David Demaree

Taking Pictures of Taking Pictures »

Mark Richardson, for Pitchfork:

“Wicked Game” covers were as common this year as covers of Springsteen’s “I’m on Fire” were in 2007. Washed Out’s tossed-off take was weak, but something about the Widowspeak’s carbon-copy version hit home when I working through Retromania. Faithful covers of existing songs are as old as popular music, but something here felt new: It was part of what I heard as the subtle Tumblr-ization of indie; music-making as re-blog. They weren’t just covering a song that they loved, they were essentially re-producing it, unchanged, and saying, “This is me.” It’s the band as lonely teenager in a bedroom: What do people think of me? “Oh that band Widowspeak? They’re the ones that are into Lynch, right? ‘Wicked Game’?”

At the end he references the “Most Photographed Barn in the World” passage from Don DeLillo’s White Noise, a passage I think about at least every other week.

(Source: theverge.com)

The "Bomb" Buried In Obamacare »

Rick Ungar, for Forbes:

[The medical loss ratio, a provision of Obama’s Affordable Care Act,] requires health insurance companies to spend 80% of the consumers’ premium dollars they collect on actual medical care rather than overhead, marketing expenses and profit. Failure on the part of insurers to meet this requirement will result in the insurers having to send their customers a rebate check representing the amount in which they underspend on actual medical care.

This is the true ‘bomb’ contained in Obamacare and the one item that will have more impact on the future of how medical care is paid for in this country than anything we’ve seen in quite some time.

Today, that bomb goes off. Today, the Department of Health & Human Services issues the rules of what insurer expenditures will—and will not—qualify as a medical expense for purposes of meeting the requirement. As it turns out, HHS isn’t screwing around. They actually mean to see to it that the insurance companies spend what they should taking care of their customers.

Ungar goes on to predict two things: that this is the beginning of the end for the current crop of private insurance companies, who simply aren’t set up to survive on just 20% of their current revenues, and—consequently—that this is the first small step toward a single-payer health care system in the United States.

Either thing would be change I can believe in. But this sounds too good to be true—there must be a loophole.

No, Siri Isn't Pro-Life »

Adam Engst, writing for TidBITS:

In a move reminiscent of how Greenpeace harangued Apple for the PR value, MoveOn.org even sent out email encouraging people to sign a petition asking Apple to modify how Siri works, claiming that Siri “won’t tell you where you can get an abortion or even emergency contraception — instead she’ll promote anti-abortion pregnancy ‘crisis’ centers.” MoveOn went on to say, “When a user asked her why she is anti-abortion, she replied, ‘I just am.’” Oh, please.

Search engines are only as accurate and capable as the data they have to work with. (And despite seeming like magic from the future, for all intents and purposes Siri is just a search engine.) Google doesn’t have a perfect algorithm for telling the difference between a vapid, spammy page about green tea weight loss and an actual article on the health benefits of green tea. Both look similarly relevant, in that they both contain a similar set of search terms—it’s the context that makes one spam and the other scholarship.

If you ask Siri for “abortion clinics” and its (her?) data set doesn’t use that word, then Siri can’t answer your question. From its perspective, there is no such thing as an “abortion clinic”, though “Planned Parenthood”, “birth control”, or “family planning” might exist, depending on your location and what clinics in your area call themselves. And if you asked Siri a different question, like “what is abortion”, you’d get a different answer because you’d be searching within a different data set.

That said it’s not impossible for a search engine to have an agenda. For instance, if Google detects a query for a local business name (e.g. “longman and eagle chicago”) they now always promote their own local listings at the top of the page over competitors like Yelp, even if Yelp’s pages are (objectively) older and more authoritative. But that takes work, and Google benefits by selling ads on those place pages. Apple doesn’t benefit by hiding abortion clinics in Siri’s search results.

Code Blog: Sexy instance_eval proxies in Ruby »

Some of the most mind-blowing language features in popular Ruby libraries/frameworks are made possible through the use of delegate proxies: objects that invisibly intercept method and variable calls in your code and forward them to one or more receiver objects.

For example this pattern is used in ActiveRecord to provide association methods that behave like Enumerables sometimes, like scopes at other times, while also providing some association-specific tasks like build or clear. One attribute can behave like three things at once — an Array, a scope, and an association — because when you send it messages, you’re actually talking to association proxy, a fourth kind of thing that serves as a kind of message bus between you and your data.

What I’m interested in right now is a simple, petty question: why is it that the block syntax in Sunspot — Mat Brown’s fantastic DSL for working with Apache Solr — lets me use variables or methods from outside its scope, like controller params or model attributes, but Karel Minarek’s ElasticSearch library Tire doesn’t? Again, the answer is a proxy.

In this gist, I go through each part of the ContextBoundDelegate proxy from Sunspot, explain how it works, then use it to describe a zoo using sexy block syntax.

Readability Now (Mostly) Free, with iOS Apps On The Way »

Rich Ziade:

With this release, Readability is available at no cost. Sign-up and you’ll have your own profile and reading list in no time. Both Readability accounts and our companion apps will always be completely free, but we also offer a premium experience for users who want additional features and an easy way to support their favorite writers and publishers.

Translation: the current $5+ per month product is still around — and still aims to pay 70% of revenues to authors whose articles you bookmark — but there is now also a lower, free tier where your reading list is limited to your 30 most recent items. This will probably get a lot of people to sign up who balked at the $5 charge, especially once the native iOS app (which will also be free of charge) is approved by Apple.

All that said, I’m not sure what Readability is trying to be other than an upmarket alternative to Instapaper, and that’s a stupid goal. Instapaper is a great product for a particular kind of user who likes to read and craves simplicity. It invented the reading-list genre, and is virtually synonymous with it — Apple even cites it in describing Safari’s new, less capable Reading List feature, saying it’s “similar to the popular Instapaper app.”

How many people are out there who would like something like Instapaper but aren’t satisfied with Instapaper itself, who wouldn’t prefer something entirely free like Reading List? What is Readability bringing to the table that’s new?

More importantly, how much money is Readability spending trying to compete with a one-man company with little overhead and a three-year head start?

Knyle style recruiting »

GitHub’s Kyle Neath has had it up to here with recruiters and their shenanigans:

Here’s the thing: recruiters do not give a fuck about doing good in the world. They do not care about making people happy. They do not care about building a good company. They do not care about treating email addresses as human beings. They only care about their percentage.

My name is Kyle Neath and I am a designer at GitHub. This takes about 30 seconds to find out. Yet about ¾ of the emails and calls I receive offer someone named Keith, Knyle, Kenneth, or Keneath a Rails / node.js job. The only node.js I’ve written is a Hubot script to put clown noses on someone’s face. I’m not alone with these experiences — it’s common practice for tech recruiters to not give .00005 fucks about their leads.

This is a pretty accurate description of the situation. Last week I was cold-called at least three times by recruiters, including calls from two different people from the same agency, two days in a row. And every single one started roughly the same way:

Hi, can I speak to David? Hi, David, this is $NAME from $AGENCY. I came across your resumé and I was really impressed. I wanted to talk to you about a couple of|several positions we have available…

Let’s stop right there. In my career I’ve been cold-contacted by dozens of recruiters. Exactly three have been worth talking to, and every one of those was contacting me not about “a number of” positions or “several” openings, but about a specific job at one specific company. The bad kind of recruiter makes money by placing as many warm bodies in as many jobs as possible. To care about either the candidates or companies involved would be inefficient.

Another thing the good recruiters had in common: every one of them made their initial contact via email, not over the phone. That’s not to say recruiter emails are always good — many of the contacts Kyle Neath is complaining about were over email — but that a good recruiter is respectful of your time. They’re also frequently diligent enough to make sure they’re emailing the right person, just as if it were any other kind of serious business contact.

I have no hope that the bad recruiters will stop calling. Clearly having a bunch of fast-talking people interrupt me and other developers while they eat dinner three days in a row makes somebody money. I’m posting this as a rule of thumb for other developers to follow, to know which recruiters are worth talking to, just in case that information comes in handy some day.

WSJ: Swiss Canton Glimpses Growth's Limits »

Zug is a Swiss canton where low taxes have brought high employment and lots of wealth. It also has real estate prices and cost of living so high that none of its workers can afford to live there:

“If you make 300,000 or 400,000 francs, you struggle” in Zug, said Kilian Borter, a public-relations executive, referring to incomes that equal $380,000 to $490,000. Living in Zug “maybe starts to make sense starting at about a half a million” francs a year in income, he said.

And while Zug is a haven for multinational businesses, the income disparity — i.e., the fact that there is literally no middle class — makes it virtually impossible for those businesses to fill low- to mid-level jobs such as administrative staff or accountants. The town can support CEOs but not the CEO’s assistant.

Mobile FontFonts: Awesome Fonts for iOS Apps »

Now this is interesting: FontFont have released a series of 14 of their best fonts — including Basic Gothic, Tisa, and Yoga — as iOS-optimized TTF/OTF files, for developers to include in iOS apps.

Including a custom font in an iOS project is not technically very hard — seriously, you just add the files to your project, update one .plist file, and you’re set. But good fonts that render well in an app can be difficult to find, and expensive to license. The best, most interesting thing about FontFont’s mobile lineup, therefore, is the pricing: a single mobile font pack costs between $150-200 for up to five developers. No time limits, no royalties, just a one-time fee.

You can use the same fonts to render HTML5 content in a UIWebView, though to be very clear, this isn’t the same thing as a web font license, which FontFont also sells. The difference is that these fonts are meant to be embedded in your iOS app directly but not served over a network, whereas with web fonts you’re allowed to serve them to web pages but not embed them.

If you want to kick the tires before you buy, there’s a sample project available on GitHub which even includes one free font, the mobile version of FF Basic Gothic Black Italic.