Re: poly: Re: Why interest rates may stay low

 hanson@econ.berkeley.edu (Robin Hanson) writes:
>OK, this is a start. But you neglect these crucial issues:
>1) Amoritized mass of systems to mine, build, energize, and launch probes.

 I intended to include all these in my Mprobe variable. Do you disagree
with my assumption that 10^20 grams/probe is likely to be the maximum total
mass a design can usefully employ?

>3) It takes time to colonize a new system, making mass useful there.
> Until a few golf-courses, scenic beaches, and five star hotels are built,
> quality of life may be consider unacceptably low :-).

 I'm assuming that molecular assemblers will be good enough to make the
time to accomplish this small compared to the travel time.

>4) Mass and energy at our and nearby solar systems can become a bottleneck
> resource, and so worth more (and priced higher) than mass far away.

 If I understand this objection correctly, you are raising the same doubt
I raised when I said:
>I have doubts about whether it is meaningfull to compare
>value across this kind of singularity.
 Can you think of a better way to compare value between the time before the
first probe launch and the time when much of the galaxy is colonized than
to treat mass as the most stable value?
 I see some similarity between the situations faced by the first wave of
probes launched from earth and those using other solar systems for leading-
edge launches that will produce somewhat comparable bottlenecks throughout
the leading edge.
 I don't forsee much trade between the leading edge and areas substantially
behind, so I don't know how to compare value between regions dominated by
bottlenecks to those with few bottlenecks.

>5) I'm not sure the subjective time enroute is the relevant number. Colonists
> might compare going out to colonize and them coming back here to staying
> here the whole time. We may have space-time interest, comparing the
> relative value of resources at each space-time event to that at some
> reference event. (I modeled this in my paper.)

 People who plan to be here long after the leading edge is gone appear
to have lower discount rates than those trying to stay near the leading edge,
so I expect the latter to be the ones whose demand keeps interest rates
up to launch.

<rant>
If you have something relevant in your "Burning the Cosmic Commons" paper,
I've given up (at least for this week) trying to follow that because the
number of arbitrary one-letter variable names exceeds what I can keep in
my head. How about replacing letters like U with names like Value for which
I already have some relevant mapping? I can understand why people writing on
blackboards or scarce dead trees are tempted to use one-letter variable names,
but in electronic documents, I find them as evil in mathematical analysis as
I do in source code.
</rant>

>> I expect that the ability to throw a few orders of magnitude more reaction
>>mass into a launch than a mass-efficient design requires will allow less
>>efficient designs to reach the target speed earlier, producing some of the
>>interest rate reducing effects you mentioned. But I expect that once the
>>mass used per probe gets up around 10^20 grams, the advantages of using
>>more mass will be out be outweighed by factors such as ...
>
>Consider instead premature investment in the sense of launching probes when
>probe speed is much less than you've assumed, and probe reliability, hardness,
>and efficiency are also much less than I think you've implicitly assumed.
>
>Even assuming people ignored en-route time when considering probe investments,
>what if people started launching probes when they first estimated ROI >~10%/yr?

 A probe that is launched at to reach a peak speed of 0.9999c a year
before the 0.99999c design becomes available will fall behind the leading
edge after about 11000 light years. I've assumed that the designs will be
improving fast enough that most of the galaxy will be colonized by those
that wait for the faster design.
 I confess I have been assuming that the first design(s) that are fast
enough to colonize a significant fraction of the galaxy will also be good
enough that losses enroute won't affect my numbers very much, and I haven't
found a way to analyze the validity of this assumption. It is clear that if
designs with many orders of magnitude lower survival rates are available
much earlier, my argument fails.
 Does your use of the term "efficiency" above refer to something other than
maximizing speed and/or probability of survival to destination?

>What if they started bidding up the price of mass and energy for probes in
>anticipation of future improvements in probe technology, and hence later high

 Started bidding them up when? I thought my numbers implied that people
should have already started to devote large fractions of their income
towards this kind of investment if it is to produce the effects you claim.
If both you and Carl missed this, there must be something wrong with the
way I presented it.

>demand for mass/energy? What if in anticipation of high later mass/energy
>prices people fought for solar system mass sources, and started fighting just
>when, with early near-solar tech, the estimated ROI on such mass grab fighting
>was >~10%/yr? Would >70% of world income then be devoted to these first
>fights? I don't think so, and if not, this fails to be a break-out technology.

 I can't follow what you are trying to say here. Please rephrase.

-- 
------------------------------------------------------------------------
Peter McCluskey  |  pcm@rahul.net  | Has anyone used http://crit.org
http://www.rahul.net/pcm           | to comment on your web pages?