zippy5

My understanding is that it was more the 9000 bank failures effectively created a credit crunch. Like if a bank closes and there's no replacement, then most small business were unable to get loans. Farmers who couldn't afford to plant new crops, factories can't improve equipment, inventory get's squeezed across the supply chain, ect. Exports were about 5% of GDP, suggesting that maybe tarrifs may have been the trigger but weren't the primary cause of the depression.

https://www.sjsu.edu/faculty/watkins/depression.htm

But what if the DJ convinces his audience to invest in a much larger and more expensive night club at an extraordinary premium. Then it might make sense to buy the original night club just to protect the DJ’s scheme.

My math is that Musk’s twitter activity very reasonably increased Tesla’s market cap by more than 5% which is what he is offering to pay.

I mean, 90% of large cap investors have underperformed the sp500. It's not about optimal returns, it's about diversification and portfolio risk. The Saudi's want some of their money in things that aren't correlated with oil, and pretty much everything in the physical word is. Technology is one of the few things that may even be inversely correlated.

https://www.cnbc.com/2020/09/18/stock-picking-has-a-terrible...

I think you can decompose a calculus course into three key components; principle/concepts, proofs, and procedurally solvable math problems.

All three have value but clearly the math problem aspect of it has depreciated in value due to calculators, wolfram alpha, etc yet it tends to remain the focus of many math curriculums. Calculus by its nature is more computationally intensive, meaning that it has experienced the greatest decline. If you really think about it, that curriculum was designed for an era when we called human "computers". There is probably opportunity to make calculus a more broadly valuable class by deemphasizing the mechanics and focusing on the principles and proofs.

This is wonderful. I’m going to see if there is anything I can build for this.

I think it could be Nobel Prize worthy. Protein’s structure often determines its effect as a catalyst. So to map the DNA to the 2nd order outcomes seems like it could be the missing ingredient to controlling the properties of cells.

Personally I’m hoping that someone smarter than me figures out how to displace existing catalysts like platinum and palladium. Seems like it could be a pretty penny and some positive environmental impact to boot.

It seems to me that there are two ways to learn well. One is to have a carefully curated curriculum and the other is to have enough experience to parse the world by yourself. In humans we might describe these as knowledge/education and wisdom. I see data preparation as improved knowledge transfer and more training data as the path to wisdom.

Wisdom is usually heavily discounted by smart young people so I expect engineers will double down on better data prep than better data acquisition. Also which looks better on a resume?

I’d say that the first premise is objectively false. I think it’s much simpler. The volatility of asset has correlated narratives of hope and fear in the mind of the asset holder. These in turn produce a dopamine response, akin to mechanisms of a gambling addiction

He replaced his marketing department with a R&D department. It really doesn’t have to work, just keep the believers believing.

I want to apologize, I definitely don’t intend to fear monger and most definitely not want to imply that I have expertise. Roughly my level of understanding is mostly that of a low level undergrad and you should treat my naiveness as such.

I recognize that what I’m engaging in is entirely wild speculation based on limited experience and data, likely very error prone and that really I’m just having fun without considering how it may impact other readers.

I understand that for many this an important issue of health and research. I did not intend to detract from these more legitimate forms of discussion.

It’s not really a comprehensive interpretation of intercalation but I think a geometric interpretation can help some non-chemists understand how intercalating molecules bind to dna.

From the purely geometric model, some of the molecules you proposed have pretty large functional groups adjacent to rings which I think may make the intercalation process less efficient. That being said, if you took those molecules and gave massive doses to rats, some may comeback as carcinogenic.

I think that your multi-ring point is fair. The multi ring structure to me suggests that the more the pi orbitals are able to delocalize their electrons the higher the binding efficiency. I have tested 1-2 molecules where non-fused rings showed some affinity but not near the potency of fused ring structures. I would also say two rings with a carbon-carbon link seem to be potent binding as well. I presume that it’s also related to delocalizing pi orbitals and extra degrees of freedom in the intercalation process but I suppose that’s just speculative.

I used to work on DNA dyes. Typically when you see a 6 carbon ring with a chain of carbons attached, there is high probability of that molecule interfering with DNA replication.

Basically the mechanism works because the hexagon ring slides between the base pairs and this leads to a lowest energy state due to a phenomenon call pi orbital stacking resulting in the molecule getting stuck there. The carbon chain is mostly valuable in the sense that it distances the rest of molecule from interfering with the stacking process.

Take a look at ethidium bromide or pretty much any other intercalating dna stain and you’ll see similar characteristics. It’s also extremely carcinogenic.

https://en.m.wikipedia.org/wiki/Pi-Stacking_(chemistry)

It’s analogous to getting some cloth stuck in your zipper. Sometimes you can zip and unzip easily enough but sometimes it’ll get stuck. My understanding is that really DNA replication issues tend to be the root cause of some, possibly many cancers but really that’s outside my expertise.

So I would say that it is internally consistent with my limited knowledge of biochemistry that aspartame is carcinogenic.

I would strongly caveat this with saying that these structures occur in pretty high frequency across many forms of plant and animal life. Chemists in my lab used to joke about how potatoes contain 17 or so know carcinogenic compounds so why buy organic. My point is, if you go looking for correlations with cancer in many forms of food, you will find them.

I think for most people, aspartame is not likely to be major risk factor unless you are consuming it in extreme quantities and otherwise live a very healthy life.

My interpretation is that is why it’s so brilliant.

It’s incredibly simple for the end user conceptually but encapsulates optimizing processing across a distributed file system, fault tolerance, shuffling key value pairs, job stage planning, handling intermediates ect.

Hadoop a big data framework that reduces the level of competence required to write data pipelines because it was able to hide a massive amount of complexity behind the map reduce abstraction.

Id even argue that hive, snowflake, and other sql data warehouses have taken this idea further, where most sql primitives can be implemented as map reduce derivatives. With this next level of abstraction, dbas and non-engineers are witting map reduce computations.

I think my point is that abstractions like map reduce have had a democratizing effect on who can implement high scale data processing and their value is that they took something incredibly complex and made it simple.

I think the example of the railroad is biased by the exact same phenomenon. From 1900, In the next 10 years you would see the mass production of automobiles and the invention of the airplane. Basically it’s the story of disruption told from the perspective of the disrupted technology.

In the 1920’s standard oil subsidiaries were still an effective monopoly for petroleum in the us market. Therefore a reasonable proxy for the future profitability of the entire industry in that market assuming that their ruthless anticompetitive behavior allowed them retain their market dominance. Additionally they were profiting off the same disruption in transportation that you are citing, which as we are both acknowledging was massive.

The book Titan is awesome context for this. Wonderful read.

Unlike the technology such as railroads, natural resource commodities and vertically integrated supply chains tend to not be disrupted as easily (very unfortunate for us).

I’m not saying that it couldn’t have gone wrong, but clearly an asymmetrical risk reward at 3-5 PE. So in general you are right, but I think if you find a company that has a great business model, is a monopoly, and is disrupting a massive market, at reasonable price, you have a recipe for outlier returns.

I do think there is some real signal in this article in addition to the survivorship bias.

1) Noting that the stock market was boring I think is real indicator of the mass psychology of that time. There is definitely a inverse correlation between enthusiasm for markets and future returns.

2) Noting the returns of standard Oil is a reasonable take. There was a massive expansion of combustion engine production in the preceding two decades and inferring that this would be correlated with increased demand for oil based products is not hot take. Also it doesn’t take a genius to understand a oil is better business that automobiles, recurring revenue and all.

3) Tax rates have historically influenced valuations.

4) I’m not sure how to extrapolate the the German currency situation but I think looking at the relative attractiveness global markets makes sense.

I feel the opposite. My understanding is that BE-4 is hydrogen based rocket engine which really has never been executed successfully before where raptors are methane. Blue origin is going from 0 to 1 and space x is going from n to n+1. Obviously the short term results are going to be worse than Space X but I’m not convinced their engines will be worse.

That being said obviously ULA made the wrong deal since it’s not clear they will alive long enough to benefit from interplanetary refueling.

So for example, the author saw that supply chain team had difficulty managing the complexity and scale of their analysis in large part due to the scalability of their spreadsheet solution. I would have pushed them to use Airtable which is basically a more scalable spreadsheet. By choosing the data pipeline route, the people who understand how to improve the supply chain model and the history of decisions that went into it, as well as previous missteps, now have limited ability to experiment with improving it. In my experience, every rewrite of a system has something lost in translation which makes me think that in the authors example that the life of the analysts got better but may have made the quality of supply chain model worse.

In the long run, there is plenty of useful logistics software that should do everything they want but the most important thing is to empower the people with domain expertise in the data to be as close to the solution as possible. Better decisions are often a result of better information/experience than better analysis. Unfortunately I haven’t studied these vendors well enough to make any suggestions though I believe that the solutions are well defined enough to write textbooks on them, which suggests to me that existing software and I would mostly implement similar methodologies.

On the marketing and product analytics tools, I think 80% of the problems boil down to measuring conversion rates and the comparing those rates across different contexts to select for the contexts which improves those rates.

Another user mentioned heap, which is great product if you know you don’t know what contextual data is meaningful but you suspect that it’s partially in how they interact with other parts of your website. Personally I’d use heap judiciously since I suspect there will be limitations to how useful the historical data will be in the future and collecting everything is expensive. One limitation is that site interactions are only part of the potentially important context. Another limitation is that startups change rapidly, so their historical data often depreciates in terms providing insight into their current problems. For an extreme example, I’m sure zoom’s conversion data before and during pandemic look completely different. But even a small tweak to google’s search algorithm could totally change what type of customer finds your site.

Personally I’d advocate talking to customers, potential customers, and other stake holders to understand what is important and measure that. Most companies, currently do the opposite where they take a lot of measurements and then try to figure out what’s important. The first approach can probably be done in google analytics. The second I might try and use Amplitude which is I what imagine a tool like heap will eventually try to evolve into.

The hardest person to help with data in the organization is the CEO because really they use data as form sales tool and reporting. The closest I have seen a tool to doing this in a way the CEO could mostly self service is Sisu data. Though it’s the CEO so it’s probably reasonable to hire some help anyway.

Lastly data warehouses were the gold standard in the early 2010s but Presto is better fit these days for companies whose data is distributed across many different places.

So my assumption is that for a given business model, like e-commerce or Saas business much of the highest value analysis is fairly standardized and can be templated. For example breaking down conversion rate by weekly cohort is something that can be pretty easily be done in google analytics.

The problem with English to sql translators or most coders in general are the assumptions we make, in particular about the underlying data. For example, say we want a join two tables, so we write a query to join on two columns and often call it correct which it is from a logical or schema perspective it is. However, null values, defaults like 0, many to one relationships vs one to one relationships, issues with instrumentation such as networking timeouts or bot detection, etc all can impact the down stream metrics. My point is that when there are 500 lines of sql in a query such as those mentioned the article, there’s a lot of ways to be mostly correct but to cumulatively be wrong.

Like many popular enough open source tools, 3rd party vendors get battle tested, issues get found before you, and they can justify devoting more resources to rigorously ensure correctness than the average analyst has the time or energy todo because their business depend on you trusting the outputs.

I’m not saying you couldn’t do all this yourself. But given the sheer number of analytics tools that are reasonably priced, you might have chosen to spend your time on something more specialized like a recommendation system.

This was wonderfully written and if your gonna start a data team, this is how you do it. But I can see that I’m the only one who thought it was crazy to start a data team in the first place.

This company makes 10M and spends 3M on the team and infrastructure to make data a core competency?

A vast majority of wins discussed were lowly differentiated web / mobile / supply chain analytics which they could have gotten and setup with 3rd party software for an order of magnitude cheaper.

I can only imagine what this hypothetical startup could have learned if they spent that money actually talking to customers, and running more experiments.

I’ve heard people talk about data as the new oil but for most companies it’s a lot closer uranium. Hard to find people who can to handle / process it correctly, nontrivial security/liabilities if PII is involved, expensive to store and a generally underwhelming return on effort relative to the anticipated utility.

My take away was that startups benefit tremendously from a data advisor role to get the data competency, as well as the educational and cultural benefits, but realistically the data infrastructure and analytics at that scale should have been bought not built. Obviously there are a couple of exceptions such regulatory reasons like hippa compliance for which building in-house can be the right choice if no vendor fits your use case.

I somewhat agree that a gritty person shouldn’t keep a fast food job, but some of the Uber drivers I meet are incredibly gritty. They work incredibly long hours and grind on multiple apps. They have well thought out strategies for how to make as much as they can per hour and have plans for how they will invest their money.

I admire the heck out of em but I can see most people don’t want that life. I have no doubt that kind of work ethic could be wasted in the wrong environment but a successful person should have both grit and the ability to find an environment where they can use it to get ahead.

I’d even argue that corporate America as whole rarely rewards grit with a couple exceptions.

SSN were never really secret. For the most part the first 5 digits are a derivative of when and where you were born (public record) and you’ve given out the last 4 to every financial institution and employer.

You last part is spot on. Basically people should setup a password at the DMV or something.

If the people looking to learning about this subject have the “dark ages” misconception, is it so bad to make it easy to find via search and then explaining the nuances of the misnomer?

At quantum scale or near the speed of light newtonian mechanics breaks down, so we know it a local approximation. However, for set circumstances it approximates it does it well. For many decades physists searched universally true theory to no avail.

My point is this, IQ seems to have signal. It probably doesn't hold up in extreme circumstances. Psychology in particular seems to have a difficult time holding large enough studies to find universal constants that we should accept that IQ might be the best local approxination for now.

Your completely missing the point..

Honda sells commodities. Maserati sells status.

Motorola sells commodities. Apple sells status.

Computers are commodities, smart phones are commodities and at the end of day none of that stops people from buying from apple

Cars make a ton of sense for apple because it’s a product which a decent percentage of people will all pay a premium to get the cool version.

It’s one thing to pay a highly competent person 6 figures. It’s an entirely different thing if you can’t fire them. I’m not saying that the government doesn’t need more competent people but it’s not going to work if there’s a bigger carrot and no stick.

The other side is that the peak of American prosperity occurred post WWII where a majority of the developed world (any country that had experienced an industrial revolution), was blown to shreds. This gave the US roughly a temporary monopoly on manufacturing and exports. It’s hard to see a future where the dominance of American prosperity happens again unless that scenario repeats itself. Aka WWIII with the US coming out unscathed and the rest of the world destroyed.

No doubt your point of mismanagement and bad actors is valid but it’s important to recognize that much of the glory days of the American middle class corresponded to the height of many others misfortune.

I think it’s worth noting that popularity isn’t always deterministic or meritocratic.

I once read about an experiment where users would listen to music and and the platform counted the number of listens and showed users. When the experiment was rerun, a different set of songs from the first experiment were popular. I believe I read this in Invisible Influence by Jonah Berger.

The point is that most content platforms select for content that is sufficiently good (or clickable) but not necessarily the best.

Very few software engineer are working on something that 27m people want to experience per engineer on the project. Any engineer who can build something that 27m people want and get it to them can make far more than average.

Hypothetically, it seems that your problem has to do surfaces being reflective in the visible light spectrum. Do you have any luck with ultrasonic, ultraviolet, or IR sensors?

Also I was also thinking about how light source estimation might be viable work around for reflective surfaces. http://www.thomaswhelan.ie/Whelan16ijrr.pdf

I respectfully disagree. New technologies don't intrinsically expand the pie, lowering the price/performance ratio does. D2C and SAAS lower the cost of distribution which historically been a significant markup on goods. The consumer pays less and thus have more money to spend on other goods and services. Therefore the pie expands.

Examples of technologies failing to change price/performance:

1) Self driving cars: can you replace people working for less than minimum wage with the expensive hardware and liability related costs? Eventually yes but even airplanes are struggling and that is a far easier problem.

The counter example might be closer to Tesla's auto pilot.

2) DC electrical grid (as implemented by Edison): expensive to transmit over distance relative to AC.

3) Fuel cell cars: the platinum catalysts drove up the price. (There were other factors as well).

Examples of technologies succeeding to change price/performance:

1) Light bulb: factories could be productive even at night. Since most of the costs were fixed, getting more productivity was invaluable.

2) CPU's: Moore's law

3) Solar: Manufacturing costs were the real obstacle, and improved manufacturing and economies of scale led to an economically viable product.

I don't claim to be perfectly right about all these cases but I think they illustrate the gist of what I'm saying.