Interview with Daniel Spreadbury of Dorico

Interview with Daniel Spreadbury of Dorico

The moment we’ve been waiting for is finally here. Three years ago Daniel Spreadbury and the Sibelius team left Avid, and shortly thereafter work began on a new scoring application for Steinberg, the German company behind Cubase, among other music products.

Last week at a conference in Helsinki the product was announced. It’s called Dorico, named after a Roman music engraver, and will be released in the fourth quarter of 2016. There will be a cross-grade discount (€299) available to qualifying Finale and Sibelius users for a short time after launch.

Hopes for Dorico are high in the field, following a few difficult releases in the other major commercial applications. We were thrilled to be able to ask Daniel Spreadbury a few questions during this busy week about his work in open source and web standards, some of Dorico’s most exciting features for composers, and how Dorico users will benefit from the larger Steinberg ecosystem.

KC: What Dorico features do you think will make composers want to become Dorico users?
DS: Dorico takes a fundamentally different approach to how it thinks about music than existing scoring programs; its music model is much closer to the way that a sequencer or DAW thinks about music in MIDI terms. In a sequencer, a single MIDI note can give rise to a variety of different appearances in the sequencer’s integrated score view, based on the display quantization in use: it might appear as a single note, or as a sequence of tied notes. As the music is edited, the notation is effectively recreated on the fly. Scoring programs don’t tend to work that way: instead, once a note is created, it becomes a rather concrete thing. For example, if you create a quarter note on the last eighth of a 4/4 bar, the scoring program will correctly interpret that as two tied eighth notes either side of the barline. But if you cut and paste that music such that those notes both end up in the same bar, those tied notes won’t be recombined into a single quarter note again.

So firstly Dorico does think about music more like the way a sequencer does, and as you edit the score, the music will be renotated to the simplest and clearest rhythmic notation. However, it wouldn’t be any good if Dorico decided to combine those two tied eighth notes into a quarter note if they ended up, say, crossing the third beat of a 4/4 bar, because as every good music theory student knows, in time signatures with a half-bar, unless you have an established syncopated pattern going, you should always show the beat at the half-bar, so in that case that quarter note should remain as two tied eighth notes. In a nutshell, this is what Dorico does: it has a deep algorithmic understanding of the way rhythm and meter works, so it’s able to produce clear and unambiguous rhythmic notation in any situation.

Add to this that Dorico also provides tools to allow notes to be lengthened or shortened, and for music to be inserted in the middle of an existing voice, all of which keeps the music notated clearly at every point, and this is hopefully a big improvement for composers. Composers who would like to be able to do some composing directly into software and who would prefer to work in notation rather than MIDI from the start will find Dorico a much more welcoming and accommodating environment for their work than other scoring software. Our philosophy is that the user should not be penalised for changing his or her mind at any stage in the compositional process.

KC: You’ve made major contributions to open source and web standards projects for music notation. How does that work relate to Dorico, and what might be possible in the future?
DS: We have been working on the SMuFL standard for music fonts since a few months after we joined Steinberg, and SMuFL is now being developed under the auspices of the W3C’s Music Notation Community Group, which is also now the steward of the MusicXML format widely used by hundreds of music notation applications. As one of the three co-chairs of the group, I’m hoping that we will be able to build upon the strong foundations of both SMuFL and MusicXML to enrich the way music notation is encoded and to open up new possibilities for interactive applications, digital music distribution and consumption, while hopefully making it easier for software developers to handle this stuff.

As for how this relates to Dorico, a lot of the W3C work is forward-looking, so it’s not yet possible to draw too many conclusions about exactly how this will impact the software. In the short to medium term, we aim to have great support for SMuFL, and good support for import and export of MusicXML. In the longer term, we’re excited to see where the community takes the work on standardising the representation of music notation.

As an aside, I’ve been thrilled to see how Bravura, the reference font for SMuFL and Dorico’s default music font, which we released under the Open Font License a couple of years ago, has taken off: it’s now being used in a variety of music applications both on the desktop and the web, and it’s also starting to be used for publications by some of the biggest publishing houses, which is a great endorsement for the aesthetics and practicalities that have gone into Bravura, and I’m excited to see what people can do with Bravura when they get to use it in Dorico, which it was initially designed for.

KC: How will Dorico users be able to benefit from the broader Steinberg ecosystem?
DS: In the first instance, the inclusion of Steinberg’s world-class audio engine as used by Cubase and Nuendo is an obvious advantage. You’ll be able to use any VST 3-compatible virtual instruments and effects in Dorico – and certain VST 2-compatible plug-ins, too, since we know there are widely-used plug-ins that have not yet been updated to the new VST 3 standard. Our vision for Dorico’s playback features is that you will have the most direct control over how your music is realised by virtual instruments of any scoring software, and we’re doing that by borrowing some of the idioms more commonly found in Cubase, such as showing the music in a piano roll view, and allowing graphical editing of real-time controller data.

Over time, we hope to build powerful workflows that allow you to move projects between Dorico and Cubase with ease, since especially in the field of music for media it is not uncommon for a project to start in the sequencer and end in the scoring program if there is a live session to record, or indeed to start in the scoring program and end in the sequencer if a mock-up or virtual realisation needs to be produced.

Steinberg also makes some great virtual instruments of its own, with some new and exciting things coming up in the near future, and the UR series of audio interfaces has both top-notch build quality, audio quality, and solid drivers, too.


KC: NewMusicBox is a new music publication. Are there extended techniques that Dorico will handle particularly well, for instance microtonality, aleatoric notation, or unmeasured music?

DS: The first version of Dorico will not have every possible contemporary composition technique covered, but I hope that what it will demonstrate is that we are willing to tackle some of these areas. Dorico is the first GUI-based scoring software to truly support writing in open meter: in fact, when you start writing in an empty project, by default you get no time signature, and so Dorico doesn’t put in any barlines. You can go back and add a time signature at any time, and soon you’ll be able to simply insert barlines wherever you like to divide up a passage of unmetered music. You can also create a system break anywhere you like – even in the middle of a set of nested tuplets, if need be!

We have pretty good support for microtonality too: although the user interface is not yet finished, our engine has support for tuning systems with arbitrary equal divisions of the octave, so you can easily work with 24-EDO quartertones, 53-EDO Turkish maqam music, or set up your own 72-EDO system. You will be able to define your own system of accidentals, including specifying the number of EDOs that a given accidental should raise or lower the unmodified written pitch, and set up custom key or mode signatures that use those accidentals in any combination. We’re not yet sure exactly how playback of these kinds of microtones will be handled, but the semantics and the graphics will be there from the start.

Dorico also has support for many of the different conventions employed through the last century for accidental duration rules – so if you want accidentals to apply only to the note on which they are first written, and show naturals on all unmodified notes, or not repeat accidentals if the same pitch is immediately repeated, and so on, Dorico has all of that covered.

We don’t have great support for arbitrary graphical notations in the first release. There is so much complexity to producing really beautiful conventional music notation that we haven’t yet been able to focus on the kinds of convention-busting notations that have been used through the latter half of the 20th century and into the new century, but hopefully we will be able to build on our foundations in this area as the application matures. I think it’s worth saying that Dorico’s focus is on music that can still largely be written using conventional staff notation (and eventually tablature) rather than completely arbitrary graphics. I think other software will provide better tools for scores whose requirements barely include any conventional elements at all.


KC: Lastly, is there a way to apply to beta test Dorico?

DS: We are not quite ready to start beta testing Dorico just yet. We have already been inundated with literally hundreds of requests from prospective users from all sorts of musical fields who are interested in helping to iron the kinks out of the first version of our software, and that is incredibly encouraging. We will certainly not be able to accommodate everybody who has expressed an interest, and might even end up with some kind of a lottery system. If I haven’t managed to dissuade your readers from expressing an interest altogether, then of course they are free to get in touch with me, either via the new Dorico forum on the Steinberg web site, or via the Making Notes blog.

*Updated 5/24 at Spreadbury’s request to clarify his orginal comment

NewMusicBox provides a space for those engaged with new music to communicate their experiences and ideas in their own words. Articles and commentary posted here reflect the viewpoints of their individual authors; their appearance on NewMusicBox does not imply endorsement by New Music USA.

5 thoughts on “Interview with Daniel Spreadbury of Dorico

  1. Engela Fullard

    Everything I read about Dorico is so encouraging and exciting. It is wonderful to know that Daniel (Steinberg) is LISTENING to what musicians and composers want. The fact that one is able to communicate direct, is such a plus: ask a question, get an answer!!

    Reply
  2. M. Ryan Taylor

    I know this is completely sophmoric, but I was wondering why the branding of ‘Dorico’ was causing me some cognitive dissonance . . . until ‘dorko’ popped into my head. It sounds like your innovations are worthy of a better name for the product.

    Reply
  3. Carsten

    Good interview! Thx. I have a small side note. Daniel claims: “Dorico is the first GUI-based scoring software to truly support writing in open meter:” This is not quite right. Columbus Soft PriMus has always supported open meter…

    Reply

Leave a Reply

Your email address will not be published.

Conversation and respectful debate is vital to the NewMusicBox community. However, please remember to keep comments constructive and on-topic. Avoid personal attacks and defamatory language. We reserve the right to remove any comment that the community reports as abusive or that the staff determines is inappropriate.