Exclusive interview with Kyle Van Essen, co-developer of Notify

Kyle Van Essen of Vibealicious recently made quite a name for himself and his application, Notify, in the Mac and Objective-C development space. With thousands of users behind Notifiy 1.x, a GMail notifier, Kyle has recently been undergoing a complete rewrite of the application. In doing so, he has been very active in voicing his concerns about the current state of paid versus free software and how much time it takes to create a beautiful piece of software.

We decided to talk with him to get his thoughts on the subject (our questions are in bold, Kyle’s answers follow).

What made you decide to write Notify? Were you just annoyed with other Gmail applications on the Mac? Did you find other options such as Mailplane too expensive or complicated?

This is basically the boilerplate answer I give everywhere else, but it’s the truth: The default Google Notifier sucks, so it seemed prime for replacement. I’ve always been surprised Google didn’t make something better, as their Notifier offerings on both Windows and OS X are sort of “meh.” They’re good enough, but Google isn’t usually a “good enough” sort of company. I hadn’t heard of Mailplane at the time.

How did you initially get your start in programming for the Mac? Was Notify your first large project?

I got started by reading this book, on the recommendation of Joel Levin. After that, basically just screwing around with silly little projects.

Notify was my first “main” project, but it’s been through three iterations. The first one ended up being a strange mess, so it got thrown out. The UI was never planned out or anything, it was sort of a weird frankenstein version of a menu-based app, and a free floating window.

The second version was Notify 1.0, which had a better codebase, although I still opted to rewrite it again for Notify 2.0. Recently, Joel has pointed out to the codebase still sucks, so it’ll probably get rewritten or refactored in the future.

I worked on / am working on a few other things, but they’re nowhere near completion, nor can I talk about them.

You’re pretty active in blogging on the Vibealicious website. Lately, you’ve been undergoing a complete rewrite of Notify for the 2.0 version, and you’ve voiced your thoughts about people complaining about paying for updates such as with Tweetie 2.0, which happened to be a full rewrite as well. Do you think the App Store has caused this feeling that every app should be free or $0.99, no matter how good it is or what platform it is on?

Yeah, the app store certainly has caused an extreme drop in perceived value of a developer’s work. When the App Store opened, most didn’t have a problem paying $10 for a game pushed out by a big name like EA—now it seems like the average person gets angry when they need to pay more than $2-5. I was extremely amused by people complaining about the $3 upgrade price for Tweetie 2. You use this app every day of your life. Is $3 too much to ask for some new, kick ass features?

I think part of it also comes from the idea that most consumers simply don’t know just how much work goes into building a quality application. Sure, you can push out pieces of crap, but if you really want to give it polish, that takes time, and time is money…developing a good application doesn’t take days, or weeks—it usually takes months.

Notify 1.0 was free and uses Gmail’s RSS feed. Notify 2.0 support IMAP and will have a free version and a pro version for $10 with the ability to reply to messages in the app itself, among other things. Do you think people will be receptive to (potentially) paying for the software although it was initially free?

We’ve banked almost 4 months of work on this idea. People usually pay for good software, so if it’s good enough, people should pay for it…in theory. No idea what will happen in practice.

We’re not removing any features from the free version, so people should just be able to go along with using the software with little change if they don’t want to pay for the pro version.

I’d love to not charge for software, but it just takes an insane amount of time to maintain current versions, build new versions, and manage support emails—far more time than I’d ever of assumed possible. Live and learn, I guess.

Do you have any thoughts about where the paid software development industry is going? It appears as if people simply don’t feel as if they should have to pay for software if it’s digital, much like what’s happening to the music industry.

I imagine we’re going to see a shift towards software as a service in the next few years, where you pay a yearly or monthly fee to use the product, with the user’s data being stored on the server (under an account name), thus making it harder for piracy to occur. This will also help with portability, as the line between desktops/laptops and phones start to blur.

It’s hard to say how much piracy actually hurts developers—I imagine not that much, since most pirates wouldn’t of bought the app anyways—for this exact reason we didn’t really put much work into protecting Notify. Where piracy does hurt the developer is when those pirates are relying on a server backend maintained by the developer, such as in a game (where logging in does not require proof of purchase)—then it’s just a straight loss.

Your application is interesting in that the Gmail service itself is free, and you will soon have a purchasable product based on it. Furthermore, Google even has their own Notifier that yours has proven to potentially be better than. What do you think Google is thinking?

No idea. Hopefully it’s making them want to build a better product to offer their customers. I can see why they don’t want to offer a more immersive desktop experience: Because their main revenue stream is based on ad revenue, which you don’t see when you’re using a desktop notifier.

In your blog posts you’ve also mentioned that you were so upset with these complaints because software takes time and money to develop. While this itself is obvious, what do you find the most time consuming part of creating software to be? Resolving support emails, adding new features, etc?

I’d say adding new features, obviously. We split that further between concepts, discussion, and then coding, as I imagine all software developers do. The other business things don’t take nearly as long—but it can still be a sizeable chunk of time.

If you’re just getting into development, I cannot stress enough the importance of sketching your ideas out on paper, no matter how silly they seem—it’s a great way to combine thoughts and make something better.

Knowing how much time and effort it takes, how much would you pay for an app like Notify had you not been the creator of it?

Sort of a loaded question, but I’ll try to answer it anyways. I usually don’t have a problem paying $20-30 for any piece of software, so I would probably buy Notify for $20 without issue. $10 seemed like a fair price point, not too high, and not too low, either. Part of the price decision was based on this article by Jeff Atwood.

What made you ultimately decide to rewrite Notify knowing how much time it would take? Was it simply easier for supporting IMAP?

Well, Notify 1.0 was written in about 2 weeks, so even if we didn’t do a complete rewrite, it wouldn’t of saved that much time. The 2.0 codebase is about 4 or 5 times the size of the 1.0 codebase, so it just made sense to start fresh from that sense too.

As you said, yes, it also worked out to be much easier to simply rewrite due to the switch to IMAP. Syncing IMAP messages with the server is much harder than just an RSS feed, due to the fact that some IMAP servers have strange implementations of certain things—I won’t bore you with that, though.

Finally, 2.0 is basically the poster child for feature creep. For those that don’t know what feature creep is, it’s basically what happens when you keep adding features because they sound cool—this is why we’re behind on the initial release estimate.

Notify for Mac has seen tons of success. Are there any plans to expand Notify to the iPhone or any other platforms? Do you have any other apps in mind?

Could be. I don’t want to say too much, but we’re certainly looking at expanding Notify to things beyond email. And we do indeed have other apps at various levels of completion. You’ll just need to wait and see