Meet the Developers: Keith Blount of Scrivener

In today’s interview, we’ll be speaking with Keith Blount – the writer and developer behind the phenomenally successful Scrivener. Created by the Literature & Latte team, Scrivener is a fantastic application for writers of all types, helping you organise your ideas and produce a piece of work to be proud of.

Keith will be talking about the Literature & Latte team, how Scrivener came about, giving back to the developer community, and sharing a few fascinating, high-profile examples of people using the app.

I hope you enjoy the interview!

Tell us a little bit about the Literature & Latte team – where are you based, how many of you are there, and what motivates you as a company?

Literature & Latte’s official headquarters are in Truro, Cornwall, in the UK. But that’s just a spare room in our house which has become my office – the team is actually spread far and wide. There are six of us, four of us full time, two part-time, and a seventh person helping out with the code on the Windows version. Three of us are based in the UK, two in the US, and the Windows team are in Australia.

As for motivation, Scrivener is our motivation and our raison d’être. A couple of years ago L&L was a team of one – yours truly – because it started out with me writing a piece of software I wanted for myself. We all use Scrivener ourselves (Lee, our Windows developer, was a user who was desperate to get it onto Windows, and our two support team members started out as enthusiastic users too), so we’re just trying to make the best writing software for ourselves as much as for anyone else.

With Latte in the name, does the company take its coffee seriously?

Ha, I wish I could go into detail about all the great coffees I brew but the sad truth is that my mornings start with a cup of Nescafé. The company name comes from the name I always wanted to use for a bookshop-café – I used that because I figured I’d never open a bookshop and I didn’t want anything that sounded too dry.

You can’t please everyone though – we received an e-mail from someone the other day suggesting we should go for something more “generic like Microsoft”. But I like it – and it is true that I can’t function without coffee, latte or not.

What does a typical day look like for you? Walk us through your usual schedule and activities!

Oh, it’s total rock and roll all the way (if you take “rock and roll” to mean sitting at a computer until your spine fossilises). My day starts – after taking the kids to school and that vital coffee of course – with technical support. I’m at my desk by 9am, and I spend the first hour replying to e-mails from users and answering queries on the forums (I’ll answer the occasional e-mail and question throughout the day, too, but I try to get most done first thing and before I finish in the evening).

I also look at issues that have been answered by others on the team, just so I always have a good overview of the most common issues users have – that’s always vital for working out what features to add and improvements to make.

After that it depends on how busy we are. If we’re not in a busy period then I’ll try to do two or three hours writing each morning. I developed Scrivener to use, so even if I never finish anything worth publishing, I make sure I set aside time for my own writing efforts – no matter how successful Scrivener might or might not become, it would seem a waste to me, having set out to write software that fits my way of working, were I never to use it in this way.

But this is also fundamental to Scrivener’s development, because as I use Scrivener myself I’m always seeing areas for improvement and a lot of the changes and updates are driven by my own needs as well as by those of other users. (For instance, the next update has some subtle improvements to outlining and a new “synopsis finder”, because I was working on an outline and wanted a better way of being able to jump back to edit earlier synopses without losing my place.)

After lunch I code for the rest of the day, and in busy periods I’ll code instead of writing in the mornings too. At the moment, for instance, I’m working hard on 2.1 so the coding is taking up all my time. Other than breaks for lunch and spending time with the kids, the coding usually goes on until 7 or 8pm, sometimes longer, and then I try to answer any e-mails or support queries that are left before wrapping it up for the day.

But it all depends on what’s going on. There are a couple of e-books coming out about Scrivener over the next fortnight, so a lot of my time recently has been taken up with proofreading those, and there’s the Windows version to test and various other odds and ends that need doing. We’re a small company, so we all pitch in and do various jobs.

What inspired the original idea for Scrivener – was there a particular piece of writing/novel that spurred the creation of the software?

Scrivener grew out of my own attempts to write a novel, and my experience of writing a thesis. I’m not the most organised of people, and I found that I would have dozens of Word files scattered throughout folders containing notes, ideas, partial chapters and so on. I’d copy and paste them all into a large Word file, then cut and paste the parts to restructure.

I would write synopses in Excel or on index cards to get an overview, shuffle them around to find the best sequence, and then go back to my Word documents and set about cutting and pasting it into the new structure once more.

I also came across an essay by (the now Booker-winning) Hilary Mantel in which she talked about her own writing methods – how she would keep notes on index cards and note paper, place them on a corkboard and move them around, then eventually move them into a ring-binder and keep arranging things and writing more in a non-linear order until the shape of the work became clear and she could see what gaps needed filling. She said that this was “really a method of growing a book, rather than writing one.”

It struck home with me because that was how my mind worked on longer texts too. And I kept thinking that there must be a better software solution for this way of working, something that would allow me to rearrange a long document using synopses, to allow me to edit the text in small pieces or as a whole and so on.

I tried out all the software that was available on both Windows and the Mac, but nothing did exactly what I wanted, so eventually I set about teaching myself to code so that I could write my own.

I was interested to see the range of components and applications that you’ve open sourced on your Free Stuff page. What was the reasoning behind this, and do you think it has impacted the success of Scrivener?

There’s a lot more stuff I keep meaning to put up there, actually. That page contains test projects that didn’t really go anywhere that someone else might be able to make something of, some utilities that I created for myself that are useful for single, rather esoteric tasks, and various code snippets.

When working on a large program such as Scrivener you often find yourself reinventing the wheel – having to create a control that seems to be used in many programs but for which there isn’t a standard, having to write word counting code and suchlike. So I try to place snippets of code that might be generally useful on there.

It’s also about giving something back – there are lots of great sites out there with Mac developers giving away code, and I’ve benefitted from a fair few of them so it would be nice if other developers benefitted from some of the stuff I give away in return. I’m way behind, though, as I say – there are lots of general-use components I developed during the 2.0 development cycle that I’ve meant to put on there but which I just haven’t had time to cut out and document yet.

I don’t think any of this has had an impact on the success of Scrivener directly, although the great nature of the Mac development community at large certainly has – but that’s the next question.

What’s the one thing you love about developing for the Mac?

I doubt I could narrow it down to one thing, sorry; there are too many things. There’s all the obvious stuff – the great hardware and OS X itself, which I love. And then there’s Cocoa – I really like the Objective-C development language, because it’s just such a human-readable style of coding.

If I really had to pick one thing, though, it would probably be the development community. So many developers are willing to help each other out and share code, and competition tends to be friendly. Of course, the Windows development community may be just as great, I have no idea.

Tell us a little bit about your Mac setup. What software and hardware do you use on a regular basis?

Until recently Scrivener had always been developed on the smallest machines – at first on an iBook 12”, then on a MacBook 13” through a couple of generations. Recently I decided it was silly to be sitting around waiting several minutes for Scrivener to build, though, so I bought myself a Mac Pro, which flies. So all of my development is done on that.

My other main computer is a MacBook Air 11”, which is my favourite machine since my original iBook 12” – it’s surely the perfect writing machine. A MacAlly IceKey keyboard and a Microsoft natural mouse completes a really comfortable set-up.

As for software, obviously there’s Scrivener – I use it to keep track of all of my development notes as well as for writing. I have a single Scrivener project with folders for upcoming and past releases, with documents on what features I plan to add and notes about them, release notes that get added to as I work and exported for release, and so on.

My main workhorse for coding is, of course, Xcode. I’m currently dipping into Xcode 4 as I’ve been putting it off, but I’m slowly getting used to it, although I’m mostly still on Xcode 3.

Skitch is one of my most valuable utilities for support – it allows me to take a screenshot, resize or crop it, annotate it and then either upload it to our servers for linking in a forum reply or drag it into an e-mail – all within seconds. It’s an amazing program, and I can’t even imagine how much time it has saved me.

I use Photoshop for creating or tweaking icons or images where necessary for Scrivener itself (although we use professional icon designers for most things these days). And then I have a lot of programs that I use to test Scrivener’s export and compatibility features – Microsoft Word, obviously, and Nisus Writer (which is my personal choice of word processor), Pages to a lesser extent (because Apple don’t make the .pages format publicly available it’s not really possible for anyone to support Pages directly, so that comes down to RTF and Word export); and then various e-book readers, such as Kindle Previewer, Adobe Digitial Editions and such like for testing e-books created by Scrivener.

What type of sales and customer response have you seen after launching Scrivener on the Mac App Store?

The Mac App Store has been really good for us, and we’ve been – what’s the US equivalent of “chuffed” or “stoked” – psyched? – to have made it as a featured app, and to have been kept on the front page for a few weeks. The Mac App Store has basically opened up a second stream of customers for us.

We were worried that it was going to cannibalise our regular sales (which would have been bad seeing as our regular e-store provider takes much less than the 30-40% Apple takes), but so far that doesn’t seem to have been the case, at least not too much. So all of our App Store sales have been on top of our regular sales, which is exactly what you would hope for.

I think it’s safe to say that the Mac App Store is still building its customer-base, too – which is natural seeing as it was only launched in January and is only available to Snow Leopard users (we have a lot of users still using Tiger or Leopard).

Aside from a couple of peak days, our regular sales channel still accounts for most of our sales, but it will be interesting to see how that will pan out in the future. Customer response has been great, though, and it’s fantastic to be able to reach more users and hear what they think.

What’s the most interesting use case of Scrivener you’ve ever come across? Maybe a particularly famous author, or unique/interesting writing project?

That’s a difficult one. We have quite a few bestselling authors using it (and were given a fantastic quote by Michael Marshall Smith which we use everywhere), and I’m always interested in how they go about their work. (In fact, one bestselling author, David Hewson, has written an e-book about how he uses Scrivener which has just gone on sale on Amazon – and he wrote and compiled the e-book in Scrivener too.)

I was excited when I found out that Scrivener was being used to write some episodes of the BBC series Spooks which I loved – and it was used by the same writer (Neil Cross) to write Luther starring Idris Elba (The Wire’s Stringer Bell) last year, so it was great to watch a fantastic series on TV and know it had been written for the most part in my software.

Recently I heard that one of the producers of Lost uses it, and I believe it was also used to write the script for the first Dead Space video game, which I didn’t realise until after I’d played it. Oh, and not long ago we received an e-mail from a witch using it to organise her book of shadows – that probably goes down as the most unique…

Which websites, Twitter users, and magazines do you follow in an effort to stay up-to-date with the activities of other developers, and the Apple eco-system in general?

AppStorm, of course! I also take a look at MacRumors occasionally, and check out all the magazine sites – Macworld, MacUser, MacLife, MacFormat and so on – and buy whichever magazines have the most interesting articles on any particular month.

In general, though, I tend to concentrate on the technical side of things, so spend more time using the developer section of the Apple site, and the excellent Apple-maintained developer lists and forums.

Do you have any interesting updates in the pipeline that you can give us a sneak peek at?

I’m currently working on Scrivener 2.1, which has some cool new stuff such as the synopsis finder mentioned earlier, and I’m working on providing a simplified view for Compile, for users who only want to tweak the most common settings rather than being confronted with customisation city when they are first learning the software.

2.05 introduced a feature whereby the user could have session targets automatically calculated based on a deadline date and the number of words they needed to achieve; 2.1 will allow the user to tell Scrivener which days he or she actually writes on, so that if you only write three days a week then the target will reflect that.

2.1 also introduces the ability to import research files as aliases, for cases where you have large PDF or video files that you want to reference in the project without bloating the project size. Lots of other stuff, too – I’m hoping to have 2.1 ready some time next month.

Along with that, I’ve started testing out Lion compatibility. While I can’t really say much about Lion for obvious reasons, I can say that I’m very excited about its full screen feature – using Scrivener’s corkboard in conjunction with Lion’s full screen mode is a superb fit, for instance.

So the next thing on the schedule is to ensure that Scrivener takes full advantage of Lion ready for when it’s released – which I personally think is one of the most exciting OS X releases for several years. This time last year I was concerned that Mac OS X was being neglected in favour of iOS; now we have MacBook Airs and Lion on the horizon. It’s a good time to be a Mac developer.

Thanks, Keith!

I’d like to say thank you to Keith for taking the time out of his busy schedule to answer our questions, and offer a little insight into what goes on behind the scenes at Literature & Latte. All the best for the future success of the application!

If you haven’t already, head over to the Scrivener website and give it a try. You’ll be impressed.