How long to keep garbage?

[Jymesion]

Posit duplicator technology where a device can:
'read' material inside a chamber, scanning it and recording all
details necessary to reproduce it exactly,
'absorb' material, reducing it to forms easily stored, and
'make' where it uses the stored material and the recorded read to
create a duplicate.

Every society produces garbage. A society with cheap and easy
duplicator technology is likely to produce inordinate amounts.

Logical development is to replace dumpsters with appropriate chambers.
At regular intervals, a truck arrives, interfaces its duplicator with
the chamber, absorbs the material, and takes it to a production
facility.

It is a staple of cop shows that when someone is found murdered,
uniforms have to search all the dumpsters in the area for weapons,
stolen wallets, etc.. A frequent plot complication is that the
dumpsters have been emptied and either garbage trucks or landfills
must be searched. This compromises the value of the evidence as it can
no longer be positively linked to a specific site.

It has happened in real life that police have had to excavate
landfills in their search for evidence or bodies.

Also, people accidentally drop things into wastepaper baskets or throw
away something without knowing it's valuable.

So -- modify procedure so garbage collection trucks read the contents
of a chamber before absorbing the material and store the information
in case a duplicate must be made.

But any type of information storage costs something.

Good practice dictates that a city of 1,000,000 has 200,000
dumpster-emptyings each week. That equates to storing 10,400,000
dumpster/chamber reads each year.

What is the minimum length of time, irrelevant of cost, the reads
should be kept in case they are needed?

Opinions on the effect of cost on how long the reads should be kept
will be appreciated. (Costs might include warehousing, maintenance
(keep-alive voltage), media cost (mainly inventor royalties), standard
recordkeeping, etc.)

[Sea Wasp (Ryk E. Spoor)]

On 4/14/13 3:50 PM, Jymesion wrote:
> Posit duplicator technology where a device can:
> 'read' material inside a chamber, scanning it and recording all
> details necessary to reproduce it exactly,
> 'absorb' material, reducing it to forms easily stored, and
> 'make' where it uses the stored material and the recorded read to
> create a duplicate.
>
> Every society produces garbage. A society with cheap and easy
> duplicator technology is likely to produce inordinate amounts.
>

A society with the type of technology you posit will not have garbage,
as such. The discarded X -- whether it's a broken toy, your own
excrement following that steak you ate yesterday, or a wrecked car --
becomes raw materials for the replicator, which you, the owner, dump
into your own replicator and ask it to produce you a new X, Y, or Z.

With perfect replicator technology, everything's raw materials.


--
Sea Wasp
/^\
;;;
Website: http://www.grandcentralarena.com Blog:
http://seawasp.livejournal.com

[Jymesion]

On Mon, 15 Apr 2013 14:38:43 -0400, "Sea Wasp (Ryk E. Spoor)"
<sea...@sgeinc.invalid.com> wrote:

>On 4/14/13 3:50 PM, Jymesion wrote:
>> Posit duplicator technology where a device can:
>> 'read' material inside a chamber, scanning it and recording all
>> details necessary to reproduce it exactly,
>> 'absorb' material, reducing it to forms easily stored, and
>> 'make' where it uses the stored material and the recorded read to
>> create a duplicate.
>>
>> Every society produces garbage. A society with cheap and easy
>> duplicator technology is likely to produce inordinate amounts.

> A society with the type of technology you posit will not have garbage,
>as such. The discarded X -- whether it's a broken toy, your own
>excrement following that steak you ate yesterday, or a wrecked car --
>becomes raw materials for the replicator, which you, the owner, dump
>into your own replicator and ask it to produce you a new X, Y, or Z.
>
> With perfect replicator technology, everything's raw materials.

That's like saying we should all have nuclear batteries in our
basements (which were promised in the 1950s) so we won't have to pay
electric bills.

Besides the fact they'll probably be huge machines and expensive to
operate, it's likely there'll be a period when they're strictly
controlled so inventors and designers can receive proper royalties.

[Sea Wasp (Ryk E. Spoor)]

On 4/16/13 12:49 AM, Jymesion wrote:
> On Mon, 15 Apr 2013 14:38:43 -0400, "Sea Wasp (Ryk E. Spoor)"
> <sea...@sgeinc.invalid.com> wrote:
>
>> On 4/14/13 3:50 PM, Jymesion wrote:
>>> Posit duplicator technology where a device can:
>>> 'read' material inside a chamber, scanning it and recording all
>>> details necessary to reproduce it exactly,
>>> 'absorb' material, reducing it to forms easily stored, and
>>> 'make' where it uses the stored material and the recorded read to
>>> create a duplicate.
>>>
>>> Every society produces garbage. A society with cheap and easy
>>> duplicator technology is likely to produce inordinate amounts.
>
>> A society with the type of technology you posit will not have garbage,
>> as such. The discarded X -- whether it's a broken toy, your own
>> excrement following that steak you ate yesterday, or a wrecked car --
>> becomes raw materials for the replicator, which you, the owner, dump
>> into your own replicator and ask it to produce you a new X, Y, or Z.
>>
>> With perfect replicator technology, everything's raw materials.
>
> That's like saying we should all have nuclear batteries in our
> basements (which were promised in the 1950s) so we won't have to pay
> electric bills.

There's a *huge* difference -- a couple of them, actually. You posit
something that can *duplicate itself*, which eliminates a lot of the
challenges of getting enough of them made for everyone. There's also
not, as far as your description goes, any indication that there are
inherent hazards of this device (other than accidentally falling into
the raw materials hopper), which isn't the case for nukes.

>
> Besides the fact they'll probably be huge machines and expensive to
> operate, it's likely there'll be a period when they're strictly
> controlled so inventors and designers can receive proper royalties.
>

A period that will last for about 2 seconds, until ONE of the people
involved scans it and duplicates it for someone else.

A replicator technology that can do what you describe can replicate
ITSELF, at which point you've got the ability to perform physically what
we currently do with electronic files. One copy can make some large
number of copies of itself, which can each make more copies of
themselves, and so on. Note that in the parallel world of media piracy,
pirated, SUBTITLED versions of new anime show up WITHIN ONE DAY of the
show being aired, and immediately are duplicated around the world
uncountable numbers of times.

A physical system might take somewhat longer, but unless all examples
of this machine are kept under constant guard, and every product from
them checked like hawks, and no one has the ability to get access to it
alone for any length of time, this technology WILL escape into the wild,
and pretty soon. The only way I can envision such a thing working (in
the real world, anyway) would be to incorporate a nanoassembler, and
that could be scaled down to a pretty small level -- allowing portable
versions that can give you any small thing you want.

Also, you didn't posit that it was too huge to move or used so much
power that it was impractical for any private individual to use. If it's
that large and power-hungry it probably isn't useful as a manufacturing
device for "anything", but only for things that are going to require
more energy and effort to manufacture the conventional way.

[Jymesion]

1) I posited everything necessary for a reasonable answer to the
question posed.

2) I see no point in responding to straw men.

[Gerry Quinn]

+In article <kkjccs$13k$1...@dont-email.me>, sea...@sgeinc.invalid.com
says...

> On 4/16/13 12:49 AM, Jymesion wrote:

> >>>garbage. A society with cheap and easy
> >>> duplicator technology is likely to produce inordinate amounts.
>

> There's a *huge* difference -- a couple of them, actually. You posit
> something that can *duplicate itself*, which eliminates a lot of the
> challenges of getting enough of them made for everyone. There's also
> not, as far as your description goes, any indication that there are
> inherent hazards of this device (other than accidentally falling into
> the raw materials hopper), which isn't the case for nukes.

If the devices have to be large for some reason (but you can make lots
of different things in the duplication chamber at once) it might be
impractical to keep them in houses.

Or maybe they can duplicate everything except the expensive Unobtainium
that makes them work.

I don't know how long they would store garbage data for, though. Which
is good, because if nobody has a clue about how long it should be, then
you can have them store it for just the time that suits your story best!

- Gerry Quinn

[Michael R N Dolbear]

Sea Wasp (Ryk E. Spoor) <sea...@sgeinc.invalid.com> wrote

> There's a *huge* difference -- a couple of them, actually. You posit

> something that can *duplicate itself*, which eliminates a lot of the
> challenges of getting enough of them made for everyone. There's also
> not, as far as your description goes, any indication that there are
> inherent hazards of this device (other than accidentally falling into

> the raw materials hopper), which isn't the case for nukes.

In fact with such a device you could *make* a nuke starting, say, with
sea water and a microgram of U235 which you doubled 30 times.

Of course if the device can transmute, a microgram of plutonium would
also do (which is all Seaborg had when he worked out the chemistry of
the new element so a production plant could be designed).


--
Mike D

[Jymesion]

On Thu, 18 Apr 2013 18:23:08 +0100, Gerry Quinn <ger...@gmail.com>
wrote:

>If the devices have to be large for some reason (but you can make lots
>of different things in the duplication chamber at once) it might be
>impractical to keep them in houses.

There are more reasons than I can relate here . . .

There may be a minimum size (which is the reason we don't have nuclear
firecrackers).

It may take hours, or even days or weeks, to turn a read into a
phantom image, but once it's done, you can make physical duplicates of
it as fast as you can empty the chamber and hit the button again (like
taking hours to download a file, but making multiple copies on your
hard drive takes a couple of minutes).

The create cycle may be dangerous enough to warrant special
shielding/isolation.

The chamber might need to be at an incredibly high vacuum.

The create cycle causes the chambers to have a very short lifespan.

etc .etc .etc.

>Or maybe they can duplicate everything except the expensive Unobtainium
>that makes them work.

If it's a duplicator where the material must be within a chamber, it
obviously can't duplicate itself because of size.

Take for example the problem with registration:
You have to have the accurate position of each particle. Being off by
the slightest amount could mean putting an electron into a nucleus, or
giving neutrons the velocity of electrons.

A few mistakes here and there will probably be no big deal. A few too
many mistakes and you've created a nuclear weapon in full flower.

Ever notice how long it takes a DVD to load? Most of that time is the
unit finding where the tracks are because no DVD will sit in exactly
the position same way twice.

For a duplicator, each 'read' is going to have to be preceded by
something similar -- the read-head is going to have to establish the
boundaries of what's being read and use those points to create the
baseline for measurements.

Unless you want that registration reading to take days, perhaps
months, all of the points the unit uses for that registration are
going to have to be in a single piece of material. No assembly, no
matter how precisely manufactured, could possibly supply the stability
needed.

A single unit is going to expand and contract uniformly.
An assembled unit won't because each seam affects thermal flow.

A single unit is going to transmit vibration uniformly.
An assembled unit won't because each seam is a damper.

etc. etc. etc.

It's going to be hard enough for the read-head to locate the molecule
it uses for the first registration point and calculate where the
second one should be, find it, and then locate the third, etc. when
the points are in a structure acting uniformly.

It'd be nigh impossible to do it quickly, or properly, when they're
flopping around wildly, from the perspective of measuring things in
nanoangstroms.

So, by logic and the laws of physics, a chamber-type duplicator needs
its chamber to be a one-piece unit.

You can't fit a chamber within a chamber of the same size.

This, inadvertently, also provides a measure of security. If you have
a duplicator with a two-cubic-meter chamber, you can create, through
duplication, as many one-cubic-meter duplicators you want, knowing
someone can't buy one and undermine your business by making and
selling duplicates of it.

(If you want to posit a duplicator that doesn't need fixed
registration points, it won't need a chamber, so that's an entirely
different situation.)

>I don't know how long they would store garbage data for, though. Which
>is good, because if nobody has a clue about how long it should be, then
>you can have them store it for just the time that suits your story best!

There's the rub -- some stories might need the evidence destroyed a
week later, other stories might need the detectives to order up
garbage from two years ago. I'd feel more comfortable if there was a
baseline established by someone else.

[Sea Wasp (Ryk E. Spoor)]

My apologies. I responded without fully reading and comprehending your
question and made myself look like an ass.

My actual response to your actual question below:

On 4/14/13 3:50 PM, Jymesion wrote:

Minimum time if we're considering legal evidence consequences --
probably seven years. That's one of the classic "statute of limitations"
lengths of time. I think that's also the "tax returns" limits.

A lot of this depends on how much you think it takes to store the data,
and how much data storage is available. If Moore's Law keeps crankin'
along, storing all your garbage data for a year becomes trivial
eventually; you store it all, forever, for $5 a year.

If you use non-volatile storage you won't need any power to keep the
data stored, so that will save you a lot. Recordkeeping (i.e., which
load was at which house when) could easily be automated and attached to
each stored garbage record as a UID, so that wouldn't be a major problem.


The major problem that occurs to me with the scenario as stated is that
if you can reduce all the stuff in the garbage to data, you can
presumably EDIT the data. You'll need some serious security to prevent
someone from removing evidence from the records, or planting it. "Heh,
just copy this "dead body" file over into the President's garbage for
Monday!"

If it's possible to edit it, you start getting the question of whether
garbage DATA can be used as evidence. Then it's back to more practical
questions of whether you keep it so that Joe Schmoe can call you up and
say "was my paycheck accidentally thrown out in this week's garbage?"

[Jymesion]

On Fri, 19 Apr 2013 18:42:22 -0400, "Sea Wasp (Ryk E. Spoor)"
<sea...@sgeinc.invalid.com> wrote:

>My apologies. I responded without fully reading and comprehending your
>question and made myself look like an ass.

No apologies necessary, and you didn't look like an ass.

I apologize for the snippy reply. My only excuse is that I spent more
than an hour writing a reply to address each point you made, but it
was virtually unreadable (causing a high level of frustration), a
migraine was starting, and I realized our exchange was heading down
the same old road which is futile, often borders on hostility, and has
given me a bad rep. So I deleted everything and wrote something so
terse I couldn't screw up it's meaning.

The 'old road' usually goes like this -- I posit a situation and ask
for help on a specific point. Rather than understanding that I might
know something about the situation I propose and double-checked each
facet I could, people attack the concept of the scene.

I respond with explanations of why their points don't apply, but
obviously I don't write in a comprehensible manner since their reply
ignores what I said, I reply again, etc. etc. etc. ad infinitum ad
nauseam until people say I'm acting like a troll and won't accept any
advice.


Example:
My protagonist is lost and wants to go home. That is her only
motivation.

A secondary character can quickly and safely send the protag home.
Instead of doing that, this character steals clothing from a corpse
and makes the protag wear it, then tells the protag to make a long and
dangerous journey to seek help from a person who can't actually do
anything.

This secondary character is blatantly bigoted (she states that ugly
people are evil). She also likes to instigate trouble (insinuating the
protag murdered a person who died accidentally and taunting the
deceased's only living relative just before sending the protag to
where the relative, now hopping mad, has a great deal of power).

Question: how do I make this secondary character lovable and seemingly
worthy of respect and even awe?


What I'm seeking is someone with the brainstorm: "Have her wear a
beautiful gown, say she's a good witch, and name her Glinda."

What I get often ranges from "it'll never be believable" to "you don't
understand anything about psychology, human interactions, or the
female psyche."

The frustration is so high I'd just quit asking for help if it wasn't
for the fact that there's almost always at least one suggestion which
is a perfect gem, and more than once, someone has had the brainstorm
which solves all my problems.

>On 4/14/13 3:50 PM, Jymesion wrote:
>> What is the minimum length of time, irrelevant of cost, the reads
>> should be kept in case they are needed?

> Minimum time if we're considering legal evidence consequences --
>probably seven years. That's one of the classic "statute of limitations"
>lengths of time. I think that's also the "tax returns" limits.

Thanks! I hadn't actually thought of that angle.

As for taxes, it was a long time ago, but I remember someone being in
trouble because the IRS found errors going back four years, which gave
them the authority to go back another ten.

> A lot of this depends on how much you think it takes to store the data,
>and how much data storage is available. If Moore's Law keeps crankin'
>along, storing all your garbage data for a year becomes trivial
>eventually; you store it all, forever, for $5 a year.

I don't think this could be considered in the same way as normal data
storage. The amount of information is so large it would be like trying
to put the whole Internet on a flash drive.

My thinking on it is this:
You can't just say: an iron atom at location 34,16,26. Without
duplicating the atom's spin, electrons in the proper shell, etc., it
might not interact with neighboring atoms in the same way. A beautiful
crystal of hardened steel might come out of the duplicator as a lumpy
line of dust.

Even if it's mostly the same, it won't be exact. One of the best
things about proper duplication is you can test an item until it
fails, so you know exactly what it can, and can't, do. If even a few
atoms in a piece of steel migrate from one microcrystal to another
because of a coarse duplication, the fracture pattern will change. The
original bolt might withstand 150,000 psi, a duplicate might shatter
at 134,000.

I think, therefore, useful duplication has to be at the quark/lepton
level. This would require taking measurements (x, y, z co-ordinates)
in nanoangstroms. For a duplicator with a 1 cubic meter chamber, that
means three 10^19 numbers, plus an identifier for flavor.

Multiply that by quarks in a proton, multiply that by protons in an
atom, multiply that by the atoms in the chamber.

Complicating matters is that two reads would have to be taken at a
very short interval apart. It is by comparison of the reads that the
particle's vector is determined.

Even with ingenious shortcuts and the best data compression, it's
still an insane amount of information.

I think it'll have to be, at the very least, some sort of holographic
storage.

The media will be essentially free, but I suspect it'll be of
considerable size.