Londoners: PM 'not ruling out' Tube strike ban

Glaucus said:
You don't need a builder the purpose is life, the program evolves by mutations and with iterations get better and thus bigger chance of surviving.
Click to expand...

Where does this equate to a machine? You are creating a situation to refer to a metaphorical concept. The flaw is that you are ascribing a literal perspective to it.

Glaucus said:
We are absolutely pre programmed and if you look at what a computer program is and does we mirror it.
Click to expand...

We do not have a program in the same literal sense as a machine or computer, you are confusing an analogous situation with reality. You are expressing a relationship based on a contextual similarity..it is not a literal comparison.

Glaucus said:
Of course i can have it both ways, you forget as an organism doesn't evolve your lifetime, the species as a whole does. You have not exceeded your programming.
Click to expand...

A machine does not reproduce, it does not replicate its original programming, it has no capability for independent thought or cognitive differentiation, it has no ability of self awareness..it's entire existence is defined by the actions of a creator, it cannot exist outside of it's defined operation or programming. If I had 40, 000 CNC machines, would they evolve as a species? No...and therein lies the metaphor...we are, if we take your analogy to its logical conclusion, a biological machine, however in the literal sense, as a biological machine we are not comparable to the actual artificial machines in reality...it is a metaphorical concept, nothing more.

You are discussing a concept, not a literal truth. And one that has gone a long way from the topic under discussion...perhaps begin a thread on the comparison of Human Cognitive Evolution and Machine Programming, it might be very interesting, but right here, right now..the conversation is getting increasingly out of context and into the realms of semantics, so finally we will have to agree to disagree...unless you start a thread with such a premise, then I'll gladly contribute.
 
Last edited:
So how about machines that can learn and can build themselfs?
Or does that not fit your nice little pigeon boxes.

Same litteral sense? Its a different format, but its litteraly the same. We stil habeeverything that defines a program.
Anyway im out.
You go back to thinking handcrafted is expensive due to orecision and that cnc parts arent as good.
 
Glaucus said:
So how about machines that can learn and can build themselfs?
Or does that not fit your nice little pigeon boxes.
Click to expand...

Machines cannot evolve...a machine that builds copies of itself, builds what a designer has programmed and built it to build..it has no awareness of what it is doing or identity of what it is producing..their learning mechanisms are artificial and only mimic the human learning processes...

Anyway, like I said, begin a thread and I'll happily discuss your ideas further...but as you can see from the "pigeon boxes" comment, this is just turning nasty for no reason.

Glaucus said:
You go back to thinking handcrafted is expensive due to orecision and that cnc parts arent as good.
Click to expand...

Well, that is a misrepresentation of what I have said, so pretty meaningless overall.

Seriously though Glaucus, start a thread at some point because the concept is interesting and I think it deserves some exploration in the right context and setting...:)
 
Castiel said:
their learning mechanisms are artificial and only mimic the human learning processes...
Click to expand...

That's the word I was trying to think of! Thanks Castiel - you've saved me an hour of trying to explain why we don't have machines that learn - we only have machines that emulate learning.
 
For what it's worth, and it is far too late to detail why and I not sure it would be too indepth anyway, but you are not quite on the ticket with how we learn and the potential for machines to learn. I think what the human often offers over a machine is an aesthetic quality and that is the key difference. As we stand we can get programs to fulfil the basic criteria of life, etc but we lack the ability to give them and they have never had the chance to evolve into something that is as multifaceted as the human mind. Therefore, a machine will produce something to do something according to a design - if you wanted it to create randomness into it with a degree of structure and feedback from assessment you could do. What I think would be very different would be to get a machine to make something and weigh the balance of its effectiveness against something so intangible as aesthetics. That is the difference between something produced and something crafted.
 
Thats assuming aesthetics are random, which i would totally disagree with, they are far from intangible. They have a basis on how we feel and think, mostly by pre learning, through peer pressure, media etc.
 
Glaucus said:
Thats assuming aesthetics are random, which i would totally disagree with, they are far from intangible. They have a basis on how we feel and think, mostly by pre learning, through peer pressure, media etc.
Click to expand...

I never said they were random they have well founded structural components some of which are mathematically defined. But the point still stands. You are adding a degree of complexity and the ability to see a total end result from partial information that the human brain is designed to do and machines are ill suited for. Machines have to calculate complexity, non-linear systems are a nightmare for them because of the range of possibilities - the brain however perceives and interprets therefore it matters not the complexity as it is performing a very different type of action.

Think of it like chess: the computer may slaughter the best GM in the endgame where the possibilities can be calculated but at the start when the patterns have to be discerned and not calculated then they perform notably better if they follow a 'book' of established patterns from human experience.
 
Well no, in both its just calculating. You aren't really adding complexity by seeing the end. Its just a bit of planing.
If youn find sweeping curves visually pleasing then you implement that. Its no more than you doing complex equations on the fly, no different to what a computer system can do..

And a chess player is also doing pre determined moves to start with, depending on their experience, so that isn't different either. If a computer has experience it can calculate which is the best move to take to start, if it doesn't you can give it the info from the chess books, no different to how human chess players, take info from the same book, then modify it on their own experiences.

The patterns don't have to be discerned at the start at all, you are simply calculating, which starting strategy you think is the best odds.
 
Hm, well, on the one hand, they should be allowed to strike, although it bothers me that they do so often... Price of trains are annoyingly expensive.

On the other hand, the tube is central to my commute, so this seems like a good idea from a completely selfish perspective.

kd
 
Glaucus said:
Well no, in both its just calculating. You aren't really adding complexity by seeing the end. Its just a bit of planing.
If youn find sweeping curves visually pleasing then you implement that. Its no more than you doing complex equations on the fly, no different to what a computer system can do..

And a chess player is also doing pre determined moves to start with, depending on their experience, so that isn't different either. If a computer has experience it can calculate which is the best move to take to start, if it doesn't you can give it the info from the chess books, no different to how human chess players, take info from the same book, then modify it on their own experiences.

The patterns don't have to be discerned at the start at all, you are simply calculating, which starting strategy you think is the best odds.
Click to expand...

There is a fundamental difference. The chess analogy may have been better for the early midgame though in fairness. The difference is how the information is stored and that answers why the human brain can 'craft' - when you store information on a computer you store it a) by its location and b) its content but when the brain stores it you get a) its location b) its content and c) its relationships.

You can see this quite easily with young children learning their times tables. When you say get to their 7 times table they are not working out 3 7s which is what a computer would do in a millisecond and do it well they are retrieving 7 3s - they see the 3 the 7 and the X and see pattern is also responsible for 7 and 3 and X and then retrieve that information accordingly.

Also it is worth noting again a chess computer is notably poorer if it does not have a book to work from. This is now becoming moot as computers are so powerful but a computer will be tactically supreme however will still look poor against a human strategically. It is also worth noting if the 'book' says to make an incorrect move the computer would blindly follow that failing to see the implications of deviating from a 'good' pattern.
 
Lol, that isn't different at all.

And humans either work times table out, or just have the result pre saved.
This again isn't different. A computer can have a table and read the result or do the maths, its the same.

You just keep putting up scenarios thinking humans are somehow special. The stuff we do is pretty darn basic. The difference is we can do thousands of these simple things at the same time. We have huge parallel processing power.
 
Glaucus said:
Lol, that isn't different at all.

And humans either work times table out, or just have the result pre saved.
This again isn't different. A computer can have a table and read the result or do the maths, its the same.

You just keep putting up scenarios thinking humans are somehow special. The stuff we do is pretty darn basic. The difference is we can do thousands of these simple things at the same time. We have huge parallel processing power.
Click to expand...

Could a machine independently create, evolve and develop the concept of multiplication tables from scratch?

Do machines have independent creativity?
 
Castiel said:
Could a machine independently create, evolve and develop the concept of multiplication tables from scratch?

Do machines have independent creativity?
Click to expand...

At the moment kind of, in the future more than likely. You can give computers a series of numbers and it can figure out the relationships, which is what we do. It all comes down to the base program and hardware, something nature had millions of years to perfect. No need to apply something special to it, with nothing to back it up. Again. This is parallel computing issues, everything we do is fairly basic, but we do thousands of things at the same time.

We have got to the stage they have got low independent thinking.
Teach robots some chairs for example, then show it some more objects and it can decide if its a chair or could be used as a chair.
Same as we do, we simply learn what dimensions, shapes etc and apply it.
 
Last edited:
Glaucus said:
At the moment no, in the future more than likely.

We have got to the stage they have got independent thinking.
Teach robots some chairs for example, then show it some more objects and it can decide if its a char or could be used as a chair.
Same as we do, we simply learn what dimensions, shapes etc and apply it.
Click to expand...

The stage we are at today is a simple mimicry of how a human would reason the use of a chair....it requires a set of coding to enable the robot to choose from a preset number of options..human do not reason like that, so it is not really compatible as such. Turn the chair upside down, or move some of its constituent parts and the Robot would not be able to reason the original use of the chair...unless you specifically programmed it to recognise such, the fundamental difference is that a human could independently reason the use of the chair from very little information whereas a machine would require greater external information the more complex the conundrum became.

The kind of robot that would fulfil your criteria would be an example such as R. Daneel Olivaw in the science fiction novel Caves of Steel....this kind of machine would effectively be indistinguishable from a human and then I would agree with you, but currently that is science fiction. What you are promoting is speculation, not reality.
 
Last edited:
Again your applying something special to a human. Its not mimickery at all, that is exactly the process we used.
You still seem to think we have no base program which then learns. This is no different.

We are taught chairs, we can than see that chairs are general x proportion and apply this.

So no, its no mimicry at all, unless you decide humans are suddenly something special and use magic. We don't, we use the same principles, memory and maths to decide if the portions, shape etc fit our knowledge of items that we have been told are chairs.
 
You must log in or register to reply here.
Back
Top Bottom