Wednesday, February 20, 2019

Systems fragility may be decreasing in spite of stagnant productivity.

Following up and expanding on the previous post.

I'm willing to agree for the sake of discussion that productivity is going to stagnate. I make gears for a living, and I see no way that technology is going to fundamentally make it cheaper to produce a 32 tooth 12 pitch 14.5 degree pressure angle involute spur gear any time soon. You still have to make a blank then hob it. The energy and materials inputs to 3d print in metal for the same part are unlikely to be cheaper except at the lower quantity limit of 1 piece, where setup and transport dominate costs.

The variety of parts that can be made by the home shop, without regard to cost, have never been higher, on the other hand. While mass manufacture won't get cheaper, it is now possible for a better networked set of outsiders to help lower the fragility of the overall system by making it possible to reroute around damage and/or collapse. As long as cheap NC control systems are available, amazing things can be made in ones garage, basement, or kitchen, but at lower "productivity".

Is productivity actually increasing?

I was browsing this thread on Reddit, and came across this quote (emphasis mine), which got me to thinking, and posting...

Finance is assumed to have been the driver of productivity post-stagflation, but there is a widely-held sentiment that the returns from the stock market are an illusion not based creating value. Put another way, there has been enormous sums of money produced in a series of financial bubbles based more on consumer confidence than any returns- think of the unsupported internet bubble in the 90's, or today for social media- a phenomena perhaps best epitomized by welfare-fraudster Elon Musk. Millions have been made off of investing in bitcoin, a currency which has no proven utility to date and resembles a ponzi schem. Or what about computers? What about Moore's law? To date, there has been no satisfactory evidence that computers have been used to improve real productivity. These machines support the internet and can simplify administration, but they haven't made our industrial technologies significantly faster
My world view has shifted considerably since I joined the blue collar, industrial workforce. I previously would have questioned the last sentence of the above paragraph, but now I'm not so sure.  In the making of gears, my specialty, it still comes down to making a blank, and then subtractively machining it by shaping or hobbing.  The machines and cutting tools would all be familiar to someone from the 1940s, though some of the instrumentation is a bit easier to use.

I work in a job shop, which means the quantities involved in production never exceed 1000. Mass manufacturing at scale doesn't make much of a dent at work. To make gears faster or cheaper would require new machines, and the capital outlays involved just aren't justified. It is possible to make bevel gears with 5 axis machining, but it turns out the process wouldn't be any faster, just a bit more flexible.  The return on investment isn't there.

Additive manufacturing might someday displace things, but the real inputs to the process in terms of power and materials make it highly unlikely.

These are my thoughts, I welcome yours.

Thursday, February 14, 2019

RSS still works

I came across this post by Semaphore and Cairn, which pointed to an article beyond a paywall that I can't read.... but I believe the point still applies in the context of RSS.   In the first few iterations of reading RSS first, some things just had to go because they turned out out be just too politically grating (why subscribe to a feed that consistently makes you angry?)
Turns out that BoingBoing is EASIER to read on their RSS feed.

Tuesday, February 12, 2019

RSS First

I've decided to change the loop I find myself in daily.... I'm going to read all my RSS feeds FIRST before doing BoingBoing, Slashdot, Facebook and Youtube.

My current RSS reader is RSS Owl.

Monday, February 11, 2019

New goal - Disprove dark matter/ dark energy on a $100 budget

Ground breaking science will happen this year, either in my garage, or at Pumping Station One in Chicago.
Dryer vent pipe, old water pipe, heavy weights, and a bit of fishing line will be used to disprove the dark matter / dark energy bullshit that has infested physics.
Hopefully I can keep the budget under $100.

Sunday, February 10, 2019

Leading edge science in the home shop?

I strongly suspect that there are some elegant and affordable experiments similar to the Cavendish experiment which could help prove Mike McCulloch's Quantized Inertia theory. 

If you were to use a servo to keep weights on a torsion balance stationary, you could measure the current required and have a very sensitive gravity balance.  The suspended weights and servo could all be soldered inside a copper pipe to all but eliminate drafts and electrostatic forces as noise sources.

A computer controlled linear stage could be used to log data, and to move the test masses external to the assembly and make many measurements over weeks or months of the forces with weights and different distances.

I hope to do this experiment in the next year or two.  I welcome anyone who cares to do it before or after me.