Evostories rely on ignorance. In the case of information, the only way they can succeed is if one is willing to forget what one already knows. It is easy to get confused when discussing information. There is, after all, more than one kind of information, and information itself is a specific kind of order. One quick litmus test is to ask whether or not the candidate "information" in question could be false. Could it be a forgery, a mistake, or just plain untrue? Those who aren't alert, or are disinclined to pay attention, should probably put off reading this, by the way.
Information is an abstract representation. It can represent a thing (or an action, status, or lots of stuff), but it is never the thing itself. It can be recorded on a thing, but it is not that thing either. Information is almost always found in patterns (sound, light/dark, dots & dashes, 1 & 0's, etc.); but this does not mean all patterns are information, any more than all order is information. The set/subset relationships must be kept in mind. The primary means of confusing folks about information involve confusing these relationships. Just because all information is a form of order, it does not follow that all order is/contains information. Just because information is found in patterns does not mean all patterns contain information. Some waves carry information, but this does not mean all waves are information. Some paper contains information, and to buy into what they say, you have to believe it is impossible for blank paper to exist, because all paper must therefore contain information.
There is a very good reason for information to exist in patterns. Information must somehow stand out from nature, in order to be efficiently conveyed. Were you named the sound the wind makes blowing down your street, you'd hear your name called all the time; how would you know when you were being summoned (assuming folks could pronounce this sound)? Likewise, it is impractical to type in black ink on a black piece of paper. The nature of the paper and the nature of the ink being the same, the message is very difficult to make out. For this reason, and others, nature and information have a basically antithetical relationship. In fact, one of the most common themes in secret communications is to disguise information as something natural, to one extent or another.
There are several kinds of information. Information can be linguistic or non-linguistic. Open or coded. Obvious or subtle. Intentional or unintentional. It can result from direct actions or indirect (a camera used to monitor a space, or even a tripwire). One needs to keep in mind which types of information one is talking about, rather than muddle them all together. In science, "Shannon information" is quantifiable, but need not contain meaning; "Gitt information" is meaningful, but not (yet) quantifiable. These are the two I'm familiar with, and there are others.
It should also be stressed that informatinal entropy is not the same thing as physical entropy. The entropy status of matter/energy which bears a message does not have any direct bearing on the information content thereof. Now the message-bearer will certainly degrade, but this may or may not happen before the message is received, copied, etc. Indeed, the copying process is one of the most likely events to degrade the message. Yet, depending on the method of copying, there may be no thermodynamic impact on the original.
Music is an intersting form of information to consider, especially music without lyrics. It may still contain meaning, but nobody seems to have much success putting it into words. Our capacity to describe music is very limited. In my experience, the only way to find out if I'll like a piece is to actually hear it. It might be argued then, that music isn't information, or that it isn't meaningful. I consider music to contain abstract emotional information. The fact that music always has in intelligent source makes it difficult to categorize as meaningless; and it certainly isn't random (when played correctly).
Properly written words in properly constructed sentences leave no room for doubt. They are always meaningful information. In fact, there are two levels of intelligence obvious in them. There is a convention among intelligent beings as to language, and a second convention as to the symbols used to convey the language. One must be a party to both conventions if one is to understand. Those who speak English, but have not learned to read it, they can't participate. Those who use the same symbols, but do not speak English, they can't read what I write here. Even this isn't cut & dried. "6 + 2 = 8" can be read by folks who don't speak a word of English. What matters is that a shared convention exists.
Such conventions can be built into machines. In such cases, one must know what convention the machine understands if one is to communicate with it. The phonograph record is an interesting case. Machines are used to record and play information. Humans attempting to scratch information with a needle onto the plastic that the machine could use would be facing no small difficulty. But in the end, it is the words and music that are received when the record is played. They're received by humans, so the machinery is primarily just a tool in the end. It does act, to a limited extent, upon the information; but this action is determined by its design parameters.
We begin to understand that there are several forms of information; but we know that the term does not include everything. Otherwise it would be a synonym for 'stuff'. It wouldn't have much value in science. As I think I mentioned, information is always an abstract representation. A painting contains information, as does a photograph. When we look at a picture of an elephant, we don't see the actual elephant. We still obtain information thereabout. Is a mind involved? If a camera took the picture, is a mind involved? If the camera was triggered by a motion sensor, was a mind involved? We could sidetrack onto the issue of minds quite easily, you see. But why?
Actions are also informative. If a boy says he hates apples, but steals all your apples the minute your back is turned, this means the original statement is very unlikely to be true. Actions are easiest to interpret when they contradict or confirm words. It is valid to assume one does things either because one desires to do them or because one must. It's rather hard to make the case that one just up'n chooses to do things one does not desire to do. Of course context is required in any case, before actions can be properly interpreted. Technically, the act of stealing apples isn't information; but it certainly conveys it. The tricky part is that although an act isn't information, almost all acts are the result of intent. But to argue that acts are information, one would then include accidents, and they don't work.
Working from easy cases to the trickier ones, we come to the footprint. Is the footprint information? It's probably unintentional, and it doesn't operate under conventions; but it is an abstract representation of a foot or shoe. This qualifies it as low-grade, unsophisticated information. It is the result of life, but that's not the key. The key is that it's not the foot itself. Another clue is that it can be false.
If a footprint is information, so then might be a meteor impact site. It isn't strictly accurate to say that random natural forces don't create any information whatsoever. But when we compare this type to the others, we see that it is the barest, most minimal form. Indeed, the footprint is generally better because it reveals directly a thing or two about the foot. But a meteor impact site requires a lot of investigation to unravel. It is abstract, and it can be false. I think it is information, although I understand that some maintain a higher standard.
Information of this type is low grade, and generally of less value. The lowest grade of information includes broken twigs resulting from a pedestrian passing through the woods. But how to tell such a twig from the other broken twigs is a lost art, at best. The more information blends in with its surroundings, the less practical use we have for it as a general rule.
When dealing with even this information, it must be detected and interpreted by one who understands it before it is of any use. What does a footprint mean to one who has never seen a foot? A better question might be "what do animals make of tire tracks?" because most people have feet. But in any case, one needs some knowledge in advance of the encounter.
And the obstacle is clearly seen when applied to nonlife. There's plenty of information on most hard drives, but none of it is useful to a remote-controlled aircraft. There's information on a DVD, but try & play it on a record player or in a cassette deck. Information must be intelligible before it can do any good. I still have Atari floppy disks which my PC will not read. Even if it overcame that, it wouldn't execute the programs properly.
When we look at the barest minimum grade and amount of information required for life, we can only laugh at the notion that it arose randomly in nature. The DNA code is a highly efficient language, well-structured, with syntax, with spelling conventions, with multiple messages encoded in the very same text. This is not the result of a meteor impact. It is not a footprint. It's not even on as low a level as an encyclopedia. The quantity and quality of this information is awesome. The machinery that processes it is tricked out to the max.
To assert that life arose based upon the types of information found in nature is about as absurd a thing as one can do. Yes, the erosion patterns on a rock could be considered information of a sort. Did the primitive, prebiotic components have years of college education in geology instructing them how to interpret it? I doubt it. It'd be as useless to them as the rosetta stone would be to me, moreso in fact. I have the capacity to recognize writing as writing, and it is merely improbable that I could figure it out, given time. The "building blocks" have no chance. It is impossible for them to recognize information, let alone make any attempt to investigate. Molecules just aren't that smart, so far as we know, and giving them more time does not help.
One thing we observe about cases of non-life using information: there is no "figuring out". Never. Either the information is properly encoded and can be used or it isn't. Non-living things never figure out how to use inappropriate information.
The types of information found in nature are of no use whatsoever in salvaging the evostory. Many draw the line a little higher and don't even include them, and one might just as well. Even if they could be read, they'd be pretty useless to the aspiring molecules. The luck god(ess), the selection god(ess), and the mutation god(ess) all put together couldn't make so much as a start on the project. Only those intimate with the imagination god(ess) and the grim determination god(ess) could convince themselves such totally unreasonable nonsense might happen.
Whenever this subject comes up, snowflakes are likely to be mentioned. We're told they contain information, but I have my doubts. Nobody's ever read a snowflake, or devised a machine which will do so. The favourite criticism of "Gitt information" is that it isn't quantifiable. I have not seen the information in a snowflake quantified either, so there seems to be a double standard in play.
Of course quantifiability is an arbitrary standard, which excludes much of what we know certainly to be information. This is why it comes up, usually. "Shannon information" is right to quantify because (quite obviously) Shannon was not dealing with generation or understanding of information. Shannon was dealing exclusively with transmitting information by use of machines. Nothing else. To employ his definition for any other purpose is to take things grossly out-of-context. As machines are involved, his definition cannot be used for identification of information. It assumes the information already exists, and is in a form the machine(s) can work with.
The question may arise: Why have I classified some information as lower grade than other information? It's pretty simple. Low grade information has less potential content, and requires more effort to properly interpret. High grade information is easily interpreted, and has the potential to communicate a lot more. Language is high grade, and spoken language I would say is higher grade than written because it's more efficient. A meteor impact requires a lot of work to properly interpret, and cannot tell one very much at all. Even a series of footprints has more potential because strides can be measured, and the potential exists to find the height and speed of the individual who left them.
This post is long enough. For those who may be interested, Werner Gitt's excellent books are available online for free
here. In the beginning was Information is excellent.
There are also some online videos he has made somewhere. I may try to find a link to them also. Even if one has the attention span for reading books, the video lectures are still well worth watching.
