RSS 2.0 Feed

» Welcome Guest Log In :: Register

Pages: (25) < [1] 2 3 4 5 6 ... >   
  Topic: Jerry Don Bauer's Thread, Lather, Rinse, Repeat< Next Oldest | Next Newest >  
GaryGaulin



Posts: 3148
Joined: Oct. 2012

(Permalink) Posted: Dec. 09 2012,20:21   

Quote (The whole truth @ Dec. 09 2012,02:55)
Quote (GaryGaulin @ Dec. 08 2012,23:27)
 
Quote (Doc Bill @ Dec. 08 2012,16:28)
GaryBillyBobDumbFuck wrote:
       
Quote
From what Google found it looks like you wrote the second to the last paragraph, not copy/past from something you found online.


Well of course I wrote it!  Only dumbfucks copy and paste from Wiki and call it "information" or an "argument" or a "rebuttal."  I sourced my information through my subscriptions to the NIH archives and I got some dates from Ancestry.com with some help from friends I have at that company.  Ya see, Gary baby, it's contacts and resources that distinguishes a Real Scientist ™ like me from a miserable dog like you.

Anyway, your model sucks pond water mainly because you haven't even attempted to add any environmental feedback loops which looks simple enough to me to do.  No organism evolves as an island unto itself.  (The Archipelago Principle of Biogeography - look it up.)


Hey, at least the whole truth is helping to provide that, and now you just look ridiculous again!

     
Quote (Doc Bill @ Dec. 08 2012,16:28)
Case in point is research done by S. Baldwin, J. Xu and Cleveland Tyler (NIH Letters, 2006, 2012) on the "Symbiotic Communication in the Distribution of Pinus taeda (Loblolly Pine) in an Extended Rhizome Field."  The trees in question are known as the "Lost Pines" of central Texas because they are genetically nearly identical to loblolly pines that grow in east Texas, Louisiana, Oklahoma and Arkansas.  The "Lost Pines" near Bastrop, Texas are curious because they have had no obvious connection to the separated population for thousands and possibly hundreds of thousands of years.  Baldwin and Xu, however, document a symbiotic relationship with a mushroom rhizome common to south Texas much like the Armillaria ostoyae (honey mushroom) of Oregon that can grow to be hundreds of miles long subterraneanly.  These rhizomes form some of the largest organisms on the planet although a handful of individual rhizome fibers weigh only a few grams.

The rhizomes in question only inhabit loblolly pine forests although this particular rhizome has been traced from the Hill Country of Texas through the San Jacinto river basin and down to southeast Texas where it joins up with the main body.  Following the devastating fires in the Bastrop region a couple of years ago, the researchers noticed a huge increase in rhizome activity not only in the fire affected area but in the eastern pine forests, too, as if the regions stress was being felt or communicated several hundred miles away.

Coincidentally, or not, the eastern pines have dropped an inordinate  amount of seeds this year, as have the live oak trees with acorns - well documented - and these seeds are being carried to the fire-affected area by migrating grackles who flock in this region every year.

Why the increased rhizome activity hundreds of miles apart?  Why increased seed production?  Why have loblolly pines in the Bastrop forest recovered faster than other plant species?  Baldwin and Xu do not propose a mechanism but their research shows a high positive correlation with a 95% confidence interval that an undocumented interspecies symbiosis is occurring not only between species but between phyla.  I'll download a copy of this paper and, copyright restrictions permitting, will make it available.


And to add to that I have this in the Multicellular Intelligence section of the theory:

     
Quote
As though they were a single giant multicellular intelligence fungi may have a form of underground communication system that wires together entire forests.  But to be considered intelligent it would need to meet all four requirements of intelligence, which here has not yet been accomplished but cannot be ruled out.


The fungi network does not have to be fully intelligent to be modeled with the algorithm, but is there in case you find a way to get all four requirements identified in the real thing, which would qualify it as intelligent.

     
Quote (Doc Bill @ Dec. 08 2012,16:28)
But, for your model, you'll need to extend it to multiple bots with interacting parameters and feedback loops, positive and negative, that you can tinker with to simulate stress and reaction.  My thoughts.


As I indicated last reply, that is coming in time, too! But for right now I'm working on the standard methodology part. Charts and such.

Gary, some things to consider:

It still depends on the definition of intelligent or intelligence. You might notice that you used the phrase "fully intelligent", and you often refer to different degrees of intelligence. And what you consider to be intelligent/intelligence isn't necessarily what others consider to be intelligent/intelligence.

Even if (and that's a BIG if) most or all scientists were to someday agree that the words intelligent or intelligence were appropriate for describing the actions/processes of atoms, molecules, cells, etc., that wouldn't even begin to show that this universe or anything in it was/is designed-created by a supernatural designer-creator-god. Intelligence isn't necessarily designed by some allegedly intelligent designer-god anymore than the color red is designed by something red.

I will go further than that by saying: Science does not care who has the best sounding “definition”, what most matters to science is which is the most useful theoretical “model” to use for investigating and experimenting with “intelligence” and "Intelligent Cause". For that very reason the Intelligent Design Lab code is now getting a good going over. It was complete enough for Planet Source Code to experiment with. But testing the model like some here want to will require superseding the last with a new version which also has what some in the forum noticed missing, by not having much for a line chart.

Memory optimization is now much better. Still 25% is unused due to using only 3 of 4 possible SubSystems. The forth SubSystem is already in part coded in. That will give it equal amount (to eye facets) of various taste/touch sensors. I expect the new sense(s) to make it like a whole new critter. The current two lobe design might even end up looking like a zombie in comparison. I’m hoping so, anyway. Having to get a version that visibly obsoletes the last one (real good) at Planet Source Code would be welcomed.

Adding a more detailed line chart here gets more complicated than at first seems. But between rushing out well thought out replies and other things, I’m working on it! Already have the LaTeX code for nice .pdf charts. With the circuit diagram all messed up from the rest working differently after optimization of the core algorithm, it’s a major restructuring. Thankfully the project is becoming easier, not harder. The Lab now saves a file for variables needed to track how it formed new memories. To take care of your needs I will add a system where line charts are drawn from files saved together in a folder that names the chart by folder-name such as “One Lobe and Two Lobe - Comparison 1A” and each line is named by its filename such as “One Lobe run 1”. Also seems to need a (for lack of what else to call it) Monte Carlo feature that seeds the random generator at the start of each lifetime, resulting in a probability distribution of various outcomes for an introduced behavior, as opposed to what one of almost infinite number of individuals would do in response. In Monte Carlo speak:  The approximation is generally poor if only a few grains (runs) are randomly dropped into the whole square (of possibilities). On average, the approximation improves as more grains (runs) are dropped:

If I have Monte Carlo right then this from Wikipedia is predicted to exist in the ratios where my earlier hypothesis (model already has its version of Monte Carlo in it) proves to be true. It doesn’t have to have that exact outcome ratio in the distribution to be methodically similar, but finding that exactly in the data would certainly be one way of proving to be amazingly true:
[quote]The ratio of the two counts is an estimate of the ratio of the two areas, which is ?/4. Multiply the result by 4 to estimate ?.[quote]

I admit to not ever having modeled a “Monte Carlo” but that’s because of sensing something like this might be true. I had plenty else in that direction to work on, that requires a good Intelligence Design Lab, with features you demanded, and more. And not to hold that up, I’ll get back to work on it!

Meanwhile, you can work on letting me know whether I’m closer or away from the Monte Carlo method. If it works for you then I will make it so filenames that start with words “Monte Carlo”, capital letters “MC ” or word like “Dist”ribution but I am not sure what it no-doubt already has for a name. I want to make it whatever makes it easiest for you to know what it is. For me it’s something that’s in the model. I have to name something and exactly what it is does not matter. Figuring out what it needs to can be challenging because needing to be all you to know what you need there. In this case that things can be named will help all of us (and again especially me) figure out how to best explain (to you) how it works, how to use (the new useful ID methodology we are now pioneering).

--------------
The theory of intelligent design holds that certain features of the universe and of living things are best explained by an intelligent cause, not an undirected process such as natural selection.

   
  740 replies since Nov. 21 2012,08:55 < Next Oldest | Next Newest >  

Pages: (25) < [1] 2 3 4 5 6 ... >   


Track this topic Email this topic Print this topic

[ Read the Board Rules ] | [Useful Links] | [Evolving Designs]