It takes a lot of patience, and good IPM, but living soil has been awesome to me. We often say we grow soil, rather than plants, because of the fact a healthy living soil will provide all the nutrients the plant needs, and all you have to do is water essentially. Speaking strictly from a marijuana perspective (the limit of my experience with living soil), it makes the marijuana taste better and have a better smell. I've seen two different growers, starting with the same clones from the same mother planet, where one grows in soil using bottled nutes, and the other does living soil. The living soil definitely was superior.
With that said, living soil is not practical for everyone. Trying to do it indoors where you also live could create issues if you do not have a good IPM strategy, and fully understand the soil food web. "Teaming with Microbes" by Jeff Lowenfels is an excellent book to learn about how the soil food web works.
Hamming (of Hamming codes) has a famous Bell Labs talk "You and Your Research", describing how to have a large impact. It covers a lot of the same ground, but in more detail:
How about having lots of `brains?' It sounds good. But great work is something else than mere brains.
The people who do great work with less ability but who are committed to it, get more done that those who have great skill and dabble in it.
The prepared mind sooner or later finds something important and does it. So yes, it is luck. The particular thing you do is luck, but that you do something is not.
One of the characteristics you see, and many people have it including great scientists, is that usually when they were young they had independent thoughts and had the courage to pursue them.
Phil Bagwell's VList data structure seems really interesting as a way to avoid a bunch of the pitfalls of naive lists performance-wise - the paper's good fun and there seems to be an implementation in racket for anybody who wants to play with it.
For anyone not in the loop, Norvig was the author of Paradigms of Artificial Intelligence Programming. This was a substantial contribution to the field of educational computer science literature, and helped to kickstart the idea that the way to learn is to read, not just write.
It's also one of the few AI books that isn't rooted squarely in Algol. It's written with fairly decent though not always portable Common Lisp, just like most Common Lisp books of the era.
For a native web copy, abuse Safari Online's free trial. It's what I assume everyone else does when they want to read a niche technical book that O'Reilly put out but doesn't print another run of.
Depending on the day, the book ranges anywhere from $2 to $60 on Amazon, used, if you want a hard copy.
(Edited to fix the very butchered title that I wrote in error initially.)
I'm a level above an amatuer in this, having studied data assimilation in college (I was a Math major). Numerical errors are a given, errors in general too, actually, especially in the measurements used to train the hidden parameters (it is very similar to supervised ML). Data assimilation is a collection of techniques used to tackle that issue: 3dvar, 4dvar, kalman filter, extended kalman filter, ensemble kalman filter, particle filters, and so many others are used to find the most likely (minimum energy
etc) hidden state (the mean and covariance) from a given set of measurements+associated covariances and then that resulting hidden state is used to run the model "forward" (generally in time). AIUI, DA was specifically developed for weather modeling. It is definitely true that weather modeling is a heavy employer in anycase. This is on top of the methods used to solve PDEs ie so that they are forward/backward stable etc.
The errors in the weather forecast are not the result of chaos: they are the result of the errors in the measurements (recorded in the observation covariance) and the sparsity of the measurements themselves vs the size of Earth, for example, and limitations in model resolution (consider a FEM grid over the entire surface of Earth). The effect of chaos just compounds these errors near bifurcations around fixed points.
Perturbations are not used in the way you think; think of Taylor series approximations around specific points of interest.
Anyway, I work on compilers/auto-vectorization now (lol), so I'll defer to The Expert, if such person wants to chime in.
> 2020-2021 has shown me that most people happen to be 'fair weather fans' of civil rights.
This is always how it's worked. The idea that protecting the rights of your enemies can be salutary relies on too many complex concepts for the majority of people to have any hope of grasping it: burning the commons, collective action, norm evolution, meta-level thinking[1], modeling counterfactual worlds (eg one a future where your favored ideology is not dominant and is in need of the protections you are currently burning down).
The average person is nowhere near smart enough to be able to put these pieces together into a coherent worldview, let alone one that they find more convincing than "they're the enemy, crush them". The periods where liberalism has been resurgent are not ones where the masses are suddenly enlightened, but ones in which they either have little power or are pacified by unrelated conditions. This is not unlike the conditions in which dictatorships are stable, as the common thread is simply "the masses can't or don't care in detail about the fundamentals of the way they're governed". It's not a coincidence that the global illiberalism surge coincides with the rise of universal connectivity: On top of the social and economic upheaval that it induced, suddenly large amounts of people can coordinate epistemically, through hashtags and reshares, without making their way through distribution chokepoints controlled by elites.
[1] I couldn't think of a concise way to phrase this, but I'm referring to the tendency to claim that a big chunk of your beliefs/preferences are incontrovertible and fundamental tenets of society while others' are simply their beliefs and preferences.
I recently read "Software Engineering at Google" https://www.amazon.com/Software-Engineering-Google-Lessons-P..., and my take away was that engineering culture matters a lot to drive any success and innovation. Those who work there have the freedom to challenge anything, to steer the product, and as a result, find opportunities as described in the previous comment and fix it. The management works on shaping that innovation.
There is a graveyard in most Telecom company. ATT bought ad tech business AppNexus and media giant WarnerMedia back in 2018 (stone age of the internet) and didn't manage to operate it well, it's not the only one, Verizon has heavily in ad tech and media companies such as Yahoo!, but none of these companies brought the same internet culture shifting in the market as Google/FB/.. etc. We all love to hate social media companies, but these companies managed to tap into the opportunities presented and -arguably- didn't stop innovating.
It is written by a mathematician who has taught in Russia, the US and Brazil. They have a lot to say about how math is taught in the US. The paper also has a lot of sample problems.
The corporate media is good at memory holing things, but here's a compilation of them calling it the "Wuhan Coronavirus", "Chinese Coronavirus" and "China's Coronavirus":
I just want to provide you with another data point. I may well be in the minority, but nevertheless I am a bonafide contributor and enjoyer of HN. I should also point out that I did not and would likely not flag posts about the Tienanmen Square Massacre. However, there is a sense in which I could be construed as being "sympathetic" to the CCP and I'm 100% British. I am not sympathetic in the sense that I support the murder and cover up of protesting students, rather I am unsympathetic to the idea that the CCP is motivated by a cartoonish idea of despotism. Whilst I do believe that the most significant component in the censorship of Tank Man is formal CCP policy, I don't believe that extends to the wider conversation. My feeling is that the majority of non-Chinese commentators are disproportionally informed on Chinese history compared to Western history. For example the Opium Wars, or why Hong Kong was a British colony, or the effect the Century of Humiliation had/has on the Chinese collective consciousness. Again, to make sure I'm clear, there is no possible justification for state-sanctioned murder. However my "sympathy" is that we are failing as intelligent and caring fellow human beings if we cannot imagine the CCP as anything other than a Hollywood-esque super villain. So whilst I know I push against these stereotypes, and I suspect so too do Chinese nationals, that doesn't mean I also push against attempts to censor The Tienanmen Square Massacre.
So I'm not trying to disprove your point that censoring is actively occurring, I just want to try to add some nuance to the situation, that there are other currents at play here. And that when sweeping sentiments are expressed and reinforced about China there's a danger those other currents get drowned out and even misconstrued. So if I'm interpreting @dang's intentions correctly, I very much support his attempts to navigate this almost impossible situation with fairness yet firmness.
We have solutions to this for most (maybe all) contexts, at least as far as agriculture is concerned, in regenerative agriculture[0]. It requires a paradigm shift from the dominant type of agriculture taking place in the US and many other parts of the world, but it is not only better for soils and ecosystems – it's probably much better for our health as consumers, potentially more profitable for farmers, and can have an enormous positive impact in reigning in climate change.
It can seem too good to be true, but consider that when we mess with one aspect of nature by say, growing mono-culture crops rather than poly-cultures, we end up needing to intervene in many other aspects of ecosystems. The choice to annually till fields and grow acres upon acres of single crops year after year means we also need to recreate (poorly) natural systems of fertility (fertilizers), competition defense (herbicides), pest defense (pesticides), and disease defense (fungicides).
This model has other negative externalities as well, such as soil erosion, poor natural water infiltration and holding capacity, leaching of fertilizers and *-icides into surrounding areas and bodies of water which interrupt natural processes and are detrimental to life and health elsewhere, poor carbon sequestration, etc, etc.
The core principles of regenerative agriculture are to mimic nature and take advantage of natural processes rather than fighting them at every step, and to avoid intervening whenever possible. In the same way that the fastest and most maintainable line of code is the one you don't have to write, the cheapest and most effective agricultural practice is the one you don't need to do.
For those interested in a deep dive with a farmer from North Dakota who is very successfully putting these ideas into practice, see Gabe Brown[1] (who I first heard about here on HN, and for which I'm very grateful). If you're interested in agriculture, farming or even just gardening, I highly recommend watching the whole video. For a higher level and more accessible take, check out the documentary Kiss the Ground[2].
I've been trying to put these ideas into practice in my own ~100sq ft of growing space in NYC and, while it's still early in the growing season and hard to distinguish variables year-to-year, my garden is looking dramatically better this year[3] than at the same time in years past. Some of it for sure is a better sense of timing on my part and early warm weather here, but there are a lot of promising signs and I'm looking forward to seeing the state of things towards the end of the season when last year I had lost most of my tomato plants to pest pressures.
Edit: The article specifically mentions regenerative agriculture, but says "pesticide companies know that these practices are often accompanied by increased pesticide use." I think it's a misnomer to call something regenerative agriculture when pesticide use is even higher than in conventional agriculture.
The flags look normal. All but one or two of the users who flagged it are legit HN contributors who have been here for years. We can only guess why users flag things, but if I look at their flagging histories it seems clear that, as usual, (a) some have a strong view about the topic which causes them not to want the story to be on HN, while (b) others just have a problem with sensational/inflammatory topics or otherwise feel the post wasn't in keeping with the site guidelines. In my experience, the A-flags usually aren't enough to win out over upvotes on a story; it takes a coalition between A-flags and B-flags.
HN's front page is mostly determined by a tug of war between upvotes and flags [1]. It's common, indeed typical, for a sensational story to get a lot of initial upvotes, make the front page, and then provoke a "WTF why is this on HN" reaction from others, who flag it. With enough of the latter, the story falls off the front page, leading to a wave of "WTF why is HN censoring this story" from the first crowd. This is the cycle of life on HN. Recent example: https://news.ycombinator.com/item?id=27240048.
We do sometimes intervene to switch off flags, but only when a story is intellectually interesting and contains enough significant new information (SNI) to create conditions for a substantive discussion [2]. I considered doing that in this case but decided not to, because (a) it's not obvious that this is SNI, and (b) there have been several lab-leak threads recently. The most important thing to understand is that interestingness decays under repetition [3, 4].
If there hadn't been major threads on the topic recently, would we have turned off flags on this one? Well, the odds of that would be higher—but I still think probably not, because the new information in the story probably isn't enough to support a substantive discussion. If you look at those past discussions, you'll notice that in nearly all the cases, the articles themselves were among the most substantive ones that exist on the topic, and in most cases included SNI.
My GP comment had two purposes: it points people to interesting relevant discussions, but it also pre-empts the objection "WTF why is HN censoring this story". Users who post the latter have usually not yet learned to use the search box at the bottom of every page:
https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu...
This is what Trump meant by "fake news". As much as the HN crowd hates him, he's right about that. We don't have a regular report-the-news media in the US anymore, we have a propaganda arm of the authoritarian left and a satellite network of reactionary right wing outlets that are about as bad. Glenn Greenwald has had some excellent writing about this recently.
My general position is that if something is reported by a mainstream media outlet, there's a good chance that the opposite is true.
A more modern analysis can be found in The Revolt of the Public by Martin Gurri.
His hypothesis is that what was once an information trickle has become a virtual tsunami with the internet + cell phones + satellite television, etc. Governments have no control over the flow of information, which they had at least a semblance of pre-2000. This wave of information has not only exposed the worst excesses of the elites, but has also exposed the enormous gap between their authoritative promises and the actual results they produce.
This has pissed off a lot of very entitled people, who don't take the fact that the gap has always existed into consideration, who for historical reasons place very high expectations on government, and as a result attribute bad intentions to the previously mentioned poor results.
Not only is the media courting those people, they are made up of those people. So you get a media that just heaps negation on even the smallest failure of government. It's not just for clicks -- they are true believers in that they think they're doing the right thing.
In graduate school, I worked for 2+ years before my first paper was published. In that time, I passed my PhD qualifying exam, took classes, wrote code, read papers, learned math, and so forth. Yet when I applied for internships, I received no interest from employers. I suspect this was because I had no concrete signal that I knew anything in my field.
While working on my second paper, I started blogging. In the language of this article, I started generating public intellectual capital for myself. I have definitely experienced the effects of this capital on subsequent job hunts. Now I can point people to my blog to demonstrate knowledge, technical skills, and communication skills beyond the scope of my peer-reviewed work. Furthermore, there is no question about who contributed to my blog, and when I learn something new, I can externalize that quickly.
> Equally right or wrong is he who says that Napolean went to Moscow because he wanted to, and perished because Alexander desired his destruction, and he who says that an undermined hill weighing a million tons fell because the last labourer struck it for the last time with his pickaxe. In historic events the so-called great men are labels giving names to events, and like labels they have but the smallest connection with the event itself.
These food delivery apps tend to execute it poorly, but multiples orders in a single run is pretty standard fare in food delivery.
If you order delivery from a pizza place during lunch or dinner rush, the vast majority of the time the restaurant is going to try it's damndest to dispatch orders in a way that allows the driver to take multiple deliveries in one run. Even if it means letting one that's ready to go out the door sit under the heatlamps for a bit while you make and cook another that's going in the same direction. That said, your order goes from oven -> metal table with heat lamps -> insulated (or with an actual heat plate in it) delivery bag. So it should still end up getting delivered hot, even if not quite fresh.
I managed a Dominos when they started their pizza tracker[1] back in the day, and 99% of complaints were from online orders, which got quoted wildly optimistic delivery times and coupled with the following the progress tracker, gave a false sense of when to expect your order. Dominos recently updated their pizza tracker with driver GPS tracking[2] so that it works on par with the delivery apps, and I can't even imagine how many complaints come from people who are second on a driver's run and notice the driver taking off in the wrong direction from the store or taking a highly roundabout route to their house.
I recently re-read Noah Smith's essay "Leaders Who Act Like Outsiders Invite Trouble" https://www.bloomberg.com/opinion/articles/2020-03-03/leader.... It's not directly on point, but one concept is: "This extraordinary trend of rank-and-file members challenging the leaders of their organizations goes beyond simple populism. There may be no word for this trend in the English language. But there is one in Japanese: gekokujo." And later, "The real danger of gekokujo, however, comes from the establishment’s response to the threat. Eventually, party bosses, executives and other powerful figures may get tired of being pushed around."
Institutions are being pushed by the Twitter mob, and by the Twitter mob mentality, even when the person is formally within the institution. And I think we're learning, or going to have to re-learn, things like "Why did companies traditionally encourage people to leave politics and religion at the door?" and "What's the acceptable level of discourse within the institution, before you're not a part of it any more?"
I think one of the root causes of this is the increasingly partisan behavior of the "mainstream" media. People are sick of having their opinions dictated to them, they're sick of media outlets that all push the same narratives 24/7 (Russiagate, anyone?), and they're desperately searching for any outlet that will objectively report "all the news that's fit to print" (as the NY Times likes to say).
In the past year we saw Andrew Sullivan leave New York Magazine[1], Bari Weiss leave The New York Times[2], Matthew Yglesias leave Vox (which he founded) [3], and Glenn Greenwald leave The Intercept (which he founded) [4]. All of them had one thing in common that Greenwald summed up well: "The same trends of repression, censorship and ideological homogeneity plaguing the national press generally have engulfed the media outlet I co-founded, culminating in censorship of my own articles."
The mainstream media has created an information vacuum, and the spread of disinformation is a direct result of that.
I am an engineering manager in a large ecommerce company, overseeing machine learning and our in-house A/B testing framework.
One of our big tenets in my organization is that null-hypothesis significance testing and any associated methodologies (random effects, frequentist experiment design, and various enhancements to fgls regression) are simply not applicable and not useful to answering questions of policy (any kind of policy, which subsumes pretty much all use cases of running tests).
We take a Bayesian approach from the ground up, put lots of research into weakly informative priors for all tests, develop meaningful posterior predictive checks and we state policy inference goals up front to understand what we are looking for (predictive accuracy? measurement of causal effect sizes? understanding risk in choosing between different options that have differently shaped posteriors?).
which discusses “type m” (magnitude) and “type s” (sign) error probabilities in Bayesian analyses, and how that can provide some benefits and flexibility that NHST methods a cookie cutter power designs cannot.
Your mileage may vary, but my org has found this to be night and day better than frequentist approaches and we have no interest in going back to frequentist testing really for any use case.
It looks to me like so many are bothered by Trump because he didn't wrap all the lying and cheating and manipulation in a likeable, family-friendly public persona, like the presidents of the past did. He certainly said a lot of outrageous or false things, but policy-wise he didn't really make any horrible decisions.
So many of you want to desperately believe that the abnormal situation in the US can be pinned on one man and that once he's gone everything will go back to normal, or if not the situation will normalize to a certain extent. That just doesn't seem possible.
Trump is just the convenient scapegoat for the disastrous globalisation which hollowed out American industry, for the race but especially class conflicts which were never really settled and for the ultra-aggressive capitalism which creates many more losers than winners and for the two party system which is fundamentally vulnerable to corruption.
We have to find a way to come together and just assuming half the country is terrible doesn't lead anywhere good.
Reasons people might support Trump in spite of his many many shortcomings:
Trump has generated a middle east peace deal with UAE, Saudi etc. recognizing Israel which helps pave the way for stability in the region.
We have not entering any new proxy wars during his reign.
Trump banned lobbyists from the white house (although this one turned out to be a net negative because then the much of the white house staff members became defacto lobbyists--for instance Michael Cohen was paid around 10 million from various fortune 500 companies to help bring their interests to the president)
His stance on not murdering babies (which is how many prolife members see abortion, your mileage may vary--I am pro choice myself).
Trump has appointed many "true conservative" judges, so if you consider judicial appointments extremely important and are conservative you might support him for that reason.
What about the fact that Richard Spencer, who started and led the Jews will not replace us rally and was the leader of the white nationalist movement at Charlottesville is now a CNN contributor and endorsed Biden. And the fact the president did condemn them, repeatedly.
Trump passed the most comprehensive criminal justice reform bills of the last 20 years.
2020 Trump has the best percentage of vote share among republican presidential candidates from Hispanics, Jews, and African Americans of the past 50 years so maybe he isn't quite as racist as he seems.
You could believe that school choice is the most important civil rights issue of our time.
You think that the government has not right to be mandating lockdowns, school closures, etc. And have seen many instances of overreach from democrats.
Very low trust in the media and other "ruling" elites whom Trump seems to be constantly embattled.
There are of course lots of reasons to support Biden.
As VP he oversaw a very successfully timeframe of American history, with economic growth, political and social stability.
Trump has a tendency to say many things which are dog whistles/outright racist.
Trump speaks about women in derogatory ways.
Instead of bailing out local governments and schools during the coronavirus, Trump was pumping money into large corporations.
Biden takes climate change seriously. We are likely to see a green new deal.
Biden is likely to increase social safety nets during a time which is likely filled with economic uncertainty.
Biden isn't doing saber rattling with China threatening a second cold war.
Biden/Obama created a peace deal with Iran with the potential to bring stability to the region.
Its hard to judge policy, but Biden's character is much much more appealing than Trump's.
Biden is likely to bring a sense of dignity back to the office.
Biden is not a divisive figure and can help heal the country.
Biden is likely to bring police reform.
Trumps response to the coronavirus has been terrible. He basically left it up to the states instead of showing leadership. Moreover he was constantly spouting whatever nonsense it was that he was spouting.
Biden is likely to try and do something about the growing inequality in the country.
Biden has shown he can lead competently.
It's hard because conservatives and liberals tend to not only have differing personality traits, they tend to have underlying philosophies, priorities and thought patterns (personal anecdote not science). But we have to find ways to love each other despite our differences. Much love, I hope Biden does great things for the country and the next four years are prosperous, full of love and happy.
As non-white person I find the average white liberal MORE racist than your stereotypical redneck/Trump Follower.The last one is in your face, but most of the time the racism comes from ignorance or fear, cure them and you will go a long way to diminish it.
The white liberal is way different, they see themselves basically as a superior person, the summit of human ethical achievement. They interject themselves into minority problems to show how much "they care" but 99% the acts are only to "virtue-signal", when push come to shove they will quickly align with any policy that makes their life easier, no matter the impact on the world.
More insidiously, the supposed tolerance and open-mindeness are very quickly thrown out of the window once a minority sub-group fail to align to their worldview. See for example the vitriolic insults received by Cubans/Venezuelans in Florida just because they "dared" to vote for Trump.
If I were to live in America, I think I would happier and more accepted among the "racist rednecks" in Idaho or Alabama that among the NY/SF liberal crowd.
> The racial resentment scale has been criticized for not separating racism from ideas like conservatism or individualism. Some political scientists have attributed Republicans' higher resentment scores to the fact that they typically favor less government intervention
Well that's not surprising considering the nature of the questions. Damn your notion of racism is strange, Americans.
With that said, living soil is not practical for everyone. Trying to do it indoors where you also live could create issues if you do not have a good IPM strategy, and fully understand the soil food web. "Teaming with Microbes" by Jeff Lowenfels is an excellent book to learn about how the soil food web works.