andrewducker | Interesting Links for 10-11-2017

andrewducker

What are the personalities line of people who share images of their pets on social media?: (tags: personality cats dogs socialmedia psychology )
Unprofessional Astronomy: (tags: astronomy space research OhForFucksSake )
Improvements in renewables are slowing. A renewable future now depends on improvements in storage: (tags: renewables batteries thefuture )
Niantic’s follow-up to Pokémon Go will be a Harry Potter AR game launching in 2018: (tags: harrypotter augmentedreality games )
Germany serial killer: Niels Hoegl killed at least 100: (tags: murder germany )
Low-status men are hostile to women to minimize the loss of status resulting from women competing with them: (tags: women games men society sexism )
UK's biggest solar farm planned for Kent coast: (tags: uk solarpower )
Model student says she was sent home from school for wearing an 'inappropriate' outfit: (tags: clothing school UK OhForFucksSake )
Rian Johnson Is Developing a New Trilogy of Star Wars Movies: Chances are there will be a new Star Wars movie every year for the rest of your life.
(tags: StarWars movies )
Law 'rebels' put spotlight on using skills to help vulnerable: (tags: law scotland )
Gender pay gap widening for women in their 20s, data shows: (tags: gender pay uk )
CEOs Don’t Steer (Deeper, and more interesting, than it sounds): (tags: business )
Bank of England looking at Machine Learning for making predictions: (tags: uk banking economics predictions ai )
The end of the housing supply debate (maybe): (tags: housing economics )
Why Theresa May can't sack Boris Johnson: (tags: politics uk OhForFucksSake )
Uber loses court appeal against drivers' rights: (tags: taxi uk rights work )

Crossposts: https://andrewducker.livejournal.com/3633403.html

Flat | Top-Level Comments Only

From:

jack

This seems to support the theory that technological progress is usually S-shaped. Fingers crossed storage gets good enough.

From:

andrewducker

Yeah, I really hope so!

From:

mountainkiss

I got muddled between your Johnsons and had Boris making Star Wars movies, which gave me pause.

From:

danieldwilliam

Boris can't make Star Wars movies - he's already got his name down to replace Chibnall making Doctor Who. He's the perfect choice.

From:

andrewducker

The thought, frankly, appals.

I'm not sure he's competent enough to actually manage a film shoot.

From:

danieldwilliam

Some thougths on the GreenTechMedia article.

Learning Curve Effects and Solar PV Factory Capacity

A slowing in the annual rate of decrease in costs is what I'd expect if the cost reduction is being driven (mostly) by learning curve effects. These are based on a doubling of lifetime production. They are not based on the absolute increases in production. The learning curve effect in renewables is probably about 16%. So you go from total installed capacty of 1GW to 2GW, the cost falls 16%. From 2GW to 4GW, you get another 16%. From 4GW to 8GW, another 16%. For some years, a few years ago, there were several doublings in the lifetime number of solar panels and wind turbines built and costs have fallen quickly. I think at the moment we a slower rate of doubling and the learning curve effects are being spreadout over more years.

My expectation is that now that renewables in good locations are cheaper than fossil fuels for marginal production we'll see a steady increase in the absolute amount of capacity added and the rate of doubling will pick back up again. But it's not going to get as fast as it was ten years ago because there is a limit on the demand for energy plant. If you have solar PV factories that can build 4% of the planet's total required capacity each year then you have factories that will replace all generation with solar PV in 25 years. Solar Panels have an estimated useful life of 25 years. So once you have replaced all existing generation capacity with solar PV, a production rate of 4% of total capacity is about replacement rate. However, the useful economic life of solar panels may well be much, much higher than 25 years. Therefore, the long term capacity of the factories manufacturing solar PV panels is probably capped at about 4% of global electricty generation capacity. At the moment manufacturing 4% of global electricity generation capacity per year would imply a doubling of lifetime solar PV manufacture this year, but only a 50% increase the following year and 33% the following year, 25% the year after that.

Cost is not the same as Price

I think the Lazard report is looking at the price energy plant developers are paying for renewable generation plant not the actual cost of producing it.

If renewables are the cheapest form of marginal electricity production and the capacity of factories to produce renewable generation kit is limited (i.e. there is a shortage of renewable generation plant) I would expect developers to bid up the price they are willing to pay for new wind turbines or solar panels until the cost of their projects is only just cheaper than the alternatives. The value created from cheaper renewable generation plant will go to the owners of the factories making the plant in the form of surplus profits.

If we are in the build out phase of renewables, where demand for them is higher than the long term equilibrium or replacement production rate (and renewable generation manufacturers are competing on scheduling and not price) then I would not expect the price energy project developers are paying to fall as quickly as the cost of producing renewable genetion plant. Building a solar PV or wind turbine factory takes time and money. Entering the market has costs and it is not obvious that over the liftime of your factory you will recover all the costs of setting it up if you are overbuilding manufacturing capacity and end up competing on price.

The Next Renewables Challenge

I disagree that the next renewables challenge is to get the Levelised Cost of Energy plus Storage down to levels that compete with Combined Cycle Gas Turbines. The next challenge for the developers of renewable energy projects is to get their LCOE down to below the cost of the gas burnt in the CCGT's. Once you reach that point then the LCOE of renewables plus back up CCGT is lower than the cost of running a CCGT. So a target cost of about $30/MWH is what renewables should be targeting and let storage worry about its own economics.

From:

armiphlage

"useful economic life of solar panels may well be much, much higher than 25 years"

Agreed. One data point - the panels that our factory made were designed to still produce 80% of rated output after 20 years in southern Canadian environmental conditions. If maintenance and property taxes are low, utilities will likely leave the panels out for many more decades until they die completely.

From:

danieldwilliam

Aye - buy 10MW of solar PV panels, get 8 MW free in 25 years.

I've seen suggestions of actual degradation factors of 1%.

Edited Date: 2017-11-11 08:43 pm (UTC)

From:

andrewducker

I do wonder if some people are currently competing on the grounds that they think costs will get even lower over the next few years, and it's worth making a loss right now in order to gain the experience. In which case the slowing of the efficiency gains is going to throw a really big spanner in the works.

However, others will be able to buy them up cheaply (or recapitalise them) and keep things going, so long as it's still worthwhile (which is really looks like it will be).

Good point about renewables + gas. That could be a potent combination - and a good holding spot while battery researchers get the next set of advances online.

From:

danieldwilliam

There are certainly developers who are forward pricing, in that they are bidding today to offer electricity in 5 years time at a price that relies on the price of renewable generation plant being lower than the price being paid today.

What is unclear is whether they have firm prices from the manufactures or are speculating on prices fallling.

My own experience of power station development suggests the former. I think typically if you are going to bid a power plant you need 1) a firm grid connection, 2) a firm generation plant contract, 3) a firm fuel supply agreement, 4) firm financing and 5) a firm off-taker for the power. You need to have 4 out of five of them signed up before you commit.

Building small utility scale renewables might be a different undertaking compared to building a CCGTbut I suspect not very much different.

So, I think that tells me that either the costs and the prices of renewables are still trending down at a decent rate, or that someone, somewhere is taking a lot more risk than is usual, or that the article and the Lazard's report on which it is based are already five years in the future and looking at the pace of change from 2022 onwards.

From:

danieldwilliam

Renewables and gas operate well together.

A fleet of CCGT's where solar PV (or wind) is cheaper to install than the cost of gas will cost less than just a fleet of CCGT's on their own. Eveyone else should be worried.

And that still leaves plenty of opportunies for batteries to develop the learning curve and economy of scale effects or the transformative technological break throughs that they need to become cheaper than building a CCGT. Opportunies in selling balancing services, selling storage to small, islanded grids, batteries for vehicles and batteries for local security of supply or where the existing market and regulatoray environment is dysfunctional.

From:

danieldwilliam

I think Mitch Benn is half right.

The other half is that if May sack Boris now then Boris launches a leadership challenge now. This might wreak Britain, it will probably wreak the Conservative Party but it certainly wreaks May.

From:

andrewducker

More infighting please!

From:

danieldwilliam

Oh there is going to be more infighting.

From:

danieldwilliam

The faster Uber is made to burn through cash the better for us all.

From:

andrewducker

I do wonder if there's a middle-ground here. Many taxi/hire-car drivers _are_ effectively self-employed. It's just the way that Uber is doing things which is causing issues. Could there be similar models which worked better for both drivers and job-clearing-houses?

In any case, I suspect self-driving taxis by 2022 at the latest, and then they're all on the dole...

From:

danieldwilliam

I think self driving cars will take at least 5 years to get regulatory approval once they work.

From:

andrewducker

I'm not so sure:
https://www.wired.com/story/waymo-google-arizona-phoenix-driverless-self-driving-cars/

From:

danieldwilliam

I think there are going to be law suits.

One each for the first dozen people killed in an accident involving a self driving vehicle in each jurisdiction.

Several each by every organised labour organisation in each jurisdiction.

That's going to tangle even the most supportive legislature up in some difficult regulatory issues.

From:

danieldwilliam

A middle-ground on the employment status of Uber drivers?

From:

andrewducker

No, a middle ground approach that a different company could take.

Hailo drivers, for instance, weren't employees - they were black cab drivers using an app to find people who wanted a black cab.

The question is, is there a minimum level of service which an app can provide which means that the drivers remain legally self-employed.

From:

danieldwilliam

It's not so much the app that makes Uber an employer as the control they exercise over the work of the employee.

So plenty of ways to organise the provision of an app that leaves the drivers self-employed. A coop or a non-exclusive contract for services (the services being provision of a ride / drivermatching service. (

From:

calimac

Theory of men suggests that high-status men aren't hostile to women because they feel little competition at that level. Therefore, as women rise, those men should get more hostile. Alternate theory: low-status men are hostile to women because they have nobody else to feel superior to. (This is also why it's poorer whites in the US South who are most hostile to blacks.) Under that theory, high-status men should not increase in hostility as women rise.

The more Star Wars movies, the less obliged I'll feel to see any of them. Good.

I'd like to test the theory of CEOs against the Peter Principle.

Nitpick, but possibly significant in comparing Boris to Winston: Winston wasn't sacked. He resigned. (From the opposition shadow cabinet, so there was no big show about it, but still ...)

From:

snippy

Hmm, imposter syndrome might interfere with high status men feeling lack of competition from women.

From:

andrewducker

Good alternate theory.

And good point on Winston!

From:

drplokta

According to an earlier link you shared, the article on renewables costs can safely be ignored, because it uses the word “baseload”

From:

andrewducker

I wondered who would spot that!

From:

birguslatro

But, but, if you give schoolgirls an inch, it just leads to...

http://www.nzherald.co.nz/nz/news/article.cfm?c_id=1&objectid=11940110

From:

andrewducker

You just can't trust teenage girls!

From:

miramon

CEOs Don't Steer is one of the most interesting and thought-provoking articles on management that I've read in ages. I keep having "but that implies... oh!" moments, and now I'm going to have to go back and read it again.

From:

andrewducker

Yeah, it didn't sound that interesting to me from the headline - and then it all made so much sense.

Flat | Top-Level Comments Only

Profile

andrewducker

My NotZen Website

January 2026

S	M	T	W	T	F	S
				1	2	3
4	5	6	7	8	9	10
11	12	13	14	15	16	17
18	19	20	21	22	23	24
25	26	27	28	29	30	31

Page Summary

Active Entries

Style Credit

Style: Neutral Good for Practicality by timeasmymeasure

Expand Cut Tags

No cut tags

Page generated Jan. 7th, 2026 09:32 am