discussion? As has been pointed out many, many times the
IETF doesn't operate a resource discovery service. What it
is supposed to do is promote activity to develop
Internet-based technologies. Are you saying that somebody
might sue the IETF to stop development of a technology
because they might be misused to infringe upon their
trademark rights once deployed? Wow, giving in to *that*
kind of self-censorship would surely shut down the IETF
pretty fast.
Remember, protocols don't violate trademarks, people violate
trademarks. When they do so, they can be sued. Meanwhile, I
don't think we're here just to make sure Disney has a good
week on the Internet...
- peterd
--
---------------------------------------------------------------------
Peter Deutsch pdeutsch(_at_)gydig(_dot_)com
Gydig Software
That's it for now. Remember to read chapter 11 on the
implications of quantum mechanic theory for time travel
and be prepared to have been here last week to discuss.
---------------------------------------------------------------------