Sometimes very active discussions about peripheral issues overwhelm a thread, so this is a permanent home for those conversations.
I’ve opened a new “Sandbox” thread as a post as the new “ignore commenter” plug-in only works on threads started as posts.
No question. What worries me is that we’re past the point where there is any doubt about it and also past the point where we can do enough to limit it.
I didn’t realise you were on Mallorca!
A worrying prediction of the consequences of climate change…
Less beer
Oh dear… pretty sure shortage of beer is one of the 10 plagues of the bible. This is it, we’re doomed. But in all seriousness, this is fucked up. Hope it’s not too late
It will have to get far, far worse before the USA Republican party will take it seriously.
But warmer temperatures means more arable land available in Canada for cannabis, hence more supply, hence cheaper prices. If I weigh beer against cannabis, I’ve got to go with the cheaper cannabis.
Especially since beer has so many more calories.
You forgot to factor in the munchies though 😀
petrushka,
Yes, I’ve never understood the visceral antipathy of some to nuclear energy. Flying commercially has become very safe because it is highly regulated, planes and other infrastructure regularly inspected and maintained.
With similar levels of safety standards and implementation, what’s the downside to nuclear energy?
And there are alternatives to uranium.
There are several viable designs for “throwaway” thorium reactors that cannot melt down. They seem like a no brainer for places where wind and solar are not feasible, and they seem like a useful adjunct for places that use solar.
I have no use for virtue signalling. Politicians have proven useless at solving this kind of problem, because it requires thinking decades ahead, and elections come up every couple of years.
In the near term — the next hundred years — we will just have to live with climate change. But within 20 or thirty years, carbon based energy will become more expensive than alternatives, and politicians can fight over regulations, which is all they are good for.
petrushka,
Now some of the mud has settled where I live, the real cost of climate change is being counted. Yesterday I drove through the next town to where I live. It’s built around a water course that doesn’t dry up in summer. The banks used to be covered with potagers (vegetable gardens) benefiting from the available water, houses backing on to the stream had gardens tiered down to the water’s edge.
Idyllic!
Until Sunday night and Monday morning.
Unprecedented torrential rain falling in the stream’s catchment area resulted in a wall of water, mud and debris hitting the town and the bottleneck of the bridge. The devastation is appalling. An estimated 400 people (just from this town) are having to be rehoused. The damage to property is enormous. The damage to people’s confidence in the future is worse. How can these people restore and move back to their homes without worrying whether another unprecedented storm will flood them from their homes again?
I guess it must be the same problem for people living in Florida.
Anybody any idea what this is? Looks like some kind of millipede. My daughter encountered it on Nias and sent me this photo.
I just wanted to complement Mung on his patience and deferential demeanor over at peaceful science.
I never cease to be amazed at how perfectly reasonable and generally pleasant folks can be so clueless and nasty when the “other side” dares to question them.
peace
thanks!
But the moderation there sucks. 😉
Gibberish. As in this post of mine is gibberish, I’m testing something.
Test
Take 3
Didn’t work.
Rumraket,
Funny, but I was thinking earlier today that a crucial thing that few people “get” is the consequence of the selection coefficient
being in exponents.
I forgot what the trick was if you want to edit a post with a latex equation in it to make it look correct after editing. I keep making some error I want to correct but it just gets worse. I need to put *something* in front somewhere right?
Wherever you put a backslash in the original input of the comment, you have to put two backslashes when editing the comment.
I’m not enough of a math wizard for it to be obvious to me what you’re getting at. But just from playing around with plugging in different values in that equation it seems to me the biggest contributor to probability of fixation is a large N. Even very weak selective benefits at tiny fractions of N get magnified.
I tried thinking about why that would be the case, and I came to the conclusion that a large N essentially means there is more opportunity for mutation to create new instances of the allele in coming generations without it having to rise through selective death and/or mating.
Does that seem wrong to you?
Ahh yes, thank you.
Thinking about it some more I don’t think that’s the case. Among other reasons because the equation does not anywhere factor in probability of mutation.
I’m starting to think that with a large N, even at a low initial frequency there will still be so many individuals with the mutation that it becomes incredibly unlikely for all of them to die by random death. So given even tiny increases on reproductive success, it’s rise becomes inevitable when there are already many that carry it.
What I meant was that people don’t understand that when we talk about one genotype having a small selective advantage over another, we’re talking about differences in exponential rates of growth.
I hate talking about stuff I don’t understand well. But it just happens that I’ve read a bit of Kimura lately, and I know for a fact that the length of his interval of effectively neutral mutations is inversely related to the effective population size. So there’s more drift with smaller effective population sizes. (I’m sure I’ve seen this mentioned by a number of people other than Kimura.) I don’t know the specific equation you’ve entered, but I thought it might be that
lay between 0 and 1, and that 
and 
were, respectively, the population size and the effective population size.
Don’t explain this to me here: I won’t be back for some time. I’ve been taking a little break from other stuff I’ve got to do.
Rumraket,
Sorry, I’m no help to you here.
No worries, I’m pretty sure my first idea was wrong and the 2nd one must be correct. And I see what you mean by difference in exponential growth, small value differences in fitness can have large consequences for the rate and probability of fixation.
Interestingly the same can be true for the physiological/morphological effects of even single substitution mutations. Replace a single nucleotide and a lot of downstream effects can cascade into major phenotypic change.
People have touched on the reasons why there is less genetic drift in large populations. Let me try my description:
1. Genetic drift happens in all finite populations (which means all populations in real life, but in theoretical models only those that have finite populations).
2. With larger populations changes in gene frequency by genetic drift are smaller each generation, since the process is closely analogous to tossing coins with a Heads probability that is equal to the previous generation’s gene frequency. With more tosses you get closer to the Heads probability.
3. If there were initial genetic variation but no further mutation, and no natural selection, fixation of the population for one allele (the descendants of a randomly chosen copy in the initial population) occurs, in a length of time in generations that is proportional to population size.
4. If there was a small amount of natural selection, in a large population it would change the gene frequency more during the longer time that is available in a large population, and change it less in the smaller time that is available in a small population.
5. With a constant input of neutral mutations at a constant mutation rate per copy of the gene, and if these go back and forth between a modest number of alleles, there is more standing genetic variability in the population in the long run in a large population, because genetic drift will fix alleles more quickly (or cause their loss more quickly) in a small population.
To get some feel for this, try downloading a good program designed(*) for teaching about evolutionary processes. The best is of course, in my unbiased estimation, the program
PopGthat our lab developed over a period of 40 years as I taught population genetics. It can be downloaded here and works in any system that has Java installed, which can be Windows, MacOS, or Linux. Sorry, no iOS or Android versions yet as these aren’t yet tolerant enough of regular Java (they lack the necessary graphics functions).(*) As to “designed”, yes, I designed it and I and many others gradually improved it. But no, that doesn’t mean that the processes that it simulates are Intelligent Design. After all, we can make computer simulation programs that simulate all sorts of chemical, physical, and biological processes such as erosion, Brownian Motion, celestial mechanics, and Mendelian segregation and no one says that this shows that those processes are processes of Intelligent Design.
No one but me. 🙂
https://peh-med.biomedcentral.com/articles/10.1186/s13010-018-0066-z
My physicist grandfather became a crackpot (or maybe not):
My Grandfather Thought He Solved A Cosmic Mystery
Tim Maudlin takes on John Horgan (the second rate Horgan again, I am afraid):
Philosophy has made plenty of progress
Sean Carroll interviews Alex Rosenberg, hero of the new atheists and (in)famous unapologetic supporter of unrestricted scientism. The interview gets around to his latest book trashing historians after it gives Rosenberg a chance to state his continuing scientismic creed.
ETA: the new book seems more about arguing for eliminativism about mental content and propositional attitudes, with historical narratives being example of why that is a good idea.
Click on the see transcript link if you prefer reading to listening.
Joe Felsenstein,
One way I look at it is to imagine the large population split into an arbitrary set of smaller populations – say 10 individuals each. In any one of these, drift can promote an allele to local fixation; it only has to get into 10 individuals. Even if it has a non-neutral selection coefficient, if that amounts to (say) one extra birth in 100, you won’t see that effect in one small trial. But in the larger population, you are increasing the number of trials. It has to keep on getting into 10 more individuals, then another 10, then another … While it’s a doddle to get into one set of 10 by drift – a small sample – running the game over the long term results in convergence on the expected value (though the knots ‘expected value’ tie some people into …!)
There are close analogies in ‘polls of polls’, and the meta-analysis of clinical trials.
Allan Miller,
I wouldn’t quite put it that way because those groups of 10 individuals would not retain their identity over time. So the 10 this year are not parents to any 10 you choose next generation, as they mate with others.
But it is true for changes of gene frequency in one generation. Just as you can view the outcome of tossing a coin 100 times as 10 sets of 10, and then point out that a single set of 10 can get a fraction 0.7 of Heads much more easily than all 100 can get to 0.7.
Joe Felsenstein,
I was thinking more of the progression through the population of the haplotypes, which do persist through multiple rounds of recombination and segregation – the allele’s ultimate fate when fixed is to colonise a ‘population’ of loci twice the census size, in chunks, granted that selection operates mostly in the transient diploid state.
Thanks. I listened to that podcast.
It’s strange. Rosenberg is skeptical about beliefs, desires, etc. And I am also skeptical of those. And, in a way, that’s a big part of what Rosenberg is arguing. But I disagree with him about almost everything, apart from that common skepticism.
Rosenberg thinks that folk psychology is innate or near innate. I have never agreed with that, even going back to childhood memories. It has long seemed to me that folk psychology is obviously wrong. My suspicion is that folk psychology is a viral disease transmitted by the philosophy class.
Rosenberg describes himself as a materialist and physicalist. But he isn’t. He is an immaterialist. His world is built entirely out of propositions, which are immaterial things of doubtful existence. I suppose we could say that Rosenberg’s propositions are about the material and physical world. But there’s the little problem that be denies intentionality (as aboutness).
For Rosenberg, the physical facts fix (determine) everything. In the meantime, I am studying how humans fix the physical facts.
(Perhaps I should mention that there is supposed to be a hint of humor in the above).
If you enjoy that style of podcast, I recommend Rob Redi’s After On.
Good to know.
BruceS,
As you know, I don’t think “progress” is being made on any heavyweight matters.
Testing.
That’s FORTRAN notation, not mathematics. Instead try
Thanks. I considered making a similar comment, but decided to let it pass. But that poor notation does stick out like a sore thumb.
People do that all the time. Also they represent 0.0001 as 1e-4 (again, FORTRAN), and exponentiation by using the up-arrow or ** (the former is BASIC, the latter is FORTRAN).
Also, they give names like BETA to variables, when in mathematics that is not the name of a variable but represents
.
For people who don’t know what FORTRAN is, who aren’t like me originally FORTRAN II programmers, it is one of the very earliest computer languages, the first one used extensively for science. The notations in languages like C++ or Java are descended from those in FORTRAN.
It would be really nice if you could preview latex output before you post.
I hope the recent elections in the US will mean a return of compromise and pragmatism.
Ditto for the shitshow that is Brexit!
Sorry, it is about to get worse in the US.
Let’s hope so.
Strange things already happening, the caravan has disappeared from the news. Another victory for Our Leader.
I’m hoping against hope the old saw “things have to get worse before they get better” turns out to be true.
newton,
Things have to get worse before they get better! How much worse can it get?