Dan Grover

Inequality and Mass Transit in the Bay Area

2013-05-07T09:58:55-07:00

A friend of mine and I made a mashup over the weekend. It shows median incomes along different mass transit routes in the Bay Area (BART, MUNI, CalTrain). Check it out!

"Launching Your Indie App"

2012-06-13T08:55:45-07:00

Here are the slides from my talk yesterday at Indie Dev Lab 2012:

Whistling Into a Tape Recorder

2013-03-24T03:49:02-07:00

A friend of mine once told me a rumor about a famous film composer. According to rumor, his process consists of watching the rough cuts of the film, whistling into a pocket tape recorder, handing the tape to a team of orchestrators, and waiting for them to compose the actual score.

It probably isn't true, but it makes me think of a lot of first-time startup founders I've met lately that are trying to find a "technical co-founder."

Forget for a moment the hackneyed truisms about ideas and execution. Ideas are easy. Execution isn't so bad. What people never seem to get is that translating ideas into execution is the hard part.

There are a lot of names for this. There's Hofstadter's Law. Merlin Mann calls it "Just-a-button guy." Ze Frank calls it "brain crack." Steve Yegge calls it "Shit's Easy Syndrome."

But no matter how you stack it, shit isn't easy. Once you've committed to an idea, if you keep pushing, you eventually fall off the precipice of think mode and land in do mode. It's not always obvious once you've stopped falling. And it doesn't matter how deeply something has been specced or how experienced the team is, it never ends up going how you thought it would.

There's a precise moment when a product is more reality than idea. It's not always when someone actually starts coding it; sometimes it can come after months of work or even days before shipping. It's an emotional tipping point where you stop procrastinating little decisions, stop thinking about what could be, and focus on what's actually going to be done for 1.0. It can feel painful, like pulling off a bandage. But once that point is reached, it's all downhill.

Anybody who makes anything — writers, musicians, architects — seems to develop a gut feeling for this and gets better at bringing about the transition with a minimum of fuss.

CODE IS WHERE THE RUBBER MEETS THE ROAD

In software companies, engineers are often (but not always) in a good position to develop a knack for this simply because the last chance to find out if an idea has been fully thought through is when someone sits down to code it. All sorts of questions start popping up. Did you really figure out all the edge cases? Did you forget to design a screen or two? When you put that one feature in at the last minute, did you really think about how it interacts with the other ones? Did you really think about whether people want this?

You can spend months planning and specifying but these kinds of problems always come up at the worst time. Sometimes they can be avoided early-on by sharp product managers and designers. Or by setting shorter milestones and using agile practices. But more often than not, code is where the rubber meets the road.

EVERYTHING IS TECHNICAL

This is why it irks me when people try to sort others into "technical" and "non-technical". Being technical isn't about technology, it's about technique. Techne. Skill in something.

Shipping a software product requires skill in quite a few fields. So many that, at an early stage, you couldn't possibly employ a full-time specialist to perform each unless you have a lot of funding and a lot of patience. There are many types of engineering, many non-overlapping kinds of design, business/partnership development, marketing, press, support, writing, finance, legal, the list goes on.

So it would seem that if you're going to form a team to take on all of this, you should have people that are good at some of these things, and willing to try the rest until they get it right. Anything else can fall on hired help or advisors.

The pervasive idea that a founding team must have one person dedicated solely to engineering and one person for "everything else" is completely arbitrary and possibly dangerous. It's one that's been made into an archetype by people like Jobs and Wozniack, but isn't all that typical in the grand scheme of things.

The only situation where this possibly makes sense is if the "technical" person on the team is the neckbeardiest of the neckbeardy – the kind of person who shuns anything except for pure code. And those people belong in cubicles anyway.

The problem with this archetype is that once you accept it as a given, it becomes not just a divide in skills, but can be conflated into a divide between ideation and execution, thinking and doing, executive and worker, alpha and beta. And once that happens, you risk introducing all the inefficiencies of large companies into your two or three person team.

HOW TO REALLY FORM A TEAM

If, like me, you're pondering what kind of people you want to work with and what role you want to have in a team, it's probably best to forget the archetypes and just find collaborators who:

Can do — that is, have definable, testable skills.
Can think.
Can switch back and forth between the two.

It's a little vague, but it's a start. I'm going to try it out for a while and see what happens.

EDIT: Removed incendiary subtitle. East vs. west isn't the main point.

IndieDevLab

2012-05-20T08:13:08-07:00

I'm going to be speaking at Indie Dev Lab alongside Jay Freeman, Nate True, and Ben Zadik. The conference is June 11-14 in San Francisco.

(see announcement on TUAW)

Etude in Apple Store Promo

2012-02-04T08:13:55-08:00

Apple's Valentines Day promos show Etude running on a white iPad 2!

It's also linked prominently on the iPad site under "From the App Store."

(via benglert)

Etude Acquired by Steinway & Sons

2011-09-11T08:14:31-07:00

I'm happy to announce that my iPad music learning startup Etude has been acquired by Steinway & Sons.

When they approached me about the idea of working together, it seemed like an odd idea. What would a piano company in New York want with an iPad app?

It's not the usual acquisition story for a software business, but Steinway is more of a tech company than one might expect. They practically invented the modern piano and have been patenting improvements to it ever since. They're pretty much the Apple of the musical instrument world.

The main reason for acquiring Etude is that Steinway is thinking about how people will learn, perform, and create music in the digital era. Etude is another step in that direction, and I couldn't have wished for a better group of people to work with to continue the app's development.

Which brings me to my next announcement...

INTRODUCING ETUDE 2.0

I'm also announcing the second major release of Etude. It's the first "Steinway-ified" version of the app, but it's way more than a re-branding.

The biggest request from users was for more music. This update brings a brand new music store with thousands of songs to learn, with everything from J.S. Bach to Justin Bieber (yes, it actually has the beibz).

Etude 2 also features a new way for beginners to see music and not to mention tons of little UI improvements that make everything much smoother.

Oh, and it's a free app now! So go get it from the App Store or check out the site to learn more.

SHOUT-OUTS

Etude 1.0 was a long, solo slog, but since coming to Steinway, I've been lucky to be able to bring in and work with with several very talented engineers and designers to build version 2.

Ben Englert and Jon Toohill tackled the technical challenge of sanitizing, converting, and engraving music originally created in other formats. They also helped build the backend for the new store. Kevin Murphy helped with app UI and with QA, noticing bugs that only a cellist would. The incomparable Jesse Benett-Chamberlain designed the new user interface, icon, and site.

Etude for iPad is here!

2010-06-02T08:15:31-07:00

The iPad version of my music app Etude is now released.

Toward a Grand Unified Theory of n00bs

2010-02-19T16:09:35-08:00

The blogosphere has been abuzz lately about a post on the blog Read Write Web. The post itself is an uneventful one about the growth of Facebook Connect as a means of authentication on third-party sites.

The reason we're all talking about it is that last weekend, it momentarily showed up on Google results pages for the phrase "facebook login." And lots of people clicked on it, thinking it was Facebook's login page (when it was clearly an entirely different site). They wrote comments on this post demanding that Facebook change its login page back. You know, to the pretty blue one that said "Facebook" at the top instead of "Read Write Web."

One blogger summed up the situation astutely:

"It’s like… Like if you asked a friend if there was a Starbucks in his neighborhood and he said, yeah I think there’s one half a mile down, maybe. And you drive half a mile and see a big carwash place, and you park and walk in and ask to speak to the manager. And you tell the carwash manager how unhappy you are with this terrible new Starbucks redesign."

It reminded me of a similarly-fated post on a site called OK/Cancel. Somehow, it got page-ranked as the top result for the phrase "cancel google", so hundreds of irate, confused people used the blog comments for that post to lash out at Google for being set as their browser's homepage.

Or take the Yahoo! Local page for the Chicago Better Business Bureau (or that of almost any city's BBB). People come to the page and leave scathing reviews of the Bureau itself, thinking the listing page is the official site and that the review form is, in fact, a generic contact form to report your grievances about whatever business has just wronged you.

I've always found this kind of thing fascinating. Since I run a software business full-time, I have to deal with it constantly.

Like a few months ago when my company recently gave away a bunch of licenses to ShoveBox as part of a MacHeist promotion. Don't get me wrong, it was worth it, but the customer support load was pretty overwhelming, despite lots of preparations. You shoulda seen some of 'em. We got everything from questions about other random Mac apps, to people who thought the whole thing's a scam because they didn't read the part about actually entering in their serial number, to people threatening to call the FBI (and for some reason the FCC) on us. Craziness.

As a computer savvy person, it's easy to forget how different the average joe is from you. This disparity is the source of much frustration and inefficiency in the world. And it's only going to get worse, unless we do something about it.

If these seem like isolated edge cases to just shrug off, consider the vast financial loss attributed to phishing scams. Any phishing scam can be easily avoided by anyone who can read a URL and tell if the website they're on is, in fact, their bank's official site, but naturally most computer users can't.

The initial reaction of folks like me, of course, is to dismiss these people. After all, they don't know what a browser is. They don't seem to understand what site they're on or who it's run by. They're unfamiliar with basic design idioms used universally across the web (e.g. headings, subheadings, navigation, comment forms). Some seem to lack even basic reading comprehension, and no amount of redesigning or copyediting will enhance the understanding these people take away from a page. They can only be described as n00bs.

How n00bs are different

But there are legitimate excuses for this kind of behavior that can be found by examining some key differences between those who build software and those who use it:

Confidence

Many incredibly bright people simply lack confidence when using a computer. To them, computers are confusing, inconsistent, and baroque in their complexity. Their IQ drops 50 points when a simple question or statement is presented to them on a screen instead of by some other means.

Try to watch anyone over the age of 50 add money to their subway card and you'll see what I mean. BART kiosks are pretty bad, but you should see old Bostonites try to use the reasonably well-designed MBTA ticket kiosks. "Come on, this is wicked confusing. Where'd ya pahk the cah?" Then the poor station agent comes by and explains that it's just asking them "Would you like a receipt?"
Disabilities

It seems stupid for someone to blindly click the first search result on Google without reading it. But what if they actually are blind? That might be a perfectly reasonable thing to do for someone who uses a screenreader. Outside of government websites that are required to obey Section 508, very few sites are designed with disabled users in mind. If they were, Stevie Wonder might not have such a hard time on Twitter.
Nested Hierarchy

Any kind of recursive data structure is baffling for most human minds. It's just a weird concept. A common example of this is a filesystem, but because many computers hide the filesystem from the user (the only notable exception is the Mac), users never get much exposure to the idea. But even sites that have too many levels of sections and subsections baffle people in the same ways, even if the design presents this topography well. It's just not a natural way for many people to think about information.
URLs

URLs are mystifying for regular people. They're background noise.

Not only are they hierarchical (which is a stumbling block in itself, as mentioned above), but their syntax doesn't even intuitively reflect that hierarchy.

Postal addresses are pretty simple — they go in order of small to big. Each part describes an actual region of space which is contained within the region described by the next part. Most people could take an address and carry it to its logical extreme, like in that one line in Thorton Wilder's Our Town:

REBECCA: I never told you about that letter Jane Crofut got from her minister when she was sick. He wrote Jane a letter and on the envelope the address was like this: It said, Jane Crofut, the Crofut Farm, Grover’s Corners; Sutton County; New Hampshire; United States of America; continent of North America; Western Hemisphere; the Earth; the Solar System; The Universe; the mind of God - that’s what it said on the envelope. And the postman brought it just the same.

URLs are totally different, though.

The domain name portion of the URL goes from the bottom-most subdomain to the top-level-domain, but then the path portion is expected to go from the root level of the site down to the specific page they're on.

Many websites have pretty crazy gobbledegook-laden URL schemes, too, so it's no wonder people just skim over the location field in their browser or even use the search field exclusively.

The movement toward clean URLs and some changes to the way browsers display them (like Google Chrome has made) may alter this perception, but I doubt it.
Abstraction Distraction It takes a while to really understand any abstraction in our modern society: how currency works, why we try to separate church and state, how a bicameral legislature acts on the will of the people, how interest compounds.

I know it always takes me a long time to grasp any new programming paradigm. I still don't quite get the idea of monads in Haskell. You see, unlike laymen, programmers are regularly challenged with new ways of abstracting information (be it entire programming paradigms, new frameworks, or just a new way of factoring their own code) and eventually become adept at this meta-skill.

But regular people don't have to learn new abstractions on a day-to-day basis. Anything presented to them in a new symbolic or abstracted way is bound to cause confusion for a long time. Especially second or third order abstractions.

Here's an example.

In the early 2000s, "blog" became a new buzzword. To understand what a blog was, you sort of had to understand what a website was (which many still did not), and how that's different than a book or a newspaper.

At some point, blogs added comment forms. But to understand comments, it helped to understand the concept of a blog to begin with, and how those comments are different than the actual posts (and certainly different than a contact form or a guestbook that covers the entire site).

Eventually sites like Facebook and Twitter came along with status updates. But to understand a status update, you kinda had to understand a blog and why the heck a normal person would write one. And to understand the comments feature that Facebook added to status updates, it's easier if you think about them like comments on blog posts (and how that's different than the person's wall). And the whole thing makes a lot more sense if you get general idea of a social network — which was something popularized well before Facebook came around — and how that impacts who can see what on the site.

When Facebook launched its "news feed" feature in 2006, it caused massive uproar. Not many really understood how it was supposed to work and threatened to quit the site out of concern for their privacy. But now most social websites have worked in news feeds. Some sites have played with the idea, aggregating multiple feeds from different sites or letting users filter them and change how things get posted on their own feeds. But to understand any given site's news feed, it'd help to be filled in on how news feeds function in general.

I could go on. This all sounds like really elementary stuff, but the point is that regular people don't follow the evolution of these ideas and see how they progress. They approach an abstraction without seeing what it's analogous to (or what that's analogous to). They form their understanding from scratch.

Sure, users don't have to have a complete history of any feature they're using (in the same way I don't need to understand why my TV doesn't have a Channel 1 to watch it). But any time a designer frames an idea as an analog to another ("It's like X, but for Y!"), they must be prepared to deal with people who need an introduction to the whole shebang.

To Facebook's credit, it has one of the best design teams in the industry and handles these challenges well (though apparently not well enough for some parents).

So there are a lot of ways regular people are different from programmers, and in ways much more nuanced than simple "computer literacy." How do we solve the problems caused by these differences?

PARTIAL SOLUTION 1: A COMPUTER COURSE FOR THE 21st CENTURY

Should computers be dumbed down? Yeah, probably a bit. Programmers can't build software expecting that their users will dork out and explore its every facet. That's got to stop. The iPad is a bold step in the right direction that might even be going too far.

Another solution lies in education. I think programmers and "digerati" may have caught onto some trends that will become important for everyone in the 21st century. The solution lies in effectively teaching to these.

Really, every subject's curriculum should be overhauled to account for the vast economic and social changes that have occurred in the information age (Paul Graham's The High Res Society is a wonderful summary).

But I'll focus my comments instead towards courses that purport to cover computers.

I think I can speak for most of my generation in saying that computer classes in high schools, colleges, and community centers are universally worthless. Courses for young people are usually taught by out-of-touch adults with a much less advanced understanding of the things they're teaching than their students. The only kind of teacher likely to be more incompetent than a computer teacher is a gym teacher. But that's not the problem.

The real problem is that these courses often teach a specific operating system or a specific office suite in an extremely facile manner. They're glorified typing courses. That means when Microsoft changes the locations of buttons in Word, students' knowledge is obsolete. Even programming courses in high school (and many colleges) are tied to specific programming languages, not general concepts. A good course teaches a mix of theory and application, but most computer courses can't even handle application right.

To create a computer course for laymen that does not do them a disservice, it should be rooted in things that we can reasonably anticipate will not change. I'm not quite sure what those are but the stumbling blocks outlined in the previous section are a good place to start. It should combine practical computer skills and general information literacy. It should be required and it should be rigorous, not a blowoff course.

Imagine how many fewer bank accounts or email accounts would be hacked if a section on the final exam gave students URLs and asked them to identify the domain name, the subdomains, the path, the port, and the protocol. This sounds like esoteric technobabble at first. But if high school students are expected to know how many valence electrons molybdenum has or how to define trigonometric functions in terms of each other, it's highly practical by comparison.

Teaching students how a hierarchical file system works would make sense. It could even briefly cover the directory structures on each popular OS at the time and where things go. I have my doubts on how long the idea will last, but I'm betting at least another 15 years.

I'm hoping email (as a means of communication) will die a horrible death in the near future. But if it doesn't, its technical intricacies should be taught in some detail.

Such a course might also include a primer in Boolean logic and proofs, plus basic set theory — nothing too comprehensive. This would certainly allow students to refine their Google queries if the first result isn't right (assuming they actually check).

Or there could be projects where students are asked to make simple relational database applications (not requiring any programming — just using a GUI) and to construct queries that join across tables to locate information. Or maybe they could use Yahoo Pipes to create mashups of multiple sites to solve a given problem.

Hell, a basic non-mathematical introduction to public key cryptography wouldn't be a bad addition. I don't think that's going away for a long time, at least not until we get quantum computers.

This course could give students research projects to hone critical thinking. They could be given news stories, Wikipedia articles, and whitepapers by interest groups and be asked to get to the heart of the matter and argue one way or another. Students should be able to recognize logical fallacies and propaganda techniques so often employed by politicians and forum trolls alike.

The crux of all of this is that it's becoming an essential skill for any capable person in our society to be able to sift through vast amounts of information of uncertain quality to find the best. Teach this and the rest effectively teaches itself.

Ultimately, if schools were completely successful in designing a curriculum for the 21st century, they might find themselves the architects of their own obsolescence as institutions.

But until then, teaching kids how to read a URL so they don't get scammed by the Nigerians would be a nice improvement.

PARTIAL SOLUTION 2: UNDERSTANDING THE ECONOMICS OF DEALING WITH N00BS

Many Web 2.0 companies have a pretty effective way of dealing with n00bs needing customer support: they don't. Hey, buddy, it's free — what do you expect? And they have enough of a user base that they can gain an intimate understanding of how to design their products through processes like A/B testing.

But for those of us actually charging users money and hoping to keep them happy, however, we have a more complex set of obligations. Here are a couple realizations I've made on that:

Define Your Audience

When designing a piece of software, it's important to draw the line at the level of expertise required to use your app and communicate this consistently in your marketing, documentation, and correspondence with customers.

For instance, if you've written a desktop app that's document-based, it makes sense to expect users to be familiar with Open/Save dialogs. It's reasonable that they not blame you if they confused their document for one they gave a similar name to in a different folder.

Yet some apps like iTunes and iPhoto are not so arrogant as to presume such knowledge. And others, still, like Final Cut Pro or AutoCAD expect users to come in with quite a lot.

Should all software condescend to the lowest level? Not at all! At least, not necessarily. By presuming knowledge of conventions and abstractions used on your platform and in other apps (or even inventing your own), you enable a lot of power and functionality to users. Ideally, you can find opportunities when designing a piece of software to lower the learning curve without hurting anything. But, like any film or piece of writing, you must have a very clear idea of your audience that informs every choice you make.
Optimize Support Processes

Before the MacHeist promotion I mentioned earlier, I set to optimize users' interaction with my software and website in two ways:
1. Lower the rate at which users write in with questions or complain about the app
2. Increase the rate at which users buy the iPhone app companion to the desktop app they got for free and increase the number of people who subscribe to our newsletter.
These goals were accomplished by fixing lots of usability flaws in the app and edge case bugs (particularly syncing bugs). I redesigned the support section of our website (this still needs lots of work), and found ways to encourage more users to use resources like FAQs and manuals. I even put a special box at the top of the site if the user came from MacHeist.com.

In addition, to handle the demand, I also switched to a ticket system for support emails and hired part-time help.

Part of the reason why I hired someone to do support (even during periods when I could handle the burden) is that I tend to take email from users personally. I try to fix that person's problem and then find a way to make sure nobody else ever has it. I even made a house call to a guy who happened to live in San Francisco and had a problem with the app. But it took a sudden influx of 300,000 new users for me to realize that you can't please everyone.

The first line of defense is to make sure that people who do have a clue can get help with their problems easily without having to contact you. If all your emails are from wackos with hopelessly dumb questions, that means you're doing things right! Then you'll have to find cost-effective and time-effective ways of dealing with the wackos, so that you can respond to the remaining emails with the full consideration and respect they deserve.

WHITHER THE N00BS?

Will n00bs forever be the nuisance they are today? Will there always be some sort of paradigm shift in the world that the Luddites born before will fail to understand in a fundamental way? Or was the computer revolution a once-in-a-century sort of thing, not the first step in the dizzying race towards the singularity? Time will tell.

But for right now, the fundamental problem is that the people who seek to popularize and commercialize a given technology (say, personal computers or digital cameras or social networks) only have it in their self interest to make the public understand the technology enough to buy the darn thing. Not to use it well. Kathy Sierra has talked about this. But ultimately, it's in society's collective interest for people to be computer-literate and information-literate. And since the big players aren't doing it, the problem is foisted on the rest of us. Hopefully the partial solutions outlined above begin to address the problem.

SELF PROMOTION

If you weren't offended by the Stevie Wonder joke and actually liked this post, you should follow me on Twitter!

"How Different Groups Spend Their Day"

2009-08-03T14:12:37-07:00

This is pretty neat! I'm a sucker for good graphs.

I'm Speaking at the "Voices that Matter" iPhone Development Conference

2009-07-31T12:23:45-07:00

In October, I'll be in Boston at the "Voices that Matter" iPhone development conference put on by Pearson Education.

I'll be giving two talks:

One on Saturday about data synchronization — all I've learned from doing the Mac and iPhone versions of ShoveBox and all the funny edge cases that come up when dealing with syncing.

The other one is about getting more done in Xcode.

If you register before September 12, you can save $200 on admission. Hope to see you there!

Bonjour API Tweak in Mac OS X 10.5.7

2009-06-03T14:43:35-07:00

Recently, Apple released an update to Mac OS X that broke syncing in my app ShoveBox. The app syncs over the user's local WiFi network and uses Bonjour in order to find the iPhone/iPod version of ShoveBox.

A few fellow developers were curious what was changed in 10.5.7, so I thought I'd post it here.

After some troubleshooting, I found that Apple had made a slight tweak to the way the Cocoa APIs for Bonjour work. As of 10.5.6 and below, if you called -startMonitoring on an NSNetService, it would immediately call the delegate method – netService:didUpdateTXTRecordData: to let you know that the TXT record changed or became available, even if it had not changed from when -startMonitoring was called.

I'm not sure what I was under the influence of when writing this part of ShoveBox, but I had (perhaps defensively) coded it to only count the NSNetService as a viable sync source once –netService:didUpdateTXTRecordData: was called — as it did seem to be called in every case.

As of 10.5.7, NSNetService (correctly) only calls that method once the TXT record has actually changed.

The fix involved checking for/validating the TXT record for the NSNetService as soon as its address is resolved (in – netServiceDidResolveAddress:). Now, only if it is not present and valid, it I call -startMonitoring. It should be sufficient, however, to only validate once if the broadcasting application does not change its TXT records.

For some reason, I had never heard this song before.

2009-04-23T15:12:55-07:00

For some reason, I had never heard this song before.

New York Times: Staff Jobs On Campus Outpace Enrollment

2009-04-23T15:05:59-07:00

“A lot of it is definitely trying to keep up with the Joneses,” said Daniel Bennett, a labor economist and the author of the center’s report. “Universities and colleges are catering more to students, trying to make college a lifestyle, not just people getting an education.”

This is a huge problem with college today. Many students have a hard enough time paying for an education, much less the expensive "lifestyle" colleges are selling.