An idea for Story Tree

The other day I was playing Dead Or Alive 5. I completed the story line in and there are a few characters in this game that I really like. However, the story is really not my taste that I cannot help thinking: “Gosh! I seriously hope they had hired some professional science fiction writers that doesn’t suck. I think I can write a much better story if I have a free weekend!” At this moment an idea came to me: there should be a website for salvaging all those ruined fictions, and continuing all those beloved but concluded fictions, in books, movies, cartoons and games.

Don’t get the wrong idea; I’m absolutely aware there are thousands of websites hosting fan fictions for every popular fiction. For example, at www.fanfiction.net there are over 4,000 fan fictions for Star Wars alone at the moment. However, in my opinion what is missing in these websites is a sincere effort to unify the world view and story lines from different fan fictions, or from different authors. As a result, most fan fictions only extend the original story (or trunk, if you will) a little but never branches off, grows, and becomes a new trunk. Let me explain.

I’ll use the example of the TV show <House>. For those of you hated the abrupt ending of the final season like I did, you might wish to see or write a story where Wilson was just mis-diagnosed, and House continues bullying his associates and saving lives. So, you, Austen,  start writing one or two your own House episodes in a universe where Wilson did not had cancer and House stayed.

Now another hobby writer Twain is thinking about writing his own House story in the PPTH hospital too, but obviously he cannot base his story on the official ending where House has left for good. Normally he will then need to cancel Wilson’s cancer in his story too. But suppose he has read your story and really loved it, and more importantly, his story is fully compatible with yours, he can simply use your story (or “inherit it”, if you will) as the “trunk” where his story “branches off”. This reduces his work and also enriched the story line for both fictions quite a bit.

In the computer world, this type of version branching, inheritance and merging happens all the time, because several code developers need to work on the same project simultaneously. A software called Subversion manages this for you. You can create your own branches of codes away from the trunk, work on it for a while and merge it back whenever you like. Or, you can decide never to merge your code back to the trunk code, if you hate it.

It is easy to see how useful such a version control website/software will be useful for managing fictions. The analogy here is that the original fiction will be the “trunk”. Any authors can create their own story line 1) from the “trunk”; or 2) from another fan writer’s “branch”; or 3) start his very own branches, anywhere in the timeline of trunk or any branches, if he want. Of course the original authors (or branch authors) do not need to accept the new story at all, and this is totally fine because this is  a new branch, not any official extension of the original ones. BUT, if they like it, the two parties can converse and merge the two “branches” of stories in a joint effort.

One can immediately see how this story management structure opens a world of possibilities. Going back to the House example, let’s say another guy Asimov was thinking about implementing the House story too. In his fantasy, Wilson was intellectually transferred into a ipod-sized quantum computer after he passed away from cancer, and since then he becomes the robot friends of House. This creates a bizarre story line that’s compatible with neither Austen nor Twain’s story, but it opens a new crazy Sci-Fi branch that a lot of people would like to follow or work on. Note at this point, we have two absolutely different fiction branches, and three good stories (from Austen, Twain and Asimov). The Subversion website that I’m conceiving will present all these possible stories with a graph, so that any new House story developer can opt to start his story at any point of the three branches. He is supposed to mark his story as “compatible” or “incompatible” with all previous branches, but this is optional.

OK, you’ll now see how this becomes REALLY FUN on the reader side. With several branches of stories, he can choose to 1) read the works of a given authors that he liked, or 2) read officially merged stories; or 3) the best of all, let the algorithm decide for him which story he reads. Because the website is aware of compatibility of different branches (this information can be provided by authors or gleaned from readers’ inputs via machine learning), the website probabilistically follows a new story line every time. This means, you can read House stories at the website many times, and most times you will get completely different stories, which gives you the experience of exploring a universe of possibilities! Moreover, when the website accumulates more information about 1) stories; and 2) reader’s preference, it can try to match your taste and very likely pick the stories you love.

If it is a show, will the show producers be fine with the Story Tree idea, or call it a copyright violation? I cannot decide for them. But, they may like it seeing people writing stories for them. If they like some stories a lot, they can buy the story ( with the copyright, of course) from the author(s). On the other hand, for a smaller game developer team with limited budget for screenwriting, they can start a story trunk on Story Tree and invite people to branch for a reward. At very least, I will probably have a better DOA 6 to play!

Why we cannot live forever？

I come from a country where long period of country-wide stability is a norm in history. Several dynasties in China last longer than 300 years with remarkable levels of economic prosperity. After fulfilling their personal/political ambitions, many emperors sought for elixir that would give them eternal youth, but, good luck with that. It’s an interesting question though, from an evolutionist point: why we cannot live forever?

There is obviously selective advantages for living longer. The length of sexually active period of an animal is generally correlated with the lifespan, which means the longer an animal lives, the more likely he will have a larger number of offsprings. But the longevity has cost too. For example, a robust immune system and good recovering/regeneration capability is probably necessary, and all these consume energy generated from metabolism. Thus instead of pursing a longer lifespan, many organisms choose a different evolutionary path by reproducing more and faster, such as most insects, and rodents. But aside from biology, statistics could also provide some insight about the lifespan of animals in nature.

An important thing to note is that even if an individual can live forever, he probably will not because of the unexpected adverse events, such as famines, floods and deadly diseases. A quick fact is that prior to the modern era, the average lifespan of human beings are below 30 years old, mainly because of the aforementioned factors. With that in mind, we can start the calculation of the expected lifespan given a natural lifespan and the incidence rate of adverse events.

For simplicity, assume the natural lifespan is always L. Further assume that the time until an adverse event is an exponential distribution, with parameter λ. In other words, an individual randomly dies according to a poisson process prior to the age of L, but never lives pass L because we assume it is the limite of natural lifespan. We thus have the expectation of the actual lifespan being:

      (1)

Simple calculus reduces this to:

     (2)

This is the actual average lifespan of an organism in the wild, despite the natural lifespan of L. Let’s put in some real numbers. Given that the pre-history average human lifespan is roughly 30 years old, I estimate for human,  λ equals 1/30 years-1. We can plot the actual average lifespan (y-axis) vs. the natural lifespan (x-axis):

It’s obvious from the plot that the curve plateaued around 70-90 years old, meaning, even if human natural lifespan is beyond 90 years, it has basically no impact on the actual lifespan of pre-history humans. Coincidentally, the modern human lifespan is around this number. This might imply that evolution purposefully designed our natural lifespan, so that as cavemen we don’t die early because of the limit of natural lifespan, and we never grow too old, which is a waste of energy that can be used toward reproduction or whatever.

This story will obviously change in the modern era, because the unexpected death rate reduced thanks to agriculture and modern medicine. Unfortunately, few of us can live pass 90 years old since this was the optimal design for millions of years. Given enough time, we can probably acquire new mutations to live longer.

Let’s get back to the maths for a moment. The maximal value of expression (2) is 1/ λ, meaning that no matter how high the natural lifespan is, the actual average lifespan can never pass 1/ λ. If we are the designer of life, we probably won’t set the natural lifespan parameter too high because it will just be a waste of resources. Since statisticians love the number of 0.95, let’s just see what natural lifespan (L) will give us 95% of the maximal expectancy:

Here L is the natural lifespan and L0 is the average lifespan in the wild. This equation means, when the natural lifespan is three times the average lifespan in the wild, an organism will achieve 95% of the maximum life expectancy. We can define this as efficient (if you will). Interestingly, many animal species seem to satisfy this equation. For example, wolf has a life expectancy of 6-8 years in the wild but their natural life span is 20 years. Feral cat has a reported median (not mean though) age of 4.7 years, and their natural lifespan is around 14 years. Pre-history cavemen generally lives up to 30 years, and the modern human can typically live to 80~90 years old, if he is healthy. So a ratio of 3 seems to be the magical number!

There are probably more factors that can be incorporated into the lifespan model, such as the correlation of reproduction capability/cancer incidence rate to the natural lifespan, but it is amazing how such a simple mathematic model is already starting to work. If I have more time in the future, I’ll revisit this topic.

An analytical review of the Snapshot program by Progressive

Progressive’s snapshot program installs a small device into the customer’s car, and if the driver rarely has a “hard brake” (which they define as more than 7 mph decrease in speed in a second), they will “award” the customer for beng a good driver. Which, I think absolutely is A JOKE.

Suppose I drive with 35mph speed on a urban road. With my speed, if I see a yellow light at the next cross and want to take a full stop without having a “hard brake”, it will take me at least 203 feet to stop. This is assuming I have normal reaction time (1.5s) and decelerate at perfectly 7mph /s.

If I decide to take the chance and pass the cross before the light turns red, assuming the cross is 46 feet in length (four lane street) and the yellow light lasts for 4 second, I need to be within 149 feet of the cross, otherwise I WILL run over the red light.

This means, if I see a yellow light when I’m between 203 feet to 149 feet from the traffic light, then I’m screwed. I need to choose either to run over the red light or be penalized by Progressive for having a “hard brake”.

How frequent is this? Assuming the average green light duration is 30s, then I have 1.5% chance of running into this awkward situation at each cross. (Even if you drive at 30mph, it is still around 1.5% in chance.) So If I drive 10 miles per day on weekdays in a pretty busy district, I WILL on average run into 3 hard brakes per day, unless I compromise my driving safety somehow. The probability that I have zero hard brakes on a typical day is less than 5%. All these are assuming I have perfectly fine driving technique, and makes no mistake anywhere.

As I said, the “7 mph/s” definition for “hard brake” is totally a joke and basically penalize anyone who lives in cities. What is even worse is that it encourages people to take chance and run over red lights. I seriously doubt the legality of the “snapshot” program and suggest it be suspected until proved safe.