As outlined on the Android Developers site, Google now uses the data collected when users visit the Google Play Store; under the previous system, any check-in to the store by the device would have been incorporated into the results, user-generated or not. The new system went into effect starting with this month’s results.
The change essentially skews the results towards those users who are actively visiting the Play Store. Google says as much on the page itself, noting that the new system “more accurately reflects those users who are most engaged in the Android and Google Play ecosystem.”
The Verge’s headline for this article: “Google changes how it measures Android version adoption, sees uptick in Jelly Bean devices”.
A more accurate title would be: “Google changes how it measures Android version adoption to show an uptick in Jelly Bean devices”.
You could argue that it better reflects the breakdown of OS versions among active buyers, but that means that Google’s statistics on version adoption can no longer be trusted to represent Android as a whole. Given that Play Store engagement seems about as disproportionally low as web-browsing marketshare on Android, these numbers now have far less relevance to the real Android market and aren’t useful for much.
It’s hard to see this change as anything but a desperate move by Google to attempt to hide Android’s poor update-adoption rates.
Think of Tapstream’s Taps.io as Bitly for apps: create shortened links directly to your app’s lander or its App Store page, and it will tell you how many clicks you get, and more importantly, how many downloads you get from each link. You can even drill deeper and find out how many in-app purchases or social shares were driven by any of your Taps.io links. Fully social-ready with HootSuite Enterprise integration, Taps.io makes promoting apps as easy as it should be.
And this week, Marco’s readers can get the Pro account for only $50.
Thanks to Tapstream for sponsoring Marco.org this week.
Tesla announced a “new finance product” that’s an interesting hybrid of buying and leasing. Supposedly, you buy the car with favorable financing terms, but with a minimal down payment (after tax subsidies) and a buyback option at a fixed residual after 36 months. It sounds like the tax benefits and flexibility of owning, but the predetermined resale benefit of a lease.
I wish Tesla would stop playing fast and loose with the numbers, though:
When considering the savings from using electricity instead of gasoline, depreciation benefits and other factors, the true net out of pocket cost to own a mid-range Model S drops to less than $500 per month.
What does “true net out of pocket cost” mean?
The middle Model S, by Tesla’s own calculator, will require you to pay at least $7,990 as a down payment — probably more — before you get any tax credits.1 Even ignoring the significant up-front costs, their calculator is quoting a $711/month “total cost of ownership”, which is actually a $1,252/month car payment on a 5-year loan.
That’s pretty far from “less than $500 per month”, so let’s give them the benefit of the doubt and assume that by “mid-range Model S”, they really meant “the cheapest Model S with zero options”. That brings the actual payment to $1,097/month (again, hand-waving away the substantial up-front costs), or a $471/month “effective cost”. This is probably the figure that Tesla based this claim on.
But that price is also based on a number of extremely optimistic assumptions:
You use the car for business 100% of the time by the IRS’ definition and deduct it from your taxable income to net a $222-month tax reduction. Without this, the “effective cost” rises to $693/month. This, by far, is the worst default assumption in Tesla’s calculator.
You live in California. If you live in Colorado, Illinois, Georgia, or West Virginia, you’ll pay a bit less. If you live in any other state, increase that “effective” price by $69/month.
You drive 15,000 miles per year and you’d otherwise pay an average of $5 per gallon of gasoline over the next few years and your car in this alternative scenario would require premium gas and average only 19 miles per gallon, netting you a $284/month effective savings after average electricity costs to replace the fuel.
But a $5-per-gallon average over just 3 years seems shaky at best, many luxury sedans get much better gas mileage than 19 MPG (and some don’t require premium gas), and the average U.S. driver drives about 13,476 miles per year. If you drove the average number of miles and would have otherwise bought a BMW 528i (the 5-series was the best-selling premium sedan of its size class in the U.S. in 2012) and averaged $4.50 per gallon over 3 years, for instance, this $284/month savings would be reduced to $140/month or less.
So if you live in most states, don’t deduct your vehicle as a business expense (most people’s vehicles don’t qualify), and would otherwise buy a 5-series, your “effective cost” by Tesla’s calculator rises to a minimum of $906/month — and that’s for their cheapest model with no options, assuming that a surprise rich benefactor comes out of the woodwork to pay all of your sales tax and the remaining up-front costs after tax credits.
Tossing around that “less than $500 per month” figure to the press and the public is dishonest at best.
The middle model, 85 kWh but not “Performance”, starts at $79,900 MSRP with no options added (unlikely) before a $7,500 U.S. federal tax credit. The 10% down payment required by the bank needs to be based on the full, pre-tax-credit sale price since that’s what the bank is paying, so the down payment for a “mid-range Model S” is at least $7,990 plus sales tax (unless tax is rolled into the financed amount, in which case the monthly payments will be noticeably higher). Average U.S. sales tax seems to be about 7%, which would put the down payment at $13,583. ↩︎
The next-gen Thunderbolt tech (code-named Falcon Ridge) enables 4K video file transfer and display simultaneously in addition to running at 20 Gbps. It will be backward-compatible with previous-gen Thunderbolt cables and connectors, and production is set to ramp up in 2014.
This could enable the first generation of desktop Retina displays: it wouldn’t surprise me if the first standalone Retina display was a 23” panel with exactly 4K resolution (3840 × 2160), run logically as 1920 × 1080 (1080p) at 2X, and driven by upgraded Thunderbolt ports in the next generation of MacBook Pros and Mac Pros.
Tom Chivers, The Telegraph, on how lower vaccination rates caused by unsubstantiated autism scares led to a recent UK outbreak of measles.
Dr Ben Goldacre, the author of Bad Pharma, who has written extensively about autism and vaccination, says: “Health scares are like toothpaste: once they’re out, it’s very hard to get them back in the tube. They catch fire fast, because they’re so seductive to journalists. And they don’t go away, because once you’ve planted fear and doubt in the mind, people become almost superstitious. Even doctors I know feel worried about MMR, knowing it’s irrational.”
It’s sad, and scary, that anti-intellectual, anti-science superstition about common vaccines has made it more likely that our children will contract and spread these diseases than when we were younger. Anytime we make the world less safe against easily solvable problems, we need to seriously evaluate what we’re doing.
To put the drive’s excellent price in perspective, the 960GB M500 has roughly the same MSRP as Intel’s 80GB X25-M had back in 2008. That’s an order of magnitude more storage capacity at the same price in 5 years time. Moore’s Law makes me happy.
After AnandTech’s usual barrage of tests and solid analysis, this sounds like a great SSD for a lot of people: it’s not the fastest, but it has more consistent performance than most, and it’s the cheapest by far among other 1 TB-class SSDs.
More important, tacking an S onto the existing model number sends a rather weak message. It says that this is our “off-year” product, with only modest improvements. If holding off on the big number change achieved some great result, I might think otherwise. But look what happened with iPhone 5.
This model brought major changes: bigger screen, better camera, greater speed, all on a thinner and lighter body. Yet its improvements were still dismissed by many as “incremental.”
I agree: if Apple’s going to keep using sequential numbers (rather than feature-based names, like the second iPhone being named “iPhone 3G”), they should just give every model the next number. The next iPhone should either be the iPhone 6 or the iPhone Something Else, not the iPhone 5S.1
The iPhone 4S was a huge improvement over the iPhone 4, but the press and fans shat all over it because it had the same case design and therefore wasn’t “an iPhone 5”.
The 4S shipped during the clear beginning of the current era of Apple pessimism. Apple reduced people’s expectations with the “4S” name at the worst possible time: just as many were starting to think iPhone innovation was slowing, Apple named the new iPhone in a way that suggested, “This isn’t a big improvement.”
It’s a move that they could have made when they were on top of the world, PR-momentum-wise, much like when they released the iPhone 3GS, a similarly substantial improvement over the iPhone 3G that was similarly underrated by the public. But when the 4S was released, Apple no longer had that much positive momentum.2
A year later, when Apple did release a model named “iPhone 5” that was far better than the 4S and had an external redesign, the inertia of Apple pessimism was so strong and the press had become such petulant children about Apple products that they shat all over it even though it was a huge update that gave them everything they asked for, plus more.
Now, Apple pessimism is even stronger. No matter what they release and no matter how well it sells, they won’t win over the press, the pundits, the stock market, or the rhetoric. Not this year. They could release a revolutionary 60-inch 4K TV for $99 with built-in nanobots to assemble and dispense free smartwatches, and people would complain that it should cost $49 and the nanobots aren’t open enough.
Since they’re not going to meaningfully improve their PR momentum anytime soon, they might as well at least avoid trying to make it worse. This is not the time for Apple itself to suggest to the world that it’s slowing down innovation on its most important product.
Regardless of how big of an improvement the next iPhone is, Apple should just call it the iPhone 6 and give the finger to anyone who questions whether the name fits.
Plus, since the “5” and “S” characters are so similar, a “5S” model would be more awkward than usual. ↩︎
To make matters worse, they heavily focused the iPhone 4S’ marketing on Siri, a feature labeled “beta” that not only lived up to that label, but is fundamentally disappointing to most people’s expectations even when it works reliably. ↩︎
But wouldn’t Google prefer “stock Android”? Well, the ability to replace the home screen is a built-in, fully supported Android feature. Google intentionally implemented it. This feature is, in fact, part of “stock Android”.
Android’s home screen is largely irrelevant to how Google makes money with Android. Google cares very much if you ship an Android phone with a different default search provider, with a non-Google default browser, without Google Play, or without Google Now. But replacing the home screen? Probably not a big deal.
An interesting counterpoint to the “Google must hate Facebook Home” assumption, but I’m not convinced Google sees this as a net neutral or win — no matter how much (or how little) it directly hurts Google, it’s a potentially huge gain for their biggest competitor, and that’s never good. To add insult to injury, their biggest competitor built this potentially huge advantage on Google’s own platform.
It’d be one thing if Facebook somehow convinced Apple to build Home into iOS. But to take over as the social layer of Android — something Google has failed to do for its own struggling Google+ product, probably because of internal conflicts — is truly sticking it to Google, regardless of whether phones infected with Home will still use Google’s apps under all of those disembodied heads.
“Apple is Big Brother” has become a default narrative about the company. Apple stands for closed systems, proprietary everything, and a level of control over the way their customers use their products that would send us all fleeing for the hills if we had any common sense.
The blanket prohibition came in the fine print of a policy made public this evening, which says “Glassware” developers may not “serve or include any advertisements” and they “may not charge” users to download apps for the device.
As a result of the vulnerability, this group gained access to a web server, parts of our source code, and ultimately, our database. … Our investigation reveals that this group did not have access to any other component of the Linode infrastructure, including access to the host machines or any other server or service that runs our infrastructure.
If you’re a Linode customer, it’s also worth reading the abridged IRC log from one of the hackers bragging about the attack to get an idea of what was taken and how. It may influence your decision whether to remain a Linode customer. I’m on the fence.
Linode claims that Manager passwords were stored as bcrypt hashes, so they should be reasonably OK, but Lish passwords and API tokens were stored in plaintext. I’ve never used Lish, but if you have, consider those passwords permanently compromised.
Credit-card info might be at risk. It was stored with public/private-key cryptography, which is very smart in theory — the customer-facing machines could encrypt the data using the public key, while only the (presumably separate) billing machines could decrypt it with the private key — except that Linode apparently kept the public and private keys together, so the hackers have both.
The only protection is the passphrase on the key. A rep from Linode says:
The passphrase is not guessable, sufficiently long and complex, not based on dictionary words, and not stored anywhere but in our heads.
It’s up to you whether you’re confident enough in the passphrase’s complexity to continue trusting any credit-card numbers you’ve used with Linode.
Passwords are no longer sufficient for protecting important logins. Two-factor authentication — something you know + something you have — is a practical, proven technique for preventing account takeover.
Duo Security offers two-factor authentication as a service for SSH, VPN, web apps, and more. The service takes only a few minutes to set up and supports a variety of authentication factors: mobile app, phone callback, SMS, and tokens such as YubiKey. The best part? If you’re using a smartphone, there are no codes to transcribe: just tap “Approve” to authenticate.
I’ve seen lots of discussion about the paid-app market recently, but most of what I’ve read hasn’t sat quite right with me.
Lex Friedman’s piece tries to persuade customers to pay more, which is a nice sentiment, but it won’t achieve meaningful change. Nobody can yell loudly enough at enough people to change the app market’s pricing expectations and behavior. The market is far bigger than any of our audiences.
Chris Adamson argues that there’s effectively no market for paid apps, and most iOS-developer income comes from contract work:
Earth to MacWorld: It’s already too late. The market has spoken, and it refuses to pay for apps, even when the toxic side-effects of that are manifest. …
Of all the full-time iOS developers I know, almost none of them work primarily in writing apps for sale to end-users. Nearly all of them, myself included, work for clients for whom an app makes sense in some other business model.
Certainly, it’s easier to make a living working for other people than trying to make your own product. But that has always been true in the iOS market, and in fact, it’s true everywhere: iOS development, Mac development, Windows development, web development, and most other industries in the world.
When you work for someone else, they’re taking the risk (or already have). You get paid regardless. (Well, most of the time.) The downsides are that your earnings are effectively capped and you have far less control over what you make.
This is a good balance in most cases, including development, because there isn’t enough demand for individual mainstream apps from every developer capable of writing them. In other words, just because you can write and sell your own app doesn’t mean that there will be a substantial market for it. Imagine if 1,336,300 people in the U.S. alone each designed a new kitchen utensil: a few thousand may be useful to a good number of people, but the rest probably won’t be different enough to be noteworthy, will solve problems that too few people have, or will simply suck.
Above all, build software to meet a need and don’t become a commodity or enter a commodity space. Not all needs are equal. I need air way more than I need another drink cozy. Weather apps and Twitter apps are fun, beautiful, and engaging, but they are also very difficult to earn sustainable revenue.
He’s right. The issue with the iOS market is about more than price: it’s about demand.
I currently have 91 non-Apple apps on my iPhone, but I only use about six of them frequently. The rest are mostly utilities for specific, occasional needs, or games that I played for a few hours a long time ago and don’t have the heart to delete. Six is probably below the average for frequently used apps, but I bet this ratio isn’t too far off the mark.
My Big Six — Instapaper, Downcast, Reeder, Tweetbot, Instagram, and Dark Sky — solve big problems exceptionally well, and with the exception of Instagram, these are problems I’ve used apps to solve for almost the entire time I’ve owned iPhones.
I haven’t always used these particular apps to solve these problems, but it takes a lot to change my mind on one. If you make another RSS reader or Twitter client, there are certainly a lot of people who could use it, but you’ll need to compete with very mature, established apps. Competing in these categories isn’t about price: it’s about relevance and attention. If you can’t find enough customers here, it’s probably not because you’re charging $2.99 instead of $1.99 or $0 — it’s because your app isn’t convincing enough people that it’s worth using over the alternatives.
For these “Big Six” apps, price is almost irrelevant. If your app is useful enough for many of its customers to use it almost every day, they’ll pay a decent price for it. (Not all of them will — but you don’t need all of them.) The challenge is either making your app that much better than the alternatives, or finding new app roles that are that useful to a lot of people.
If you eschew common app categories and try to address a smaller niche to give your app a better chance of standing out, you run a different risk: your niche might be a lot smaller than you think. Usually, this happens:
Paid app isn’t gaining traction.
Developer lowers the price to boost purchases and climb the ranks.
App doesn’t sell much better, and developer just makes less money.
I’ve seen so many developers fall into this trap, even with good apps: an app can be good but not compelling enough to attract many customers. A game can be beautifully crafted but just not fun.
This isn’t a pricing problem.
Pricing does skew the market in other areas, and the App Store’s poor design exacerbates much of the dysfunction in its dominant top-list culture. And pricing can sometimes hinder an app’s chances of success by its nature: if your app has a substantial social component and it isn’t very useful without it, you need as many people as possible to start using it as quickly as possible, so social apps must almost always be free.
That’s not really a pricing problem, per se. It doesn’t matter what your price is — it only matters whether you charge at all, because that slows the rate of new users. This is complicated because you’re not giving people much value in the app itself: you’re relying on the customers to provide most of the app’s value to each other, and you probably need a lot of them to give substantial value to any of them. Social apps must therefore almost always seek indirect, deferred revenue because having as many users as possible is worth much more, long-term, than hindering growth with a paywall.
Since a few successful social products have been so big and high-profile in the last few years, many people mistakenly assume that the must-be-free rule is the new norm for all web and software markets, but that’s not the case. Even in social, it’s a massive gamble — since they rely so much on being extremely widespread to have much value, most social products don’t succeed. There’s not much of a “middle class” of social products. The big ones get bigger, and unsuccessful attempts become nearly worthless. Chris Adamson’s theory is actually correct in the social market, but these rules don’t apply to non-social categories.
In most categories, if you either solve a new problem that a lot of people have, or solve an old problem in a new and better way, you can sell a paid app today just as well as you could in 2008. In fact, the market is much bigger now. But, as with any maturing market, you’ll need to do more to get noticed since so many problems have already been solved so well.
From High Scalability, condensing lessons from a talk by Pinterest engineers Yashwanth Nelapati and Marty Weiner:
When you push something to the limit, all technologies fail in their own special way. This led them to evaluate tool choices with a preference for tools that are: mature; really good and simple; well known and liked; well supported; consistently good performers; as failure-free as possible; free. Using these criteria they selected: MySQL, Solr, Memcache, and Redis. Cassandra and Mongo were dropped.
This is one of the smartest, most dense collections of web-scaling wisdom I’ve ever seen.1
Even at smaller scale, these lessons apply. An independent developer or small company can’t afford to waste time messing with flaky, overly complex, or high-needs server infrastructure.
At Tumblr, I used PHP, MySQL, and Memcache. I then built Instapaper with PHP, MySQL, and Memcache. When I needed to build a server back-end and CMS for The Magazine, I used PHP, MySQL, and Memcache. I’m now building a little sponsor-management tool for myself that will probably only ever be used by a maximum of four people, and I’m using… PHP, MySQL, and Memcache.
Using mature, reliable, widespread tools isn’t just about scaling more easily — it’s about being as low-needs as possible so you can spend more of your time and attention on things that matter more to you.
Every well-read site gets comments. (For people like me who don’t let others publish comments directly on our articles, we still get plenty of commentary — just in other places such as email, Twitter, and Hacker News.)
Every app with a nontrivial number of downloads is likely to have comments, too, in the form of App Store reviews. This goes far beyond the app market these days, with stores like Amazon allowing anyone to review products and media, and sites such as Yelp that will publish anyone’s review of any restaurant, doctor, or church.
If you put yourself out there at all by offering a product or service, you’re going to get comments, usually anonymously and outside of your control, potentially inaccurate, malicious, or not credible, yet available for all future prospective customers to read.
Now, think of something you saw recently that had a lot of comments or reviews. They’re not all going to be positive. What could its creator have done to please all of the negative commenters?
Go ahead, this isn’t a trick question. You know the answer.
No matter what you make or how much you charge, some people will find things to complain about. If you drop your app’s price all the way down to free, people will still complain — just not about the price. They’ll move on to the features, the implementation, the design, the updates, the way you look, or what kind of dog you have. They’ll complain about every facet of your app, and then they’ll complain about unrelated topics just to pile on. They’ll say they use your app every day and love it, then give it a two-star rating until you add their pet feature. They’ll drop you from five stars to one star after an update that broke their edge case, then never come back to update that review after you fix it. They’ll complain that your productivity app isn’t a very fun game. They’ll scream at the world that your app is “useless” because they don’t like something about it. They’ll jailbreak and install a bunch of hacks that make your app crash, then leave you a one-star telling everyone that your app crashes constantly without mentioning their hacks.
You will never please everyone. You will never win that battle.
When evaluating complaints, we need to consider whether the complainer is credible, whether they have reasonable expectations, and whether a significant number of others have made similar complaints or are likely to have experienced similar problems. For many complaints, a reasonable outcome isn’t possible or pragmatic, and the best solution is to ignore them.
Developers are usually able to maintain a healthy balance of knowing which feature- or design-related complaints are worth paying attention to and which aren’t. So why is it that we lend so much credence to every person who complains that our $3 app is too expensive?
Someone saying they won’t buy at your price is just one data point. Each sale of your app is another data point. If you sell 100 copies of your app and get 3 comments on Twitter from people saying it’s too expensive and they won’t buy it, I’d say you’re doing great.
It’s impossible to get every customer. Fortunately, you don’t need every customer. If you sell a 99-cent app to just 1% of the people who bought new iOS devices in the 2012 holiday quarteralone, you’ll clear about $519,750. Not a bad quarter.
The most important input for your pricing is quite simple: are enough people buying it at its current price? If not, you might have a pricing problem, but not necessarily. If enough people are buying it, you face the interesting question: can you charge more?
If the traditional band business model is to generate hype through the media and radio airplay, and then monetize that hype through album sales and tours, Phish doesn’t fit the model at all. For a band of their stature, their album sales are miniscule and radio airplay non-existent. And so when the “music business” cratered in the 1990s because of file-sharing and radio’s importance declined because of the internet, Phish remained unaffected and profitable as ever.
What fascinates, puzzles, or annoys most non-fans about Phish is that their appeal doesn’t seem to make sense, and the fans who try to convince them to like the band often can’t make a very persuasive case. I’ve even tried to explain the appeal myself, but it rarely sticks.
Explaining Phish’s appeal in the context of mainstream music doesn’t make sense, because Phish doesn’t make mainstream music. When they’ve tried to condense their music into the mainstream formats of short, flawless, studio-recorded songs grouped into dozens and crammed into 50-minute albums every 18 months, most of their appeal and personality have been lost. The mainstream only sees these mediocre adaptations and, understandably, can’t figure out why so many people like the band.
If you’ll permit a pretty rough analogy, imagine a world in which the vast majority of published fiction was in the form of 3,000-word short stories, and most people had never read anything longer. Phish is the one outlier publishing novels, and they’re pretty weird, complex novels. No effort to condense such novels into bite-sized short stories will truly capture the appeal.
But if you’re one of just a handful of novel publishers in this rough metaphor, you’re going to slowly accumulate a hell of a fanbase from the people who actually like novels, even if yours get a bit too weird sometimes, because almost nobody else is creating what these fans want and love.
Crashlytics addresses a huge hole in mobile app development: deep insights into an app’s performance to pinpoint and fix issues quickly and easily. Built by a hardcore team, Crashlytics is the most powerful, yet lightest weight crash reporting solution. We find the needle in the haystack, even the exact line of code that your app crashed on, so you can quickly scan and trace an issue. Crashlytics has been deployed in hundreds of millions of devices and powers thousands of today’s top applications, including Twitter, Square, Yammer, Yelp, and GroupMe.
Dr. Drang on the widespread coverage and frequent overanalysis Apple’s quarterly earnings:
I don’t use a Mac or an iPhone because of Apple’s balance sheet. I use them because they (usually) work for me, and I save my anger and frustration for when they don’t. I’m far more affected by an iCloud screwup than by the iPhone’s market share, by Lion’s removal of Save As… than by Apple’s stock price, and by Preview’s lack of AppleScript support than by anything Rob Enderle says.
Glenn’s overview of this excellent issue, which includes articles on vaccines and “personal science”, the legality of BuzzFeed’s photo usage, the origins of the widespread Couch To 5K running program, a bizarre coin-op museum as a metaphor, and small-batch distillery stills.
When I launched Instapaper in 2008, it was a very basic web app. It quickly expanded to define the pillars of the read-later market: a one-click “read later” bookmarklet, a web sync service, an adjustable text view optimized for reading, and an iPhone app with offline saving. I did almost everything myself, which worked well for the first few years, but for the past year, I’ve had a lot of trouble keeping up with it.
Instapaper is much bigger today than I could have predicted in 2008, and it has simply grown far beyond what one person can do. To really shine, it needs a full-time staff of at least a few people. But I wouldn’t be very good at hiring and leading a staff, and after more than five years, I’d like an opportunity to try other apps and creative projects. Instapaper needs a new home where it can be staffed and grown, but I didn’t want to give it to a big company that would probably just shut it down in six months.
A couple of months ago, at 1:30 AM, I suddenly realized who should take it over. I jumped out of bed, tiptoed downstairs (no parent wants to wake a sleeping baby), and sent an email. It didn’t take much convincing, because we both knew it was a great fit.
I’m happy to announce that I’ve sold a majority stake in Instapaper to Betaworks. We’ve structured the deal with Instapaper’s health and longevity as the top priority, with incentives to keep it going well into the future. I will continue advising the project indefinitely, while Betaworks will take over its operations, expand its staff, and develop it further.
I’ve known Betaworks for years, and I’ve spent a lot of lunches at their office. They have great engineering talent, great product direction, and plenty of experience running services at Instapaper’s scale. I wouldn’t put Instapaper in just anyone’s hands, and I know that they’ll do right by it.
With Betaworks’ drive and resources now behind it, I’m confident that Instapaper has a very bright future. I’m looking forward to seeing what they can do.
This week: Laptops in school, getting tech jobs, mail servers and spam, Steve Jobs’ direct commentary, WWDC tickets, going ticketless or watching WWDC videos at home, Cook’s hints on the earnings call, bored consumers, and Marco’s mom’s first smartphone.
(Recorded on Wednesday night, after WWDC was announced but before tickets went on sale.)
John Siracusa’s excellent “The Lottery” summarizes most of my thoughts on the unexpectedly quick (and unexpectedly random) WWDC sellout.
What made WWDC so special to my friends and me in the past is that you could see the same core community there every year, because we were the ones who cared so much about getting tickets that no surprise or inconvenient timezone would stop us. The last few years, we’ve been on high alert for the entire month of April, and when tickets went on sale last year (and sold out in 45 minutes), most of us easily bought our tickets within the first five minutes. We were ready.
Every year, that community got larger and larger, but the conference’s size has been bounded by practicality and its venue. Apple has faced a significant problem growing the conference: every year since 2008, some people haven’t been able to get into WWDC who wanted to. And every year, that number has grown substantially.
Before, it was effectively merit-based: whoever cared the most (and had the money and ability) to get a ticket could get one. Now, no matter how much you care and no matter how early you get there, it’s a lottery.
Many people complained that Apple’s sudden availability in previous years was unfair, and they were right. Everyone didn’t have a fair chance at getting a ticket — only the people who devoted the most to getting one had a good chance of it. Most people who weren’t expecting tickets to go on sale soon, didn’t create or use alerting systems (or slept through them), or just needed time to get approval from their bosses were locked out last year.
This year, by giving everyone advance notice for the first time, Apple did fix that problem: it is more fair. A similar percentage as last year of interested developers got tickets (or didn’t).
But now, that core community that I’ve enjoyed for years is fragmented. Our elaborate hacks and dedication previously allowed almost all of us to get tickets, but now, we have the same success rate as everyone else: some got tickets, most didn’t. Apple eliminated our unfair advantage.
It’s hard for Apple to come up with a better solution if fairness is the goal (and it probably should be). This is the new normal.
I’m going to miss what we had. But like everything else in this business, it was temporary. That’s what makes it such an exciting business. WWDC will go on, but it’s going to be very different for my friends and me. Here’s hoping that the community finds creative ways to replace what will be lost. Fortunately, it usually does.
Every year, Apple tries to address the excessive demand for WWDC in new ways.
They already make the WWDC videos available promptly after the conference to all registered developers — and every year, they come out faster. They already periodically send the small Developer Evangelism team around the world doing small “Tech Talks” (free, one-day mini-WWDCs, effectively), and they give attendance priority to people who haven’t attended WWDC.
It hasn’t helped.
Developers usually suggest expanding the conference or having more conferences throughout the year, but both of those have big problems. Jeff LaMarche explains why and proposes a sensible expansion idea, but I don’t think Apple would do it, and it would probably only increase the number of available tickets by two or three times at most — the additional scale may ruin the conference, and when you’re already selling out in less than a minute, it’s not going to make a huge difference to availability.
Daniel Jalkut came up with the creative proposal of Apple ending WWDC entirely, then addressing our demands by significantly expanding the online developer resources and documentation. This sounds plausible on paper, but would crush the spirit of our community. Apple should certainly expand the developer resources, but WWDC provides a lot that online resources can’t.
WWDC has an energy. It’s a huge rally to juice developers’ confidence and enthusiasm for the platform. Every year, I’ve been filled with an insatiable desire to just make something the whole time, and that energy gives me a boost for months afterward.
The reason we all want to go so badly is because it’s great.
When you go, you’ve allocated that week to learning this stuff. That’s your only job. You leave most of your workplace and family obligations behind for a week so you can spend all day in sessions, meeting great friends and colleagues in the industry, and immersing yourself in the technology, community, and culture.
Some of the academic and technical benefits can be had outside of that environment, but watching the videos alone at home doesn’t bring the spirit of actually being there. And since watching those videos at home is never going to be your all-day-every-day-for-a-week job, how likely are you to actually watch many of them?1
And while the usual community has been disrupted and we’re likely to see continued growth of the smaller, independent conferences, WWDC is still going to be the big one: the one that most of us try to attend every year, the one that matters most, the one with the best access to the most people (Apple or otherwise), and the one that Apple kicks off with an enthusiastic keynote that excites us with new OSes, new features, new APIs, and potential new markets for our apps.
No matter what Apple does to address the extreme demand for WWDC, that demand will always be there as long as so many people want to develop apps for Apple platforms, and no amount of videos or documentation can replace the real thing.
Personally, I always miss a few sessions that I tell myself I’ll watch on video when they come out. I almost never have. The videos serve as a reference when I look something up, but it’s just never “the right time” for me to sit down and watch a session video just to learn about something cool that I don’t necessarily need right now.
I’ve asked other developers, and this seems to be the case for most of them. ↩︎
I invited myself onto Dan’s Quit! show today to talk about “quitting” Instapaper: how I knew it was time, feeling burnout and guilt, weighing the options that a single-person company has when the person wants to move on, and being selective about who buys it.
This was effectively a public therapy session, which greatly helped me unload the feelings I’ve had for the last six months or so. (It’s long, but my part mostly ends 57 minutes in, before the sponsor break and phone calls.)
Essentially, rather than having comments at the bottom, Medium’s “notes” are annotations inserted in the margins, and they’re private by default — authors must explicitly make each one public.
It’s a great idea. It still suffers from some of the problems of comments — for instance, cultivating the expectation that anyone’s thoughts deserve to appear to your audience and on your article — but by making them private by default, it removes a lot of negative incentives.
I’m still not sure I’d want them on my site. But I also wouldn’t write on Medium, so I’m probably not the target audience.
Pull back the curtain on one of the most annoying aspects of the app economy: not knowing where your downloads come from. Tapstream helps you grow your userbase by identifying sources of new users, engaged users, and high-value users. A single, lightweight SDK is all you need, or choose the SDK-less integration.
Follow which ads, ad networks, social networks, or SEO efforts are driving user growth. More importantly, learn where your highest-value users come from instead of acquiring a bunch of lookie-loos that run your app once and never come back.
Join other great companies, including WebMD, Bleacher Report, and HootSuite, by signing up today.
Tapstream works for iOS, Android, OS X, and Windows apps.
Thanks to Tapstream for sponsoring Marco.org this week.