An Interview with Daniel Jalkut
Daniel Jalkut works at Red Sweater Software, which he founded in 2002. Prior to this, he worked at Apple from 1996 until 2002, before venturing off as an independent developer. He is best known as the developer of premier Mac blogging application MarsEdit, which he acquired earlier this year from NewsGator Technologies and Brent Simmons. I interviewed Daniel over e-mail to talk about working at Apple, going independent, MarsEdit, and iPhone development.
Austin Heller: Let's just start with the basics. For those here who don't know much about your background, you worked at Apple from 1996 to 2002. How did you get started writing Macintosh software?
Daniel Jalkut: Before I got my first job as a tester in Apple's Quality Assurance, I had been an Amiga and Unix user for most of my (young) life. I had a friend who worked at Apple and when I saw his PowerBook Duo, I was hooked. I knew I wanted one, so I scraped together the money and got one. Using this teeny grayscale Mac I slowly learned my way through the Carbon developer APIs, and wrote a Mac version of the UNIX game "robots." It was a pretty sad little game, but it gave me my very first taste of programming on the Mac.
Austin: You were a System Software Engineer while working at Apple. Specifically, what kinds of things did you develop for Mac OS X?
Daniel: It's always kind of hard to explain, because a lot of the work I did was in the deeper areas of Mac OS. The stuff that supports applications running on the system, such as memory management, process management, etc. On Mac OS X I worked a great deal on the support in the system that let older Carbon applications run on Mac OS X.
Finally, before I left Apple in 2002 I spent some time working on CFNetwork, which is a framework for low-level access to networking protocols such as FTP, HTTP, etc.
Austin: What inspired you to become independent? Would you ever go back to working for a big company, if the offer presented itself?
Daniel: Honestly when I left Apple in 2002 to go back to school (for a Music degree), I thought I'd probably end up coming back when I was done. I joked with the people who work in the Apple gym. You have to sign up for a waiting list to get a locker, and it can take quite a long time. So when I gave up my locker I signed up again on the waiting list, figuring I'd be back in a couple years - just in time!
While I was going to school I picked up some consulting work here and there and sort of fell into the indie developer world. I slowly got to know more and more developers on the "outside," whereas when I worked at Apple I only knew people on the inside. By the time I graduated in 2005, my now wife and I had decided we wanted to spend some time on the east coast, so we moved to Boston, and that pretty much sealed the deal that I wouldn't be going back to Apple anytime soon.
At this point the momentum is so great with Red Sweater that I don't think I'm likely to go back to working at another company soon. I'm really excited to see what happens as Red Sweater's brand recognition increases and as I further develop the product lineup.
Austin: Red Sweater is, to my knowledge, a one-man operation. Do you have to work on the programming, design, and technical support of your software by yourself, then?
Daniel: You're right in identifying the solo-nature of my business. For the most part I do everything myself, but I do outsource graphic icon design, and would like to outsource more tasks as long as I can find the right people to help out. It's a challenge for a small business to organize things in a way that lends itself to outsourcing. For instance, it would be great to get some help in the customer support area, but right now all the emails come straight to me! I hope to build more structure that will allow me to, for instance, hand off customer support responsibilities for a week or two while I'm on vacation.
Austin: How long have you been a Mac user? What's your current setup?
Daniel: My first Mac was that PowerBook Duo I mentioned, which I bought in 1994. I've been using Macs exclusively since then, and loving (almost!) every minute of it. My current setup consists of a MacBook Pro and a PowerMac G5. The G5 has been great - basically keeping up with the times for a few years now. But I think I'd like to get a Mac Pro at some point, and will probably trade my MacBook Pro for a smaller MacBook, since I mostly use it when I'm traveling.
Austin: Now, something I've always wondered: where does the name "Red Sweater" originate?
Daniel: It's funny how often people ask this question - I guess it begs to be asked. Unfortunately, the answer isn't very amusing. I had an old red sweater I was wearing when I got the notion to start a company. Actually, at the time I was young and ambitious, and thought it would be three companies: Red Sweater Records, Red Sweater Press, and Red Sweater Software. Well, at least one of them survived!
Austin: Your three-company plan certainly fit your gifts: music, software, writing. I'm dying to ask, though: given your music degree, what instruments do you play?
Daniel: I play very little these days as I've been spending so much time in the software business! But I started out hacking around with a guitar as a teenager, and ended up buying a piano-sized keyboard and teaching myself rudimentary piano, which I studied for my music degree.
MarsEdit
Austin: Starting with Safari's feed integration in 2005, Apple seems to be slowly embracing news aggregation in Mac OS X, and to some extent, blog/site publishing with iWeb. Mail has also picked up RSS support, using the Mail-like interface Brent Simmons achieved with NetNewsWire; obviously, there's still very little comparison between these two applications, but do you think this paves way for Apple to build their own blogging application?
Daniel: It definitely crosses my mind sometimes. When Apple announced the sessions for their developer conference earlier this year, there was a session all about blogging from Mac OS X, and it got my heart racing a little bit. But when I step back and look at how Apple has approached RSS, and how they would be likely to approach blog publishing, if they did, I end up not being very frightened by it.
I think they would probably tackle the problem in a pretty well-defined but relatively feature-less way, that would open customers' eyes to the possibilities that only a more powerful app like MarsEdit will provide. At least, that's what I hope!
Austin: Back in January, you challenged Mac developers to sell you their software, and you acquired two applications: Black Ink (originally MacXword), from Advenio, and MarsEdit, from NewsGator (originally created by Brent Simmons). What attracted you to acquiring the latter?
Daniel: Imagine you're a developer who runs a program almost every single day, loves it, and only wishes it could be made even better. When Brent and the guys at NewsGator approached me about possibly acquiring MarsEdit, I was really excited. It would have been almost impossible to say no, because I was already among the most passionate users of the application.
But even more than that, I honestly believe we're on the tip of the iceberg when it comes to desktop blogging. Things are going to change quite a bit over the coming years and as an avid blogger myself, I'm really excited to be able to play a leading role in how things shape up on the Mac.
Austin: What were your main goals for MarsEdit 2? Was there anything you wished could have made it into 2.0 that didn't?
Daniel: There were a few "must-haves" for MarsEdit 2, practically before I even sat down to do any planning. I knew that the user interface needed a refresh. I loved the simplicity of the UI that Brent designed, but we both agreed that it was starting to look a little dated. With Leopard around the corner, I thought it would be the perfect time to give the application a new look while keeping Brent's spirit of simplicity in tact. What I'm really proud of with MarsEdit's user interface is that its actually less noisy and in some ways simpler than MarsEdit 1, but offers a great deal more functionality.
Apart from the user interface, it was a question of working with a gigantic list of feature requests and paring it down to something manageable. I tried to pick things that would affect the everyday experience of the most users, and make the application easier to understand and get started with for new users.
Finally, I added the Flickr support because I recognized it as a hugely popular web service that lots of bloggers are using. Pulling direct access to it into MarsEdit was a way I could easily differentiate MarsEdit and make it the obvious choice for a large number of bloggers.
Austin: The user interface refresh is important to note; MarsEdit even feels a bit like NetNewsWire 3, from an aesthetic standpoint.
Daniel: I think Brent and I have a very similar world-view when it comes to how software should behave. It made it really easy for me to appreciate the design goals that are "deeply set" into MarsEdit's character.
Austin: Brent Simmons has always maintained a free version of NetNewsWire, but MarsEdit has forever been a shareware app, even when Brent was developing it. Of course, it's very much worth the money, but have you ever considered providing a "lite" version of MarsEdit for those who only do extremely casual blogging?
Daniel: I've thought about it, but it would require a bit of careful thinking to decide where to draw the line. It also complicates things somewhat, since you suddenly have two products to maintain, instead of one. For the time being I'm happy with users getting a 30 day no-restrictions trial of MarsEdit, and deciding whether desktop blogging is for them.
Austin: I agree that the trial is generous; it's nice to be able to test the waters easily and see how every tool works, something that's lost when a product is stripped of features until the user purchases it.
Daniel: I take a pretty casual approach to copy protection, and I want to err on the side of letting the user have a care-free trial experience, as much as possible. It's also just incredibly boring and unrewarding to work on copy protection or activation schemes. I'd rather add new features.
Austin: About a week before Leopard was released, Panic co-founder Steven Frank wrote on how developers don't receive the final build of the OS for testing. Interestingly, it seems press members do receive the gold master. Any predictions on why Apple would do this? Was this the case for previous versions of Mac OS X when you were still working there?
Daniel: When I was working at Apple you can probably appreciate that whether developers were getting a GM of the operating system wasn't exactly high on my list of concerns. I think that's basically what it comes down to, is that making sure "life is great" for developers is only the concern of a relatively small number of people at Apple.
It's easy to see from the outside how keeping developers in the loop and able to serve customers well is good for Apple, but when it comes down to these specific issues it seems Apple doesn't see a compelling enough advantage for the specific task of making sure we get the release ahead of time.
Austin: Did you anticipate any problems from the gold master with your own software? Were there any?
Daniel: I didn't anticipate any problems because I had tested pretty extensively against the release Apple gave us a few weeks prior. But as with any software, there's always the concern that some minute change could cause unpredictable, nasty consequences. After Leopard was released, I was relieved to find that everything basically works as expected for my applications (so far as I can tell!). Users haven't been reporting anything major so I am hoping for the time being that it has been a smooth transition.
Austin: Perhaps this is a silly question, but do you think there's a market for MarsEdit on Windows? Or rather, do you think MarsEdit could be popular on Windows?
Daniel: I don't think it's a silly question. I have to confess that I don't really know all that much about the Windows market. I have a few prejudices, as many Mac users and developers have, but I've never seriously analyzed the market in terms of whether it would be a good business plan to develop for it. The reason is simple: I like developing on the Mac, and I doubt I would enjoy developing on Windows.
I do get requests sometimes about the possibility of MarsEdit on Windows, but it's always from Mac users who have to use a Windows machine for some reason or another. I suspect, and this goes back to my prejudices, that most Windows users would not be tuned in to the types of things in MarsEdit that set it apart from other applications. One great thing about the Mac is lots of users really get and value highly "the little things" that make a product more usable and refined. I know there are Windows users who value these same things, but I think overall the market values feature-heaviness over refinement, and that's not a playing field I'd be interested in competing on.
All of that said, there are a couple really great options for Windows users who want to run a similar application to MarsEdit. BlogJet from Coding Robots, and Windows Live Writer from Microsoft, are two great options (and are actually quite refined).
iPhone
Austin: iPhone jailbreaking has become especially popular among more technically-oriented iPhone owners, and is becoming a more tantalizing choice every week with the emergence of easier installers and more useful applications. Apple announced an official iPhone SDK back in October, which will be released in February, which has seemed to quiet those who doubted such an announcement.
But iPhone web apps have proved to be moderately successful; Apple has published a Web Apps directory, which encourages developers to submit their creations, and applications for things like VoIP (IM+ for Skype), word processing (Google Docs), and even money management (ExpenseView Gadget) have been created for the device. Do you believe any of your products would be worth developing for the iPhone, or are you going to wait until the official SDK is released before you make a decision?
Daniel: I think the "web apps" solutions are great for things that lend themselves naturally to the web. For everything else I really think it's an uphill battle. I applaud the people who have creatively managed to come up with applications that don't feel like a web site, but I don't share that interest in trying to eke out the maximum functionality from a web site, when native APIs are so much easier to work with.
I see a real SDK for the iPhone not so much as allowing developers to make better applications than we can through the web, but more as allowing us to make applications that are completely impossible through the web. When you consider the ways that desktop applications on the Mac leverage the power of the desktop and host operating system to provide a stellar user experience, it's easy to see how native iPhone apps will offer similarly unbelievable functionality on that platform.
I haven't seriously considered building web-based versions of my applications, because I don't think I could offer anything compelling enough to be worth the considerable effort. When you consider that both MarsEdit and Black Ink are essentially desktop alternatives to web-based applications, it would be pretty silly for me to backpedal and assume all the limitations of the web!
Austin: What difficulties do you believe will arise from developing iPhone software, from a business perspective?
Daniel: It's really hard to speculate because we don't have any hard details, yet. It's uncertain whether Apple will require businesses to be authorized to develop for the iPhone, whether all sales will have to go through the iTunes Store, etc. I think it remains to be seen, but assuming the SDK is based on Cocoa and Objective-C, many Mac developers will have a significant head start at building incredibly great iPhone applications.
Austin: That brings up an interesting point- the iPhone is compatible on Windows, and I'll bet, when the SDK is released, there will be some Windows developers interested in embracing the device as a platform for their own software. Assuming the SDK is based on Cocoa and Objective-C, could that encourage Windows developers to release products for the Mac?
Daniel: I think that could happen, though I think the lure of Objective-C and Cocoa is pretty strong even without the iPhone. From my vantage point, the developer world at large is trending toward switching to the Mac, just as everyday users are. The number of Mac developers I run into -- even really good ones! -- who have only switched over in the past year or two is pretty amazing. I'm sure we'll continue to see more transplants as the incentives get greater and greater.
Austin: After the iPhone SDK was announced, there was speculation that Apple might not keep the iPhone completely open to development, and restrict its SDK to select developers ( i.e: iPod Games), or keep it closed to ADC members only, which would stop totally open development. The latter makes the most sense to me, but what are your thoughts?
Daniel: I think we're very much in a beggars can't be choosers situation. The iPhone as a platform belongs to Apple. None of us has the intrinsic right to develop for it, so we just have to hope for the best. I'm excited to develop for the platform and will jump through hoops to do so. While I think a completely open platform has obvious advantages, I can understand why Apple might want to take things slowly at first and see how the process works with some restrictions in place at first.
A huge thanks to Daniel for this interview. You can download MarsEdit and other Red Sweater software at the company's website.
Copyright © 2007-2008 Austin Heller.
Powered by Movable Type.