Before you buy: how Apple software is born

MF: What input, from whom, and at what stage, do you take on board?

RP: Till is a very creative mind. He's definitely the most important source for input.

TS: And of course our developers. We're working very closely with the Technical University here in Munich to find motivated new developers. Our entire development team is located in Munich, Germany.

RP: But to answer the question, we collect the feedback of our users from previous versions – the management team at equinux also has an ear to the market. We try to find interesting discussions in the community, try to find interesting tools and so on. The longer a project runs the more we depend on feedback from our own staff. And we even depend on feedback of first users especially regarding the user interface of an application. Real life tests are very important to new products – our iPhone apps all went through several such tests.

MF: What tools and techniques do you use, and do they have their limitations?

RP: We're using paper prototypes, Photoshop is also a common tool, Interface Builder, and code. Of course, they are all limited. The perfect tool for user interfaces and usability engineering still has to be invented, but we would certainly be the first one to buy it. Tools which push around existing user interface elements don't help, because we want to make something new, therefore we have to draw it ourselves.

MF: How much time is spent on the initial planning stages, the main development process, and the finishing and polishing stage? Do you wish you spent more time on one particular part, and what part do you enjoy most?

RP: It's 33% for planning, 33% for main development and 33% for finishing and polishing. We really love all stages and try to always spend the same effort on each.

TS: Usually polishing gets a bit less because of the time-to-market. My passion is the user interface and I'm well known for making clean sweeps at the very last stage of a project. But I think this is what makes the difference: at the end you understand the processes, the flow and the solution to the original problem.

MF: How long overall can you expect a project to take, and how quickly can you react to change in the market caused by, say, Apple introducing a new feature or hardware revision?

RP: The development of CoverScout 3 took about six months all in all. Reacting to changes in the market really varies; it can take from two weeks up to approximately six months.

TS: A completely new approach developing applications for us was, for example, the iPhone app market. Due to the specifics of that market it's hard to tell if the investment really pays off.

ALTERNATIVE APPROACH: A concept design for CoverScout, putting controls on top of the cover artwork

MF: If you can, please share some ideas of the cost of developing a product. It could be overall, big-picture costs, or some specific examples, such as the cost of creating a high-quality icon or physical prototype.

TS: Creating a new application or even a major release is always a big investment. equinux has about 40 employees, with offices in Munich and San Francisco, there are product costs like boxes, you have to spend money for marketing the product and much more. The investment of six months of development needs to pay off for us to be successful with the product. The big advantage for us is to have a broad product lineup in several markets, which helps to finance the investment.

MF: How do you test your products? What typical – or specific – outcomes does this process generate?

RP: We use a combination of planned unit testing, automated testing and ad hoc extreme testing to produce reports about regular bugs and problems with the flow of the software. We have a lot of different newer and older Mac models in our testing lab to simulate the setup of different users. But you can't simulate each setup. Users always have the strangest combinations of hardware and software, but we try as much as possible to find any glitches in common setups.

MF: How often do ideas run almost to completion, just to be axed at the last minute? What decisions factor in this?

RP: Actually, this would be the worst case scenario. But on the other hand this doesn't happen that often. We might strip unimportant things late because we maybe notice that it is not necessary or confuses the user, but the scrum approach helps us not to go too far with risky features.

TS: More common is to release a 1.0 application with a feature set that is basic but to have a feature set in mind that might be available several updates later. The vision of a product also means to know the direction of an application.

MF: Working in an market dominated by a single company, to what extent do you feel inspired or hidebound by Apple's design aesthetic?

RP: The relationship is two-fold. Needless to say, we are interested in new things from Apple, look at them closely and decide if we want to incorporate them. In terms of the user interface, we don't always agree with the things Apple does.

TS: Apple's software approach always has the user in mind and sets standards. But we believe in our own approach for how user interfaces should look. Of course, sometimes you'll find parts of our software in features of Apple as well, look at the cork pinboard of iSale we built in years ago and the new Faces feature in iPhoto.

MF: Talk us through the process of creating retail packaging, brands and online presence.

TS: If you ever stand in front of a software shelf, let's say in an Apple Store, you get a picture why the front of a retail box needs to attract the attention of a customer. The box should tell the story about the product in just a few seconds. We had to learn to pay the attention to the customer and not only the product itself. On the other hand, the box should tell the idea and the vision of the product. Our first boxes had a huge application icon at the front, today we try to attract customers by telling them what the product actually does in combination with conveying a lifestyle. It's all about the message of a product.

CR: Presenting a product online is different from the standing in front of the shelf approach. When a user visits a website he actively types in a URL, searches for the solution to a specific problem via Google, or reads something about the product on the web. The website is the best place to present videos, tutorials, screenshots, pictures and more details.

TS: When we created CoverScout 3 we wanted to play with the Cowboy/Western look of the previous version and wanted to have a really cool application with a black background color. CoverScout should really be a 'rockstar' application – easy to use, and fun when in action. We jumped into the same picture and the same story when we created its companion application SongGenie. We wanted to have a girlfriend for the cowboy, fitting into his world. This is how one brand can be extended into a second product.

MF: Tell us about how you support a product after it's on sale, and what factors in your decisions about when to revise or discontinue a product line.

MS: In a perfect world, support isn't necessary. But we don't live in a perfect world. The big challenge is to react very quickly to requests on the one hand and to reply very individualy on the other. We always try to improve things: for example, we actively support customers via Twitter, have an iChat support where users can ask questions live and much more. But I think the easiest way is to provide users with problems with a FAQ system that really helps to clarify the most frequent questions.

TS: Our Customer Relations team is really working hard to give customers the best possible user experience. Direct contact to the customer is most important, especially when you're looking for new and useful features for minor or major updates for an application. We really try hard to support our community and to improve every application.