fromthestart

A couple issues with the science here:

>Between 2005 and 2015, global rates of fracking went from producing 31 billion cubic meters per year to 435 billion

>In the last half of the 20th, century levels of methane in the atmosphere rose. They then plateaued, and spiking in 2008.

So the shale gas revolution started in 2005, but the spike didn't occur until 2008? Also it isn't clear if this was a transient spike or the author is referring to a sudden and sustained increase.

> While methane released in the late 20th century was enriched with the carbon isotope 13C, Howarth highlights methane released in recent years features lower levels. That's because the methane in shale gas has depleted levels of the isotope when compared with conventional natural gas or fossil fuels such as coal, he explained

First, all conventional gas is "shale gas," except for a tiny minority of highly atypical cases where reservoir sections have experienced prolonged high temperatures and matured further. Thus almost all natural gas ultimately comes from shale - fracking is merely extracting it at the source, where it does not significantly differ in composition from gas which has migrated to reservoir in conventional extraction. So, second, I'm suspcious of the use of c13 to differentiate fracking and non fracking sourced methane, but I'm not a petrophysicist. It should be related to the age of the fluid, and I don't expect shale gas to be substantially younger than reservoir gas in most cases.

If we allow entry of students from underperforming schools into colleges with historically high academic requirements, one of two things will happen, neither of which are particularly good:

1. These students will struggle and be much more likely to fail out, depending on how much of a boost they're given by the so called adversity score

2. Standards will be lowered and as a result graduates will be less competent and schools lose prestige and/or society suffers in the long run from workers who previously would not have qualified for their degrees.

You cannot undo the damage of 12 years of poor schooling by simply throwing a student into a more rigorous curriculum, no matter how admirable the intent.

And arguably point 2 has already been happening for some time anyway given the swelling of college ranks over the last few years and the corresponding devaluation of the college degree. More people may hold degrees, but the normal distribution of intelligence has not changed, which must indicate that the rigor of the average degree has dropped substantially in the last few decades given the proportion of the population that is now educated.

Well, no, and here is the Crux of the issue with affirmative action, and possibly one of the strongest base drivers of the cleavage between the American left approach to race and equality, and the American right approach to the same.

These policies attempt to achieve equality of outcome based on the implicit presumption that given equality of opportunity outcomes will be equal among various demographics. Thus any systematic inequalities are, without much in the way of hard evidence (a limitation of the related fields of psychology, economics, and the like) immediately explained by uncompetitive social norms of successful groups, i.e. racism/discrimination - unlocking the justification for penalizing those on top to equalize outcomes for those on the bottom.

The result of this implicit and fundamentally unproven assumption is the conflation of equality of outcome with equality of opportunity by the modern American left. This is a point of consternation for those near the center and leaning right primarily because it punishes success without allowing for the possibility of personal/internal explanations for discrepancies, i.e. cultural factors independent of poverty which limit economic success, and/or things like proven heritability of intelligence. Further, one could argue that such a conflation optimizes economic equality over merit, which is probably worse for society overall.

Perhaps because vegans and vegetarians are substantially more likely to be thin and, well, less traditionally manly, both in appearance and in values. It's a different set of cultural values and it isn't invalid just because it differs from yours.

The big deal is the fact that being told that the "flavor won't be missed" or that I won't be able to tell the difference is condescending.

>we use a wonderful Italian caterer, so flavour isn't going to be missed

A non meat dish is still going to taste like a non-meat dish. Perhaps your colleagues were just being polite and avoiding treading into politically charged territory after you brought up the Amazon.

Personally I'm a picky eater and vegan/vegetarian dishes simply do not compare to meat. I'm sure others feel similarly. Though for one meal I suppose it isn't a huge deal - but acting like it's just as good simply isn't true for the majority of people - that's your personal taste.

Not to add to the sensationalism, but it is worth noting that U-238 is probably dangerous to ingest moreso because of its behavior as a heavy metal rather than it's radioactivity.

In which case because of the bioaccumlative properties of heavy metals, drinking 10x the limit as you say may not actually be so harmless. But I don't know what safe limits really are.

Wow, side note, compare China/SE Asia PM2.5 to the rest of the world.

>The fact that you go from one no-context quote to "silencing conservative perspectives" is not good. Please keep thinking critically

There's more context than that single quote, including the other strong evidence of overt bias in other sources, like OPs article. I get the feeling you didn't watch the video.

You don't debunk something by saying it's debunked. So her claim is that because she was no longer in the trust and safety team at the time of the video, her opinions are irrelevant?

Of course she's going to claim that the video was "heavily edited." But the validity of the video comes from the fact that there is no context which makes the clips of her speaking any less damning. They clearly indicate a willingness by Google to curate search results.

>The video then goes on to stitch together a series of debunked conspiracy theories about our search results, and our other products.

I'd love to see this debunking.

>Google has repeatedly been clear that it works to be a trustworthy source of information, without regard to political viewpoint. In fact, Google has no notion of political ideology in its rankings. And everything I have seen backs this up. Our CEO has said ”We do not bias our products to favor any political agenda.” He’s somewhat more powerful and authoritative than me.

Oh, well, if the CEO says it, it must be true. This is just a weak attempt at damage control.

The hidden camera captured an interview with Jen Gennai, Head of Responsible Innovation, Global Affairs at Google. You could at least watch the video and find out for yourself.

I think the real problem here is that people are aware of Google's bias and OK with it, but unwilling to admit as much.

Where do you go for official sources when the entire modern media establishment has an undeniable (and rather open) liberal bias?

Ignoring bias, do you believe that every given situation that some group may identify as a problem is guaranteed to have a reputable source reporting on it?

I'd love to find a more reputable source, but one is not currently available, and I'm not in a position to create one. That does not imply that the problem being reported on is illegitimate. You should be able to assess the contents of the video with PV's bias in mind - and most of the content really does stand on its own, especially in the context of relevant quotes from employees in OPs article.

There was also a massive trove[1] of internal Google documents dumped on PV related to, among other things, AI development and arguably politically motivated guidance by internal definitions of "fairness", designed to shape search results. Should these documents also be ignored because they're only being hosted by Project Veritas? You can see them yourself...

Ironically, this is the danger of trusting only select biased organizations with curating the flow of information - as you suggest, to most people, if a "reputable" source hasn't reported about it, it doesn't exist. How many years has 4chan been screaming about epstein, as another example, without major media investigation?

The bottom line is that no entity that acts as the front page to the internet should be curating information flow without a massive, blinking disclaimer.

1.https://www.projectveritas.com/google-document-dump/

People are rabidly dismissive of the project veritas expose, but articles with quotes like this repeteadly show pervasive bias among Google staff which aligns exactly with the accusations and candid management recordings within the video [1]. The video is admittedly obnoxious and overly dramatic, but the candid clips with exec Jen Gennai strongly suggest that these political biases are leaking into technical work with an intended effect on society.

1.https://www.projectveritas.com/2019/06/24/insider-blows-whis...

Because Google is not a publisher, and selectively choosing how to present information according to the predominant internal politics is unethical without a disclaimer, which currently does not exist, given Google's status as the defacto gateway to information on the internet.

Curating information in such a manner is, ironically, marginalizing the groups who's views are being suppressed. Information should flow freely for people to make their own decisions. This is a dangerous amount of power for an increasingly biased private entity, particularly given certain choice quotes in this article and others regarding googlers' views on the current presidency.

Well here's an interesting thought. Let's say you train a generative network on illegal imagery. Is the combination of network and weights illegal to posses? What about images it generates?

>What is new is the hype around deep learning that took off after 2012, and because Google and Facebook decided to champion it.

Have you seen what neural nets are now capable of? Speech synthesis/transcription, voice synthesis, image synthesis/labeling/infill, style transfer, music synthesis, and a host of other classes of optimization problems which have intractable explicit programmitic solutions.

The hype is justified, because ML has finally arrived, thanks primarily to hardware, and secondarily to the wealth of modern open research, heavily influenced congregations of leading researchers enabled by funding at Google, Facebook, etc.

The problems being solved by "curve fitting" ML were simply unsolvable by any practical, generalizable means before recently, and the revolution is just getting started.

Whenever possible I use the burner app if I need to provide a number to sign up for a service that I'm not particularly worried about losing access to, that sends a confirmation code on login. Especially when its something I only need to use once, like anonymously downloading a file from a website that requires a Google or Facebook login, for example.

Project veritas leaked a candid, hidden camera chat with Google's Head of Responsible Innovation, Jen Gennai, where, essentially, she and an engineer shamelessly describe using ML to manipulate search results to better align them with the reality that they would like society to adopt, i.e., a strongly left leaning agenda. Among other topics, she specifically mentioned that Google is working against "another 2016 situation" (regarding the election of Trump). The clips in this video stand on their own-there is no context which mitigates the fact that Google is injecting political bias into its search results. In an interview with a whistleblower in the same video[1], claims of blacklisted autocomplete terms are also implied to be part of search curation.

Given Google's status as the defacto gateway to information for the majority of the first world, this kind of meddling with queries and query results should be cause for concern, regardless of one's political leaning. Ironically, the video was taken down from YouTube.

Note: I don't appreciate the bias of the following source either, but the content is damning on its own.

1.https://www.projectveritas.com/2019/06/24/insider-blows-whis...

>Your reply says that you meant to say that Climate Science "cannot be experimental". This, it turns out, is also not true. There are cases where nature runs the experiment for us.

>For instance, the large methane leak recently in the LA area allowed us to understand the effectiveness of the inversion algorithms that are used to turn observations of concentrations into observations of the emissions that caused those concentrations. In other words, to understand the correctness of wind models.

This is exactly what differentiates model driven from experimental sciences. You cannot go out and define an experiment to confirm your claim, you can only make observations of natural phenomena. Which can take decades when you're considering sweeping, global changes, particularly when you're measuring changes which occur on scales of decades-centuries.

>You said that "no one knew" about a possible future crisis in the 1980s. I supplied links showing that this is not the case -- the science establishment, the Congress, and the readers of the front page of major newspapers knew.

Once again you are twisting my words. My point is to differentiate between knowing that something is happening with certainty, e.g. a consensus that climate is indeed changing due to human activity, and knowing of the possibility of something significant occurring, requiring substantial evidence to match the claim. It took decades to gather enough evidence to match the scale of the claim of global climate change.

>Your reply counters with "what should we have done". That's another question, isn't it?

Again I've done no such thing. My point is that we did what we should have - it is not rational to make sweeping changes to society, culture, and economy for every alarmist prediction uttered by a scientist, because such changes, big or small, are not free. One requires a minimum degree of certainty and, considering the magnitude of the implications of climate change, from the perspectives of both consequences and mitigation, it isn't unreasonable to say that 30 years is a relatively short time to establish a rigorous scientific consensus given the age of the modern scientific establishment and the amount of data that needed to be collected and analyzed because of the purely data driven, non-experimental nature of climate science.

I didn't see it defined on the page, so I'm wondering, do zero click searches count frustrated users giving up because the first two pages are entirely full of SEO optimized commercial websites when searches are made for informational content?

Try searching anything related to researching foundation repair. You get pages upon pages of foundation repair services and biased one pager summaries about certain procedures.

Long gone are the days when search queries retrieved educational content - when was the last time you saw results for engineering toolbox, or hyperphysics, or anything other than ads, commercial services, and stack overflow for software related questions?

SEO killed search and Google is complicit. I think this has massive ramifications for society - the sources of information are shallower, and people are increasingly unable to notice differentiate objective knowledge sources from commercial websites. The people who are interested in generating and curating true knowledge repositories are probably not interested in paying for SEO, so their efforts are effectively hidden from what amounts to the gateway to the internet.

>terraforming" as a feel-good substitute for unplanned global climate change),

Not true, please don't put words in my mouth. I use the term terraforming to emphasize the magnitude of the claim. We are fundamentally changing the properties of an entire planetary atmosphere, inadvertently.

>and claims that are very strong, sweeping, and incorrect ("totally model-driven field").

Also not true. The point is that climate science by nature cannot be experimental. We can only make predictions and then proof amounts to measuring how well decades of measurements line up with predictions/models.

>Another correction, relating to "no one knew" about a future crisis in the 1980s. Here is a paper [3] from James Hansen in Science magazine, in 1981, that clearly warns of crisis-level consequences. From the abstract -- In the 21st century, we will see "creation of drought-prone regions in North America and central Asia, erosion of the West Antarctic ice sheet with a consequent worldwide rise in sea level, and opening of the fabled Northwest Passage." For whatever it's worth, this appeared on the NYTimes front page in 1981.

Again, given the non experimental nature of climate science, this was a massive claim which requires decades of data collection and modeling to verify before committing to sweeping social and economic policy reform which has costs of its own.

People underestimate the fact that climate change is one of the most radical ideas ever proposed by the modern scientific establishment, in terms of the scale of the changes, the disastrous predicted consequences, and the scale of societal change required for mitigation. We're talking about a potential irreversible, runaway global extinction accidentally triggered by humans 100-1000+ times faster than any measured natural rate. Due diligence, in a non experimental field, limited by 80s modeling hardware, took time. Meanwhile we all benefited from the fossil fuels we (and our parents) burned.

>I still find the fact that Exxon internal researchers knew about the incoming crisis since 1982.

I'll keep posting about it untill this hysteria dies, though I fear that given social pressures regarding anything representing "denial" will keep it alive indefinitely. No one knew about an incoming crisis in the 80s. People were aware about the possibility of climate change in the 80s. It took 30 years of data collection, research, and computational modeling on increasingly modern hardware, which allows for exponentially larger and more precise models, before we had amassed enough evidence in a totally model driven field to conclude that we may be terraforming on the scale of 100 years.

I wonder if this is a tractable reinforcement learning problem. The objective is somewhat clearly defined-minimize spatial deviation over time from the input path.

>Who on Earth is thinking that it is a good idea to take an EBS snapshot and make it public?

Maybe they're trying to reproduce functionality of docker? It would actually be extremely useful for research involving modeling/AI because you could trivially reproduce the results by bundling the exact code and data.

Edit: actually maybe I'm confusing EBS snapshots with AMIs...

I have read them. These was the theoretical work of a handful of geoscientists working from the computational limitations of 70s-80s technology, and substantially limited data compared to what has been made available in the decades since.

The historic certainty of climate change is being vastly overstated, neglecting the critical role that petroleum has played in the development of modern infrastructure and society in a blind rush to find some evil to blame. Science is built on skepticism.

And the head of the Manhattan Project had concerns that the nuclear tests could ignite the atmosphere.

Science is far more than theoretical speculation. Grand claims require development of supporting theory and evidence. Even the consensus regarding climate change has been rather quick to form considering the enormity of the claims and how young the field of modern science is.

>It's not ridiculous. I was taught about the carbon cycle, greenhouse gasses, and climate change in a United States public high school in 1993. 26 years ago.

Climate is a complex, chaotic system. Again, there was no certainty regarding warming without the decades of research, data gathering, and modeling that followed. Knowledge of the carbon cycle and greenhouse potential is not enough alone to prove that the planet is warming from human activities. And, again, 20 years ago the consensus was that of global cooling, to emphasize the degree of uncertainty in the historic consensus. You're underestimating the gravity of the notion that we are terraforming, modifying global climate, which normally changes over 10^4-10^6 years, on the scale of 10^2, 2-4 orders of magnitude difference. No time in history have humans even been able to do such a thing, let alone had they reason to believe that it were even possible.

>Sure, but society has provided massive subsidies (both direct and indirect) to carbon-based fuels. Had we not provided those subsidies, society would be much better off.

Much better off without fuel for local and global shipping and power generation and raw material for plastics?

>Exxon and other energy companies knew as long ago as 30 years that carbon emissions were becoming perilous to the planet.

This is ridiculous.

No one knew about climate change 30 years ago. A small number of scientists were aware of the possibility of climate change - it took some 30 years of research, heavily dependent on modeling capability which was only really unlocked in the last decade or so, to come to any sort of alarmist consensus, not to mention less than 30 years ago the consensus was one of global cooling.

Sure, in hindsight, it all seems obvious, but extraordinary claims require extraordinary evidence, and we're talking about terraforming a planet on the scale of 100 years. You'd be crazy to believe in such a thing without some hard data to back that up.

Meanwhile let's ignore that society, globally, has advanced - technologically and socially - on the back of petroleum fuel. We all may be paying a price, but we are all partaking in the reward.

The source you linked discussed reproducibility of modeling and analysis of historic data. It makes no mention of the problems responsible for the experimental replication crisis that I'm describing in other fields - it wouldn't make sense to because climate science is not and cannot be experimental. Which makes the science more vulnerable to bias because there is fundamentally no way to prove beyond statistical estimated whether it is right or wrong.

Why do you refuse to admit the possibility that political and social pressures in such a strongly politicized field can bias climate science? All of the ingredients are there, and the only reason such an assertion is contentious is because of these very same political norms. It begins to resemble dogma, when any criticism is treated with such disdain.

>You've failed to establish how the existence of gender studies departments means we can dismiss the findings of the physical sciences

I am not advocating for dismissal, I am merely suggesting that social and political pressure for certain results from the hiring and financial appropriation practices of a politically biased administration can introduce aggregate bias in published results. And on the subject:

>The majority of academic scientists identified as Republicans, up through the 80s

You've failed to establish how even a republican leaning scientific establishment is immune from the whims of administration resulting in, say, only publishing results that support the politically correct positions. It is a fact that the overwhelming majority University administration's lean strongly left - and when science is politicized, there's a strong chance that, again, such administrative bias will affect results in seemingly innocuous ways. Not to mention that most environmental scientists have personal left leaning biases and experience social and professional pressures which also may be reflected in results.

Significance testing, publication of only positive results, and model design are three methods by a which slant may be unintentionally introduced and, again, we know that these problems have lead to the replication crisis explicitly identified in other empirical sciences - why do you think no one is willing to ask the same, legitimate question about climate science, particularly when climatology by nature is not a reproducible discipline?