DalekBaldwin

On the other hand, the end result is released in polished form for a reason. As Llewyn Davis said, "In the entertainment business, you’re not supposed to let your practice shit out. It ruins the mystique."

This breaks my workflow -- I almost never open the forgot-password email on the same machine I used to initiate the request. Usually I need to briefly access a personal account from somebody else's computer or my work computer, and when I'm told I need to check my personal email, I only want to open that on my phone.

I once took a filmmaking class from Robert Mehnert who directed a documentary about the SR-71 at a time when its production was at risk of being discontinued. At first only a few hundred people in Congress saw it, but based on the size of the order for new planes, he called the highest-grossing opening weekend of all time. (I haven't been able to find the film online anywhere.)

I know this from audio samples used in the Music for Programming podcast. I always want to find out where various bits of dialogue in the podcast come from, but they're generally really hard to Google. It made me laugh to see that this one looks exactly the way I imagined.

> Using Yale economist and recent Nobel-prize winner William Nordhaus’s $31-per-tonne estimate of the social cost of carbon, it amounted last year to $107 billion for energy-related emissions from oil and natural gas in the U.S. Within that, emissions from transportation — the biggest source in the U.S. and the only one still growing — enjoyed a free ride worth $59 billion.

But what is the net tax/subsidy in plain monetary terms? I just did a cursory bit of Wikipedia research so I may have gotten something wrong, but the article neglected this aspect completely:

https://en.wikipedia.org/wiki/Energy_subsidies#United_States

> On March 13, 2013, Terry M. Dinan, senior advisor at the Congressional Budget Office, testified before the Subcommittee on Energy of the Committee on Science, Space, and Technology in the U.S. House of Representatives that federal energy tax subsidies would cost $16.4 billion that fiscal year, broken down as follows:

> Renewable energy: $7.3 billion (45 percent) > Energy efficiency: $4.8 billion (29 percent) > Fossil fuels: $3.2 billion (20 percent) > Nuclear energy: $1.1 billion (7 percent)

$3.2 billion in fossil fuel subsidies (including categories besides motor fuel), versus...

https://en.wikipedia.org/wiki/Fuel_taxes_in_the_United_State...

> Federal fuel taxes raised $35.2 billion in Fiscal Year 2014, with $25.0 billion raised from gasoline taxes and $10.2 billion raised from taxes on diesel and special motor fuels.

That's not counting state taxes, which are even higher:

> The United States federal excise tax on gasoline is 18.4 cents per gallon and 24.4 cents per gallon for diesel fuel... On average, as of January 2017, state and local taxes and fees add 31.04 cents to gasoline and 31.01 cents to diesel, for a total US average fuel tax of 49.44 cents per gallon for gas and 55.41 cents per gallon for diesel.

So about $80 billion in net taxes, which exceeds the estimated social cost of carbon from US auto emissions. Of course, fuel taxes go toward the maintenance of infrastructure for motor transportation, circling back and resubsidizing it, but electric vehicles enjoy the same benefits from that maintenance (as well as other subsidies for the generation of the electricity that powers them).

He starts with a simpler problem to elucidate the principle that is key to the original problem. Do continue reading.

The expected number of sons is 1, and the expected number of daughters is 1 (by the framing of the problem, in every possible scenario, there is exactly one daughter), but the expected value of the ratio is not 1:1. E[X]/E[Y] = E[X/Y] is not a valid identity.

http://www.thebigquestions.com/2010/12/22/a-big-answer-2/

Christopher Date and Hugh Darwen, in their work on type systems and the relational model, like to use the Principle of Incoherence: "A principle, sometimes invoked in defense of a less than fully successful attempt at criticizing some technical issue, to the effect that it's hard to criticize something coherently if what's being criticized is itself not very coherent in the first place. Occasionally referred to, a little unkindly, as The Incoherent Principle."

And for more on the history and philosophy behind it: http://www.nhplace.com/kent/Papers/Condition-Handling-2001.h...

But it doesn't encode Maybe/Option in a composable way. Say you have a Map<T,Option<U>> and you look up a key. If the key is present but maps to None, you get Some None, but if the key is not present, you get None. A plain untyped null/nil results in a lossy encoding of the two distinct cases. Lispy maps tend to paper this over by returning multiple values or returning a tuple (e.g. using Clojure's `find` instead of `get`), but I almost never see a codebase maintained with enough discipline to handle these situations consistently. If you have multiple developers working on different parts of a record-munging pipeline, you will typically have to deal with endless bug regressions involving the silent disappearance of key-value pairs that come and go depending on the order in which you compose the transformations.

I've run into the opposite situation making me hesitant to trust legitimate correspondence with my own banks. The past few times I had to take care of something over the phone, they did not ask for anything that could reasonably confirm my identity or account. One bank only asked for the last four digits of my account number. When I called another bank in response to an email alert about a fraudulent transaction, the representative asked for a phone number to text a verification code that I had to repeat back to them ("You want me to give you a ten-digit number?" "Yes"). Looking back on it, the first bank may have figured that few people will ever have the same account status problem at the same time and would ask for more information in the event of a collision, and the second one may have required me to name one of the phone numbers they already had on file (I'm used to representatives telling me a few digits of the number they're going to text based on what they have on file). But without knowing the entire workflow ahead of time, it seemed just as likely that this was a bunch of meaningless ceremony meant to give the appearance of bank-scale IT infrastructure in action so that I'd feel more comfortable revealing sensitive information later.

The fundamental reason for worsening subway service in recent years may be more mundane than we assumed: https://www.villagevoice.com/2018/03/13/the-trains-are-slowe...

It was slowly driving me insane at my last job. Now, I just have to dial in to the morning meeting, then I can work from home for about an hour before heading in to the office, and we've been asked to stay home even for fairly light snowstorms this winter. I wouldn't be surprised if other tech companies in the city are adding more flexibility to their work hours.

And its ultimate goal is, of course, to convert all of the mass in the universe into paperclips: https://wiki.lesswrong.com/wiki/Paperclip_maximizer

My laptop can run computationally-intensive physics simulations and play streaming hi-def video with no problem, but if I click a link to a mainstream news article it just about keels over and dies. How did our grandparents ever manage to load the news on an underpowered machine called "paper"?

When the experience is amplified by homoiconic data structures and structural editing, yes -- exploratory, ephemeral metaprogramming is a breeze. Compound Clojure data literals have a fully readable print representation that can be generically manipulated by built-in core library functions. Say I have a DSL that takes a bunch of nested maps and vectors as an input spec and compiles them to nested closures, and I want to create a large spec derived from some other source satisfying certain constraints, but it's not worth defining yet a higher-level DSL for a one-off definition that I intend to commit to source control. I can generate that data structure any way I like and spit the result out at the REPL or in an interactive buffer, then effortlessly plop that structure or any of its substructures into an EDN file or just inside a `def` or `let` binding in a regular Clojure file, which I then pass to the DSL compiler. It's much simpler than manipulating things like GADTs in Haskell. Even in Common Lisp, it might require some ad-hoc parsing code to accomplish the same thing.

It's included in disc 2 of the re-release.

My default is Fiona. It seems like the cadences of a Scottish accent help mask the shortcomings of TTS technology.

For the curious: it's difficult to find that particular demo on the web now, but the system didn't look too much different from its earlier incarnation at Microsoft in the '90s: https://www.youtube.com/watch?v=tSnnfUj1XCQ

> It won't be seen as a safety net to follow their ambition

Well then, a basic income only given to the ambitious wouldn't be a universal basic income, would it?

And yet, when it comes to cartoons that have been on the air for multiple decades, they almost always create something more entertaining with a production schedule of six days than The Simpsons can create with a production schedule of six months.

As I followed the links in the article, and the links those links linked to, I noticed that none of them report the gender breakdown among those who have never had sex.

In the surveys I've seen before, the average number of reported female sexual partners per man did not equal the average number of reported male sexual partners per woman. Men overreport their number of sexual partners; women underreport. This effect could skew the aggregate reported results among the entire population one way or another, depending on other factors.

If the underlying trend is one of increasing inequality in male sexual activity -- fewer men having sex with more women, while women have the same number of sexual encounters overall -- then because fewer men have had sex, the proportion of all people who have never had sex would increase compared to previous generations, even if the average amount of sex across the board actually remained the same as in previous generations.

I think this kind of response reinforces the cult-like image brought up elsewhere in this thread. The social rules within the Recurse Center have little bearing on an external discussion about the Recurse Center on an independent forum. There's room to talk about it here. It's not Fight Club.

> So what are macros? All they are is read packaged up nicely into formal system.

Macroexpansion happens during a different phase altogether, and would work exactly the same even if programmers had no access to the reader at all. The reader transforms an unstructured character stream into structured data, while macroexpansion transforms already-structured data into different structured data. (Read-macros are another story, but it's clear they're not what the author had in mind.)

Too often these kinds of posts only add to the confusion they're attempting to clear up.

It might have to do with the low resolution of the background image. The small drops produce stronger distortion, compressing a large region of the image into a smaller area, while the large drops faithfully reproduce the blurriness of the image after it's already been blown up to the size of your screen. Everything looks more uniform if you load it in a smaller window, but if you go too small, you'll notice aliasing in the smaller drops.

CL is designed so that code written for earlier Lisp dialects can run with little or no modification, which means there's a lot of bloat and historical baggage in the core `common-lisp` package. But even if that weren't the case, you'd still need conventions like you do in any other language. Clojure is a young language with a BDFL. It already has historical baggage that prominent members of the community will tell you to avoid.

The fact that so many people write their own cleaned-up wrapper libraries like CLUEL isn't a sign that CL especially needs them; it just means it's easy to do. If a language doesn't have macros and doesn't let you shadow built-in operators, you have to play the hand you're dealt.

It's trivial to deal with cryptic function names and inconsistent parameter ordering in a language's core library. I can read code with `car` and `cdr`; I write code with `first` and `rest`. What makes a poorly designed language is something you can't paper over, like broken lexical scoping. There are definitely some things worth complaining about in CL as a language, but I didn't see any in this post.

How do they get so confused about it? Is it because Java also allows overloading, which is a sort of static multiple dispatch?

I thought they were trying to allay people's fears of their self-driving cars going full Skynet, but now I realize I probably misinterpreted what they meant by "the other way around": "We leave no person unturned, to create industries that serve bits and atoms."

Funny, as of 5-10 years ago, I heard they refuse to entertain anybody affiliated with Caltech due to repeated trolling incidents. I guess you didn't cut him off soon enough.

It was largely designed by Bret Victor while he was an undergraduate. It had its own operating system! You can find the documentation here: https://web.archive.org/web/20080229205618/http://www.cs.cal...

The vending machine would occasionally be replaced so students had to create new hacks from scratch, but none of them were nearly as impressive as that one. During my undergrad, its only feature was a "random soda" button. More recently, it had a space cut out of the front panel which housed a flatscreen monitor so you could surf the web on it while eating your lunch, but nothing was hooked up to the vending machine's control system.

I guess we were more inventive before we had the modern internet.