Meet the Developers: Greg Scown and Jean MacDonald of Smile Software

Macworld has always been a great place to network with your favorite app developers. There were even times when apple attended the conference and announced its own products. Now that those days have come and gone, it’s important to focus on the smaller companies, like Smile Software.

While at Macworld 2013, I spoke with Greg Scown and Jean MacDonald about the company’s history and latest developments. It’s the perfect follow-up to our previous interview with the company.

TextExpander is one of Smile’s most popular products.

How has Macworld been for your company this year?

It’s been good. It’s busy at our booth and that’s what we really care about. We’re happy with the event this year from that point of view. I haven’t had much time to go around and look at the floor, but they’ve changed things up a lot since last year. The conference has definitely shrank a lot since Apple was here, but it’s still going strong for us.

The Mac App Store has become the standard for software distribution on the Mac. Why does Smile not use it to sell TextExpander 4?

We like the Mac App Store and we participate in it, but we also have to live within the rules that Apple has for it. As you know, Apple introduced the requirement of “sandboxing” one’s app in order to have it approved in the Mac App Store. We spoke with the folks who are involved in that and worked with them, but it was determined at the time that TextExpander was a key-logger. (It reads every keystroke, but does not retain them all. In fact, it only retains the last 15–30 keystrokes, depending on the circumstances.) Since it is easy to create a malicious type of keylogger, Apple has decided that they will allow no such apps into the store. It’s good because they’re protecting users.

I respect Apple’s opinion about what they choose to allow and not allow into the store. At the same time, I’m grateful that they introduced the Apple ID system for developers. With that, we can sign our apps with an identity that’s provided by Apple and is trusted by Gatekeeper. So, when you install our non-Mac App Store TextExpander app, the system knows that it comes from a trusted vendor and allows you to install it without complaint. I think they made a good accommodation there.

PDFpen and DiscLabel are on the Mac App Store and we expect them to remain there. PDFpen is sandboxed as of version 5.9. It took some work, but it’s done now, and future updates will be easier.

How has not selling TextExpander 4 on the Mac App Store impacted business?

Interestingly, we have still been allowed to sell TextExpander 3 on the Mac App Store. We even upgrade those users to version 4 once they’re ready or when they so desire it.

So why are you keeping TextExpander 3 in the Mac App Store for now?

I don’t know why TextExpander 3.4.2 is still on the Mac App Store. Apple doesn’t pull stuff and it’s still grandfathered, until the day they decide to remove it. There will come a time when they say they will only carry sandboxed apps in the Mac App Store and when that happens, it will inevitably be removed.

And did you submit version 4?

No. We had already worked with people at Apple and we knew that it would be rejected out of hand. There’s no point in antagonizing Apple and we felt that’s what such a move would do.

If a new developer asked you which would be the best way to distribute his creations, what would you recommend and why?

I would say to distribute through as many venues as one can manage. You certainly are wise to have a distribution mechanism that is under your own control because at least you can satisfy your customers in the event of some unforeseen circumstance. But I would certainly also encourage them to put their app on the App Store. It’s convenient for end users and the developer can be fortunate enough to get featured.

What are the good and bad things about each method of distribution?

Well, the App Store is convenient, it’s right in the dock, it’s trusted by users, and updating is integrated well into it. However, we use Sparkle to provide users with an automatic updating mechanism, which makes updates not an issue with our apps. But the one-spot “update all my apps” feature is nice. It’s simple and users like it.

One of the challenges of the App Store are that reviews take time. You can be well tested and ready, but end up waiting half a month for your app to be published. Another issue is that the cut is high: Apple takes 30%, which is something you wouldn’t have to pay if you published the app yourself. Of course, the argument for the App Store is that an affiliate fee can be 20–30%, so in a way you’re paying Apple for both hosting the app and promoting it.

The upsides of direct distribution are that we’re in control of it, so we know we can distribute a release to our customers right when it’s ready. If we’re adding something of value, at least they don’t have to wait for it. Overall, we control the experience. While the App Store has a very nice experience, there are some things that we can do which you cannot do on Apple’s venue. We can provide users with information about updates in a way that we prefer and we can emphasize something. You can’t even bold things on the App Store descriptions, which makes sense because developers could easily abuse it, but it is a frustrating limitation.

As developers, we’re quite confident in our ability to provide information for our users in a well-designed form. It’s a trade-off. We can balance things better on our website than we can with the App Store. Everybody is different though.

While Smile is centered on Mac apps, it does have some interest in iOS. If Apple agreed to add any feature you requested to iOS, what would you want to see?

Greg Scown: I would love to see core event filtering on iOS. That would allow us to develop TextExpander for the platform. Our customers would really appreciate this kind of an experience on an iOS device because typing is much more awkward than on a Mac. The problem with the built-in keyboard shortcut feature right now is that the interface to manage the shortcuts is very flat. You can really only have a few of them, they can’t be multi-lined, and they can’t have intelligence, so anything like TextExpander is impossible.

Jean MacDonald: I want developers to be able to respond to reviews on the App Store. Right now, a good half of the reviews on almost all apps are complaining about something that’s wrong with the app. The problem is, the users don’t know that the feature is there sometimes. We, as developers, are not given the ability to moderate any comments. The only thing we can do is report it, but even then it must be libelist or very harsh.

It would help people if we could speak with them. It would be good for us as developers to talk to the users, but mainly it’s about helping them out with their problems. Even though they can get in contact with us using our website, it’s more likely that they will leave us a lower-star review complaining about something.

Somebody made a suggestion that when a user submits feedback for an app, there would be a checkbox that says “Allow the developer to respond.” That would be perfect. Then, of course, the user could choose not to let the developer respond, and that’s perfect for those who just like to rant.

How do you get inspiration for new features on your apps?

It depends. It’s a little bit of what’s possible, what people want, and how much pain can we take in terms of trying to get something done. With updates, we take a look at a chunk of the interface on a product we have and say “Gosh, do you think we could orient this in a way that’s better for users?” We don’t have pitchforks and torches saying “This is so hard and we need to change this.” But we know that if we make this better it will make the experience better for our users, and for us — we use the product all the time. That particular one came out of “we can do better”.

Another one is a realization. Sometimes we’ll say “Wow, we’ve been asked about this for so long and we’ve finally found a workable way to do it.” Then there’s the “People have been asking us this for so long and there’s no excuse for us not to do it, so we’ll take the time now to get it done.” We have a prioritized list of what our customers ask us for, alongside a list of where we’d like to see the product go. You can’t do all one and all the other; you have to blend the two in order to be successful.

As an example, we do get a fair number of requests for Japanese optical character recognition in PDFpen. It’s not surprising, but it does have a couple of challenges. For one, the licensing on the character recognition is more expensive than Roman language. We might have to make a Japanese version, which makes things more complicated. The other thing is that Japanese text can be columnar and single-letter columnar text is not something that’s PDFpen’s forte. We could spend some time and become a lot better at that, but it’s a very limited item to spend that amount of time on. That’s a case where the user requests are high and the chance of fulfilling it are low.

While we’re on inspiration and stuff, what caused you to start Smile?

Historically, Philip and I (Greg) were independently working on Mac software. I shipped a piece of software called PageSender which is fax software and I was exhibiting that at Macworld in 2004 when I met Philip. The reason I built PageSender was that I was living abroad and didn’t really like the way eFax had worked — it just didn’t fit my needs. Instead of using something I didn’t like, I built my own solution to the problem and after a while, I thought I must not be the only person who needs this.

So I released PageSender to the public and immediately people said “that’s great, now make it drive my modem and get fax from my computer.” It wasn’t really my plan to get in that area. I thought it was more printer dialog, but that’s not where it went. As soon as I got fax modem sending, they wanted fax modem receiving. By the time I met Philip, the list of requests was pretty long and it was clear that I would be working on PageSender for the next year or two just to get through another major version with at least a portion of what customers wanted. Unfortunately, I also had another idea and I really wanted to work on that.

Then Philip came and met me at Macworld and we talked a bit about partnering. Philip was excited about my idea, which ultimately became DiscLabel. Over a period of three months of insanity, we wrote it, shipped it, and released it at Macworld New York. Our drive for that was to build a very specific product. I wasn’t a huge disc maker or anything, but I watched people using Microsoft Word to create their disc labels and it aggravated me, so I made something better.

It’s a good thing that I met Philip when I did because there’s no way I could have released DiscLabel when I did, but I didn’t know that at the time. Also, meeting him taught me that there was no way I could build everything on my own. I needed expertise in graphics, which is what he had. The end combination was very good.

In 2006, we acquired Peter Maurer’s tool TextPander, which we renamed TextExpander and made one of our main products. Maurer created the little utility with a lot of potential, but he was really busy so we thought we’d ask him about acquiring it. It’s significantly evolved since 2006 and Peter is still a good friend of ours. That’s the basic history of our company.