I am in Seoul right now, my first visit ever. The car I’m in has a GPS with a live traffic feed showing which bridges are jammed.
In the press on the launch of a new shopping app called Smoopa, you might have noticed a remarkable fact: one of the developers, Derek Langton, only learned to code a year and a half ago, and before that he was an 18-year veteran of the Massachusetts State Police.
I met Derek and his business partner (and husband), Mendel Chuang, the other day, and heard Derek’s story – perfectly reported since then by Kim-Mai Cutler. She captured all his tips for learning to code. I love that he mostly used YouTube videos from kids, not educational materials produced by professionals. She also nailed his fundamental insights about the main ingredient – motivation.
But I found myself asking one big question: is Derek an anomaly, or could there be millions more like him?
Yes, he had some generally relevant, and uncommon, background — Derek came from a family with a technical bent (his dad repaired VCRs), took a BASIC course in high school, had some trade school, and was always tinkering with technology. There were early signs, he told me, of him doing engineering-like tasks even in his work in the police. “I was the person automating payroll in Excel.” But is that all it takes?
And, true, he did need extraordinary motivation (“I saw the world was going to change”) to piece it all together. But most of his learning was about how to solve a detail. He wasn’t learning rocket science, he was repairing a 21st century VCR.
As time passes, that learning, today scattered in a thousand corners of the Internet, will become easier to find and use. More people will grow up familiar with how technology works, less fearful of it. The programming frameworks will become more powerful, and abstract (i.e., simpler). Maybe it won’t quite take Derek’s massive store of willpower, and then become much more accessible to millions.
So if Derek can code, can you?
Applications are due today for IGN’s Code-Foo program, where if you’re passionate about IGN and know how to code — even if you taught yourself, never worked a day in your life, and don’t have a degree — we’ll hire you.
(Side note: having more developers “trained” in this way may make for better teams, because they’ll bring different approaches to bear than traditionally-trained engineers. Derek said, “I build it and then break it 10 times,” gradually watching a consumer-grade app emerge. It will be interesting to see the companies built by this new generation of recent-learner hackers.)
Or, the return of the human.
There was (once upon a time, in the early 2000’s) a generation of Internet products built on the premise that if you mastered Google’s search rankings, you could grow traffic. Services like IMDb, About.com, Wikipedia, etc. provided value to their users, but were also able to accelerate their growth by playing the SEO game. Making your Internet service findable was, in some ways, as important as making your service great.
Roger McNamee and Mike Maples wrote on some fundamental ways today’s Internet is changing – and these changes have real implications for how new services get found.
They point out that the iPhone, by taking us away from the SEO-driven world of URLs and websites, with their freely available but uneven user experiences, “did more than change the smartphone market: Its success serendipitously changed the architecture of the Internet.” And they talk about the growing power of brands and content in this world – which would be great if it turns out to be true (and I agree it will).
If making a great Internet service is only as important as the number of people who discover that service, how exactly will new digital services get found in this new world?
Search won’t come first, as Roger and Mike point out. There is already a new generation of Internet services that was born without ever really needing to master search: everyone’s i.e.-of-the-moment Pinterest, Angry Birds, Flixster (where I got to observe the action for a period), the social networks themselves, Tumblr, and many others. (Though of course we’ll all still have to practice good SEO.)
So how do the new kids get discovered? Some think social media optimization will replace search engine optimization – and I think that’s part of the answer, but it will go further. Some of it flows from the nature of the platforms and devices that are connected to the new Internet. The keyboard is no longer the primary input device, and so text (and search based on text strings) declines with it.
There are advantages and disadvantages to the SEO era: on the one hand, it’s a fair system in that the rules apply equally to everyone and it is tough to game through status and relationships; on the other hand it’s an algorithm that rewards sites that learn the tricks without necessarily providing other value. Quirky sites with occasionally-middling content can thrive. (Google “buy a lawn chair” and see what you get.)
In the search era, you get websites with millions of users who arrive predominantly through search, and who never form much of an ongoing relationship – or sometimes even realize what site they are on. (I call this the “if you’re so big how come I never heard of you” problem.)
In the next era, dominated by apps and (as Roger and Mike think) HTML5 web experiences, it’s different. There’s an entirely new set of skills required to get to #1 in an app store (or the App Store). Of course, a great service helps – and is still the single most important factor. But copycats can thrive (at least until they are taken down), as can masters of the viral loopholes left by platform creators (which are much bigger than the imperfections of the search algorithms), or those who find marketing vendors who help them work the system with incentivized installs and other methods. And, of course, black hat techniques still work, for those willing to use them (like the bot farms found downloading iOS apps en masse a few months ago to boost App Store rankings).
Maybe the single most significant difference, though, is that in the next era, you can get discovered merely by having a strong relationship with one of the platforms of discovery: Apple, Facebook, YouTube, Android, maybe even the carriers and television manufacturers again, etc. You couldn’t do this in the search era. These platform companies pick early partners to test and premiere their new services, humans at these companies choose apps to feature, etc.
That’s why, in the new world, business development (BD) may take the place of SEO. This could be great for companies with well-known brands based on a quality service.
A relationship with the platform can get you early access to the APIs and other technical services (see the early partners for Facebook’s Open Graph, for example), the PR boost of being first, and the opportunity to learn the ins and outs of the new distribution systems. That’s awful for new services with great content but no way to get an “in” – but it’s great for wonderful brands who might otherwise have struggled with the strange world of search algorithms.
Like farming was in the 17th century, factory work during the industrial revolution, construction during the Great Depression, and manufacturing after World War II. Better, because writing code is a creative act which can be done with or without a traditional (antiquated?) office-based job, and can create enormous personal and economic value.
Most young people start in jobs that don’t have much of a future. Most don’t get higher education – only a third get any advanced degree. In the past, students who missed out on a higher education learned vocational skills – but this stuttered as we moved to an information economy. Today, students without a higher education generally enter service professions or trades where employment, if they can get it, doesn’t offer much career growth.
There is a new opportunity emerging for young people to do productive, entrepreneurial, satisfying work: they can learn to code. Code isn’t that hard to start to learn – one outsourcing firm takes people with no training and makes them full-time Java programmers in 3 months. (Of course, mastery takes tremendous talent and craft.) Coding isn’t expensive – with netbooks, cloud hosting and storage, and open source software. Beyond a certain point, coders are self-taught, and can continue to advance their skills.
They’re handing out Gutenberg printing presses out there: with services like Treehouse (I’m a dues-paying member) and Codecademy (and its expertly-timed year of code), countless university courses free online, Google Code University, the warm embrace of Stack Overflow, in-person courses like Dev Bootcamp, summer camps for kids, even the promise of a one-day result with Decoded (the six-minute abs of learning to code), and great organizations like CodeNow (which I’ve been supporting) reaching out to teach code in underserved communities. I’m sure I’ve left many out.
Yet very few high school students learn to code. Almost no high schools teach code as part of the curriculum. Though of course they should — code is literacy, not (just) a specialist skill. And kids can get started coding early. Many students who would be terrific at coding, a creative, tinkering act, also may not thrive in institutional (school) environments.
There is real demand for coders – even despite overall unemployment – so learning to code produces rewards quickly. Online marketplaces like oDesk and Elance hire starting programmers at rates as high as $15-20 an hour or more. Learning to code is one of the best paths to entrepreneurship. Coding also offers students the joy of creation and mastery of a complex skill. Code may one day be a basic workplace expectation – like emailing, or “proficient in Word.” Young people are also willing to learn: coding now has a brand. The kid who writes an iPhone or Android app, these days, gets the girl (or boy!).
It might even be possible to do more than just learn to code – but also to become an elite coder – without necessarily going to college. We are in the early days of teaching code as a profession. Most academic training is focused on teaching students theory, not practice. (One Ivy League computer science program only required one course where students actually write code.) Imagine if students who might not otherwise even attend college could become elite coders.
In the U.S., the STEM line of thinking is about creating the next generation of scientists. In computing, this is even reflected in what we call the study of programming — computer “science.” We could be doing something different (and complementary), teaching students to be makers, not scientists: creating the next generation who can hack, beget, get paid right away, and maybe become entrepreneurs. Learning this would make the high school experience more rewarding, because it would have an immediate result. (I went to a high school with a vocational tradition, Stuyvesant in New York, and wish I had more courses like the architectural drafting class I took for a year.)
I’ve become personally passionate about this idea over the last couple of years. I think it could be a path to helping fix a lot of what doesn’t work right now: our ways of teaching students, powering our economy’s future, and making work a creative and fulfilling way to spend time.
I’m sure there are many more out there working on this — if you’re one of them, hit me up and let’s find a way to make common cause. And if you think I’m crazy, tell me why.
Spent today at Minecon, the live event for the game phenomenon Minecraft.
If you don’t know it yet, Minecraft is an open, world-building game. Built, initially, by one guy — goes by Notch. Sixteen million people have registered for it, even though it was technically only released today. You download the game, a standalone client, from the Minecraft website. (It is coming to Xbox, and you can play part of the game on iOS and Android.)
Minecraft confirms some long-known truths:
Anyone can create a hit game, distribute it themselves, and entertain millions. Even today, after 4 million units sold, Mojang (the maker of Minecraft) only has 14 employees.
Games don’t necessarily need to “launch” to reach a big audience — they can dribble out, and flood millions.
Games can be put to unexpected, wonderful purposes (like the educational uses popping in schools, promoted by Minecraft Teacher and others).
But Minecraft matters because it demonstrates some powerful new things:
Indie games can be for anyone — not just for traditional core gamers. Families are all over the place at Minecon, some with three generations playing. Apparently someone registered a baby for the event. The traditional game platforms no longer have a monopoly on making games for families.
The browser isn’t there yet, for playing emotionally immersive games. Even a relatively visually simple 3D world like the one in Minecraft, gets built in a separate client.
A game can be shaped by its community. The community doesn’t just talk about the game, they jointly make the game with its creators.
Curious what other ways it matters; I’m sure there are many. Feels like the beginning of something that may last a long time.
Pageload times are a problem for some online media companies – for good reason — heavy content, with video, high-res images, and features jammed in. (Let’s set aside the separate problem of content/feature bloat, and assume for a second that it’s a good thing to have all that content and all those features.)
IGN sites need to load much faster — and we’re down to 6 seconds or so on a typical site, from 7.5 seconds a year ago. But it still takes too long to enjoy the site.
But as much as we improve our content and features, about 40% of the load time for IGN.com is still in the ads — even though the ads take up less than a fifth of the real estate on the page. We introduced an open source JavaScript library to call ads simultaneously and ensure content loads, but there is much more we need to do.
Why are ads slow? First, the creative is getting better and so it takes longer to load. Ten years ago, online ads were mostly just flat images and everything else was called “rich” media — anything animated, interactive, and therefore heavier. Last year, two-thirds of the display ads on the Internet were rich — so rich is the new standard.
And the delivery of online ads is a complicated handoff from third-party ad servers to creative agencies to clients and to publishers. Add in targeting services and a daisy chain of ad networks and the delays stretch longer. This relay race happens under constant deadline, and creative is often discarded after a campaign is over — so there is less re-use of code and tools than there should be, and a lot of invention on the fly.
Advertisers can’t solve this alone. Advertisers don’t spend their time focusing on shaving tenths of a second from load times, but publishers do.
So online publishers need to help advertisers by making it easier to speed up ads. We need to not make ad calls when ads are unsold, we need to treat it like a shared issue — not like someone else’s problem.
This is an opportunity for advertisers — users would enjoy their marketing campaigns, and probably pay more attention to them (though we need to do the research on that), if they only loaded faster. In fact, online ad blocking tools promote how much faster it is to surf the web if you’re not looking at the ads.
The advertisers still need to have a full palette of creative choices: explosions, video, animations, all that stuff. We need to apply all the tools of speeding up the web — compressing files, minifying, reducing requests. Imagine a toolkit creative agencies could use that would create faster ads out of the box. We should see how far we can get.
(And thanks to Mediapost for sharing this with their audience as well.)
I run a company whose product is written in code, and I don’t yet speak the language. I sometimes feel like a newspaper publisher who has to take his editor’s word for it that the articles are good. You trust your people, you know you could never write the way they do, but it would still be good to be able to read.
Coding, no surprise, is also a different kind of thought from what I do all day every day. I have to eliminate distractions to do it, and – if I can ever get that flowing thing going – I bet it could even be meditative. I’m enjoying the variation in ways of thinking.
I love the imagination of it – I find myself putting as much work into thinking what to build as how to build it. And only some of the “how to build” work is actually coding, more of it is envisioning how the different pieces of a project should relate to each other, what they should each try to do.
I love how alien it is to me. Every time I open my editor, I’m reminded that I can’t just “jump in for a few minutes” the way I can with my usual work – either because coding isn’t like that, or because I’m not good enough yet. It takes me time to rev up, remember what I was supposed to be doing and how it fits together. Forced syntax, ugly websites from the 1990s that still seem to be the sources of record for a beginner’s questions, the sense of a whole world that I’ve only just begun to explore…
Steve Jobs claimed Apple TV was always a side project, because there was no way to get people to buy a separate set-top box. (Ignoring all the Xboxes and PlayStations folks have bought, natch.)
There’s one other obstacle to making a great Internet set-top box: the user controls stink. It’s really hard to get at all the content available on the open Internet (or even on Netflix or iTunes) while keeping the controls simple. You end up either using a wireless keyboard and trackpad (what I do at home, even though it’s ridiculous), or scrolling through the alphabet using a regular remote, which takes forever.
The verbal controls on Xbox Kinect are a step in the right direction, but they only help with certain commands (“Pause”) and it doesn’t feel quite natural yet.
Siri could fix all that. An Apple TV, with Siri, plus an app store so you can play games, will be something that could belong in every living room.
My first job was an internship at New York’s old Chemical Bank in 1994, that I got as a prize for winning a debate tournament – don’t ask. I was surprised because one of the top guys didn’t know how to use a computer, and we’d had one in our house for 10 years (my dad was an early adopter before they had a name for it). This Mr. So-and-so thought he didn’t have to use a computer, because he had “someone else to do it for him.”
A few years later, at another job, one of the top guys knew how to use a computer for word processing, but didn’t do his own email – his assistant printed out his emails, and he wrote replies by hand on the printouts (don’t ask) which she then typed back to the sender. He had the same logic, he didn’t have to email because he had someone else to do it for him.
The expectation of computing skill in a run-of-the-mill office job keeps going up. There was a time when a job applicant’s resume would say “Proficient in Microsoft Office” – but now nobody really says that anymore because it’s like saying you’re proficient in breathing. And these days if I see a resume from a recent college graduate that says “Proficient in HTML,” I think the same thing.
This will continue. It’s not hard to imagine people in any role writing a simple script for themselves to automate a task (the line between coding and doing a macro is blurry), or testing an idea with a simple web application. Eventually, maybe everyone will truly have to be able to code to effectively do any office job.
IGN has people who code in almost every group – people doing “editorial engineering,” “sales engineering,” “finance engineering,” “operations engineering,” and of course product folks who will mock some things up themselves or just go ahead and build working features, designs, and services. Granted, we’re an Internet company, but not all our work is unique to our industry.
I don’t yet know how to code. Years ago, I took classes (BASIC, when I was in elementary school, and a C++ starter class in college where my grand achievement was a card-counting program for Baccarat that told me even a computer can’t get much of an edge), but I can’t yet build anything useful.
So I’m learning. I don’t want to become the guy who thinks he doesn’t need to know how to code because I have someone else to do it for me. Never too late.
I’ve heard a few speakers in the last couple days talk about the state of education. The usual litany: failing grades on basic skills, and how college is necessary for an upwardly-mobile career yet only a third finish college.
But I’ve been wondering lately, hang on a second, is college really necessary for an upwardly-mobile career?
I learned a lot in college. Mostly from my friends. A lot from working at a nonprofit. Some from great teachers. It informed my outlook, and broadened a love of learning. The but: very few classes actually prepared me for the world of work (even on soft skills like discipline, communication, thought processes).
And if two-thirds of us don’t go to college, can our best hope really be reversing that? Or, instead, is it to make sure that high school can give people the skills for an upwardly-mobile career?
The manufacturing jobs that once created a reliable path to the middle class are, despite the best efforts of city and state governments across the country, going away. And only a third of the jobs to be created between now and 2018 even need a college degree.
The data on the effect of college on income are marred by all kinds of correlation-not-causation issues. They’re basically inconclusive, as I understand them (though I may not).
Employers are complicit, too. We use college as a signal of someone’s success — but what if it isn’t? Dropouts who succeed — Gates, Zuckerberg, Hilleman — are treated as an exception, but what if they’re just a sign of things to come?
My high school, Stuyvesant in New York, was once a vocational school, and I wonder if it could be again.
More to come on this, but I think coding could be an answer to this — a trade that could be a mass profession, at scale. But, whether it’s code or not, it feels like it should be something. And that students should have a chance to learn something useful, long before college. As Paul Graham says, “The only real difference between adults and high school kids is that adults realize they need to get things done, and high school kids don’t.” But they could.