Innovation, Trial and Error

Good read for anyone interested in why some projects do better than others. I think I can give this a slightly different spin. Innovation is all about trial and error, emphasis on error. Your ongoing development needs to tolerate errors to make substantial moves forward. Nukes don’t let you fiddle around with techniques they way other technologies do. So innovation (i.e cost savings) seldom happen. On the flip side, small scale tech like solar allows for all sorts of tweeks at all levels. This leads to large, compounded improvements. Back to the energy grid portion: a nice quote below.

Well, the data on that is in. Wind and solar have plummeted in costs and grids with higher penetration of renewables are actually more reliable than coal, gas or nuclear heavy grids using industry standard metrics for outages per customer per year.

The Nuclear Fallacy: Why Small Modular Reactors Can’t Compete With Renewable Energy

That’s My Home

Was thinking about Louis Armstrong’s private tape collection. I read about it a bunch of years back and thought I would see if they had made it on line someplace. It a combination of music and conversations made in his home in Queens, NY. Ran across this pretty amazing site, from the Louis Armstrong House Museum. Will have to spend some time out here looking around.

That’s My Home – Louis Armstrong House Museum Virtual Exhibits

Louisiana is building electricity hubs to power communities after a disaster

There are two outrageous facts here. First, why wasn’t this done decades ago, as part of Civil Defense? Second, why is the private sector handling this? Every public facility, especially fire departments and hospitals, should have had this already. I am also perplexed as to why the military has been so behind on various new power technologies, but that is another subject. From PBS.

Louisiana is building electricity hubs to power communities after a disaster. Here’s how they work

Building a Computer

In one of my jobs I had to select the processors for the shared compute resource for a major US bank. I would load up all the processor family information from the web along with some benchmark data into a spreadsheet. I would graph a “bang for buck” chart that was usually a pretty simple arc. Smaller processors were cheap, but somewhat weak. Bigger processors with more cores were pricey and you usually paid a premium for the very top end. The answer, some what boringly, was always something in the middle. But it was good to have data to back this up. After all, we were buying thousands, if not tens of thousands of units.

With a home PC it isn’t so simple. I’ve been building PC for decades and I’ve settled on a Rule of Thumb. Spend about $100 each for the motherboard, CPU, RAM and disk. It doesn’t have to be exactly $100 and you can move this around depending on your needs (and how much you want to spend). I’m also recommend last year’s tech. It’s usually more solid and you can read reviews and you don’t pay the premium for the Latest and Greatest.

Upgrading my sons old gaming machine. It has an MSI B250M PRO-VD motherboard. Today it goes for well over $100, but being made for 6th and 7th generation Intel CPUs, you can guess this is just pricing of an old item in stock for repairs. The CPU is an Intel i7-7700 3.6 GHz. Today this sells for over $200 refurbished and even almost $350. Again, I didn’t pay this much new a few years back. This is just old inventory for people that need to make repairs or have other unusual legacy needs. Otherwise there is 16GB of DDR RAM and 500 GB of SSD disk.

I may be upgrading this, keeping the case and power supply. Here’s what I would pick today. Going to stick with a similar MSI motherboard, but one with on-board Wifi. Here’s the list:

Doubled the RAM and disk for less than $100 each but went a bit over on the motherboard. As far as Intel CPUs, they seem to keep getting more and more expensive. But this isn’t the place to skimp.

Oh, the GPU. Used to be optional, but this is for a gaming PC. I’ll save that for another post.

FTC intends to ban noncompete clauses that bind 30 million US workers

These have a long history in tech. My understanding is they were illegal, or at least ignored, in California. I always found it bizarre that someone could make you sign a document preventing you from working where you want, in your chosen field. This is simply a ploy to keep wages down. If people steal intellectual property, then punish them for that crime. Stopping a person from working for a competitor out of fear of a potential crime being committed has always been ridiculous.

FTC intends to ban noncompete clauses that bind 30 million US workers