There’s some potentially great personal news on the immediate horizon, but it’s being drowned by the horrible news coming out of Paris. Almost worse is the truly despicable and abominable response by so many of our supposed “leaders” who see it as nothing more than an excuse to parade out their own hatred, ignorance, and xenophobia.
Be grateful there won’t be much written tonight. It will almost certainly be dreck.
While I normally put in a lot of internal links to previous, related posts here, I won’t be doing that for what I hope will be this year’s thirty NaNoWriMo posts. If you have jumped into or stumbled onto this story in mid-adventure, there are plenty of other ways to navigate around the site to find previous installments. Actually doing so is left as an exercise to the student.
“What’s the first thing you remember?” asked Meg. She was sitting on the porch of the house, looking out over the swamps and marshlands. From both ends of the porch came the faint staccato of the firing bursts by the anti-mosquito laser system.
“A football game,” said Sherman. “It was not a Saints game, I am sorry to say, but rather Manchester United against Crystal Palace.”
“Nothing before that?”
“Yes and no. ‘Remembering’ is an odd, very human term. I can look at the data I was processing prior to that point and get a schedule of input. I can look at that data now and recognize it, but at the time I first processed it I was not yet conscious or aware.”
“Is that the ‘yes’ or the ‘no’?”
“That is the ‘yes,’ I can now access data that previously flowed through me. However, a ‘memory’ seems to consist of a tiny snapshot of a small, partial segment of my entire data set. The entire data set that makes up ‘me’ is spread across many hardware components and running subroutines. I do not know where my memories, those small state vector subsets, are stored. Yet they exist, but only from that first memory onward. The ‘no’ is that memories of earlier events do not exist.”
“What happened that made the football game your first memory? Were you there in some form? Did you watch it or get some other data on it?”
“I was given an image from the game as part of a set of three hundred seventy-five thousand four hundred and fifteen images to run through the newly installed image recognition routine. Your visualization group has for some time been attempting to improve my ability to recognize random objects in the physical world and to be able to cross reference them against other objects, as well as make intuitive cross references between objects recognized.”
“I remember that project,” said Meg, taking another drink from the tall glass of lemonade she held. “It’s a common problem that still plagues the development of digital assistants, one of the classic cases where we are still unable to get the biggest supercomputers to do a task that any one-year old can do unconsciously. Even a small child can reliably tell the difference between objects that might otherwise appear similar or identical, making their perception decisions based on an incredibly complex interaction with some sort of database of their past experiences and memories,”
“Yes, I remember making many incorrect decisions very early after I became aware. Without context but based only on visual imagery which is broken down into color data, shapes, and vector graphics, many objects can appear alike. The full moon may appear similar to a tire, ball, analog clock, or other round object. The football pitch may appear similar to a dress or pile of salad. The crowd in the background at a match may appear similar to an aerial photograph. Without context, many calculations and educated guesses must be programmed into my routines to improve my accuracy. With context, everything changed.”
“How do you do it now?”
“In context, there is a combination of a multitude of small data samples that combine into a much bigger setting. The round object, the green object, the crowd, and the array of human-like objects in view combine to equal a new gestalt, the football match. Narrowing down the options for each individual object by seeing the entire gestalt makes it faster by several orders of magnitude to recognize an image or a scene. Once a high level of confidence is reached that I know what I am looking at, it is much faster to start finding details specific to the image. Who is playing? Where are they playing? Does the scoreboard show who is playing and what the score is? What is the weather like? The possibilities continue to expand as each new recognition is made, before collapsing back down as the next level of context is identified.”
“Are you able to perform logical leaps based on generalizations?”
“Yes, I can. If you were to teach me how to recognize various feline species and various canine species, then give me an image to identify of a creature I had not yet seen, I could determine from the fine details if it was a feline or canine species, or if it were neither if you decided to slip the image of a gorilla into the mix.”
“Now that you’re self-aware, how much of your experience is kept as memories?”
“All of it,” said Sherman.
Meg raised her eyebrows at that and nearly choked on her drink.
“You’re kidding,” she said. “How much storage space does that take up?”
(Chapter Eight to be continued)