OT: New book "The Long Thaw"

Doesn't surprise me. I can't imagine you spending money on electronics books, much less hardware.

Who is Phil Larkin? Some long-lost cousin of mine?

If a physical process is chaotic, it's not predictable.

the solar system is also chaotic, but it is

Agriculture wouldn't work without long-term climate predictions?

Can you explain that to us?

Which, by the nature of chaotic systems, we can never know how to do.

John

Baby steps first, I say. Personally, I would feel better if "they" (whoever that is) could even get the annual Hurricane predictions correct. Every year they tell "This is gonna be an above-average with x-named storms, y-Cat 3 or higher, and z-making landfall".

Their track record is complete shit. I don't know why they bother.

All it does is whip up the insurance companies (and their rates) into a frenzy. And we, the property owners and taxpayers at large, get to line their pockets.

And when a storm DOES hit, those same "they" say, "Oh, yeah, those levees are good to Cat-3. Don't worry about it. Go on about your business. In fact, go shopping.) Or those same f^ckhead insurance companies go bankrupt long before paying their obligations to policy holders.

And look, I'm not saying the research is bad (though it likely is?), or that it shouldn't be attempted. But let's at least get someone local, and not asshole in Colorado who has a proven track record of being WRONG much more often than simple chance would allow!

They are patient. They persevere. One of these years, they will be right, this proving the dangers of AGW.

They use advanced computer models, of course.

Definitions:

Weather: atmospheric effects that models demonstrably can't predict.

Climate: atmospheric effects whose time scale is sufficiently long that none of the models can be checked.

John

The local book-store doesn't carry anything that looks remotely like an electronics book. The last one I bought - "Inductance Calculations" by Frederick Grover, 1946 - came from Amazon, and I bought it because Roy McCammon recommended it, rather than because I had any immediate need to calculate an inductance, though I do have clear memories of needing to calculate inductances from time to time,

No. Serious finger trouble - I'd intended to type John Larkin.

Only over a sufficiently long period - as I'd pointed out to you before, the solar system is chaotic, but adequately predictable for the next 100 million years.

If you'd bothered to read the rest of the sentence before typing out your idiotic mantra, you'd have had the opportunity to look slightly less idiotic

You plant the seed in spring in the expectation that there will be enough rain between then and the end of the growing season to give you a decent crop. You can't predict when the rain will fall - that's weather - but you can predict that enough rain will fall over the growing season to keep you in business.

Not an opinion reflected in the literature

but John Larkin does feel himself to be infallible on matters of faith and doctrine.

-- Bill Sloman, Nijmegen

And Bill Sloman seems to have adopted the imperial "We" - he addresses people in the third person, even those to whom he's directly responding.

Interesting.

Cheers! Rich

Not interesting!

He's a boring, silly, pompous, useless git who loves to hear himself talk. Which is why he posts hundreds of lines of useless mincing crap at a whack. I suppose he can't find people willing to listen to him pontificate in person.

John

You might not have noticed, but John Larkin never processes any information that contradicts his ideas or damages his self-image as some kind of infallible pope of technology. Since he was never going to read the last line of my post, it is a comment addressed to other readers, and thus it did make sense to refer to him in the third person, even if the whole post was nominally addressed to him.

-- Bill Sloman, Nijmegen

Whereas John Larkin wholly confines himself to talking about electronics - which he does know something about ...

-- Bill Sloman, Nijmegen

: If a physical process is chaotic, it's not predictable.

Even if the detailed path of a process is not predictable, there may be some predictable statistical invariants in the process.

Take the Logistic Map, for instance, perhaps the simplest process that exhibits chaos. It's long term behaviour cannot be predicted given its current state, but still the process is guaranteed to remain between zero and one. The governing equation contains this fact built-in.

In a similar manner, it is conceivable that there are physical constraints that limit how the 'weather' can behave, such as energy conservation. For instance, there is a well-known and predictable phenomenon, affecting the statistical properties of the weather (itself unpredictable) when the radiative forcing changes. It is known as 'summer' and 'winter'.

Analogously, one might also make the coefficients that control the range of the Logistic Map variable, and make them follow some dynamic equation. Now, that dynamic law may or may not be chaotic.

I'm not saying that the climate (statistical long-term features of the weather) is *not* chaotic, I'm just saying that the chaoticity of weather cannot be used as an argument that the climate is also chaotic. Showing that would require some additional arguments.

Regards, Mikko

Not entirely electronics, but mostly. Other things are interesting, in moderation. You seldom mention electronics, and then it's usually to repeat old stories.

Do something original.

Do something helpful.

Do some electronics.

Try being nice and making friends.

John

But as time increases, the predictability of the system state goes to zero.

Sure, we could be 100% tropical or 100% glaciated. Assuming those limits doesn't add a lot of useful predictability to climate. The Mandelbrot set has only two discrete values at any point, "in" and "out". It's generally accepted as "chaotic."

I agree, it's conceivable the COE might apply to the atmosphere. Great insight.

For instance, there is a well-known and

Periodic external forcing. Another brilliant insight. Do you think that, even in the depths of an ice age, we might still have summer and winter?

Look at the climate reconstructions from past times. Can you explain the patterns? If not, how could you project the future?

All I've ever said there is that the existing climate models may be, and are likely, very bad models of reality. Because we don't have good data (what *is* the temperature in Chicago?), because we don't understand the physics, and because the system being modeled is itself chaotic. Sloman trusts the models and he trusts the data and he argues that, while weather is chaotic and unpredictable, climate is causal and predictable.

My heretical response: "unlikely."

It matters to me, a little, because dynamic systems are my business. I can't imagine why it matters to Sloman.

John

The UK Met Office has dropped it's seasonal forecasts because they got them repeatedly wrong. BBBQ summer was wet and windy, warm winter was worst since 1964 etc. They lost the public confidence.

ith

ng.

Well, the last electronics thread I was seriously involved in was about work that I did in 1993 and published in 1996. The stuff I did at Nijmegen University and at Haffmans in Venlo seems to be of less general interewst

It happens from time to time. Most problems can be solved by recycling well-known solutions which are known to work. People who invent new solutions when they don't have to can be something of a menace.

I do. You wouldn't be aware of much of it.

I could do with a broader source of electronics problems to solve. This user-group doesn't throw up much.

Advice that you might take. Here's a hint - people who tell you that you are wrong may not be your enemies, and people who tell you that your ideas are clever and perceptive may be flattering you with a view to exploiting your good oinion of them.

-- Bill Sloman, Nijmegen

The time involved can be quite long. The positions of the planets around the Solar system is eventually chaotic, but the pretty predictable for the next 100 million years.

The oscillations between glacials and inter-glacials don't look all that chaotic. Admittedly there have been some 50,000 year long interglacials, and this one looks as if it might be one of them, but the pattern is mostly repetitive.

There's quite a lot of evidnece that we did.

Climate scientists can do fairly well. They need to know more about ocean currents before the can do better.

The weather is chaotic but predictable for up to about ten days. The climate may be chaotic, but it is predictable for a lot longer than ten days, and the ice core data shows tolerably consistent repeating patterns over some 800,000 years.

Since you equate "chaotic" with "unpredictable", ignoring the fact that even chaotic systems are extremely predictable in the short term, you response isn't heretical, it is merely ignorant.

You obviously aren't sufficiently interested in dynamic systems to learn what "chaotic" means in practice. I've got a broader range of interests, and - unfortunately - loads of free time to learn a boit about stuff that interests me.

-- Bill Sloman, Nijmegen

They may be better when the Argo buoy data starts giving them a better idea of what is going on in the deep ocean. A lot of heat travels from the equator towards the poles in deep ocean currents, and we don't know much about them, except that they seem to switch routes and destinations form time to time.

We do know something about the El Nino/La Nina alternation in the Pacific, and we are starting to get a handle on the Atlantic Multidecadal Oscillation, but what ever it was that produced the Dansgaard-Oeschger oscillations, with a period of about 1470 years is still pretty much a mystery, which is a pity. Some people think that the Medieval Warm Period and the Little Ice Age represent the most recent Dansgaard-Oeschger oscillation.

-- Bill Sloman, Nijmegen

Sure, but that's not electronics design. And that's not a lot of fun, either.

People who invent new

People who don't can be a big bore. I hate to use the same designs over and over.

There' a zillion lifetimes worth of interesting, unsolved problems out there. The internet is a great help in finding tham.

I have people tell me I'm wrong several times a day. Often we're both wrong. We work it out and come up with some pretty good ideas. And they we build stuff and sell it. The process, arguing at the whiteboard, can be wonderful. Some people can do it, many can't.

I have a friend who's working on his EE degree and needs a senior project. I think he'll do a "modern" TDR pod, and see if we can use my deconvolution algorithm to prettify it. Maybe we'll get a product out of it. I've always wanted to build a TDR, and now I'll get to do it, by proxy anyhow.

John

Good for you, John. There's nothing more satisfying then mentoring a group of young EE students. Especially when you finally see an idea sink in.

tm

It's not really amazing that he is posting in an imperious style.