I’ve been awfully restless lately. Fair warning: this is a highly self-indulgent post. Read on at your own risk.

I belong to a tortured class of individuals you might refer to as “creatives;” I have a passion for making, for producing. It’s really more of a sickness than anything else, especially combined with my self-imposed and -appointed “profession” of software engineer, another incurable disease.

There comes a time in every creative person’s life — indeed, it’s a nearly monthly phenomenon; consider it the Menstrual Cycle of the Imagination — where you’re struck with the desire to go make something awesome, or beautiful, or useful, or just neat. This phase is, in almost every case, accompanied by a complete absence of anything that resembles inspiration. It’s a kind of restlessness of the mind, and it’s downright depressing.

Sometimes, it’s not even a lack of inspiration that’s the problem, but rather a lack of resources to execute. When I sit here and look at my {13bold} Dropbox folder, I see no fewer than 29 folders, each representing a project we’ve started, many of them even dating back to before Laurent started working at Apple. So let’s see: {13bold} has released all of two software products (three if you count Bowtie for iPhone); that leaves roughly 26 projects unattempted. Sadly, there just isn’t enough time in a day to work on all of them.

I find ways to keep myself entertained, of course: when I come off from working on a large project (like Bowtie), I’ll usually bang out a couple of mini projects with varying success. Quite frankly, though, it always ends the same: a boredom that only creation can fill. Perhaps it’s time to try something different.

Incidentally, if you didn’t chuckle or at least groan when you read the title of this article, go back and reread it a few times until you get the pun (spoiler alert).

Note: this article is part of a series of articles on my real-world App Store experiment: App Power: An Experiment, App Power: The Submission, and App Power: The Reveal.

I woke up this morning to find an email from appFigures containing yesterday’s sales reports and figures; with that, I officially have data for my first full week of Bowtie for iPhone sales. And the numbers were… somewhat depressing.

Now don’t get me wrong, I’m not really complaining: I don’t do this for the money, I do it because I love it. I had, however, hoped to use my App Store profits to visit a friend in Europe. While I may eventually earn the money to do that from the sales of my app, I’m fairly certain it won’t be in time (she’ll only be there for about three more months). Either way, it looks like there’s a little more to succeeding on the App Store than meets the eye.

The Results

I’ve always believed that graphs are incredibly useful, so allow me to present my week-one sales graph:

As you can see, sales started pretty promisingly: 170 downloads on day one, 202 on day two. Since then, however, sales have been steadily dropping as they look about to converge on 0. Yesterday, a mere 22 people downloaded Bowtie, netting me a grand total of $14.70 (accounting for conversions from other currencies, etc.).

So what does it all amount to? In all of week one, I made (estimate by appFigures) $454.78.

Now don’t get me wrong, that’s not a bad sales figure or anything — if it was an average weekly sales figure. If trends persist, though, it looks like my average weekly sales figure will be somewhere between $40 and $70, if even that, which is hardly the goldmine I was expecting or hoping for.

Surprises

I ran across a few things that were a bit more than surprising during this first week. The first was that Windows users really don’t read. On day one, I received over 1,200 downloads of Bowtie Remote for Windows, an application whose sole purpose is to control Bowtie for iPhone. That would imply that each download of Bowtie Remote should be paired with a purchase of Bowtie for iPhone roughly 1:1. I was ready and willing to discount 30% of those downloads for stupidity, but not greater than 90%. As many of my friends (and Twitter followers) will tell you, I can now use this as anecdotal evidence as to why there are so many viruses on Windows: Windows users will download anything. In fairness, a couple of days after release, I changed the description on the Bowtie website to make it considerably clearer, and downloads have slowed (the rate is still considerably higher than the iPhone app download rate, however).

The other surprise was just how plain unreliable App Store reviews are. I’ve been very closely monitoring every review that gets written about Bowtie on the App Store: they range from one-star reviews calling the app “too limited to even be worth 0.79€” (translated roughly from Italian), and others alleging that my app is responsible for problems it simply couldn’t possibly cause; to glowing five-star reviews heralding Bowtie for iPhone as the Messiah of iPhone Apps. Personally, I don’t think either extreme is merited — I’m most fond of the four-star reviews, personally, many of which contain valuable feature suggestions (though I do appreciate two separate five-star reviews that say “this app solves a problem I didn’t realize I had;” that’s always the goal ). Anyway, I digress: I “grew up” in NewEgg culture, where the reviews aren’t just reliable, but they’re (in my opinion) some of the best indicators on the Internet of the quality of a tech product. On the App Store, that just doesn’t appear to be the case. It’ll certainly make me reconsider opting not to download the occasional two-star app that looks pretty cool.

What’s Next?

For starters, the app is really cool, but also very feature-light (mostly by design). People want more, and the biggest request is some sort of search feature that allows you to switch playlists or albums from your computer, rather than just skip forward/back and play/pause. That one’s already in the pipeline; it’s slated to be released with Bowtie 1.2 (Bowtie 1.1 will be out either today or tomorrow, with fixes for a number of bugs that cropped up during the first week) and Bowtie for iPhone 1.1. You’d be surprised how hard it is to design a good music chooser.

Also, as a part of my experiment, the only “advertising” I did was via my websites, my blog, and my Twitter: any third-party advertising that occurred (like a feature on Mashable) was on a completely voluntary basis, often unrequested by me. The fact is, that doesn’t generate enough exposure: I would have loved to believe that the App Store magically put your app out there for everyone to see, but it doesn’t appear to be the case. As soon as Bowtie/Bowtie for iPhone get a solid search capability, I think I’ll start pursuing some more conventional advertising routes.

Probably the most important thing I learned though (for next time) is that Bowtie for iPhone is still too niche. Certainly, the app isn’t quite as niche as Signals is, nor is it even as niche as TuneConnect is (my older, free app for Mac-to-Mac iTunes remote control, rather than Mac-to-iPhone like Bowtie). Regardless, not everyone wants to be able to control their iPhone’s music player from their computer; the app simply lacks the universal relevance that would cause it to become an “overnight app sensation.” If I ever really want to make money off of an iPhone app, it’ll have to be something considerably more appealing to a broader volume of users (or at least more entertaining).

Anyway, I’ll keep plugging away at making Bowtie for iPhone as awesome as possible, and we’ll see if it changes sales figures in either a positive or negative direction.

What about Bowtie 1.0? How’s that doing?

I’m so glad you asked! Downloads of Bowtie 1.0 have actually been going considerably better than those of Bowtie for iPhone. As of this writing, Bowtie 1.0 has been downloaded 14,158 times (7,497 by automatic update, 6,661 by manual download). That puts me only about 5,000 downloads short of the all-time download numbers for the first beta of Bowtie (though I still have quite aways to go before I catch up with beta 2). Either way, I’m fairly certain that this is the fastest I’ve ever hit this number of downloads with one of my apps, and I’m definitely very pleased with the results.

If you’re interested in keeping track of Bowtie downloads, you can view our public Mint. All of the files that make up the current version of Bowtie’s downloads (currently 1.0, will be 1.1 as soon as it’s released) can be found on the “Watched” tab of the “Files” pane.

Anyway, thanks to everyone who’s either downloaded Bowtie or purchased Bowtie for iPhone, and to everyone who’s been so supportive of this endeavor. I’ll continue to keep you all posted as things change. Until next time, enjoy Bowtie!

Note: this article is part of a series of articles on my real-world App Store experiment. The first one was App Power: An Experiment, and the second was App Power: The Submission.

Just moments ago, I revealed to the world — at long last — just what this super-secret iPhone app I’ve been developing is. What is it, you ask? Why, it’s Bowtie for iPhone, of course! And it’s coming out in tandem with the final release of Bowtie 1.0 this Saturday.

The app, which will be available on the App Store for just 99¢, will allow users of Bowtie to control the iPod application on their iPhone or iPod touch (or iPad ) from their desktop. Windows user? We’re releasing an app called Bowtie Remote for Windows, which gives you the same level of control (without the themes).

Once the app goes live, all that’s left to do is wait. I’ll post a summary article here as soon as I have some sales numbers. ’til next time!

When you initialize an object with its -init method (or equivalent), the default NSObject implementation initializes all instance variables to nil, as a convenience. However, whenever you declare a variable elsewhere, this auto-initialization does not occur: rather, like with any C variable, it creates a block of memory in the stack sizeof(id) big. If you don’t initialize that pointer with a value, it still points to that address in memory.

So let’s take a relatively common example:

// ... exciting code ...
if (myData)
	[myData release];

Assuming myData is an instance variable (and every other time you release it, you set it back to nil), this should go off without a hitch. If the value referred to by myData is a non-false-evaluating quantity, release it.

Note: this is usually superfluous, since it’s safe to call -release on nil — it’s safe to call anything on nil, it always returns nil. As long as you’re good about nil-ing out your instance variables after you release them, you don’t need to check if they’re non-nil first.

Now let’s modify the example slightly:

NSData *myData;
// ... exciting code ...
if (myData)
	[myData release];

In this snippet, myData is a local variable rather than an instance variable, and believe it or not, this will possibly cause an application to intermittently crash. The reason is that we didn’t initialize myData, which was assigned an address anyway. If we were to insert a few lines of code in between the variable declaration and the conditional, that space could become filled with something else. Something else that might not belong to you. And we all know what happens when you try to dispose of memory that doesn’t belong to you.

The correct way to write this would be:

NSData *myData = nil;
// ... exciting code ...
if (myData)
	[myData release];

It’s a simple fix, but it’s one that’s easy to forget, especially since it’s not a problem you need to worry about with auto-initialized instance variables.

This particular one caused me some massive headaches, because it was being called on the return trip of a Distributed Objects call, so the debugger wasn’t able to pinpoint exactly where the problem was occurring. And for that, NSLog (and selective commenting) is now, and always will be, your friend.

This part of my application has three potential entry points, and two “endgame” states; I broke up each “chunk” (many represent subroutines) onto a separate index card, marking entry routines with an arrow in the corner, and endgame routines with “#END#” at the bottom. Now, I just trace execution from each entry point with every possible environment permutation. If they all eventually hit an “#END#” card, my logic is sound. Don’t you love index cards?

Note: this article is part of a series of articles on my real-world App Store experiment. The first one was App Power: An Experiment.

So, a mere 14 days after I proposed my idea to enter into this real-world experiment, the application is done: today, it heads off to the App Store for approval. Barring any delays, it’s slated for release mid-February, with the formal announcement of exactly what it is coming a few days before.

A few days ago, I sent the application out to a select group of private beta testers to get some feedback, and it was generally pretty darn positive. Here are a few soundbites:

“woooooo – you’ve got a winner on your hands, Mr. P – I will gladly buy this app for everyone I know ” Elle Wong

“I am yet to experience any network lag or any bugs within the application. It responds well to furious clicking on the desktop client too. So far so brilliant!” Tim Davies

“Only this app can prevent forest fires!” Smokey Bear

Stay tuned for more updates!

Update (1/28/10): Just received word from Apple that the app has been approved for sale on the App Store! You’ll hear more from us on a release date (already scheduled in the App Store admin section) within the next couple of weeks.

I’m making a simple attempt to fill the gap with what I’m appropriately calling mini-blogging: short blog posts with all of their content “above the fold” (ie, no “(more)” link). This is one right here, and you can certainly expect to see more in the near future.

We’ve all (maybe?) heard the stories: brilliant young developer writes an iPhone app, puts it up for sale on the App Store, becomes a millionaire overnight. Now I’m not delusional enough to think that that could happen to me — I’m not brilliant enough, nor do I have the time/money to dedicate to marketing (the part they probably left out) — but the prospects are certainly alluring, and I think it’s time to put it to the test.

A few weeks ago, I wrapped up development on my first iPhone app, Signals (iTunes link). Signals is a reference for rail signals (the colored lights they have overhead/next to train tracks): you tap the little pictures of colored lights on the screen to change their color, optionally making them blink using the buttons on the bottom row, and Signals tells you what that configuration of lights means (tap on the name or the “Info” button in the bottom right corner to pull up the full description). It’s currently available for free in the App Store, and looks a little something like this.

So why on earth did I write an app for looking up train signals? Well, basically, because I felt like it. Every once in awhile, I get hit with an inexplicable desire to learn more about a certain subject. Last month, I wanted to understand how train signals worked, so I started learning, and eventually decided to compile my research into an app in order to get my feet wet with the whole App Store process.

The app hasn’t had many downloads (from what I can tell: iTunes Connect reports are somewhat difficult to follow), and that makes sense: not only is the app incredibly special-purpose, but until now, I didn’t tell anyone I released it, with the exception of a few close friends. If anyone thought that publishing an app on the App Store was a magic ticket to success, that simply isn’t the case.

However, the fact that the app has been downloaded — about 15 downloads per week since release — does indicate that simply being on the App Store generates a little magic mojo. The App Store is a powerful tool, and it strikes me that it’s much easier to leverage than other sales arenas.

Update 1/13/10: I just checked some App Sales figures using AppViz (decent app, not terrific UI-wise, but does the job), and Signals has been downloaded 207 times as of 11 January. More than I suspected, curiously enough.

Thus, I have compiled a hypothesis. I believe that, if you can create an iPhone app that is…

• Of more-than-marginal utility to a decent sector of the population (ie, not niche)
• Priced right at that “gee, this app is cool, and $___ is nothing! I’ll buy it!” price point
• Sexy

… the App Store puts your software in the hands of enough users that the marketing effort you have to expend is next to nil.

And like any good hypothesis, it needs to be testable.

The Test Case

I have an idea for an app that I think is both relatively cool, and relatively useful to a decent number of people (no, I’m not telling just yet ). I’m going to price it around $1.99, and I’m going to market it exclusively by posting it on my website, my blog, and shooting out a few tweets: the rest is up to the App Store and its user base.

The goal: I have a very dear friend who’s studying abroad in Granada this semester, and I’d love to go visit her. Unfortunately, not only is international airfare outrageously expensive (seriously, BOS-LAX is about 2,700 miles, BOS-LHR is about 3,200 miles, it shouldn’t cost $400 more), but I also have a handful of extra courses to pay for which have left me in a bit of debt. Therefore, my goal is to earn $10,000 over the course of the next two months exclusively through the sale of my application on the App Store.

If you do the math, not accounting for taxes but accounting for Apple’s 30% cut, that works out to close to 7,200 sales, which, judging by many of the reports I’ve seen from other developers, isn’t entirely too ambitious a target. Can I make the magic of the App Store work for me? I guess we’ll just have to wait and find out.

Look for my new app (name and purpose to be announced in the near future) coming within the next few weeks.

Oh, and as a side note to fans of software under the {13bold} label, namely Bluebird and Bowtie, never fear! These two little apps have seen more development action in the past two weeks alone than they have in almost all of last year, and new versions of the apps will be available in a very big way within the next couple of months.

“Now what will become of 2009. I fear that 2008 went far too well for 2009 to possibly be good.” (Odes to ‘08 and ‘09)

It really is unfortunate when my predictions turn out to be accurate.

I’m going to be honest with you: contrary to popular belief, 2009 was not the best year of my life. In fact, if I can be quite frank — which I can, it’s my blog — it was downright crappy. Proverbially speaking, the shit hit the fan, and it was pretty messy; you really can’t expect much out of a year that begins with a massive month-long fight with your best friend. And then business kinda’ failed. And then grades kinda’ tanked, too, and I’m currently taking a (forced) year off from Brown. To summarize, 2009 was the ultimate personification of Murphy’s Law.

But you know, it wasn’t all bad: I’ve got a great new job at Jerk.com, I’ve got a great set of friends that (for the most part) don’t make me want to tear my hair out, I’m dating someone again for the first time in quite awhile, and if nothing else, my life is a whole lot better than a lot of people’s. When in doubt, just remember: it could be worse.

My friend Becca ruminated on the year from a perspective slightly different from mine:

“Today I realized that I am at a point in my life where I hardly identify with the group of fine young people I was so close to in high school. Rather, I have new friends with different goals and interests, and we ‘click,’ just as my old friends and I ‘clicked’ five and a half years ago as freshmen in high school. We have all changed. We have grown apart. I don’t know when this all began… It’s alright, though, that we have grown apart. The key is that we have grown. Growth is important in all aspects of life. As time passes and we grow older, we also grow in maturity and stature (well, for most, anyway), and into the people we were meant to be. We meet all types of people in college, stick to those who either challenge us in a way we cannot ignore or are so like us that we are convinced we were meant to be… Some of these people will remain with us beyond college. Others will not. All, however, will somehow impact us.” (Growth, Becca Hammond)

And ya’ know, she’s right. I realized a long time ago that my own life isn’t about me. Rather, it’s about the connections that we make with other people, and the meaning that’s created when people collide. We are a social breed of animal, and none of us would survive long if we had to live without the other people in this world that make life just so damn amazing.

So to 2009, I therefore confidently say “You sucked, but I love you anyway. ” The time has come to look forward to 2010.

I’m more than slightly optimistic about the year 2010 (though that’s nothing new). If all goes according to plan, I should be back at Brown at the start of the fall semester, which in and of itself is enough for celebration in my book. Prospects with my work — both my official and unofficial jobs — are very promising, and I can’t wait to spend even more time creating, writing, and giving back to the Mac community in ways that only we developers can. For the most part, my social life is smooth, and I don’t anticipate any major hiccups in the near future. Every single person in my life is incredible, and it would be insulting and downright stupid for me to think that the products of their labors this year will be anything but astounding. I don’t expect 2010 to be a particularly significant year in any greater sense, but that doesn’t matter, it doesn’t take much for something to be meaningful.

Plus, anything’s better than 2009.

Happy New Year, friends.

This post is part of a series of ongoing posts I write every year on New Year’s Eve. You can check out the previous posts on my old blog.

• Odes to ‘07 and ‘08
• Odes to ‘08 and ‘09