Re: topmind: ID is potentially testable

topmind wrote:
neverbetter wrote:
topmind wrote:
neverbetter wrote:

but is anybody claiming
that this is the divider?

Well, put me down as somebody who claims that somebody must actually be
doing something if something is to qualify as science. It's not a
sufficient condition because people are doing a lot of non-science too,
but if you're not doing anything you're surely not doing science
either.

This issue at hand is "testable". Does an "official" research project
need to exist before something is "testable"?

Well, you'd probably have to demonstrate "testable" before you can get
the funding for your research project so no. However, you need specific
hypotheses and predictions to do that.


SETI's hypoth is not "specific". At best, only the detection procedures
are.

Which is more than DNA-ID has.

Plus, do you really want to define "science" in terms of money?

I was not doing that, I just mentioned that you usually need to be
testable before you can be an official research project because those
need funding and you won't normally get that if your testability is in
serious doubt, to illustrate that testability usually comes before
research projects. I suppose there have been some projects that start
by trying to develop technology that would make the primary motivating
hypothesis testable where it wasn't before but I can't think of any off
the top of my head.


SETI started out small also. If there is a
size or paper-work threashold volume, then please tell me what the hell
it is becaus if it is only in your head, nobody can do sh8t with it.

This still applies (although you in particular don't deserve the harse
language, it was a reply to somebody else.)

Can anybody answer what point in SETI's history that SETI became
"testable" or "scientific" or a "hypothesis"? Anybody? And why those
points?

A game of Alias? These concepts are all tied up together and hard to
explain without using each other in the explanation. Once there is a
specific hypothesis that makes precise predictions, it is testable if
we have the technology and the resources to test them, and if this is
done with sufficient methodological rigour and the results interpreted
with a critical eye we can call the enterprise scientific with fairly
good conscience.


SETI is not at the point of testing a good signal. Does this mean they
are not currently a "scientific endevour"?

I don't understand. Do you mean that the signals SETI runs through
screensavers are too low quality to count or that they haven't found
good quality candidate messages?


2) Failing to find artificial signals supports (fails to falsify) the
hypothesis that there are no intelligent agents using radio in the area
searched in the timeframe searched.

Failing to find DNA messages/patterns supports the hypothesis that
there are no intelligent agents fiddling with DNA for the
specimen DNA and pattern types searched.

For whatever that's worth. How many species among the millions of
possibilities and how many patterns among the infinity of possibilities
do we need to search for before giving it up as fruitless?

Ask SETI.

I thought SETI was just looking for one kind of pattern but I could be
wrong.


If they changed it to two, would that boot it out of the definition
park?

No but if they had billions of wavelength patterns they were seriously
considering and then found that "Oh well, there are no messages on this
channel, never mind, there are bound to be some on the following
9807455770033 channels we have lined up for inspection", some might
despair. What is the likelihood of picking the right place to search if
there is an infinite number of equally plausible or implausible
alternatives?

BTW, they have kicked around the idea of looking for laser-based
signals.

Oh no, one more thing to be afraid of: aliens with death rays.



3) There is no scientific hypothesis of intelligent design.

It is very similar to SETI's. Humans broadcast radio waves, so other
intelligences might also. Humans have fiddled with DNA, so....

Yes? So...? Humans eat Big Macs so it's conceivable that other
intelligences might also. So what? This is not a scientific hypothesis
until the 'so what' part is specified.

I am not sure what your point is. It is not "testable" or not "science"
until somebody cares? Again, I feel that "science" is continuous, and
I think you agreed. We have high probability things all the way to low
probability things, and we have things that are high importance to us
all the way down to things that are low or no importance to us.

I did not intend to imply "So what, who cares, it's not even remotely
likely.". I meant it as "So what next? What does it mean to us? What
can we do with this hypothesis? What predictions does it make? How do
we tell if it's true or not?"

Which of these does SETI answer sufficiently in your mind?

Well, they can tell if there are narrowband patterns or not. DNA-ID
can't because any actual tests somebody might have done are off topic.

Look, I think that this is a counterproductive strategy for you. You
want to show that DNA-ID is science. You won't be doing that by showing
that SETI isn't science. I'm the first to admit that SETI is an iffy
proposition and there are many problems with it, but currently it is
ahead of DNA-ID in specificity, parsimony and methodological rigour. If
any problems that SETI has make it unscientific, DNA-ID has them too,
only more so. So, in order to convince me that DNA-ID is science you'd
be better off if you convinced me that SETI is a really good thing, not
that it sucks.

Sorry, too much shorthand.

> We search for life on Mars not because the universe gives a sh8t,
but
because humans give a sh8t.

Well, I for one would like to know. Just curiosity I guess. Or too much
sci-fi.

If it is descrete, and thus there is an official threashold of
probability, then what is it?

What are you talking about? Probability of what? Us giving a s**t? Life
on Mars?

Of intelligent designers/fiddlers. Or of SETI's radio-capable aliens.

I don't think that it's any use to talk about probability estimates
there. We simply don't have enough information to do anything but
guess. Against what should we estimate the probability of intelligent
designers? Not to mention that it depends on the definition. If
humanlike genetic engineering qualifies, the probability is one, we
know that genetic engineers exist. But what's the probability of alien
life? What's the probability of aliens sufficiently like humans to have
the same capabilities? What's the probability of supernatural beings?
Robot civilizations? I don't have a clue and I don't think that anybody
has much that is definite to say about it, except that I expect that
the probability decreases the more conditions you pile up on each
other. The probability of all alien life should be greater than the
probability of alien life with one or two specified features and the
probability of that is probably greater than the probability that alien
life with one or two specified features have done a number of specified
things such as send us prime numbers via radio or genetically engineer
bacteria to spread on earth.

4) Because of 3, there are no predictions that can be tested to falsify
intelligent design.

As much as SETI.

Well, not quite. SETI has defined the parameters of the patterns it
predicts that we'll find if radio-using aliens have directed some
radiowaves our way. The nearest thing to a prediction DNA-ID has is "if
someone inserted some patterns we'll find some patterns" where
'patterns' is undefined and untestable.

So far we have Images, Pi, Primes, and the language algorithm.

These are possibilities, certainly, but are they predictions? Which
theory predicts that aliens insert images in genomes? Which theory
predicts that they insert Pi? Are they based on anything else than
"well, they're something that we humans might come up with"?

That is about all SETI is. Perhaps one can argue that SETI qualifies
because humans are *currently* sending radio waves but DNA-ID is only
something we are capable of doing, but haven't thus far. But that does
not seem like a sufficient divider. SETI may even start looking for
laser signals before we use lasers for space communication.

The analogy with humans is a problematic thing. Aliens may or may not
be using radio, but at least radio waves are a fundamental property of
the universe which they could find out about. But we don't even know
that alien life is DNA-based like ours. If it isn't, how the heck did
they find out about E. coli DNA?

Anything can be an inserted
pattern if we're not more specific about what we're looking for and
why.

Again, if SETI had to turn to content analysis, which it might, would
it stop being "science" at that point? This is the 5th time I've asked
that I think.

This is the second time you've asked me that in response to something I
said that had nothing to do with content analysis. Anything can be an
inserted pattern *during the search* if we don't specify what we're
searching for. Is the whole genome an inserted pattern? Maybe A and T
are inserted patterns and the original had only C and G. If CAG
repeats, is it an inserted pattern?

Some patterns would be very strong evidence, some patterns would be
medium strong evidence, and some patterns would be weak evidence. It is
a matter of whether a lead bar or a penny goes on the "evidence scale".

So we need some way to tell what we're looking for.

5) Because of 4, searching DNA for undefined patterns cannot possibly
support or falsify intelligent design.

Why then is Zach claiming that past searches were good enough to reduce
the idea? You two settle that one before bothering me with it.

Two different things. Searches can falsify the idea of the kinds of
*patterns* that were searched for. The absence of such patterns cannot
falsify *intelligent design*. Even the presence of many patterns might
have other explanations.

Please clarify. SETI not finding anything does not falsify the
existence of aliens. At best it reduces the chances for the areas
searched.

Yes, exactly like SETI. If you understand that I don't see why you
replied to this "searching DNA for undefined patterns cannot possibly
support or falsify intelligent design" saying: "Why then is Zach
claiming that past searches were good enough to reduce the idea?" which
implies that the antecedent of "the idea" is "intelligent design" from
the previous sentence. I wanted to say that I did not think that Zach
claimed that, I thought his claim was that the results of the searches
done in the past reduced the possibility that there are certain kinds
of images inserted in the studied genomes. Intelligent design cannot be
falsified by not finding images in a genome, the idea of images in a
genome can, at least for the specific encodings that the search
procedure can detect.

I am having trouble interpreting this. It is unclear to me whether you
are using "falsify" as a synonym for "test", or "reduce strength of
hypoth". People seem to use them both ways.

While there may be some contexts in which falsifying and testing can be
used interchangeably (when clearly talking about tests with results
that falsify the hypothesis) falsification isn't a full synonym for
testing since testing can also provide positive results and support for
the idea. But I'm not sure why you're asking.


6) Because of 2 and 5, your claim that "Having not found anything so
far is no different than SETI not having found anything so far." is
disproved.

Before it is even possible to discuss what types of patterns might be
found in DNA, you must specify a scientific hypothesis. This
hypothesis must:
A) Be based on empirical data.

1. Humans have fiddled with DNA
2. Some of biology resembles technology

What exactly? And what has that resemblance to do with logos in the
genome?

I am not sure what you are asking.


It's unclear to me which parts of biology resemble technology and how
this resemblance leads to the prediction that there are prime numbers
or pictures in some genomes.

Here are some similarities:

* Capable of movement on its own

There are both technology and lifeforms that aren't.

* Consumes energy that must be replenished for movement to continue

Isn't that rather trivial, a consequence of the way the physical laws
work? Any object, be it animal, plant or mineral, needs energy to start
moving and to continue movement if there are forces that slow it down.
Life or technology isn't going to contradict laws of physics.

* Has sensors that allow it to respond to stimuli

There is some machinery that is rather impervious to external stimuli,
if you don't count cutting off electricity.

* Has pipes or tubes that deliver fluid or fuel

Are you thinking of blood circulation and gastrointestinal system? Lots
of single celled lifeforms do without, although there may be some
microscopic tubular structures for some purpose or another

* Has "wires" or strands that are used to propogate signals from one
part of the "body" to another

Not all machinery has that and microbial life does without a nervous
system.

* Has valves that regulate the flow of liguid, energy, or air

Plenty of technology does without valves to regulate the flow of liquid
or air.

* Produces waste products

Isn't that just a side effect of the need to convert fuel to energy? It
is difficult to make it happen without some waste. Anyway, this is
somewhat different from your other points, since it appears that
they're more "Look, it must have been designed that way because it's so
similar." Our technology is not specifically designed to produce waste
products, it just happens because we don't know how to avoid it. If
there is a divine intelligence behind life, he shouldn't have had that
problem.

* Has "organs" or sealed or semi-sealed chambers to hold fluids, air,
or fuel.

Isn't that more or less the same thing as the valves? It's no use
having valves to regulate the flow of stuff if you don't have separate
compartments for the stuff to flow from or to.

* Has filters to clean or sift product

Some machinery has that, I'm happy to report. In the olden days they
were often quite happy to dump the waste unfiltered in the air, water
or on the ground.

* Wears out over time

That's also rather trivial, isn't it? Everything wears out over time.
Stars die, rocks erode etc. It's sometimes called entropy.

* Logic/info-processing circuits based on many-input-single-output
"nodes" and "strands".
* Etc.

What else would you suggest? What alternatives are there? What would we
see if life wasn't designed like technology and it existed just because
of a series of random cosmic accidents? As far as I can see some of
these are rather trivial, consequences of the facts that the physical
laws work in a certain way in this universe. Anything requires energy
to initiate movement and everything wears out over time. And if you're
to be capable of interacting with your environment and have a
metabolism to convert fuel into energy you need some mechanisms for
sensing stimuli and responding to them and regulating fuel conversion.

As far as prime numbers etc., it is one way to test for ID/IE. As far
as direct engineering, something like logos would be a better fit. Our
technology has logos and graffity in it, per that "silocon zoo" chip
link I often post. Different kinds of creators/fiddlers are more likely
to leave different kinds of patterns. The type of intelligence that
would leave a logo may be different than one that leaves prime numbers
behind. We test for different kinds of patterns to cover more grounds,
to cover more types of creators/fiddlers (types of
motivations/situations they may be in).


Well, here's an avenue for development then: specific hypotheses about
which kinds of fiddlers would tend to leave which kinds of patterns.
What would God leave? What about cyberpunks?

Similarly, some SETI aliens may be sending signals inadvertantly, such
as our radio stations leaking into space. Other may intentionally send
messages as a kind of cosmic "hello". The second would be far more
likely to contain prime numbers than the first.

Possibly the dumb ones like us are sending leakage and the real bright
ones send nothing at all :)

B) Explain why such patterns might be reasonably expected.

Please clarify "reasonably". DNA-ID is a long-shot. Is there a
probabililty threashold? SETI is also a long-shot according to many.
You have to cut down on such wiggly words.

You know what reasonably means. I think you're being a bit hypocritical
here. You're saying that reasonable is an undefined wiggly word, but
yet you are quick to say "no, that's a silly version" yourself if you
don't like some analogy. Well, AFAICS, an unclarified 'silly' is just
as much a wiggly word if its opposite 'reasonable' is.

But YOU are the ones using the "silly" test, not me. I cannot define
your tests for you. You imply there is a threashold of silliness. I was
just asking for specifics. In short, it is your metric, not mine.

IIRC, you introduced the silly season test in response to something I
said using obviously made up examples, trying to make a point that one
assertion not supported by evidence is much like another assertion not
supported by evidence when it comes to science. The point was not so
much that aliens inserting messages in DNA is a silly idea but that we
have no evidence-based reason to think that it has happened, any more
than any of an infinite number of other scenarios which might have
happened but which we have no evidence of, even though some of them are
quite possible and reasonable events in some circumstances.

SETI is sci-fi. Which sci-fi qualifies for this threashold X that we
are trying to define here?

You know, I'm quite confused about the threshold you keep going on
about. It seems to keep shifting. First it's a threshold for estimated
probability, then it's a threshold for specificity of hypotheses, then
it's a threshold for being scientific, then it's a threshold for
silliness, then it's testability, then sci-fi and then something else
again. I'm going to ignore the threshold stuff from now on because I
don't think that there is a single threshold that covers all that and
going on about it is most likely going to be pointless. Some projects
may be low on one scale but high on another. For example, there may be
some things that are quite low probability, even sci-fi like, but
testable, and tested in a proper scientific manner to clarify doubts
about a specific hypothesis. And things may change so that things that
weren't testable are and things that weren't improbable are or vice
versa. Finding aliens might be improbable but narrowband patterns are
testable and any two citizens on the street will probably disagree
about the silliness co-efficient since we do not have enough data to
make even an educated guess of whether it's silly to think that we're
alone or silly to think that we aren't.

In any case, I don't think that silliness is metric, it's more
heuristic, just a rule of thumb. Ideas are evaluated as more or less
silly according to their power to explain things and the things that
they need to postulate to explain things:
-are there observations that are well explained by this idea?
-are there observations that make it look counterfactual?
-is there other evidence that points this way? That is, is the idea a
reasonable continuation of some well-established previous idea or has
it been pulled out of thin air?
-how far fetched is it? That is, how many entities or events that we
have no evidence of does this idea need to explain things? How unlikely
does the existence of this sequence seem? (this is a form of Occam's
razor)
To account for subjective bias let's throw in also the general
nuttiness co-efficient which makes people more likely to think that
good ideas expressed by silly people (or those who are considered
silly) are silly ideas by definition and the familiarity co-efficient
which may make them think that familiar silly ideas are less silly than
previously unknown silly ideas.

Much of it appears to come down to real or percieved probability
estimates. Bigfoot is based primarily on witnesses as a metric. One's
silliness rating for Bigfoot may depend on how reliable they judge
witnesses to be. In other words, does witness testomony put pennies or
lead bricks on the evidence scale?

Well, witnesses are rather unreliable and may be more so if there is
sensational value. We have no way of being absolutely certain that all
eyewitness accounts of Bigfoot are lies or errors but eyewitness
testimony would be more convincing if we could find some objective
evidence to corroborate it. Still, there is no way to falsify the
hypothesis that there is Bigfoot hiding somewhere we just can't observe
him except if we're in the right place when he goes out cruising to
find a girlfriend.

C) Predict exactly what patterns would be found if the hypothesis were
correct.

Exactly? Why does it have to be exact? Do you mean supply explicit
algorithms? What is an example that would make you happy, and please
justify why it is a scientific standard by quoting references. I don't
trust your word. Identity thieves have no integrity in my book.

Predictions do have to be exact enough for us to be able to test them.
It is a scientific standard because testing hypotheses is an integral
part of the scientific method.

Specific algorithms *can* be created to find *candidates*, just like
SETI.

Yes but precise algorithms are different from precise predictions. We
can create the bestest algorithm in the universe to find out if there
are RHCP lyrics in the E. coli genome, but once we've used it and know
the results, we're no wiser than when we started if we don't have
precise predictions that tell us exactly in which circumstances we're
likely to find RHCP lyrics and in which circumstances we're not.

Wouldn't this be a problem for SETI also? The Drake equation is so far
a very lousy predictor.

I'm not very familiar with the Drake equation so won't comment on that.
But if you mean that SETI finding narrowband patterns doesn't tell us
what the aliens who sent them are like you're certainly right.

In fact, Zach's means deviation approach could be such a tool
perhaps. One just has to pick a threashold.

Something to do in the name of science :)


Do you really need references for this? It's not an original idea but
you don't normally have to quote references for something that is
common knowledge.

In this case perhaps we should. You guys have these seamingly weird
arbitrary threasholds and boundaries in your heads with regard to
"testable", "science", and "hypothesis", but have not been able to
articulate where they come from. None of my university science books
laid out specific threasholds. The best I can tell they are continous
metrics or concepts.

Even if they're continuous concepts, don't you think that there are
some things that are clearly science and testable and some that clearly
aren't? If you do you have some sort of category boundary in your own
thinking as well, regardless of exactly where the grey area between the
two may be situated.

Not really. I mentally approach it similar to that probability scale
that I posted a few times before. Maybe below a certain threashold
there are things I feel are "not worth" bothering with, such as Santa
and unicorns. But this is mostly a human time management issue. I
don't want my time or tax money wasted on really really long-shots. But
science truth is not really about our economic choices. It just makes
us decide which truths to persue.

In short, science is a technique, not a ranking.

Ah, but it is my understanding that some truths are not pursuable by
the scientific technique. Long shots and short shots are equally
capable of not being testable by the scientific method.

Anyway, I suppose your tax money is safe. There are so many useful high
probability projects that lack funding that if there some day is more
tax money allocated to science they will probably cover those before
the search for Santa.

I've told you what I think. We can with fairly good conscience speak of
science once there is a specific hypothesis that explains existing
observations and makes specific predictions which we have the means to
test, and if we do the testing in a methodologically rigorous way, and
interpret the results in a critical manner, not too much in love with
our own creativity, and seek to confirm the results or expand on them
by doing further and further testing, and are prepared to discard our
brilliant idea if the results are the opposite of what we hoped for.

If this is too arbitrary for you, I'm not sure how to resolve this.
Maybe it can't be resolved. Would the biblical approach work for you?
That is, you'll know them by their fruits? If something produces
repeatable results, makes predictions that come true in further
experiments which is objectively observable by other people who don't
share the same faith in the grandness of the idea, leads to new
discoveries and practical applications, it's scientific or close enough
and even if it's not, in any case it seems worth people's while because
it's useful to them?

At this point how about lets try to pinpoint why you think SETI is
materially different from DNA-ID such as to disqualify DNA-ID.

How did I guess you'd ask that? No, I'm not aware of any practical
applications that SETI has brought us, unless you count the
screensaver@home innovation. (Was it SETI's idea anyway? That's the
first I ever heard of it.) I think it has great potential, lots of
other fine projects could use the same kind of resource for common
benefit and get cheap computer time if only they could involve the
public. Maybe some already are.

But we're back at what I was talking about before. You can't prove ID
by disproving evolution. You can't prove DNA-ID is scientific by
showing that SETI is bah humbug. It is iffy, has low explanatory power,
is somewhat disconnected from our practical reality and has other
problems but if it means that testing its specific claims can't be
scientific it means that DNA-ID has even less justification to call
itself science at this point. At most, you can convince people that
both SETI and DNA-ID suck. This appears counterproductive if your aim
is to show that DNA-ID could be a science.

Thank you for your patience and thoughtfulness. It has been a pleasure
exchanging ideas with you even if it does not lead to definite
conclusions or agreements. You have a gift for focusing on ideas
instead of your personal feelings or summary judgements about
participants.

Thanks. I've been known to have strong personal feelings about some
topics but possibly this just isn't one of them.

.

Relevant Pages