"So when the Republicans take control over the executive branch... are we going to keep hearing such great boosterism?"
WHy are you asking him? I don't think anyone knows the answer to this. Obviously the major factor would be whether the next president (be they Republican or Democrat) decide to continue support for these kinds of teams, both in terms of funding and autonomy to do their job. This seems like something we should be asking the candidates and their parties about.
Because the topic of this thread is the large scale system of project delivery and changes to it. The poster made some sweeping statements about how the article was not accurate and provided great reference material to what he considered more relevant material.
Given the topic of how these large systems operate and could be improved, and given the poster's claims to have special insight, seemed reasonable to me to continue the dialog. After all, this was an example of behavior associated with the topic at hand.
And my question was about enthusiasm and public support. Why would presidential candidates have anything to do with this? If it is partisan, then my question has been answered.
The US Digital Service started with healthcare.gov, but nowadays they work across the government on projects that matter to people all over the political spectrum: VA, immigration, department of education, and more.
That said, I have no way to predict who the Republican nominee will be or what they'll do. I think the impact of USDS relative to its budget is huge, and I'd like to believe making the government competent at writing software is a bipartisan issue, so I would hope this initiative crosses over into the next administration, whoever it is.
Right now, I think everybody at USDS is just focused on getting results over the next 18 months.
soup kitchens and bednet distribution programs are trying to solve two different sets of problems, so I am not sure what you mean when you say that one is 'orders of magnitude less effective' than the other
I feel compelled to point out, as a native English speaker familiar with this sort of mathematics, that the phrasing to "go infinitely" doesn't make much sense. Or, if it makes sense, it is at least ambiguous (with plenty of room for the interpretation "diverges to infinity"). The clarification that the author makes about looping (or even better, repeating) is much clearer.
As a non-native English speaker who also interviews candidates, I try make sure to see if the candidate understood and always say it in another way -- For example after saying "go infinitely" (which is a silly way to phrase it) I would also say "it basically loops around", "creates a cycle". Maybe even draw out a little example on a whiteboard.
I have to agree. Reading this article, made me think that the interviewer is either a dick, or bad at conducting interviews. The whole thing seemed to turn on using a word in a nonstandard way to deliberately obscure the problem definition. That's just stupid.
This is really great. I was pleasantly surprised to see that there was information available about household income at the "tract" level. I had searched some months ago for a data set that exposed high resolution income/poverty statistics at this resolution for Chicago, but could never find much beyond neighborhoods, or as anonymized sets. But I wasn't until now aware of the firehouse that is http://factfinder.census.gov/.
I assume the lack of an API for automating calls to your wonderfully structured data is to comply with the TOS of the census.gov site. Or is this for some other reason?
Thanks. We may build an API at some point, but it is not to do with the TOS. The raw data is so big (and we use memory mapped files, not a DB) that it is a pain from an infrastructure perspective to have available on a web box.
I don't really take any issues with your comments about the outlook of the political situation in Turkey in relation to the EU, but the hypothesis that this is related the influence of rural populations can't be right. The percentage of the population living in urban centers is roughly 70%, on par with Germany, Austria, Czech Republic et al. . Furthermore about 1/3 of Turkish people live in cities with over 1 million people , which in a sense makes Turkey even more urban than most of the big European countries. To reiterate, I don't think this means your points are wrong, only that the reason that they are right must lie somewhere else.
Well "rural" wasn't the best choice, i didn't want to use "backward" or "traditional" because it really doesn't reflect the exact situation either and i didn't want it to come out too negative or condescending.
Turkey has the lowest educational attainment in the OECD, only 32% of adults have high school education, and amongst the younger generation it's still bad as less than 40% of it's male populations and less than 30% of it's female population completes a high school equivalent education, and only a fraction of that get higher secondary education.
Much of it's population is really on opposites sides of current western values and the government crackdown on free press, opposition parties, taking control over the courts and legal system, dismantling the (for once in a life time remarkably positive) influence of the military, and just the complete mockery of all attempts at democracy just put it on a completely different course than it used to be only 2 decades ago.
Ok, fair enough: "send them into a rough hockey game" or something like that.
If there's one thing that sets a lot of Canadians off though, it's being considered like some large, deep-frozen 51st state of the US. For one obvious example of the differences, Canada has universal health care.
>We have to start with a discussion about how this progress towards doomsday AI should be measured, and then formulate policy directly those scenarios
Here is my proposal. Since the whole of AI is a rather complicated and messy business let us just focus on a small part. That is, let's just talk about linear regression and simple decision making schemes based on estimated parameters. While one can debate whether or not linear regression counts as "actual" AI, there is little doubt that (a) it forms the conceptual basis of many more complicated AI schemes, so that if we can reason about doomsday scenarios for these we should also be able to do the same for linear regression and (b) despite its simplicity there is a substantial intelligence value in linear regression, to the point that successes and failures of linear regression can cause major effects on scientific/business/whatever process.
What are the examples about how the set of decision/action possibilities increases before and after linear regression technology was applied? What examples are there of how the failure modes of linear regression tended to increase the risk to humans (e.g. moving up the doomsday axis)? In each of these cases what policies have been or could be implemented to mitigate these risks, including technological audits as well as regulation of deployment of technologies.
My sense is that these questions actually can be answered. In all likelihood professional statisticians and operations research types already have opinions on these things. I would be curious to see what the public opinion is about this limited version of the problem, since the technology is something that many people (including Sam Altman and friends) should have a good understanding of.
The conclusion seems to be that even this level of machine learning decreases controllability substantially. "We note that it is remarkably easy to incur massive ongoing maintenance costs at the system level when applying machine learning."
One oftentimes hears arguments along these lines about Murakami. That his writing is just pop commercial stuff. I am not sure I ever really grokked what these complaints were all about. Can you say more about what makes his writing like a "cheap knock-off"? At the very least what is the argument that it is a knock off at all (deferring considerations of its valuation)?
>Bill Haast figured he had handled more than three million poisonous snakes over the years, and he had the hands to prove it.
If we assume that it takes no more than 10 seconds to "handle" a snake. This would mean he spent no less than 23 years continuously handling one snake right after another, without any breaks for food, sleep, or anything at all. Unless we are imagining an even more liberal version of handling than my calculation allows for I find the claim doubtful.
Your math is incorrect. 3,000,000 snakes * 10 seconds / 86400 seconds in a day = 347.22 days, which is less than a year.
If he was spending 3 minutes per snake, 8 hours per day, every day of every year, that's 51.37 years of snake-handling, which is less than what he actually did. So, his claim may actually be plausible.