I think the policies you cite as bad ideas do in fact enjoy broad support in the areas that enact them. "Hearts and minds" campaigns led by outsiders are notoriously difficult and prone to failure. Can you suggest a way to start the conversation about these policies in a way that you think is likely to succeed?
This is an extremely good question and very difficult to answer, to say the least. There has been quite the disinformation program going on in many conservative states; actually not simply disinformation, but demonization, so that its difficult to even get many people to listen to liberals.
But I think Al Gore probably has the best solution to this, which he talks about in his new documentary: if you focus on real economic benefits of certain "liberal" ideas, then you seem to have a chance. e.g. if wind energy is shown to be cheaper than natural gas, people will adopt it.
Democrats (and fact-believing republicans) really need to include more economics in their messages. Like how expanding Medicare actually helps small businesses and rural communities. Like how higher taxes on the rich help fund education and infrastructure projects that lead to greater prosperity.
I understand it is a somewhat skeevey way to do it, because if you're a democrat, you probably also believe in other non-economic democratic values (e.g. diversity, womens' reproductive rights etc.) and to win the support of people who don't believe in these things might seem uncouth.
"I think Prof Taleb did get annoyed when I said that I had read his ‘pop risk’ book, not the others. But I was actually trying to make clear that I had some knowledge of his work, though not a lot."
We could try to regulate away the good fortune of the people who bought the house, or we could try to convince the tech industry that they actually don't all need to be colocated, alleviating competition for resources in that area. What I've observed about human nature suggests that neither would be very popular.
I agree with building more housing. There are a decent amount of empty lots in Berkeley... and a smallish downtown area that could be revamped. However, even in best case scenario with a bunch of new housing, unless people start selling their single family houses to get replaced by higher density housing, I don't see housing prices going down a ton.. it won't be cheap but it may be more accessible to the middle class.
That actually does happen. And then we have the twisted situation where the economists say the economy is great, practically zero unemployment, even construction workers are scarce in California; and at the same time workforce participation rates for working-age males is the lowest it has ever been. Given how terrible idle males tend to be, this is worrisome.
http://www.latimes.com/business/la-fi-labor-force-men-201611...
People used to move to cities for the opportunities. Now it's increasingly difficult to do so. For lots of reasons we need to make it possible for people to move to the cities and thrive.
Suburbs in most places = medium-low pay but fairly cheap housing
country = cheap housing, no [tech] jobs
Metro areas (more dense than suburbs, less dense than cities...e.g. most of the Bay Area) = expensive housing, medium or high pay
Cities = extremely expensive, high pay (for tech jobs)
Pretty much every scenario seems to be "optimized" to have techies paying 30-50% of their monthly income to housing. You can't really win, but you can attempt to maximize the real dollar amount you end up left over with with each month. Working remotely is "cheating the system" and is therefore largely disallowed, or "work from home" is allowed only to quell some degree of outrage. Literally the industry that invented video chat doesn't [want to] understand how to use it to work remotely.
I know I'm in the minority about this, but I still prefer callbacks. In one common case, you're only doing one or two async things in a given area of the code, and in that case the async primitive you choose doesn't really matter. In another common case, you're doing a lot of async things that depend on each other's outcomes. In that case, I find that it's easier to build a reusable pattern for handling the specific type of async interaction you have using callback-based functions, whereas with promises and async/await, you tend to try to write straight line code that can't be reused and hides the asynchrony in a way that makes it difficult to debug.
I'm just too hooked on the async.auto() pattern to give up callbacks. The ability to break a large flow down into functions and then use a dependency based structure to automatically execute the flow with the most optimal concurrency is incredible. And the resulting code is very clean, with each functional step in the flow living on the same indent level. Refactoring and adding new steps is very easy as you can just add a new dependency to one of the existing functions or add a new function that depends on one of the other functions.
That looks interesting - as long as you stay in JS land. Once you move into typescript this kind of system will be impossible to typecheck. Promises and their compositions will typecheck just naturally compared to that.
Yep, there is the autoInject which injects previous dependent functions return values as parameters instead of as properties of a context object, so that may help some, but it still breaks typechecks.
I wish that promises had a concept similar to this. It's simply too annoying for me to manually refactor a bunch of Promise.all() calls to manually adjust the concurrency of the logic flow compared to using the dependency pattern of async.auto().
And based on what I've seen this is a common problem in both async functions, generators, and promises. Since it is difficult to properly execute a tree of functional steps with the maximal concurrency I see code all the time which just runs async functions or promises in a series instead of taking advantage of the ability to concurrently execute two branches of the tree at the same time.
Promises are easier to compose because you can pass them around as values. Meanwhile callbacks are used only for their side-effects, they do not return anything.
This is the problem my startup is going to solve! It's going to be a service connecting muggers and burglars who steal your physical stuff with criminals who want your passwords! I'm going to call it Robbr!
I think that the author is defining memory leaks as permanently out of scope but not deallocated memory. In that sense I don't know of anything in vanilla Python, or Java, that would qualify as a memory leak. In the more intuitive sense of a memory leak being any failure to make objects available to garbage collection, (such as by retaining references to them in an unexpected place) leading to unchecked increases in a program's memory footprint, memory leaks are possible in either language.
I really wish I could get 16gb RAM without paying for a touchscreen. It was possible to get an Ubuntu XPS configured that way as recently as two weeks ago. Should have pulled the trigger then.
For the reference of people commenting here, the definition of quandary is "a state of perplexity or uncertainty over what to do in a difficult situation"
I imagine that if you are Sylvania, or if you are thinking about making large investments in LED manufacturing capacity, then yes, this is a quandary.
