Tax in Australia

A robot can not make decisions. Think about how many decision need to be made to repair the engine bed suggestion.

 

Why can't it? Just what unique, complex decisions need to be made that can not be reduced into multiple and simpler forms? (And can't be made by a shop foreman who is overseeing a large number or robots, and gets prompted for control input if a machine becomes unsure what to do?)

 

Grind or sand , what grade, what size to get into the corner, when is grinding complete, Is it prepared properly.

Clean dust preparation, how?. How to continue, what resin what mixture, what size matting.

How could a robot roll out air bubbles.

Oh come on, no robot can do that without some one standing at the side of it driving, then what would be the point?

I don't think I have started to itemize the decisions required.

All that for one job.

 

How do YOU pick what grade of sandpaper/grinding wheel to use? You look at the project, guess at how much material you need to remove, and use a suitable grade for the job. If you have a lot to remove quickly, you use a coarser grade. If there is a little and you want a fine finish on the job you use a fine grade.

Where in that process do you see something hard for a computer?
IF amount to remove is X
__Use Grade A
IF amount to remove is Y
__Use Grade B

How do you pick how much to remove? Image processing. Smoothing something is easy, you use finer and finer paper till it is smooth, and smoothness is simple statistical analysis on the variations in height. (Something a computer can 'understand' by 3D scanning, which is only getting better with more research.)

How do you clean up dust? You vacuum-brush till no 'dust' is left. Dust being something easily changed in an image, so you let the robot go over it till it looks basically identical to the last time it did it, then you let it go over it again to be sure. (Its a robot, you're not paying it over time. A good saying from computer sciences: You can waste your time, or the computer's time, which would you rather?)

How do YOU pick your resin mixture? How many types of resin do you have in your shop? What are their different properties? How do you know these things? These are simple questions answered in a Work Order Form: "Grind section X, use Y matting type and Z resin." The same way you would tell a junior employee to do a job.

And come on, you really think it is hard to tell a robot how to mix resin? Mix X parts of A to Y parts of B, and add Z of C. A computer can calculate the numbers needed to do 1oz of resin to 2000 gallons of it, in 1oz increments, in the time it would take you to read the label.

How do YOU roll out air bubbles? Personally I start at the center of a sheet, and use a suitable tool and roll to the edge until all the little white bubbles are gone. Are you telling me that something that can be programmed to detect faces tilted at any angle can't be designed to detect a different in colour, the location of these colour differences, and the distance to a pre-known edge?

No robot does it currently, but 20 years ago a robot couldn't look at a table and pick a red ball from a blue ball. Most of the details required to do a robot like the one described above already exists, however they have not yet been combined into a single package with the modifications needed to process the data.

Then keep going, step back and look at any task you do and break it down into small steps. How do you type something at your computer?

1. You need some way to know where your computer is.
2. You need to figure out where you are.
3. You need to begin finding a path between you and the computer
4. Begin moving toward the computer, keep in mind where you've already been, and see step 3.
4a: Raise your right or left foot, but not both at the same time, while balancing on the other.
4b: move your right foot forward to a clear looking area of floor
4c: place right foot down, and begin shifting weight to it, lifting left foot up and forward. place down in a clear spot, see step 4b.
5. Position self in front of computer (Possibly sit.)
6. identify keyboard (searching for a rectangular object with projecting squarish objects on it, often of a different colour, usually with contrasting markings that correspond to letters of the alphabet.)
7. Move and position hands above the home row of the keyboard.
8. Begin pressing down on the keys in desired order to press keys you want.


And each one of those steps is broken down into smaller and smaller steps, and many of which are 'reusable' for other functions. What is the difference between walking to your computer, and walking to your fridge? What is the difference between identifying your computer and your fridge? What is the difference between placing your hands over the keyboard, and opening the door to the fridge? What is the difference between moving the hands/fingers to type something, and moving them to grasp a beer?

Robotics: How do you eat an elephant? One bite at a time.

How do you build a pyramid? One brick at a time.

 

First case scenario:
You have a job that pays you $180,000 a year yet you have a mortgage and 5 kids so you take a second job at the service station down the road on weekends. Pay is $200 per weekend x 52 = 10400 dollars gross
Joe Blow is on a part time job and needs more money, his first job earns him $20,00 a year he gets the same job at the same petrol station and will get also 10,400 a year.
However you will be netting $5,720 and Joe Blow will nett $8,840 after tax
You get punished for earning more than Joe Blow even if you do the exact same job.

Second case with flat rate tax (assuming it is 15% but it is the same any rate you care to use)

You earn $180k you pay 15% in tax, second job 10400 you pay 15% in tax
Joe Blow first job is 20000 and pays 15% tax, second job 10400 15% tax

Flat rate tax is the most equalitarian tax you can come up with. "Progressive" tax is discriminatory and patronising

On flat rate tax you of course pay more tax as you earn more but it does not discriminate on your overall income. If you earn 100 you pay 15 dollars tax, if you earn 100,000 you pay 15,000 tax if you earn 1,000,000 you pay 150,000 as simple as that. Everybody pays and the system is revenue neutral if you get the right rate wich I dno'nt know exactly but according to some studies would be for Australia around the 15% mark give or take. It may turn out to be less because a lot of people will de-corporatise and go back to direct income tax and therefore there would be more revenue to distribute.

As for your question whats the point to tax the low income earner, there is a point of course there is. It stops patronisign them and pretending they need the governemt to mother them and protect them from the bad world. If a student earns $2000 at MacDonalds and he pays his $300 in tax he should be proud to have made a contribution to society just like an adult does with a full time job. I am sure you can see the point in doing that.

 

Luckless,

You obviously by your comments have NEVER worked on even a simply 5 minute boat job........everyone that has would know that it takes at least 2 hours to do............not worth my continuing on with this nonsence.

 

Luckless,
With all your enthusiasm and support for the capabilities of computers, to simply shear a sheep (remove its wool), is still not achievable consistently and reliably... Several organisations have been trying since the mid 70's or thereabouts, (I was doing computer studies then), and writing in assembler.... End of interest, bye-bye...

Computer introduction is NOT a labour saving device - it can assist in automating some highly repetitive fixed functions... Like transferring keystrokes from a keyboard onto paper via various repetitive processes... Like search through huge quantities of appropriately stored stuff to find "pattern matches" as in spell-check, grammar parsing, translation, numerical evaluation in some cases...

 

All for it .
Never going to happen.
Work out why ....I`m sure you can.

 

No it is not.

 

It appears so.

Robotics have their place.
Production runs.
Limited by the " state of the art " we are at.

 

Luckless, your are only correct in theory, not in practice. One of my associates is a robotic engineer and he talks about robots until the cows go home.
To begin with, an of the shelf robot to market the job described here will cost something like 50 yearly salaries and so will take a long time to recover the cost. Yet since the market for repairing engine bases in boats is somehow limited there is no robot developed for this task and so it needs to be made. Add about 200 yearly salaries to the previous price.
Now lets say that someone with more money then common sense actually pays the millions required to have his robot, he will now have to pay an expert to set it up in a boat and take it out every time to do a 5 minutes job once a week or so if they are busy. I am sure you can see the absurdity of the proportion.
Robots are very good in doing a lot of precise and repetitive task in an identical environment. To get a robot to perform in an environment that changes every time and to make repair assessments will cost more than the shipyard. It does not make any sense to build such robot.
The task of repairing an engine is not as easy as you think and is clearly harder than for example driving a car. However there is yet to be a robot builder that can build a car that drives itself. Not because it is so difficult but because no one would be able to pay for it if built with current technology.

 

In all due respect to Luckless the job being discussed was an example and not necessarily the job to be done.

I think we all understand that the job can be broken down until it is simple

 

A robot to replace a worker has to be cost effective.
Jobs can go to robots if they save money do a better job or both.
Cars are spray painted in factory by robots, yet not one smash repair has a robot to spray paint cars and you can not get a simpler job than that.

There are always three sides to this type of debate. One is technical, the feasibility. The other is economical, the viability. The third is political, the ********.

In Europe there are hundreds if not thousand of factories that operate like 300 years ago particularly in the food industry, musical instruments and traditional arts like blacksmithing and woodwork. They use 10 times the amount of workers and almost no modern equipment compared to a modern version. Yet they don't even debate the need for modernisation since the product is optimum, the labour is welcomed and the cost is comfortable included in the price obtained.
Can cheese be manufactured by a robot? You bet!
Is it worth while investing 50 millions in robotics for factory that makes one million a year? No.
Is it even desirable? not even that. The learning curve to adopt a manufacturing process to automation is so high and the years of lower quality product so damaging that it is not worth even discussing.

However we can debate the feasibility any time, it is the toy side that excites the technically minded, and the politically motivated will use this as leverage to drum up a revolution against modernisation.
Yet it is only the economical side that matters. If it is worth while it will become fact as soon as technically feasible.
If it is not all will stay the same for the next 1000 years

 

I must say that I agree with Marco1 on this one. It isn't enough that a robot can be programmed to do a job; it has to be cost-effective.

Looking at it from a cynical Southern California viewpoint, there are a whole lot of jobs a robot could probably be programmed to do. Unfortunately for the roboticists though, they're already being done by illegal immigrants.

Mexicans or Central Americans desperately wanting to feed their families are definitely cheaper and more flexible than robots....

 

What about a robotic fence guard complete with nocta vision and machine gun? Perfect!