jhundal

I would pay good money for a field guide/itinerary to accompany "Assembling California".

More directly related to the Green River, I found Wayne Ranney's "Carving Grand Canyon: Evidence, Theories, and Mystery" an accessible/engaging intro to deep geological mysteries.

Same amount of food yes. But if a mathematician was just sitting on the sidelines before this new field, they may decide it is more productive to trade time they spent on preparing food for more time to spend on math.

The data cited on the summary page of the study does not support the conclusion that those units are unoccupied, which I think is the key question here. It does say that 64% do not claim a residential exemption on property tax, but [1] indicates that could just mean those units are rented out since the exemption only applies to owner occupied properties.

[1]: https://www.boston.gov/departments/assessing/filing-property...

One of the authors here was the lead on the early Stanford antibody prevalence study that had numerous issues: https://statmodeling.stat.columbia.edu/2020/04/19/fatal-flaw...

It also did not report a possible funding conflict: https://www.buzzfeednews.com/article/stephaniemlee/stanford-...

Have not picked through the statistics cited in the article yet, but I'd be careful with the conclusions derived from them.

Edit: Direct link to Buzzfeed article.

At least one guess is that this helps Softbank increase the value of its other investments in Uber: https://www.bloomberg.com/opinion/articles/2019-03-14/don-t-...

This paper was discredited. It could not be replicated using publicly available data and it was eventually found that there were large scale computational errors: https://www.bbc.com/news/magazine-22223190

I have not formed an opinion on MMT, nor do I consider myself to have the right background trying to fully understand it or any other economic model, but even at my level of knowledge, the first four paragraphs of Sumner's rebuttal seem to be willful ignorance on his part. The views he cites as inconsistent easily fit into the MMT framework, which asserts that inflation is the only constraint that matters, and other factors (like large deficits) only matter if they affect inflation.

Later on he cites low overall GDP growth in Japan to counter Japan as an example of large deficits/low inflation. This: 1) Doesn't invalidate the point. 2) Totally discounts that Japan has had _some_ GDP growth even as an aging population has led to a _smaller_ workforce over that period.

Always worth it to delve into how a statistic was computed. In this case, the organization cited as the source of the 70% figure itself disclaims it:

""" Over the past couple of years several news organizations have attributed a statistic to the National Endowment for Financial Education (NEFE) stating that 70 percent of lottery winners end up bankrupt in just a few years after receiving a large financial windfall. This statistic is not backed by research from NEFE, nor can it be confirmed by the organization. Frequent reporting—without validation from NEFE—has allowed this “stat” to survive online in perpetuity. """

(https://www.nefe.org/Press-Room/News/Research-Statistic-on-F...)

I'm particularly curious about why you prefer Reinhart over Krugman, specifically because of the disagreement over the famously erroneous Reinhart-Rogoff results (Krugman comment here: https://krugman.blogs.nytimes.com/2013/04/16/reinhart-rogoff.... As far as I know, Reinhart-Rogoff never republished or responded further than what Krugman discussed in the linked post). Specific examples supporting/disproving the case of either would be helpful.

I tend to ignore Krugman when he's talking politics (eg. his prediction of a Trump recession). However, when he goes through the process of laying out the assumptions of an economic model (happens in many of his blog posts labeled 'wonkish'), his discussion of economic consequences of various policies seems to map very well with reality.

When the stock units vest (or are exercised in the case of options), you are taxed on their current value at the regular income tax rates. What would be taxed at the long term capital gains rate are any gains that happen _after_ the stock is in your possession. The former component is likely the one that will dominate within the timespan of a year.

I would not use the Time article as a reliable source, it cites the 70/90% figures a study by the "Williams Group wealth consultancy", which of course is trying to sell its services to wealthy clients.

I've seen the statistic quoted in so many places, yet have never found a link to the actually study. No info on sample size; we don't even have a definition of 'wealth' used in the study.

I'd like to add that a first cut as infrastructure-as-code could be as simple as putting your manual setup tasks (including SSHing into a host) into a shell script and committing it to your repo.

But it takes a only trivial amount of additional time to set up an Ansible playbook (which abstracts away SSHing to a host) and then runs a shell script that's only concerned with actual setup tasks. Enough for shipping code and a good starting point for using the declarative aspects of Ansible to make your setup/deploy processes more automated and less error-prone.

It is possible to cargo-cult infrastructure best practices and make the mistake of investing too much engineering resources in making your deployment practices perfect (for some definition of perfect). Infrastructure-as-code should be understood as a continual process, not something to get completely right on day one, but that applies to any part of your code/product/infrastructure.

Edited: typos

Somewhat agreed, maven is a build tool and packages it downloads do not execute code through maven. This does not preclude malicious typosquatting packages making it into applications built using it, but does provide some option for reducing the attack surface.

In practice I think most developers would be running their project on the same exact box as they use for building it, which nullifies the separation of build/runtime environment. The reason that we don't typically see egregious typosquatting in the Java ecosystem is that Sonatype has a manual check on the claimed namespace for the organization publishing a project (among other checks). npm, Inc. could do this, but they so far have chosen not to.

Edit: Typos/clarity

But if you object to joining a union, then why not work a different job that is non-unionized? Why not look at all possible jobs as your market, as opposed to contracts for a specific job?

One possible response to that question: Why is it ethical or fair force a worker to pay union dues as part of a job?

In that case I'd argue that because the union's negotiation affects all employees (it at least sets a baseline) it is ethical and fair to ask every employee to contribute. Allowing every employee to opt out greatly reduces the amount of resources the union has available to reach its aims.

I'm assuming a few things here:

- That the negotiation performed by the union does affect all workers in the job, whether union members or not.

- That collective bargaining enabled by unions can in many cases be a positive thing for workers because it helps fix an information asymmetry that exists between employers and workers. (Such asymmetries break the assumptions of ideal free markets, which indeed are a good thing.)

- That right-to-work laws do indeed adversely affect the resources of a union. (The data shows that public sector union membership greatly dropped after Wisconsin's public sector right-to-work act passed: https://projects.jsonline.com/news/2016/11/27/for-unions-in-... . As an aside: I'm not the biggest fan of public sector unions, but I think it's likely private sector right-to-work laws would have a similar effect.)

I've participated in debates like this previously on HN without a satisfying response, so I'd ask the following for a response:

Which specific parts of my response/assumptions do you disagree with, and why?

Technically, I don't think there is anything to prevent an employer for forcing you to attend company parties as a condition of employment. Of course the vast majority of employers wouldn't do this, but in this case your only choice is to go to the parties or take a different job.

Now an argument could be made that the employer-employee relationship is different from the union-employee relationship, but is there a reason that any such difference implies that a union can't set conditions for employment?

EDIT: To try to keep the discussion focused, and the level of discourse high, let's try to find out exactly where we disagree:

You asked: Why is ethical or fair force a worker to pay union dues as part of a job?

My answer: Because the union's negotiation affects all employees it is ethical and fair to ask every employee to contribute. Allowing every employee to opt out greatly reduces the amount of resources the union has available to reach its aims.

The assumptions behind my answer:

- That the negotiation performed by the union does affect all workers in the job, whether union members or not.

- That collective bargaining enabled by unions can in many cases be a positive thing for workers.

- That right-to-work laws do indeed adversely affect the resources of a union. (The data shows that public sector union membership greatly dropped after Wisconsin's public sector right-to-work act passed: https://projects.jsonline.com/news/2016/11/27/for-unions-in-...)

Let's move this debate forward: Which part of my response/assumptions do you disagree with, and why?

(Aside: I'm actually not the biggest fan of public sector unions, but I think it's likely private sector laws would have a similar effect.)

(Original comment preserved below.)

------------------------------------------------------------

The union asks for that contribution because they are negotiating for all employees, and no one can claim that they'd negotiate themselves because the union has fought for the baseline. If the union has won the right to require that contribution for all workers, what is the issue with this requirement? Again, why try to debate whether the employee should have this choice when they had the initial choice to work a different job?

That said, I think that the right-to-work debate is a red herring. The first question we need to answer is whether collective bargaining can be a net benefit, and if so, how we can make it work.

Stepping back even further, I think folks tend to forget that model of an optimal free market carries several assumptions, perfect information being the one most notably broken by a large employer to single employee relationship. That doesn't mean that free market model isn't useful, in fact we should be trying to aim for the results of a perfect free market in most cases. But it does mean that there can be areas where a bit of intervention does a huge amount of good.

At some point the majority of workers voted to organize. After that, anyone working there has made the choice to work a unionized job. I don't think it is fair to allow a single jobseeker to break the structure of a union this way. Why not consider asking these individuals to choose to work a different job if they object to paying union dues?

In general, there's an information asymmetry between workers and employers that collective bargaining presumably helps alleviate (especially in the private sector), and union advocates would say that the value of this outweighs the drawbacks of unions. That's the right question to be debating rather than the fairness of closed shops.

I see this figure around a lot, does anyone have a pointer to the actual study? I got as far as a book called "Preparing Heirs: Five Steps to a Successful Transition of Family Wealth and Values", no detail on the actual study.

I certainly agree for trials for treatments for specific ailments. But insurers, hospital administrators, and others have always been making decisions that balance clinical outcomes and cost of care without patients knowing such a decision has been made. The experiments I refer to are similar to those choices and typically the funding will go testing choices that haven't been made because they are more costly in the short run.

Not sure how many folks are aware of this, but the ACA included funding for an 'Innovation Center' to run experiments on potential approaches to improve medical outcomes and cut costs: http://www.npr.org/sections/health-shots/2015/01/14/37719262...

I think that the A/B tests you propose would be great, though there will be plenty of ideologues still finding excuses not to accept the results.

Also, hi Nate! You probably don't remember, but you interviewed me sometime around December 2011.

I don't have anything to do with the plan (just keeping my head up about how the business runs :) ), but from the FAQ (http://www.employeefiduciary.com/401k-faq/) it looks like they use a third party (MG Trust) for custodial services, paying them out of the asset based fee they already charge. Nearly certain that the recordkeeping is included (we're using it anyways), so the fees listed may in fact cover everything.

Employee Fiduciary is at 8 bps: http://www.employeefiduciary.com/401k-plan-pricing/

Don't know if they would come out lower than Captain401 though (no mention of any cut of the overall assets on the Captain401 page, if that's correct, they're very likely lower priced).

Definitely agree here, want to add a link to this paper which shows how far we still have to go (and questions whether the current models will ever replicate human vision): http://arxiv.org/abs/1412.1897

As mentioned elsewhere on this thread, you'd use Swift (and other object storage systems) to store larger pieces of unstructured binary data (eg. images/videos/backups). The easiest way to do that in Cassandra without rolling a whole bunch of new code for yourself would be to store these things as gigantic rows in Cassandra, which isn't where it performs well. Typically object storage systems also come with a whole lot of features such as being able to control who views your files, the ability to upload files in piecemeal fashion, etc.

From working at an Openstack startup for a while, I can say that customers definitely prefer the S3 API since it works with their existing tools. That being said, they are largely compatible until you get into the authorization models, and the fact that S3 is globally namespaced, whereas Swift scopes names to tenants.

Couple of thoughts:

- Even if you provide resources to students for finding jobs/internships, they don't come with the same connections or experience as their wealthier peers so I'm wondering how much these help. For example, technology companies are typically going to hire those computer science students with the most programming experience, outside of a few small programs run by the larger companies (Google, Microsoft, Facebook) specifically for disadvantaged students.

- Often you see stats on what percentage of students attending a university are low income (by some proxy like Pell grants), and what percentage of them graduate, but I have yet to see stats on career fields and income for these students. As someone from a blue collar middle class background getting an engineering degree I could count on one hand the number of students I met with backgrounds similar to mine (and this is at a fairly large school). I would expect these sorts of stats aren't very flattering to schools, but would love to see the numbers if they exist.

I think the large retail pharmacies definitely have the advantage here. They already do blister packs, which is not too far off from PillPack's solution. They can also provide other compliance solutions (eg. http://www.glowcaps.com/) that can relay information back to caregivers, among other things.

They would seem to have a few more hard to replicate advantages, a couple off the top of my head: It is probably easier for them to provide medications in form factors other than pills (eg. inhalers), and I would guess that most folks don't want two different sources for their drugs. If you happen to misplace your medication only a physical pharmacy is going to be able to cost effectively provide you medication on short notice.

It's hard to know what the outcome of your education is going to be, so the risk of not being able to pay off a loan may just be too great. I am not aware of any elite colleges publishing statistics on things like the job placement rate and post-graduation income of lower income students. (On that note, if anyone knows where to get these numbers I would love to take a look.)

I'm not sure this article gets the definition of the zero lower bound right (though I'd love to be corrected if it does.) As I understood it, the ZLB only applies to central bank lending rates to other banks, which if they become negative the other banks only have the incentive to hoard cash.