Head To Head
Log In
Register
U-Know! Forum »
animal rights campaigners jailed
Log In to post a reply

68 messages
Topic View: Flat | Threaded
stray
stray
2057 posts

Edited Jan 25, 2009, 19:14
Re: animal rights campaigners jailed
Jan 25, 2009, 19:12
handofdave wrote:

Come one man! You're applying contemporary limitations in computing power and knowledge to a future that hasn't happened yet.


NO. No I'm not. God this is so hard to get accross to you without resorting to some horrible impenetrable jargon but I'll try.

Okay, I developed an elegant abstraction model. It is capable of modelling system of any complexity, theoretically, I have the proof and the proof reviewed. Though I didn't publish.

To achieve this it has to be truly polymorphic, as in completely flexible. Such as, to use the terms of relational databases:

A record can become a table become a transaction become a report become a relationship etc, etc. From any state/structure to any other state/structure and possibly back again elegantly (the mathematical use of the word elegance here, as in ermm... best practice.. kinda). So ? Here is the unsolvable part of the problem. How do you leverage the model ? how do you, in fact, leverage any model ? The answer is you make a series of informed decisions, but these decisions remain subjective. Sure, you may have identified a very strong causal relationship between any given pieces of data, but you will never be sure if you're right. It only takes that one in a billion odd occasions for that assumption not to be actually true for the whole model to go to shit.

However, as the model is insanely flexible, you can tweak it, capture that exception and move forward. But it is never finished, got it ?

All application of mathematics is in fact subjective, to simplify it.

Coming 'bloody close' frankly imo is really, really not good enough when we're talking about something that could result in somebodies death. Sure, as said, drug trials still kill volunteers, but do we really want to move to a purely systems based trial ? That is unacceptable. Now yes, in a lot of models 'bloody close' is good enough, I wouldn't have had a career at all if that wasn't the case. It isn't acceptable in the case we're citing, particularly as finding the exceptions we need to capture requires conducting highly risky observable experiments on people.

There are plenty of examples in modern physics of theories for which we will never be able to create observal experiments that can prove them, although the maths itself does in fact work. Of course the maths work, maths like any other semiotic, regime of sign, is internally consistent. However just because the math works doesn't make the conclusion fact. Mathematics is actually a collection of internally consistent fields, and quite a lot of them do not actually relate to each other.

It is the application of my model, and other directions (particularly in the social fields) in IT that made me quit the game.

Have I made sense of what I'm actually saying now ?

It has fuck all to do with current computing power, or knowledge of the future. These issues are not going to go away.
Topic Outline:

U-Know! Forum Index