vobios

> just preserve it or lose just a little

Housing generally underperforms the overall market.

Does the data presented in that article really suggest that the apartments are actually empty or is it simply that it takes a while to sell out a new development? It's going to take a while to move a hundred units, especially if you want to increase the profits.

The StreetEasy report the article is inspired by actually shows that in 2019 there were more condos sold than completed.

> It was much better (lighter, simpler, quicker to load) than Java applets from the mid 90s.

People often forget about those.

Are there really that many empty apartments? I am sure they exist, but are there really enough of them to make a dent in the market? Having lived in the city for years, I've never actually seen one.

I also wonder if you have enough cash to buy and hold NYC apartments, wouldn't your money do better in some sort of a REIT?

There is a section that addresses that specific point:

> The Landmarks Preservation Board protects the facades of historic buildings, but not their interiors. Thus landmarking led townhouses once divided into multiple apartments to be converted into single-family mansions for the incoming super-rich—serving only to further inflame the speculative energies that skyrocketed the prices of the buildings they preserved.

Technically, Queens is part of Long Island.

Speaking of distinguishing subtle differences in blues, there is actually a linguistic component.

English and Russian color terms divide the color spectrum differently. Unlike English, Russian makes an obligatory distinction between lighter blues (“goluboy”) and darker blues (“siniy”). We investigated whether this linguistic difference leads to differences in color discrimination. We tested English and Russian speakers in a speeded color discrimination task using blue stimuli that spanned the siniy/goluboy border. We found that Russian speakers were faster to discriminate two colors when they fell into different linguistic categories in Russian (one siniy and the other goluboy) than when they were from the same linguistic category (both siniy or both goluboy). Moreover, this category advantage was eliminated by a verbal, but not a spatial, dual task. These effects were stronger for difficult discriminations (i.e., when the colors were perceptually close) than for easy discriminations (i.e., when the colors were further apart). English speakers tested on the identical stimuli did not show a category advantage in any of the conditions. These results demonstrate that (i) categories in language affect performance on simple perceptual color tasks and (ii) the effect of language is online (and can be disrupted by verbal interference).

Source: https://doi.org/10.1073/pnas.0701644104

I am looking here and it's great: http://ggplot2.tidyverse.org/reference/aes.html

This is super controversial, but no need to use Jupyter.

I personally find that Jupyter feels like a hack compared to something like RStudio. You have to open a terminal and launch a web server?

A non-insignificant fraction of the R community only exists because of Tidyverse.

> less well documented than most R packages

I on the other hand, find most R packages provide barely readable documentation. I can just hope that the vignette exists and actually explains the inputs/outputs.

By that theory, you should diversify away from those that are not going to be "superstocks" (in the post, Dell grew 550x and pushed up the entire index). For example, HP or Exxon or UPS will probably do okay, but will not outperform the rest of S&P by a huge margin. Therefore, they will only weigh down your big earners.

> you can research growth stocks (e.g. pharmaceutical companies undergoing FDA trials) and pick the promising ones

Those FDA trials are expensive. Those companies have many experts in the field that believe those trials have a reasonable chance of success. Otherwise, they would not proceed with them. Even the FDA itself believes those trials have a good chance of success. There are plenty of Wall Street analysts who are also evaluating these trials. Regular Joe is not going to outsmart all those people.

Each doctor offers hundreds of services/procedures. Each one has a different price. The insurance company needs a certain number of doctors in the network to be competitive and they can't really have hard cutoffs for every single item. It's not that surprising that there are outliers.

Additionally, lots of doctors often bill incorrectly on purpose so that a particular services/procedure gets covered, so I am not sure we should keep "justified" and "validated" in these discussions.

For the doctor, a great benefit of being cash only is you don't have to do billing, a non-trivial task.

Do you think the locals don't already have that info? The soldiers are not invisible.

But aren't the enemies the only ones that matter?

Or you could just not start tracking right at your house. Wait until you get to an intersection, for example.

"If you know ggplot2"... but you need to make a lot of plots to get the hang of ggplot2. The "+" syntax (not sure what the proper name for that is) alone is completely foreign and intimidating.

If you want to make great graphs in R, you will need to learn ggplot2. If you just want to learn R, why not keep it simple at first?

Without experiencing base R, you won't appreciate the tidyverse packages, which tend to have more of a learning curve.

For example, you can just run boxplot(x) in base R and it will make you a plot. Only after trying to make any modifications to it that you will see the benefit ggplot2.

As you mention yourself, "I almost quit R completely in frustration". I believe that is exactly why you appreciate the other packages you mention.

This is written for random internet people. Science papers are written for other scientists working in that particular niche. They are not trying to be approachable to the general public.

Shouldn't this set off some alert at Facebook since their likes per visitor rate will be much higher than normal?

Even commercial products (where they have a financial incentive to provide clear and comprehensive instructions) often take a lot of training before they work properly. How can we expect a small research lab to be better?

> we see PhD students and post-doc working 70h hours week on experiments with seemingly random results until the randomness goes their way.

There are known and tested protocols that can fail. Not every step can be accurately recorded. It's very common that an experiment will not work well the first time it's performed (even when supervised by someone experienced). Over time, researchers improve their skills and achieve better results following the same exact protocol. Does that mean that the science behind the experiment is bad?

> Only about 10% enroll in clinical trials in the US, which only test one drug at a time.

Clinical trials in the US test drug combinations and not just single drugs.

> With 90% of the patients not enrolled in a clinical trial we wanted to design a commercial solution that could help patients today rather then relying on the standard of care.

In other words, you are enrolling them in your own clinical trial without calling it such. Clinical trials are not perfect, but it's the best way we have to guarantee safety and efficacy.

> The pathway for us is as a CLIA certified service similar to other personalized genomics companies in cancer like Foundation Medicine.

Actually, it's very different. Foundation Medicine looks for known genetic variants and offers suggestions based on those. It is not offering novel or untested solutions. Their report is actually very poor if you are looking to make new discoveries.

> I think the problem is that Notable Labs would be testing combinations of FDA-approved drugs

I assume that only previously used combinations are suggested.

> This requirement could turn an 'easy comparison' into a lengthy clinical trial.

But at least you would have a fairly reasonable idea of how a clinical trial would go. It's not like going from an animal model to human trials. I assume they can get a large enough sample size as they can probably recruit patients easily for what can be viewed as a second opinion.

> Two drugs could be very safe on their own, but in combination could cause serious adverse events

That is true. My assumption was that only single compounds or previously tested combinations are suggested.

> Unless you are running a controlled clinical trial, I'm not sure anyone would trust the data.

I agree. I just meant it would be relatively easy to collect enough preliminary data to have a reasonable idea of how a proper clinical trial would go. It's not like going from an animal model to human trials or from a tiny cohort of late-stage patients to a larger more heterogenous group.

> You're just testing for efficacy, not safety

The drugs being tested are all already approved, so safety would not be really a concern.

> Insurance companies (Medicare/Medicaid included), won't pay for cancer drugs unless there is proof

After some time to collect the data, you could fairly easily compare outcomes of patients who undergo this test versus those who do not.

That's a big leap to go from in vitro to in vivo. That would usually not even be sufficient to enter clinical trials.

Besides, given GBM's heterogeneity, I am not sure it's clear which cells you are growing and how representative they are.

Is this FDA-approved?

If you look at the actual data, poor people are less likely to move out after gentrification than before.

Nice article that covers this: http://www.slate.com/articles/news_and_politics/politics/201...