There’s Markdown writing apps, and there’s rich text editors. Then there’s Ulysses III, the app that combines the best of both into one of the nicest writing environments on any platform. It looks sharp and works great, and I use it for a good portion of my writing these days — something I never would have considered back in the days of Ulysses 2.
Recently I had the chance to talk to Max Seelemann from The Soulmen team, and was able to arrange an interview with him about his team and their work. Here, for your reading pleasure, is their thoughts on OS X Mavericks, iCloud, building the best apps for each platform, and the story behind how Ulysses III came to be.
Thanks so much for taking the time to do the interview with us, even with WWDC going on. Since OS X Mavericks and iOS 7 are fresh in everyone’s minds, could you tell us your thoughts on the changes in Apple’s software and design strategy?
Apple is up to something big this year. iOS 7 is the long-needed overhaul we all have been waiting for. The interface is absolutely beautiful. Once Apple works out current inconsistencies and listens to some of the feedback it is getting right now, nobody will be looking back. We could not be more excited about it.
Mavericks, on the other hand, is a solid update to OS X. While it’s coming with little surprise, the changes are more than welcome. The removal of textures, addition of long-missing features like Maps integration in Calendar and finally fixing multi-monitor support are all great. And, of course, iBooks for Mac.
Your team already seems ready for iOS 7 — at least most of the way — since Daedalus Touch already has a flat design and unique document interaction gestures. Why’d you go with the document and gesture based layout when you did, while everyone else was still focused on menus and traditional interactions?
At that time, everybody was just porting over desktop metaphors. File lists, drill-down folders, etc.
Marcus wanted to focus on what made the tablet so wonderful: Multi-touch interaction, the instant manipulation of objects. The first sketches had no UI at all, just a text view and a loupe, which would be used for zooming out and search, with pinches and swipes used to access different parts of the text.
That was highly experimental, but we had lots of such ideas. At one point, we even had a truthfully recreated typewriter/page metaphor, with the iPad fixed in portrait, a fixed page height, and auto-swapping of pages, once you reached the bottom. Even five-finger pinch to crumble the paper. ;)
The final product has none of that, but what we are always trying to identify is a system’s unique appeal, its underlying concepts, then go from there. We will do the same on iOS 7.
Going back even further to 10 years ago when your company started, what originally inspired you to create Ulysses I?
The plain lack of an app designed for just writing longer texts. There were word processors designed for secretaries and code editors built for programmers. But no one had done a plain text app for book writers on any platform. We wanted to fill this hole.
What were the biggest changes in the world of writing apps that prompted you to start over with a whole new program in Ulysses III?
None, really. We started in October 2010 with discussing a UI redesign only. That was after Marcus saw the OS X 10.7 announcement and instantly got hooked on the new Mail.app interface. He started work on a modern look for Ulysses, and the more we discussed, the farther we went, and we eventually just realized that we wanted to build a completely new editor.
See, the editor component in Ulysses 2 was showing its age. There was very little room left for expansion, and we just couldn’t do a lot of what we wanted to do.
For example, Ulysses was started way before Markdown was publicly known, and it never was built to be truly semantics-aware. We made it look as if the app knew, but unless you told it during export, an emphasis was basically identical to an h1 or a link-tag. This was tedious, complicated, and we wanted it fixed.
We also always felt that while plain text editing was great, the file format itself has its own fair share of problematic limitations. For one, it’s overly technical, even with so called minimal markup. Images, Finder based file structures, the fact that you need to learn, master and remember a predefined markup in order to insert links – it’s no wonder RTF is still popular.
So instead of fixing what we had, we decided to take a shot at re-imagining plain text editing as a whole: Not so much as in “plain text, the file format”, but more in the sense of interaction – still a single font, tags instead of formats, a strong emphasis on semantics (sorry, no pun), but at the same time allowing advanced stuff such as annotations and footnotes without the clutter. All the stuff standard plain text/markup is really horrible at.
Out came the Ulysses III engine.
Once we swapped out the UI and the engine, there was little left to be bring over from old versions, of course. This is when we decided to really start anew and free version 3 from all dependencies on previous releases, going as far as opting out of any upgrade path for any of your previous content. We just “ended up” writing an entirely new application. From scratch.
What was the most difficult thing to get right in Ulysses, with all the custom features you’ve included like hidden Markdown and the Pure mode.
Definitely the editor. It’s the part we started coding first about 1.5 years ago and we are still tuning. The problem is, that once the basic stuff is sorted out and actually working, there are just so many subtile issues to be taken care of. Think of indentation, alignment, broad font support, text selection with objects (links), copy & paste behavior, drag & drop, implementing undo, redo, versions and so forth. It all must work, of course, but it must work in a way that feels natural to the OS.
As an example, we literally spent months on and off making text objects look and behave like they do. You can now look at them and ask “what’s the big deal, they work as I would expect”, but getting there took countless iterations.
You’ve spoken elsewhere about the difficulties you had in getting iCloud sync working. What would you most like to see Apple improve in iCloud, and how hopeful are you for its future?
The most important thing I want from Apple is to just fix the issues. There are a number of things that do not work as expected and that took us weeks to figure out and ultimately work around. From what I have seen at WWDC, and what I’ve been told, this is exactly what Apple is doing for iOS 7 and Mavericks regarding iCloud: No new features, just fixing bugs.
How’s the App Store worked for your team? Would you ever want to go back to selling apps directly on your site?
We switched all our sales to the App Store in the month it launched and have not looked back since. There is just so much stuff we no longer need to care about: licensing, sale transactions, taxation, download traffic, updates, user discovery, restoring previous purchases, international sales — the list goes on. The App Store is so much more convenient for us and our users. Without being on the App Store, Ulysses III could not use iCloud.
Plus, the App Store really helped boost sales, and in no small manner. The occasional Apple feature sure helps, so long story short: no, we would not want to go back.
It seems slightly odd that Ulysses III doesn’t sync everything with Daedalus, its iOS counterpart, when other writing apps like iA Writer and Byword do. Why’d you implement the syncing you did in Daedalus?
There are two aspects to consider here: First, Daedalus was there before we even started working on Ulysses III. It also had iCloud sync before Ulysses shipped. We added support for Daedalus to Ulysses, not the other way around. As a result, syncing between the two is currently limited by what Daedalus can do.
Second, the data model of Ulysses is much more complex. Unlike the apps you named and even Daedalus, texts are no longer just “plain text”. Ulysses embeds images and videos and allows for additional markup options like comments or annotations. There is just no way to make standard iOS text components suite to our needs.
So we only had two options: Either limit Ulysses III by what iOS has to offer – bad idea. Or to go for partial support at launch and then port Ulysses’ engine, data model and backend to iOS.
Hint. Hint. ;-)
We’re excited over the upcoming Alfred-like search in Ulysses that you’ve previewed, as well as additional import/export tools that you’ve promised. Is there anything else we can expect to see in Ulysses III in the near future?
We will begin by addressing the most requested features first. In addition to the open panel you already mentioned, Ulysses III 1.1 will gain something similar to a global search bar. We will also add typewriter scrolling, ePub export, full theming for all exporters and a bunch of commonly requested localizations.
From there on, our plans are neither fixed nor disclosable yet. We have a big enough idea backlog to keep us busy for the next couple of years, though, and we’re very happy to now have a solid foundation to build on, extend and improve.
At AppStorm, we always curious about what apps others use to get their work done. What apps do you rely on in your work and more?
I don’t need many apps for my everyday-use. I think I spend most time of my day in Xcode, Mail and Safari. For Twitter I use Tweetbot, for graphics I use Pixelmator and Sketch, our Git management is mostly done in Tower, issue tracking is handled via Jira. Marcus relies on Photoshop, Coda and Balsamiq Mockups, all of which I barely touch. ;)
Do you ever go back and use Ulysses II, or do you swap around with other writing apps, or is Ulysses III your only writing app?
Except from code and short mails, all my writing is in Ulysses III now. In part because I believe in “eating your own dog food”, but it’s pretty easy here, because I really just like it that much. (BTW: This is different from Ulysses 2, which I admit to have used only rarely.)
We mentioned Ulysses III’s Pure Mode in a recent poll on our site about skeuomorphism versus flat UI. Where do you stand on the great design debate of our times?
I think we’re rather ambivalent about it. While I (and I think my team as well) do not like leather textures and animating cassette decks, not every relation to real-world behavior is bad. There are situations where the adoption of real-world behaviors can help to convey the use of an app. Take the gravity simulation in iOS 7 or Daedalus’ sheets.
Also, design is so much more than just the look of things. Some people tend to forget that.
That’s All for Now
We’d like to extend a special Thank You to Max for taking the time to do this interview with us. If you haven’t tried out Ulysses III yet, you should go download the trial and give it a shot. The Soulmen’s passion shines through the app, and we can’t wait to see what else they cook up going forward … Ulysses III for iPad?