Hacker News new | past | comments | ask | show | jobs | submit login
Hummingbird – A fresh take on music notation (hummingbirdnotation.com)
472 points by pie on Apr 10, 2013 | hide | past | web | favorite | 276 comments

I'm very skeptical that this is an improvement (but kudos for thinking outside the box). Here's something that was intended as constructive criticism, but maybe ended up more as just criticism:

Removing the key signature is not a good idea. When playing in G major, the sharp accidental on the Fs is not put at the beginning of the line just to avoid printing it in the score. Rather, it fits there because when I play in G major, I put my brain in G major mode, in which case it would be distracting to have an accidental on every single F.

Similarly, writing a special symbol for each pitch seems it would get heavily in the way of transposing on the fly. The position already encodes the pitch, and the ABCDEFG names kind of get in the way of understanding the melody, which is more about relative intervals than absolute values.

And what does the little parenthesis on the length line mean? For half- and whole notes it seems to mean it doubles the length (a quarternote with one or two parentheses), but for sixteenth-notes it seems to indicate that it halves it (an eighth-note with a single parenthesis mark).

I also question removing the stem of a note. I have a feeling that is one of the stronger queues for reading rhythm. Spacing is not very important, and indeed especially in dense scores for solo instruments, that need to have few page turns, notes are often just spaced as tightly as possible.

The author also recognized that the beams on eighth- and sixtheenth-notes (e.g. in the left hand) are very important rhythmic cues, and replaced them with that horizontal thing with the arrow on the left. This is a bit hard to read though when there are no stems to link them to the note and the pitch interval is big.

The part about it being easy to write by hand looks good, and made me feel good at first. Then you realize that hand-written traditional notation is quite different from typeset one, just like handwritten text is very different from printed text. Drawing all the little balls and filling in the halfmoon C, up and down thingies seems tedious, when traditionally one writes a simple dot or a little slash instead of the note head.

What you mean by this is that the key signature tells you what the tonic is, and, conversely, what the supertonic, mediant, subdominant, dominant, etc. are.

Whenever I teach key signatures this is the biggest thing I try to impart: key signatures are more than just accidentals, they define the roles each pitch plays.

Agreed. Hummingbird seems like a solution in search of a problem. I could see someone trying to learn to sight read music, get frustrated with how it is "broken" and then invent something like this, which seems far less clear to me than the old way.

sight reading is hard because of the cognitive work involved with translating, say, eight notes at a time into the appropriate finger-patterns, rhythm and pace; or translating fewer notes but at a faster rate. i don't see this making it easier.

i'm not sure why they would bother. anyone who finds sight reading too tough, might not really ever need sheet music, whether they do it by ear, or just listen to music. my friend has a disklavier - incredible. robot plays the piano better than i ever could. it makes me wonder why i still bother to try and play, and then i remind myself that every feeble attempt i make is intrinsically satisfying to me, so i keep trying.

I could see someone trying to learn to sight read music, get frustrated with how it is "broken" and then invent something like this...

I felt exactly this way learning to read music. The score just didn't present enough information fast enough for me because I hand't learned to build the context in my head and spot the relations. I fumbled about with alternatie notations, extra notations etc. I finally settled on colors. I wrote a small program to color the dominant, subdominant etc. and then printed it on my color printer. This helped me immensely and I didn't take the dead end road of learning alternate notations that would shut me out of all written music everywhere.

Some windmills should be tilted at, some should just be left alone.

I am a neophyte at reading music, but I think that the pitch symbols are almost entirely noise. They would help me personally (oh, that's a 'C' there, I don't have to count the lines!) but for anyone with a modicum of sight reading, surely they already know where the C is due to its location on the staff? At that point, pitch is already represented and the filled figures are noise.

I wonder if a suitable analogy would be like underlining all capital letters in a selection of prose - useful for the neophyte, chaff for the slightly experienced onwards.

robot plays the piano better than i ever could

Apart from the personal enjoyment angle, robots also can't detect the mood of the audience (yet) and play to suit.

Agreed. As a pianist, I not only look at the position of individual notes, but also the positions of the next 3 to 6 consecutive notes. This way I can think in "groups" of notes - which is very helpful for sightreading broken chords quickly.

The one place I could see pitch symbols being slightly beneficial is when notes are either way above or way below the staff. But even then, after some deliberate practice these can be recognized pretty quickly as well.

I've read for piano, bass guitar, flute, and now ukulele. I can read treble clef, and bass clef just fine, but stick them together as you do for piano and my brain has a conniption. The exact same figure at the exact same place on each staff represents two different notes. I suspect this is because the gap between the two staffs would have required too many intermediate lines had they re-used treble clef for the bottom staff, but this doesn't mean it is a good readability choice.

Regarding relative position. On a piano the relative position of notes on the staff almost directly correlates to position on the keyboard. On a stringed instrument this is VERY far from the case. As i proceed up the scale on a simple instrument like the bass i will proceed right on the neck on one string, then drop down a string, shift left, proceed right on that string, drop down a string, shift left, proceed right on that string. So you've got back and forth, and up and down motions to keep going in one direction tonally. On a guitar the strings aren't all tuned evenly so the back and forth on the neck changes depending on which strings you're switching between. On the Ukulele the top string is higher than the string below it so you end up jumping UP strings rather than down, AND the strings aren't tuned at even intervals. So you've just got a big jumble of movements.

Don't even get me started on the totally unintuitive nature of woodwind fingerings or how they are almost totally disconnected from what's going on on the staff.

In short, the relative positioning of notes on the staff is good from a tonal perspective but crap from a finger perspective on most every instrument except piano.

>In short, the relative positioning of notes on the staff is good from a tonal perspective but crap from a finger perspective on most every instrument except piano.

That's not really a problem once you build the muscle memory, though.

It's the same thing as touch typing. With enough practice, you know that finger placement X will produce note y, the same way you know where the keys on the keyboard are.


About hummingbird, it's amazing. Really, I found it strange in the first minute. Then I got it

Why? My major peeve with traditional notation is the lines above and below. And I never know what are they talking about after adding some lines above the stave.

About the key signature, 100% agree.

As pretty much all students, I thought the key signature just meant which notes are always sharp or flat ("oh these guys want us to use more black keys - BORING")

It took me some time to get it. It's not about flat or sharp, but as they said it, it's 'G Major mode'. Then you feel everything is clearer. It's like going from understanding words in a phrase to understanding the phrase.

ehm, I would disagree. The key signature tells you what key the piece is in, period.

What the particular notes in that key do (their role, as you put it) is a property of that key.

How is "tell[ing] you what key the piece is in" different that defining tonic/dominant/leading tone relationships? Those relationships are the product of the whole/half step pattern that the key signature communicates. It isn't just important that G has 1 sharp, and it is F. That sharp actually defines the leading tone.

That's only if the tonic is G. I play tunes all the time with one sharp, but a tonic of D. (ie D mixolydian.)

Sure, but I'd argue that's because the key signature is wrong. The correct way to notate that is with either a modal key signature a la Bartok or notate it as D with a constant natural on the C, showing that the tonal center is D while clearly notating the departure from the traditional major scale.

By this standard probably something like 30% of the music ever written has the wrong key signature -- and I've never seen a piece of non-major music with a correctly notated key signature!

Spoken like a true musicologist.

I believe he was referring to the "put my brain in G" comment.

It's all about recognizing when to build tension and when to resolve tension.

Yes, the roles that the notes play is defined by the key, but if you don't mentally make the switch, you'll often perform things poorly because you don't anticipate where things are going correctly.

Or at least that's how it works for me, at any rate. Take anything I say with a grain of salt. I can't sight-read much at all.

All I can say is this, I play guitar for 15 years, learned piano since I was 7, and I never got the classic notation. I can't read notes. And I tried learning a few times.

I looked at hummingbird, and it took me 30 seconds to memorize the symbols, and now I can read any hummingbird note sheet easily.

I would say this is a big win. Same feel I got when I saw Tau, mathematicians might argue it's incorrect, but for me, I don't care, Tau made math easy for me. Hummingbird makes music easy for me. That's all I care.

About transposing, you got a point, I think there should be apps that do it for you (e.g. convert MIDI files to hummingbird notation, I'm sure we'll see a few of these as Show HN sometime soon)

You can now sightread pages of hummingbird after 30 seconds of cursory inspection?

Hummingbird feels like a crutch to me. Hiding key signatures stunts your understanding of how music works. Forcing time to be marked explicitly as well as by altering spacing places odd constraints on typesetters. Humans use several subtle visual cues¹ to read music quickly and efficiently, and hummingbird just throws them all away.

I'm not sure this is worth the overhaul.

1: http://lilypond.org/web/about/automated-engraving/software

Yes, I can, very, very slowly, but much faster than with the regular notation, and mostly because I'm not used to sightreading regardless of how it's written. I think it will never appeal to classic / professional musicians or anyone who already feels comfortable with the regular notation, but for people lbike me, that for whatever neurological reason just can get the old notation to stick, this is priceless. The Above Below C Dot Empty Full Groove is such a great wiring that I will never be able to forget it. I might not read it fast as you read regular notes, but I will at least be able to practice. Reading regular notes to me was always counting lines, always missing the global # or b, and simply giving up serious music. I recorded music, had songs on the radio, I can play anything by ear, but not being able to read notes always bothers me. Now I can.

Hm. Thanks for replying; this does sound interesting.

If this had that much of an impact on you, and if this has the potential to encourage others to learn notation, I'd like to see a full study of how hummingbird compares to the regular notation.

How were you taught notation in the past?

Looking at the first part of God Rest Ye Merry Gentlemen, hummingbird makes it look far more complex than it actually is. A simple series of quarternotes only changing in pitch, but hummingbird is changing two things - visual complexity (filled notes) and location.

I think it's bold and interesting, but ultimately I think it hasn't got legs.

Yes, machine transposing of hummingbird is trivial, but what about transposing while sight-reading? I have to do that regularly, and the removal of the key-signature makes that impossible.

Right, key signatures aren't just shorthand, they're useful information to give the performer up front.

In fact, it's so useful to state the key up front, I might even suggest the opposite; base the notation on relative pitches, as in "movable do" solfege. Then we could replace "all F's and C's are sharp" with just "do is D". Of course, there'd be tradeoffs to this as compared to the traditional notation, but it's fun to think about.

That would be the ideal notation for a diatonic harmonica player like me. I never really have an idea of what letter note I'm playing, unless I think about what key I'm in, and then I have to count. But I always know exactly where on the scale I am.

It's funny, before I read your comment I wrote a longwinded harmonica player's lament elsewhere in the thread, and I was thinking the same thing: A relative notation would be perfect.

Why would you go to the trouble of designing a new music notation and not make it based on relative pitches? What are the tradeoffs?

My thought was that a relative-pitch notation might present a steeper learning curve. But I don't really know.

I've worked with intermediate-level musicians who are skilled at regular sight-reading, but who struggle to learn to transpose on the fly. A relative-pitch notation might be like requiring everyone to "transpose" all the time, forcing them to learn this skill before they can proceed.

On the other hand, maybe you'd teach it differently. Instead of learning transposition as carrying around an offset to add to each pitch, you'd learn it as carrying around a scale to which the relative pitches would be mapped. I wonder whether this would be easier or harder to learn.

You're going to stay in one scale for a long time anyway (C for piano). For that duration, pitches in the notation correspond obviously to fingerings (keys, frets, whatever). This includes other instruments which have a different "natural" key (D for the viola, my other sort of instrument): their notation maps just as nicely.

When you get beyond that, "mapping relative pitches to a scale" is, IMO, the correct approach. The "offset to each pitch" thing obscures what's really going on, as if they were all weird derivatives of the key of C. I can see how it might be harder, among other reasons because you have to understand how a scale works more clearly, but I suspect it would be worth it.

Of course, it's hard to really know without getting people to switch for a couple generations.

The musical world has already tried out relative scales. It was in fact the first thing tried in notation. Look at the notation device used for Gregorian Chants to see it yourself.

We already use relative notation for most instruments. Trumpets, Clarinets and Tenor Saxophones typically read notation where Bb is equivalent to C on the page. What you'll note we Don't do, is change our relative pitch between songs.

No. Relative pitch is Very Hard, because as a musician, my hands don't play relatively. When I see an Eb on a notation sheet, my body knows how to play the note without me thinking about it, and it has to, because sometimes you can have notes flying at you with a rate of 32 a bar. This does not leave time to translate between the relative abstraction on paper and the absolute pitch of my instrument. It's not just harder.

It's much, much harder to train your brain and body to be able to map the same object on the page to twelve different body positions, depending on some variable that changes at random between songs.

Also remember, that different pitches have different textures, otherwise we'd never play in a keys other than C.

Absolute pitches and are the concern of the performer, but relative pitches are the concern of the listener. Who's more important? ;-)

This is why I mainly worry that a relative-pitch notation may present a steeper learning curve. Advanced players typically need to be able to think at least somewhat in both relative pitches and absolute pitches at the same time anyway. Absolute pitches because that's what many instruments demand, and relative pitches, because that's where a lot of the meaning is, which informs other aspects of the performance. But it typically takes more practice to learn to think in both ways at once.

> Also remember, that different pitches have different textures, otherwise we'd never play in a keys other than C.

I did propose that pieces could indicate their intended key.

Absolute pitches are the concern of the performer. Correct. Also consider that the performer is the one who is looking at the sheet music, and not the listener.

This is just painful. Instruments do not have scales. Music is written in different keys, and often modulates to another key and back again during the piece. In order to cover the range of pitches, different clefs are used to make the notation easier to read. Piano music is written in two clefs, bass and treble, centered around "middle C". This is called the grand staff. Depending upon an instrument's range different clefs are used. Violoncellos use bass, tenor, and treble clef. Violas use alto clef.

Scales are a function that acts upon the set of pitches used in music. A C major scale has the pitches C, D, E, F, G, A, and B. A G major scale (function) has the pitches G, A, B, C, D, E, and F#.

"Mapping relative pitches to a scale" is nonsensical. A scale (function) is an absolute set of pitches, immutable.

There are two forms of solfège (do, re, mi, ...): fixed and moveable. In moveable the first pitch of the scale is always do. In fixed the pitch called C is always do. Watch "The Sounds of Music" to learn the solfège for a major scale, it's the do-re-mi song. There are other syllables used for the three minor scale forms. To my knowledge (admittedly not complete) Italy, France, and perhaps England used a fixed-do system, whereas the rest of the world uses a moveable-so system.

Instruments do not have scales.

Actually, most trumpets, harmonicas, and many classes of flute are restricted to diatonic scales. You'll have a hard time finding a chromatic trumpet outside of a museum.

There's more to music than western chromaticism or indeed harmonic theory.

You've just used a bunch of words that sounds smart, but provide deceptive value. The typical Bb valved trumpet that exists now does indeed play chromatically. And so will any other class of modern brass instrument with three valves or more.

The chromatic trumpet is an invention that allowed us to escape the trumpets natural harmonic series (Those being the notes a trumpet can play in a single valve position) before valves were invented by drilling holes along its tubes.

Wikipedia is not the best place to learn music theory, and knowing music theory is as useful as knowing a list of rules for programming best practice.

I was incorrect about trumpets, but not about other instruments. My point that many instruments are restricted to a particular scale is entirely correct.

Also, I was playing and recording music since before wikipedia even came into existence, so dial back those assumptions.

You'll have a hard time finding a chromatic trumpet outside of a museum.

This is kind of a confusing statement to me. Most instruments called "trumpets" where I come from are valved instruments which are fully capable of producing chromatics. (Maybe this is technically incorrect and we should be properly calling them "cornets" or something, I don't know.)

My point is, you certainly don't have to go to a museum to find one, and I expect, in a museum (of history anyway), any trumpet you'd find would be more likely to to be valveless and non-chromatic.

This wouldn't really work for anything not in a key. How would you notate Boulez?

You would pick an arbitrary base and do everything relative to that. This is pretty easy compared to the shoehorning already required to notate such music in existing notation schemes :-).

On your last question, it took me a minute to get that those various "Above" or "Below" correspond to the notes they represent, A and B. The pitch is revealed both in the symbol and in the position on the staff. I found that pretty useful, certainly easier to read than homogenous black dots where you only have spatial information.

I read through the whole page and didn't get that. It didn't click until I started looking at the second page on the site (which goes more in depth).

My immediate thought once I figured that out was "this depends on English". For speakers of Japanese or Russian or French, can you find easy to remember words that can go with the first symbol, and that start with an "a" sound?

I actually found the terms extremely confusing. At first I thought the above/below referred to sharp/flat, but that didn't turn out. I couldn't even come up with concepts for dot and groove.

I know how to read music, but I'm not very good at it. I'll say that the idea of giving each note it's own shape seems like it could fix the problem of losing my place and having to count lines/spaces to figure out what a note it.

The thing I've never liked about musical notation is that the X axis has nothing to do with time. A staff can have 8 32nd notes and then a dotted half note. Those first notes take up 1/4 the time but 3/4 the space. While it would take more space, I find a consistent relationship much easier (such as piano roles or Guitar Hero note charts).

As a native French speaker, we don't use the CDEFGAB notation but rather the "do ré mi fa sol la si do", so the symbols make no sense whatsoever.

Thanks, turns out I totally misunderstood those at first - and edited my comment in the meantime. Sorry about that.

Your critique is spot on for those trained in music. Is that his target audience?

What about for those being exposed to music and its notation for the first time? What if you had never had a concept of putting your brain in G major mode.

But if complete beginners are his target audience, it begs the question why stop there? Why maintain the use of the staff at all.

No matter what people come up with and who the target audience is, there needs to be a good bridge between the traditional and the new. Staff or not, great for beginners or not, this will only fail because of the total lack of sheet music and other materials with nothing in sight to automate the translation of even a limited subset of music and materials.

(Of course, caveat, I've been playing piano for 20+ years as a hobby. Maybe the ties to the traditional are really not a big deal.)

I certainly hope that in the near future, digital transcriptions are easy enough to come by that you could effectively just change the font from one notation to another.

I know there are a lot of judgement calls in typesetting music, but that's true of text too, and we have some pretty good algorithms for doing so these days.

All that said, I have a lot of reservations about this particular notation, but I think there's certainly room for improvement even inside the current system. Letting the back catalog of work hold notation back isn't healthy.

The whole point of any new, better musical notation is avoiding the need to deeply understand the current traditional notation.

It is enough if machines can translate the existing sheet music to the new system.

> target audience

This is a very valid point.

Certainly the learning process for children can be improved a lot. Really, a lot. But I think it is important to integrate it with the actual notation from the beginning, as otherwise you will hit a wall when you want to play a bigger variety of music that's only available in that format.

I don’t think acquiring a sense of key signature is a dispensable part of the music learning experience, unless you plan to stick to twelve tone music.

If you look at the guide, the key signatures are still included, but written in plain English. Look at the Fur Elise example:


Not all musicians work in plain English. There's the German note H, for example, or half of Europe using solfege: http://en.wikipedia.org/wiki/Solf%C3%A8ge#Fixed_do_solf.C3.A...

Musical notation needs to be spoken-language-free.

Musical notation has never been spoken-language-free: the scores I play are all full of Italian words. I have never learned any Italian; these words are simply part of the notation, much as English words are parts of the notation in many popular programming languages.

Fair enough, I stand corrected.

> Musical notation needs to be spoken-language-free.

Tempi and volume are usually given in Italian, aren't they?

Here's an interesting alternative notation that emphasizes relative pitches: http://muto-method.com/images/fig11-e.png.

It was invented for the Chromatone, a keyboard that eliminates black and white keys and treats all keys the same way, similar to a guitar fretboard. http://muto-method.com/en/index.html

Transposing then becomes easy because every scale has the same shape.

Thanks for that link, that is very interesting indeed. Being able to read the intervals more easily is certainly valuable, especially if you are playing on a chromatic keyboard.

However, I do like the fact that accidentals in traditional notation say something important: This note is out of the scale you are playing in. This is usually audibly very noticeable, so it makes sense to have it very noticeable in the score as well.

The three line system could be enhanced to show the notes that are in and out of the scale slightly differently, e.g. with color, or the size of the note-head. This would still have the benefit of being easily transposable.

The mnemonics are also anglocentric visual puns (e is for empty). I'd suggest a set of symbols with natural ordering that can be distinguished by shape (vs. all circles)

That's actually already been done: http://en.wikipedia.org/wiki/Shape_note . The hymnals at my grandparent's church actually have shaped notes.

Yes, I realized I misunderstood those, and edited my comment a bit.

Solid points, although I think you're zeroing in on too many low-level details too early even though your criticisms are valid.

Also, maybe you could make your feedback constructive by providing ideas, alternatives and suggestions instead of just ripping it apart but adding no value or solid suggestions beyond that.

The point here is much bigger and more significant than any of your individual criticisms.

The point is this - these people are rethinking the problem of notation, and redesigning it from the ground up.

I think their approach is totally badass, and I love the fact that someone is tackling this!

> Also, maybe you could make your feedback constructive by providing ideas, alternatives and suggestions instead of just ripping it apart but adding no value or solid suggestions beyond that.

Pointing out flaws IS constructive criticism. This happens all the time and it bugs me:

person A comes up with an idea person B gives good reasons why it's flawed person A - the one who wants to innovate - says that fixing those flaws isn't his problem, and that person B should fix them.


The person who generates an idea, the person who wants to do something new, is the one who is responsible for brainstorming, finding problems, finding solutions, and pivoting as needed.

To say "your criticisms should be bundled with solutions" just raises the bar on criticism, and when it costs more to generate criticism, you get less of it.

We should all encourage criticism, and then we should make finding solutions OUR problem.

Thank you, my thoughts exactly.

I much prefer someone to criticize my work, rather than sitting quiet just because they can't think of improvements. After all, I'm probably much more likely to come up with a solution than they are, and the reason I didn't is just because I didn't think of that problem.

Thanks, yes I realized my lack of suggestions for improvements after writing my comment. I simply didn't have any (it was also quite late), as I think there are some fundamental choices in that system that are not beneficial. My "low-level details" were meant as arguments against those fundamental choices, rather than nit-picks that could be easily solved.

In the end, I judged that (valid) criticism is better than no criticism, despite the negative feeling of being the guy that just "tears it down".

My first observation was that it was difficult to discern the notes on my screen. Imagine that on paper farther than arms length away in a darkened room. That's plenty enough to disqualify it.

But yes, classical notation could certainly use some refinements.

I had the same reaction. No matter how small the notes are in classical notation, I can tell if they're filled or hollow, and whether an accidental is a sharp or flat. Distinguishing between hollow, partially and directionally filled, and filled notes, with diacritical accidentals seems more difficult.

You don't have to omit the key signature.

It's pretty clear the person(s) who created this is/are not (a) very proficient musician(s).

The current notation has been in use for hundreds of years because it works. The notes are large, bold, and easy to recognize and also easy to write. Memorizing GBDFA, EGBDF, ACEG, and FACE is not that difficult.

This new notation has many egregious flaws. Removing the key signature is one of them. Not only does the key signature allow for instant recognition of the pitch and tones used and a general idea of what the piece should sound like, but it also makes writing sheets that much easier for arrangers and composers.

Second, who thought it was a good idea to replace the accidental signs with squiggly marks? A huge step down in usability, I'm afraid. Maybe in sevenths and chords with very close note spacing, but unless you're playing Death Waltz, it is not a problem (usually).

Also, the Consumer Reports-esque notes are also distracting and don't serve any purpose. If I saw what is an "E" on this note, I would play it for 4 beats -- that is, assuming that this is in x/4 time. It's more confusing for longtime music readers than musicians, but it would still throw off many, I'd guess. At any rate, however, if you can't memorize the staff lines and spaces, you aren't a musician. Period.

The uselessness of this notation is compounded by the fact that practically no instructor will be willing to give up a notation that has been in use their entire life, and also for centuries.

Do I see this being successful? Maybe, in small circles (no pun intended). But the harsh truth is that the current notation is easier to write, easier to read, and more efficient.

Case closed.

I'm not going to comment on their musical abilities but I noticed their pain points were common pain points that students have.

Perhaps not surprisingly its similar when you learn a foreign language, for a while you see a word, translate it into your native language, and then understand it. But once you "get it" or reach a certain level of fluency, you read the word and you just know what it means.

When I started playing Trombone I would see a note in the stave and count lines or spaces to figure out what note it was, and then play that note on my instrument. There were actually two translations going on, one from music to note 'name' and then from note 'name' to instrument configuration. At some point however it changed and I stopped seeing a 'B' or an 'A' or a 'C' and instead saw instrument positions (and alternates) so that playing stopped being a translation exercise and simply became execution.

The other interesting thing is that looking at the sheet music I heard music. And that was when everything clicked together because initially I could play things "by ear" by recreating the same sounds in my instrument that my ear was hearing, and now my eyes would see the music, my ears would "hear" it, and my fingers would make it real. Notation stopped being an issue until I tried to play keyboards :-)

Funny I know plenty of proficient musicians who can't read music at all. I don't disagree with what you are saying per sey but saying the creators can't be good musicians because they want to try a new approach to teach music feels awfully elitist.

They are only "proficient" in a very limited sense, then. I'm sure we can argue semantics here, but a musician needs to understand the concepts that govern music. And these concepts can only be understood and visualised by using notation.

The circle of fifth, musical modes and any advanced form of transposition, alongside the basic theory involved in cords, their components and their succession are necessary for a musician. [1]

Technical proficiency does not make you a musician, just like being able to type really well doesn't make you a programmer if you can't...read code.

[1]: This may not apply for percussionists, but even there, playing together with others necessitates the same understanding.

Side note: it's per se

> And these concepts can only be understood and visualised by using notation.

That's not true at all. One can have a deep understanding of all the elements of music you mentioned without knowing any traditional notation. Maybe it's understood in terms of modular arithmetic and fractions and visualized as such, or visualized in terms of a piano or an abstract spiral structure or not even visualized at all. Traditional notation is only one possible representation (an a fairly arcane one at that); others are certainly possible and don't preclude deep understanding.

I am well aware that Western notation is not the only one, however, I cannot see how music theory works without any notation. Whatever form of notation you choose to encode information in will still have to be decoded, i.e. read. The reason why some notations have been more durable than others is the relative amount of information that can be stored.

Tabs for guitar and bass are just one example of a deficient mode of notation, since they cannot provide all information necessary to play something.

If you want to forego all notation, however, understanding theory becomes even harder than with a deficient notation. For that, you'd have to solely rely on somebody's ear in order to explain roots, scales and afterwards more complex topics. The only other option is teaching visually, which ist somewhat possible with a guitar, and somewhat possible with a piano, but is not transferable to another instrument afterwards.

All forms of notation are deficient. If you want proof, just listen to a muzak (elevator music) cover of your favorite pop song. It is correct according to a professional transcriber's notation, but it loses nuances and imperfections that classical music notation doesn't cover.

You are right that tablature by nature does not cover rhythm and that is a gap that won't even get you to muzak. However, there is nothing about classical music notation that makes it an ideal form for representing all forms of musical sound. How would you notate a Jay-Z or Skillrex song?

those are actually pretty easy. The only issue with skrillex would be identifying an instrument to play the notes, but other than that he's low on the polyphony scale (pun intended).

When I clicked this I was hoping for a notation that describes midi values against timbre (filters, distortion etc), and blends it with traditional notation, I think that is a missing link between old and new - but rewriting accepted notation for generations? That's like replacing a-Z with some sort of base 26 number and saying its easier to learn.

I was hoping for a notation that describes midi values against timbre

Dream on. You could show controller automation lanes as digital audio workstations do, but those are only as informative as they are consistent with some scheme like General MIDI and a standard sound architecture. Once you start resampling or using modular routings any kind of timbral notation schema goes out the window.

I've seen attempts at this, and if you email I'll try and dig you out a reference but the particular book I'm thinking of is in a box in my attic right now. However I can't say I've found them very informative. When I think about timbre I think about the whole synth architecture and program it in my head, for many timbral ideas it's just a matter of walking up to the synth later and dialing it in.

Most electronic musicians stick with simple diatonic scales or within modes, at most switching in and out of relative minor. Harmonic complexity and timbral complexity don't go very well together.

I would absolutely love if you could post the reference of this book here - and I'm sure a couple of other folks would as well! It sounds quite interesting. Thanks a bunch.

I went up and had a look, but haven't turned it up yet. I'm supposed to be getting new shelves in the next week or so and getting all those books up onto them, so I'll keep an eye out.

> They are only "proficient" in a very limited sense, then. I'm sure we can argue semantics here, but a musician needs to understand the concepts that govern music. And these concepts can only be understood and visualised by using notation.

That's pretty narrow. Some people aren't visual thinkers and don't need to see something on the written page to understand it. I personally know several musicians who have a deep understanding of music theory, but don't read music. You can explain and understand modes, transposition, chords, etc without having to write it down.

> I personally know several musicians who have a deep understanding of music theory, but don't read music

"don't" or "can't" read music? I think this is a critical distinction. If the answer is the latter I would still argue that they cannot have a deep understanding of music theory, since a major component of music theory is generally accepted to involve understanding notation.

Edit: I can accept that there are many aspects to music theory that go beyond just notation, but I am skeptical of 'deep understanding'.

So would you consider the various blind musicians (such as Ray Charles) not real musicians? After all they can't read music.

Edit: I do otherwise agree with you that a full musician should understand much of the theory of music, and not just be able to repeat a group of notes that they heard (that is, they should be more than just a human tape recorder).

>these concepts can only be understood and visualised by using notation.

That attitude makes you an ass. That might be blunt, but it's true. I can't imagine someone who is actually musical thinking that for even a second.

That is very wrong... You are implying that the concept of music can only be understood through theory and knowledge of the musical notation. And that's just plain wrong because there has been thousands of famous musicians who cannot read music.

One doesn't need to know musical notation to understand any of those ideas. Related, one can learn how to read music and never learn any of that stuff.

You can understand music without being able to read music. I know a fairly famous guitarist who can tell you all about voicings, harmonization, and any other bit of music theory you'd care to know about. He uses all this to write and improvise music. You put a piece of music in front of him and he'll take five minutes to figure it out. He'll be counting lines and spaces and humming notes to do that.

How is "counting lines and spaces" not reading?

Technically it is "reading music", but it's so slow that learning a piece takes much longer than it should. Over the years I've had a couple motivational spurts where I felt like playing piano again, and lack of frequent exposure to sheet music slowed my reading skills down to a crawl.

Reading music is the ability to look at a line of music and being able to hear it in your head, or at the very least play it back almost immediately on the instrument of your choice. My buddy's music "reading" ability is more on the order of a kid who doesn't really read but knows a few words. He's effectively illiterate, but he's not stupid and can figure it out with difficulty.

You confusing musicianship with musicology. I'm a better musicologist than I am a musician (which is to say I understand music theory quite a bit better than I can play it), and I really think your claim that 'these concepts can only be understood and visualized by using notation' is utter bullshit.

You need to be able to read code to write it, you do not need to be able to read musical notation to play it.

Musicians are not composers. Who cares if they can run scales if they dont know how read/write music. That's like memorizing Numa numa and not speaking the language.

> The current notation has been in use for hundreds of years because it works.

It is not necessarily just because "it works". Learning conventional music notation is optimal only in the same sense as learning a conventional language is optimal - i.e. it lets you communicate. That it is convention is more important than "it works". The human brain is adaptable enough to learn strange things like touch typing on a QWERTY keyboard.

Very much this. Tradition is a terrible way to measure something's merits. And even if it did 'work', that's no reason to not try and improve it (though, I'd agree that just throwing something out there that might fragment things isn't terribly helpful either).

There are some good things about traditional sheet music notation, but that doesn't take away the fact that it's bloated with historical baggage. It's really not terribly efficient at communicating the 'essence' of music, it's just that we've all gotten used to translating piano notation into practical [insert_your_instrument_here] implementation in our heads. It's neither a useful 'source code' for other instruments to interpret, nor is it an efficient 'byte code' to describe the fundamental elements of music.

That's not to say that I have a solution to these problems of course (yet -- been working on my own project for a while that will eventually try to address this), but the issues aren't hard to spot, and they certainly seem like areas that can be optimized. And there will always be the problem of adoption, but I don't think that's even an argument we should bother raising until somebody proposes either a system that's either truly an upgrade, or a system that starts getting so much traction and starts fracturing music notation without any added benefit.

Personally, I sympathize with the aims of this project here, and think some of the ideas are clever, but considering how many things it changes, I don't think it really adds much efficiency when parsing/writing it in return. It actually strikes me as odd why they wouldn't just move to a chromatic grid system if they were getting rid of key signatures, that way they'd just eliminate accidentals all together -- or if not, then just combine them into one, since the key is a big reason for why they even have 2 designations in the first place. Which brings up the point that those symbols are all really quite complicated to be able to parse very quickly; the controversial 'note names' especially, but the accidentals really did not have much of a reason to be changed other than to just look 'fancier' -- which just makes them harder to parse, and to draw (IMO, but I have terrible hand writing).

Oh well, I wish them the best anyway. This isn't a terribly progressive field so I'm just glad to see anything happening.

> Case closed

You overvalue the importance of your analysis.

So true - whenever I see that phrase my bullshit detector jumps up into the red.

I couldn't tell; is the key-signature completely elided in favor of putting explicit sharp/flats on all the notes? If so, it's a huge fail, as I play the clarinet, and regularly have to sight-read oboe parts, which means transposing.

This means that accidentals need to be distinct from sharps/flats in the key-signature, and in particular an explicit natural on deviations from the key signature is also helpful.

> The current notation has been in use for hundreds of years because it works.

But it doesn't. And that's why tablature notation is far and away more widely used these days. Old-style musical notation is horribly designed and is gradually dying a natural death.

Tab is only used by guitar players.

Music notation is not designed to be easy to learn, it's designed to be efficient for proficient musicians. The key issue is chunking - just as children learn to read individual letters, then words, then whole phrases, the rapid sight-reader has to be able to see chords and phrases rather than individual notes. This requires a strong understanding of musical theory, to be able to anticipate what's coming next and why.

Reforms to musical notation are like the frequent attempts we see to create visual programming environments "So anyone can program!". They're based on a fundamental misunderstanding of what the hard part is. Learning the syntax of a programming language is trivial compared to learning the abstractions of programming; Likewise, learning to identify note names and durations is trivial compared to learning to think intuitively about music theory.

A point you are apparently missing is that musical notation has evolved from its origins in a reflexive manner. Music theory evolved from and with notation; I don't see why this new, progressive, more approachable form of notation would negatively impact learning and maybe even contributing to the evolution of music theory.

Just because something is new and unfamiliar does not mean it has to be scary. Just like fountain/feather pens are no longer all that widely used to write musical notation in a format that made sense for utensils, why does musical notation have to remain stuck in the past and cannot evolve?

Also, as a few other people have noted here, it's actually less expressive than standard notation.

There's no way to show phrasing(especially important for figuring out how to phrase 5/4 or 7/4 time), triplets, and I'm not sure how slur/tie notation works(especially because dotted notes are presented as tied notes... not ideal since they're technically not the same thing).

In the detailed guide it said that slurring, triplets, pretty much anything not specifically mentioned is unchanged.

Maybe it's easier to learn, but it's definitely not simpler. There's a difference [1].

Probably because I'm used to reading the tradional notation, I had a hard time deciphering theirs. There's a reason why, after centuries, the standard notation is still relevant. You basically need 3 elements to play an instrument: height (pitch), length (rhythm) and power (dynamics). And I can't imagine a better way to translate these informations than a traditional score.

But I appreciate any attempt to revisit musical notation, like the one that spawned the guitar tablatures, which is incredibly simple and easy to learn.

I'm concerned by Hummingbird's readability. Though each symbol carries multiple (and sometimes redundant) informations, I feel like there's a lot of noise. Also, drawing these symbols requires some high precision and I fear that handwritten versions will render some confusion, especially the small rest and rhythm symbols. I often scribble some music lines on a piece of paper, and I rarely have issues re-reading myself.

On a side-note, using English-based mnemonic hints ("Empty" for E, "Full" for F...) will hinder its portability across other countries, especially Latin ones where Do-Ré-Mi-Fa is more widely adopted.

[1] http://www.johndcook.com/blog/2011/11/11/simple-versus-easy/

Guitar tabs are simple and easy to learn, IF you're playing a song you already know. Most tabs I've seen just tell you which fingers to press down and which strings to hit, often excluding time signature, note duration, note intensity, rests, and other sightreading essentials.

You're absolutely right. Tabs are most useful when you already know the song.

As you say, most tabs don't include much information besides pitch and order, but can include guitar-specific ones, such as bends or slides.

It's possible though to write good tabs, such as the ones available in Guitar Pro (http://www.guitaring.info/uploads/software/Guitar%20Pro/Guit...).

Tabs' major appeal is that there's no learning curve: what you read is almost a physical representation of what you play. And it only requires a text editor to write, and can easily be published and shared on a website.

The best tabs are printed under a staff that contains the missing rhythm information (along with the actual pitches, etc., of course). Not exactly concise, but very info-rich.

This is technically true, but is approaching notation from a perspective that everything needs to be specified (I’d call this a “classical” perspective, but even in classical music, this was not traditionally true).

In e.g. Jazz jam sessions, good sight readers can definitely figure out what to play from lead sheets consisting of chord symbols + melody, and in terms of information, those are equivalent to guitar tabs + melody.

Admittedly, there are also books only giving tabs/chord symbols + lyrics (or sometimes not even those), and for this form, I agree that your assessment is valid.

There's also a certain difference between jamming with a few other players, and getting a whole concert band of 40+ players to sound good together.

My understanding is that in the big band era, you got fairly large bands (though not as large as 40 members) to play tightly together without undue precision in notation (and often without all members being good readers of sheet music).

This is exactly the problem we're solving with Soundslice: http://www.soundslice.com/

It syncs tabs with real audio recordings so that you get the usability of tab plus the rhythm and phrasing cues from the real recording.

"You basically need 3 elements to play an instrument: height (pitch), length (rhythm) and power (dynamics). And I can't imagine a better way to translate these informations than a traditional score."

(1) Hummingbird translates length in a trivially superior way to standard notation. No contest imho.

(2) I would say it does a somewhat better job on pitch as well. Both have huge shortcomings so I can well imagine far better systems.

Dynamics is a push so I'll say Hummingbird is an improvement in translation. I don't think it's a big enough improvement to climb the mountain of inertial standard notation has in it's favor. I'm typing this on Qwerty keyboard too, yep.

Hummingbird is fatally flawed for anything beyond beginner level stuff. It's not going to deal well with any sort of complicated rhythm. I'd love to see you accurately notate something at all complex, like say this, a snippet from a Beethoven piano sonata:


While it might have some minimal value for beginners, you might as well teach them real notation.

Let me present a metaphore.

You're in charge of the foreign language curricula in, say, Japan. You can offer either Esperanto or English but not both. Esperanto is easier to learn, but English is infinitely more useful.

The example picture you provide is a great example of how horribly, horribly flawed the current notation is for annotating rhythm.

I mean, many people have gotten used to that language, but there obviously can be a far simpler way to represent the same timing information than the first line of that pic.

And the example you give is not THAT complex - try to represent any interesting polyrhythm drum pattern in traditional notation, now that will be an excercise in futility.

The hummingbird is not a good enough solution - it's a step in the right direction and another step sideways; but we need to go much further to improve both ease of reading and ease of learning music.

To use a programming analogy, the current music notation is like if everybody had to learn programming through COBOL - it works, it's usable and widely used... but still, using a better, cleaner language for the same concepts would obviously bring benefits to all new users; even if it's of no benefit to all the current experts who already know the current notation by heart.

I think you misread me. I clearly said Hummingbird was not better enough to triumph. I was taking apart the idea that standard notation is so awesome that it can't imaginably be improved on.

Standard notation, QWERTY, English, Facebook, etc. are all flawed in many ways but to replace them you can't just be 1% or 5% better; you need to be an order of magnitude better. No one cares about learning Dvorak for 5 more WPM but if people could double their WPM the world would switch.

I'm not convinced length is translated in a superior manner. Any spacing based rhythmic notation quickly gets ridiculous when you combine long and short notes. Space your demi-semi-quavers out far enough to read them (especially if you have to annotate them with hummingbird accidentals) then see how far apart the minims and semibreves are. Traditional notation tends to be set so the rhythm is suggested by the spacing but isn't prescribed.

What level musician would you rate yourself?

>Maybe it's easier to learn, but it's definitely not simpler. Probably because I'm used to reading the tradional notation, I had a hard time deciphering theirs.

If you just found out about this from HN, then you didn't have ANY time to get familiar with it.

Plus, your knowledge of traditional notation put you already at a disadvantage. You could only compare it with a control group study, or after you have spend as much time in this, as it took you to be proficient in standard notation (e.g 1-2 months at least).

I've actually just spent my afternoon writing out some music for the first time in about 15 years, so this is quite interesting.

However I must say that I just don't get it. Every example I look at appears significantly more complex than the standard notation, and harder to discern at a smaller size. One place I can see it really struggling is on copies. Music tutors spend a lot of their time copying music sheets, and I suspect this would be quite difficult to read on a low quality reproduction.

Standard notation has survived for hundreds of years. I'll be the first to admit it's not exactly easy to get your head around to begin with, but once you understand the rules it becomes apparent as to why it is the way it is.

The embedded visual cue to the name of the note, plus the proportional sizing, seem really nice to me. The sharping, flatting, and lack of ascenders and descenders I'm not so sure about.

You also lose the "wall of black notes" warning you of deadly fast notes up ahead. ;)

To me, the visual cue of the name of the note is idiotic. VERY idiotic.

He choose to use the C, D, E... system that is not the norm (the norm, maybe not in US I guess, is Do, Re, Mi...)

And then create graphical representation of words starting with those letters.

Except this works only in english.

How a portuguese speaker for example would associate the dot thing with D or Re? It looks like neither, at most it looks like a dot (that in portuguese is "ponto", thus starts with a P)

Or the above and below? Below in portuguese is "abaixo", thus starting with a A, so you have to teach someone that A actually means B.

To me this new notation might make sense in english (maybe), but in other languages is even more arbitrary and silly (and tedious to hand-write)

>He choose to use the C, D, E... system that is not the norm (the norm, maybe not in US I guess, is Do, Re, Mi...)

A musician is usually able to use both. The A/B/C/D/E/F/G is used even in European notation, for chords and stuff.

>How a portuguese speaker for example would associate the dot thing with D or Re? It looks like neither, at most it looks like a dot (that in portuguese is "ponto", thus starts with a P)

He would either have to learn 7 words in English, that almost everybody in the planet under 30 already knows, or just learn the visual shapes, which are distinct and take about 10 minutes to memorize. It's not as if "Do, Re, Mi" means anything in Portuguese either.

The average US musician knows the C/D/E, and possibly moveable Do/Re/Mi (where Do is the tonic of whatever major key you're singing in).

The fixed Do system isn't even much covered in basic college-level music theory in the US (to my recollection); I doubt most performers will know much about it.

About asking the Portuguese musician to learn 7 English words, and switch from fixed Do to the C/D/E system... well, the problem is not that it's hugely difficult, but that it will seem like a foolish choice to music teachers. Who would teach this new system?

The benefits of the new system have to be huge and obvious if it's going to gain any ground, because the existing system is everywhere.

Tell a teacher that "here's a new system! Only a miniscule fraction of extant sheet music is available for you and your students, you'll have to re-write all of your teaching materials, and you'll have to force your students to learn the American C/D/E system with English-language-only mneumonics!"

It doesn't sound like a winning argument to me.

i took undergrad music theory classes at two american universities. they both used fixed-do solfege.

or just learn the visual shapes

Which is different from learning position on the staff, how?

Did you read TFA?

It's different in that the visual shapes stay constant in all "clefs" as well as up and down the pentagram.

So a D below middle C looks exactly the same like the D above middle C -- something which is not true for the regular notation, where you have to count the pentagram lines the note is in, or how many lines below or inside the pentagram.

So the new system retains the position-on-stuff and ADDS another visual cue for the same information (the shape of the note).

I did read TFA and... pentagram? Do you mean the staff/lines?

Adding more visual cues isn't necessarily a good thing. 'Counting lines' is what you get when you're a neophyte at reading music, and is chaff for the more experienced. And when you're playing a complex piece of music, you want to minimise visual clutter.

>I did read TFA and... pentagram? Do you mean the staff/lines?

Hah, yes, sorry. In my language it's called "pentagram" (which means "five lines" literally). For some reason I had the idea it was the same in english. Well, after all, other words like "harmony" and even "music" were borrowed in English as is ;-)

> C, D, E... system that is not the norm

For those studying music theory in the English-speaking world, A-G is the norm. (German has H as well.) That being said, I've only ever met musicians from the commonwealth... I will not comment on other languages and whether or not they use solfège since I've no experience in that area.

This was my thought as well. If you're going to create a symbol association language, why not just use the original symbols? You're just adding an extra layer of learning otherwise.

If it's because the symbols (ABCDEFG) aren't universally recognized, then you shouldn't be basing your language off of them in the first place.

These symbols are fussier than necessary, because they are trying to be cute about both the mapping to the alphabetic scale and slavishly sticking with circles.

What is required is a sequence of length seven (for convenience) that allows sharp/flat prefixes and time length suffixes. Circles are merely one option.

If written by hand, there will be ambiguities between C, Dot, Empty, and Full. I think a graphic design artist had too much fun with this project...

So, that's just the mnemonic for remembering the names, which is only marginally interesting, and which I already called out for being anglo-centric in another comment. What I like is how easy it makes to spot, say, an octave, at a glance.

The embedded name of the note seems to be the only thing this has going for it, and only in high-quality prints. I feel like the stems help me to distinguish where notes start and end, and the sharps and flats seem a little... small.

Also, properly typeset standard notation is proportional anyway.

Additionally beam groups convey a huge amount of information-- there's a reason, for example, that 6/8 is typically beamed in 2 groups of 3, whereas 3/4 is grouped in 3 groups of 2. Understanding the micro- and macro-pulse relationship make sight-reading much easier, in addition to subtly informing performance.

I've often thought that if there's space for something to be reformed in musical notation, it's the fact that different wind instruments are notated in different keys[1]. I realize there are historical reasons, but it just seems like such an artificial barrier between musicians in a modern band or orchestra.


It's a standard part of musical training to be able to read a part written in either concert pitch or in your instrument's pitch, or even to be able to transpose on sight into any key. It's not easy but learning to do so pays off when you're on a gig and the singer insists on playing Lush Life in B natural.

We can debate how large an obstacle it is, or whether learning to overcome it is valuable, but clearly it is there, even if it's just an annoyance. I just think it would be really nice for (e.g.) a clarinetist to be able to sub in on an alto sax part on sight without having to go the extra mental work of "Ah, right, up a perfect fourth --"

This is mildly annoying for composers, but it doesn't really matter for the players. They play the notes they see on the page. I suppose if they have perfect pitch it might be a bit jarring.

I play mainly clarinet (which is written Bb transposed), and the transposition thing is not a problem at all. It's the relative intervals that matter anyways. I don't feel it creates a barrier when communicating with others in the orchestra (we talk about concert pitch anyways), and not having to count five extra staff lines makes up for the small inconvenience.

This makes it much it much easier when switching between different instruments of the same family - a clarinetist (when playing any clarinet part) associates one note on the staff with one fingering. If all instruments were in C, the player would need to associate the same dot with multiple different fingerings, depending on the family member being played.

Initial impressions:

* The pitch shapes encoded in the notation are fun and probably help learning, but I think they'd be distracting past a certain point. They don't convey any extra information that the staff doesn't, and if they didn't match it's one more thing to trip up on.

* The sharp and flat signs are way too subtle, compared to the traditional accidentals. I'm not going to see those when I'm sight-reading.

* This is also true for the eighth notes and shorter. Those flags are tiny!

* No key signatures?!!

I love that someone's playing with ideas for notation, but this notation is worse for experienced musicians because it makes important information harder to see at a glance. (Yes, sure, I'd get used to new shapes, but the distinguishing marks on the page are smaller?!!)

I can't comment on whether it'd be easier to learn, but this is notation you're going to be using much longer than you're going to be learning it. Optimize for long-term usefulness.

Learning musical notation is not the hard part of playing music; getting your instrument to make the right sounds is. Changing the notation doesn't make that any easier.

Also, if you've only learned this new notation, you'll be unable to read any of the 99.99999% of music that has been published in the conventional notation over the last few hundred years. It would be pretty limiting, somewhat like learning to speak a language that's only spoken on a small island in the Arctic Ocean.

It depends on the instrument. For melody / monophone instruments (woodwinds, brass and to a lesser extent strings) musical notation is easy. For piano, it's easy once you master the bass clef. For guitar, it just never gets easy because the score has to be full of fingering information to make it even remotely clear where on the fretboard your fingers should go.

Musical notation can definitely be improved upon.

the only problem with this to me is that this new notation simply seems to add redundancy to the score, repeating information which is already there,but not adding much new.

One advantage traditional notation has over this is that the modifiers are a lot larger and more visible. For sight reading, notation needs to be easily scannable and irregularities (like sharps and flats) need to be highly visible. Connecting the beams on eighth and sixteen notes also serves to group the notes according to beat, and that makes parsing a measure much easier (also easier to skip ahead when you mess up). Neat idea though.

I applaud your efforts to update something so anachronistically designed. I think it looks very cool, I like the duration symbols.

On the down side, I am very surprised you decided to redo music notation, but keep to a 7 tone diatonic graph structure.

One of the most counter-intuitive things about sheet music is it assumes a 7 note scale, making the graph inaccurate: the space between B and C are displayed visually as the same as the space between C and D when that is not the case in any physical realm, and it makes transposition harder than it needs to be.

In my view, the 7 note scale assumption is a horrible, frustrating, legacy, like having to learn DOS before being allowed to operate an iPhone.

7 note scales are a misleading assumption not only for any "modern classical" composers, but also for any blues, rock n roll, north indian classical music ...

There may not be an easy solution to address all this, but I am curious about the reasoning behind your approach.

I once struggled trying to create a notation for tap dancing, so I applaud the attempt and hate to shoot it down, but... My eyesight is pretty poor, even with glasses, and I find this difficult to read. I find traditional sharps and flats easier to scan than the little note prefixes. They're bigger, and look substantially different. Same wrt some of the smaller rhythmic values.

Also, in the first measure, beat 1 of the base clef, it's hard for me to tell whether the bar sign for sixteenth notes applies to the very first note. I have to try to gauge the vertical alignment without the help of the traditional vertical bars extending from each note.

Finally, the mnemonic symbols for each pitch seem superflous to me. Once you've learned the staff they just amount to irrelevant visual adornment, IMHO.

"Easy to learn" might not equate to for "easy to use by trained musicians."

I definitely agree about readability - I occasionally struggle to read music when there's more than about 3 people to a stand, and I had trouble reading this notation on a tablet, fairly close to my face.

I'm fascinated by this kind of cultural technology. I think we ought to be experimenting with notation of all kinds—numerals, alphabets, languages, measurement systems, calendars and so on. We need to make it easier to learn new systems of thought so that we can actually adopt better ones!

This comment reminded me of an episode of Radiolab that I recently listened to. It's about a man named Charles Bliss who attempted to create a set of symbols that would let us think and communicate in a pure way across cultures and languages.

If you want to give it a listen: http://www.radiolab.org/2012/dec/17/man-became-bliss/

Interesting, thanks.

I tend to be a little skeptical of attempts to make "pure" forms of communication, because I think there are bound to be trade-offs, as in any engineering endeavour. I'd rather see schemes that are optimized for a particular purpose. What would a language oriented toward scientific and technical communication be like? How would currencies designed for ecommerce work? How would a calendar for global organizations work?

I like the idea of working to make music notation better -- it's certainly not flawless -- but I'd strongly discourage music teachers from using this system with any actual music students.

I'd much rather see methods of improving the readability of standard notation, by adding colors, interactivity, or anything along those lines -- but not replacing standard notations (for accidentals, note flags, etc.) without really, really good reason. The further your system departs from standard notation, the less valuable it is automatically, so for a departure even as far as hummingbird (which is still obviously related to standard notation), the value it adds already needs to be huge just to break even....

Think about the choice you're making for your student -- instead of getting started learning standard notation, you're starting them down another path of reading music; the moment they leave your studio or classroom and walk into a music shop (or even another music class), they will be completely lost.

There's a lot of sheet music freely available online -- whoops, not for your students, though.

There are also a ton of apps, interactive sites, online tutorials, software, etc. that can help music students master all aspects of music performance, analysis, and even composition. Well, some students. Not yours.

I know this sounds harsh, but it's a bit like attempts to fix the English language. Everyone knows it -- English is horribly irregular; every rule of thumb for spelling has a million exceptions; there seem to be more irregular verbs than regular ones; there are obsolete tenses only used in some common phrases and nowhere else. But if we fixed the problems -- even if we just regularized spelling and nothing else -- the first generation of students using the new system would be a bubble in a world that used "old" English. If we successfully rode out the change, after a century or so all new documents produced would be in new English... but anyone interested in reading anything before the switch would be at the mercy of automatic translators.

It sounds like a dystopian novel where an autocratic government wants to cut their population off from all knowledge of history.

I like the idea that someone gave alternate musical notations a shot.

I looked at the sample pieces and liked that they had simple and complex pieces to look at. As I went through the simpler pieces the notation was easy and fun to pick up. When I looked through the more difficult pieces I felt like I spent more time analyzing each symbol to figure out what exactly it was saying. They were "overloaded" in a sense to me.

I feel that if I'm sight-reading music (or haven't practiced it much which is the more likely case) that this would fatigue me having to parse so many pieces of information for a note. It made me realize that one of the things I appreciate about standard notation is that you have a defined set of symbols with minimal overloading and that you're marking it ("annotating?") it up to make changes to it.

It's an interesting experiment. I just don't think I could get used to it for complex pieces.

I think the visual reinforcement of note names is a poor idea. When I play piano, I play best when my brain and hands are reading the music spatially. When I start thinking of note names I become much more clumsy and slow, because it's interrupting my spatial thinking. I always recommend that people think about intervals as opposed to note names when learning a piece of music. It encourages various good habits, like being able to identify overarching patterns in the music and play in different keys easily. Intervals also correspond more closely to how your hands have to move. Because of all this I don't think it's helpful in the long run to have the visual reinforcement of each note name. It might be easier for children or beginners at first, but in the end it may be a crutch that prevents the student from "seeing" the music..!

In my opinion while this notation may be easier to learn (I have my doubts about this, but as someone who is familiar with standard notation I'm going to reserve judgement on this), the notation misses a lot of nuances and details that would be necessary to play more difficult pieces.

Some of thing I feel like it fails to capture:

1.key changes- eliminating key signatures makes key changes less obvious which is important to realize as it is important as to how you play the piece.

2. Phrasing- the visual shape of the notation seems very vertical to me, which works for some pieces but would drastically change the way I play certain pieces.

As well the following don't have any example and their current notation would likely conflict with the proposed notation: ornaments, tone of a note(staccato, marcato, slurs, formatta etc), dynamics

I'm intrigued by a new kind of notation for music, however I feel like while this maybe more approachable it won't work for high level performers, and having to learn a new notation if you reach a high enough level kind of renders the notation kind of useless to learn.

Here, I propose yet another take on music notation:


Instead of difficult to read/write symbols which map to letters, which map to pitches, why not just use the letters themselves?

I would totally love to just learn with letters written on the staff. And over time, maybe my sheet music app could randomly replace letters with black filled circles, and then eventually with standard music notation.

And the most natural notation would simply be "piano roll notation" which when oriented vertically looks like this:


The drawback is that this notation is less compact. But, it's the best notation for teaching beginners.

I'm trying to figure out how Hummingbird notation is any better than "piano roll" notation or simply putting letters on the staff. It's definitely a cool idea (infovis-wise), but I think it needs to be user tested: standard vs. hummingbird vs. letters-on-staff vs. "piano roll"...

I've come across Synthesia before and it's well-thought learning tool. But its major drawback is that it's only aimed at piano players.

I think Hummingbird's goal is to become an alternate standard of global music notation, based on the fact that it resembles the traditional one (and actually uses it as a base). But Hummingbird also has its flaws (that I mentioned in another comment).

jesus this would have been so helpful back when i was playing music at a kid. i'm sure after a while you can swap out the letters with notes without missing a beat, just like you can remove the letters from a keyboard after learning to touch type. but, along the way you never actually are forced to "learn" the positions. (which is incredibly challenging when you are trying to learn to play at the same time.)

honestly this makes me want to re-try to learn piano with this type of notation. perhaps even a notation that would let me focus more on the piano and less on deciphering the notation.

brilliant in its simplicity, thanks for sharing.

Yup. If I was teaching a 6 year old to play piano, I wouldn't start by teaching them how to read standard notation.

The key is to make it fun and easy in the beginning, so that piano players don't give up! Then, make it harder once they are motivated to learn.... and you can show them why standard notation is more expressive/compact/better than the letter notation they started out with.

This already exists for beginners. See http://s.ecrater.com/stores/159585/51208eb2d0e25_159585b.jpg

Yup, I think this would be the second step, once the beginner is used to the bare letters on the staff. This introduces extra visual noise, but begins to teach the player what standard notation looks like.

With software, the view can adapt over time as the player becomes more comfortable reading notation. If the sheet music was displayed on a tablet, eventually the little letters could disappear.

This could be a nice set of "training wheels" for beginner pianists.

Maybe I come from a different musical background as the authors of this, but none of this make any sense to me.

"Long notes are longer; sharps point up and flats down." - Having notes take up more space is about the worst possible thing in the world for me; as a pit musician, the last thing I need is more wasted space on a page, giving me more page turns to deal with while I'm changing instruments and key signatures.

"...rhythms have the same spacing" - I don't know what this means. Rhythms don't have spacing. The spaces between the notes has nothing to do with the music that's played.

Then there's the fact that all current musicians would have to re-learn how to read music. Perhaps someone can tell me what's drastically broken about the current system? I'm not saying it's perfect - I don't believe any system is perfect. But it's worked pretty well for the last few hundred years.

I would expect page turns will be going away. You can already use your iPad as a music stand, though it's kind of small.

If you're using a real device, page turns can be automated or eased. The device can listen to what you're playing (tempo shouldn't be too hard to pick out) and turn pages appropriately. Or, you could use a foot pedal or what have you - perhaps your phone.

Apps for this sort of thing already exist, though I don't know how good they are.

I believe that argument is that the current system is difficult to learn. But I personally feel this notation gives up way to much in order to try and be easier to learn. (which I still don't think is not a good enough reason to create a new notation since if you are putting the time and effort to learn how to play an instrument I think you are likely ok with taking some time to learn the notation)

In handwriting it will be much more illegible than conventional notation. Also, small details are very bad for nearsighted people.

Don't get me wrong. I am always fascinated by new approaches for doing things and new musical notation is a fun idea. But not in this case. Practically speaking, they don't improve anything at all.

Conventional notation is not broken or something.. Imho, it is actually looking pretty great typographically, and works fine in practice.

They could make new notation that works better on computers, to be used in music software, such that it is easy for typing using a keyboard. That could be a real improvement.

Interesting. As a player of piano, woodwinds, and others I've gotten used to different systems of scribbling over the staff to convey something that's not metadata but not primary info. Well-tempered Klavier is a good bad example, there's all kinds of scribblings about what Bach intended, including argumets about incidentals (is a note flatted or not?) (and what's the umbrella term for trills, grace notes, flourishes like that?). It's actually much harder for wind and strings, where infinite pitch/tonality /attack/decay combinations are possible, e.g. lipping up or down on a single reed, squeaking, honking, sibillant, and I'm pretty sure there's no way to write down the loops i get on fretless guitar, bass and cello.

Also I've been trying to get used to Don Ellis quarter tone system, and work thru haskell school of music (fantastic book, for anybody interested not just in notations, but production, composition and capture(A/D conversion/DSP etc. Also shoudl read books by Gould and Read someday:




I lost it at making a symbol to represent a note letter -- a terrible idea. Ex. D for Dot.

One thing to understand about music is that the letters are meaningless, as learning by letters restricts you to playing in a certain fashion. For some reason letters got introduced, I don't know why, but all that matters is the distance between notes.

Not learning letters first allowed me to easily transpose into any key since the LETTER DIDN'T MATTER.

Cool idea, but has some serious limitations in real music, the obvious one being transposition.

I'd rather learn byzantine notation than this 'Esperanto' of music notation. (example: http://stanthonysmonastery.org/music/KarasSample-with%20head... )


Do you know of any more resources about this system? One of my side hobbies is learning more about alternative representation systems for languages, math, music etc. Finding resources on alternative music systems is kinda hard in my experience.

The following page contains a link to a beginning Byzantine Chant textbook that explains the system: http://www.stanthonysmonastery.org/music/LearnByz.htm

The pdf is in Greek, but has been annotated with English notes throughout. It also contains embedded mp3 resources so you can hear what the notation is showing.

More details can be found: http://www.byzantinechant.org/notation.html

Including video tutorials: http://www.byzantinechant.org/tutorials.html

Thanks so much!

Hummingbird notation: lots of little dots, vaguely grouped together

Traditional notation: 16th notes are tied together, with the tying actually serving to reinforce the duration of the individual notes while indicating that they should be played as a group.

I applaud the attempt, clearly the creator(s) put in a lot of time and effort, but I don't see most of the changes as more intuitive at all. It's also attempting to solve a problem that doesn't really exist, the current notation is not difficult to learn - and anyone who thinks learning to read music is the tough part about learning an instrument is in for a real surprise once they get much past "Mary had a little lamb."

The discussion here reminds me of the longstanding debate over editors like vim/emacs vs (whatever you want to call the other editors). Or semicolons vs indentation, to make a slightly more "reading" oriented comparison.

There's obviously a difficult learning curve for many people, but others who are already proficient argue that the rewards of learning standard music notation are worth an extended effort, if that's what it takes.

The thing I've noticed most about sheet music as I'm learning it is that it's compressed. It uses different symbols and techniques to say the same thing in a smaller space, and reuses space and symbols more efficiently by applying modifier symbols at the beginning of the staff and elsewhere. It allows notes on lines and spaces instead of just spaces or just lines. Tighter, smaller, more on the page. Changes in pitch can also be indicated by modifier codes next to the notes, sharp and flat, allowing further combinations with only 2 more symbols.

So I have to learn to decompress the information at the same time I'm interpreting it. Tricky! I do stop and wonder if this compression algorithm is the right fit for humans. I don't buy the argument that continuity with the volume of existing sheet music is a good reason to never develop an alternative. But any alternative needs to be much better on some metric that outweighs continuity. Otherwise we should just keep hacking the learning process with color coding and mnemonics and whatnot.

It's funny, I was talking to a friend about my struggle to learn sheet music and music theory, and he said "It seems hard at first but you'll start to get it pretty soon." And I said "Yeah, it's sort of like math in that way." To which he replied "oh, I don't know about that, I can't do math. I've never been good at it."

The difference between "editor wars" and a new musical notation is that choosing a particular editor does not limit you in reading or writing things that people have already done. If I learn to read music via new notation, that's great, but I'll still have to learn traditional notation if I want to play a piece by Chopin (assuming someone hasn't "translated" it to the new notation). If I use vim, I can still read a program written in emacs. Editing it may be a different process, but consuming it isn't.

That's true, but the comparison I'm getting at is the learning curve. The process of consuming an algorithm or procedure is more of a tabs-vs-semicolons thing, a language war.

I mention editors because the arguments about learning one often have to do with the trade-off between expert efficiency and expressiveness and the painful learning curve (for many). And because that's honestly what the discussion reminds me of, whether or not it's a perfect match :)

Why can't notation be shown "on the fly"?

I mean, any notation editor knows the pitches/timing/etc encoded in the score, and it should be able to trivially show it to the musician in a customized way - transposed to the instrument, if neccessary; with or without fingering information where applicable; and in custom/wierd notations like this one.

Of course, printed/photocopied scores can't do that, but we're not in stone&paper age anymore and can fix things to improve functionality.

I immediately lost all context of key and mode trying to parse the examples. Also, the grouping symbol looks like an arrow pointing backwards in time.

Like many have mentioned, this seems to add a lot of noise to notation. This isn't really demonstrated in the pieces they have on the site. I'd like to see what it looks like with something a bit more complex like a Bach Fugue. I feel like the rhythm notations, in particular, would become more difficult to parse as the rhythms become more complex.

Also, how do you notate tuplets?

This isn't a new notation, it's just a new font on the old notation.

>Pitch symbols are obvious, barely requiring memorization. There’s no need to count lines, and treble and bass clefs are the same.

First, why not just put A, B, etc inside the circle? That's easier than these hanging-chad symbols. But mainly, do musicians read music by translating the symbol (position) to a letter and then the letter to a fingering (for example)? I expect instead, they translate directly from the (relative) symbol (positions) to a fingering. That's what I did after just a few days while self-teaching piano.

>There are multiple cues to the same information. Everything has both a symbol and spatial element, for all kinds of thinkers.

Generally, I think this kind of redundancy is a bad thing. While programming computers, we usually agree that there should be one and only one clear way to accomplish base tasks in a language. When storing data, consistency is essential.

> We usually agree that there should be one and only one clear way to accomplish base tasks in a language.

You are obviously not a Ruby programmer. ;-)

I expect all languages have redundancies. But do you tout them as a benefit?

All string functions contain options in the method name and as arguments, so everyone is happy!

  s.match_i("foo", insensitive=True)
  s.split_r(",", direction=-1)
  s.replace_ig("dog", "cat", insensitive=True, global=True)


They should look at adding rendering support to http://www.lilypond.org, which is basically TeX for music rendering.

If they had that, all the music in http://www.mutopiaproject.org would be rendered in their format for free.

" basically TeX for music rendering."

Tex for music rendering does exist: musixtex http://icking-music-archive.org/software/musixtex/musixdoc.p...

I've long held a fascination with the myriad ways humans have invented for writing things down. Sites like Omniglot [1] get a semi-regular visit from me. In a side quest, I spent a little while looking for alternative systems for representing math and music and came up unbelievably short. There are remarkably few alternate systems for them.

Wondering why that was, for fun I decided to try and come up with some different approaches to representing math and music to see if I could better understand why there's so little variety.

It turns out that it's incredibly hard. With math (at least basic arithmetic and algebra), after you mess around with the basic symbols, there's not really many other places you can go with how actual equations are structured without losing lots of the easy-to-use mechanical features that modern notation supports. In a few ideas I essentially recreated a parse tree, which made reducing the sides of the equation relatively simple, but moving things across the equals turned into a nightmare.

With music the obvious alternatives fall into a couple categories:a system for each instrument, something that can succinctly capture the expressive bits of a given instrument (fingering, bowing, vibrato, etc.) and throw out bits that don't work on that instrument (vibrato on a piano, pizzicato on a wind instrument etc.), but it gets impractical stupidly fast. The other alternative is a universal system like we tend to use today, but you end up with all kinds of space wasting piano roll-a-likes or hard to read while playing encodings like A2--B#2--C2--

This will probably not replace current notation, but it represents quite a bit of creativity and at least a noble attempt at doing something which most of the people on earth haven't managed to do (almost all musical traditions in history essentially exist in a state of verbal transfer). I think it's cool and has lots of great ideas.

[1] - http://www.omniglot.com/writing/alphabets.htm

As someone who is currently unable to read music, but wants to learn, this doesn't seem any more inviting than traditional notation. I'm further discouraged by the fact that even if I were able to somehow learn this notation faster, I wouldn't be able to read anything that hasn't been translated into this notation, which would really limit the practicality of going to all that trouble in the first place. My first impression is that learning this notation would be duping me into becoming dependent on a single source for all my sheet music.

Maybe it would be helpful if instead of the front page demoing the most complicated excerpt you can conjure up, it demonstrated some simpler (well known?) music in both your notation and traditional notation.

I would draw a parallel between this notation and guitar tablature. Unfortunately, all professional guitarists (and especially classical guitarists) that I know prefer traditional notation to tab. I personally find tab terribly hard to parse (disclaimer: I'm a classically trained musician). I think that the proponents of Hummingbird will need to somehow address and overcome this idea that "real" musicians only read "real" music. Overcoming an entrenched standard, especially one with hundreds of years of history is a terribly hard journey, even if you're only aiming at a tiny fraction of users.

I also have quite a few quibbles with the specifics and usability of this notation, but this is not the place for that.

How do I write triplets? Percussion? Tremolo? Why are all of the "critics" 25 or younger? How do I write parts with multiple voices?

I also wondered about triplets, quintuplets, etc. Also what if I want a dotted quarter note, how is that notated? Extra pitch information for microtones?

This is neat, but the mnemonics are pretty anglo-centric.

considering I grew up with do-re-mi-fa-sol-la-si I find the mnemonics impossible. But also, I don't understand the point: why not just write "A-B-C-D" rather than strange half filled circles?

That would make your do-re-mi problem worse, no? They're designed to be compact and visually distinct, and such mnemonics are really only useful at the very beginning anyway. I like the design, it's just a case of probably unconscious "English privilege".

sure it would, but if one is trying to make a breaking change it's worth going all the way and giving up medieval names too, I think.

If the leading tone is "si", what do you call sharped sol?

"sol diesis", but I believe you refer to reading it while doing solfege with chromatic alterations, where you would read Sol# as "Si", right?

The answer is: I have no clue, I think we do not actually account for flat and sharp when doing solfege.

Or better put: "si" was changed to "ti" in UK to be able to read chromatic alterations without ambiguity. Most of the latin and slavic world ignored this, AFAICT.

But full discosure: I know very little about music, music theory or music history, I just have vague memories of reading about it.

Doesn't look easier to me. More importantly this will only cause whoever teaches it as a first language to some child will make it that much harder once they are around other musicians or in a school system.


There are multiple cues to the same information. Everything has both a symbol and spatial element, for all kinds of thinkers.

That doesn't strike me as a positive, really. I look at it thinking I must be missing something, especially witht he above, below and C symbols.

Quite a lot of musicians have their own 'private' notation and I'm no exception, I have shorthand for ideas that go in my notebook and which would make little sense to anyone else. Traditional notation is pretty awful in a lot of ways, but I'm afraid I don't see any real improvements here - it's still uninformative rhythmically, and still promotes chromaticism over scale degree.

I really can't see what the benefit is. Although I don't like traditional notation and can't sight read, at least it's consistent and reasonably easy to learn. This is no worse, but it's not so much better that it's going to cause any significant number of people to switch. People who play acoustic instruments that need to sight read will still need to be able to do so with traditional notation so I am having trouble seeing how this will get traction in that market. Guitarists already have tab as an alternative, drummers have drum grids, and electronic musicians use piano rolls or things like hex maps, to the extent that they use written notation at all.

How am I supposed to represent multiple voices in this? Something like a 4-part fugue would be much more difficult to represent in this notation than in traditional notation.

Holy cow. About 20 years ago i wrote a similar proposal for "improving" notation. My primary aim was for easier sight reading. It didn't go as far as yours. But three key ideas where:

* Add 1 line before bass clef (E) and one line above treble clef (A). That way, both treble and bass clefs become the same (EGBDFA). There's really no reason for your brain which just learned that the 3rd line on the treble clef is B, to suddenly be D on the bass clef... WTH?! This way, the 3rd line is /always/ B!

* Make note-head shapes include its flat/sharpness. Again, there's no good for your brain to have to remember that F is reaaally F#, or be scared of that key sig with 6 flats. Instead let's say a round note is natural, square is sharp, triangle is flat (or some such). Suddenly the note-head /shape/ carries all the info needed for immediate recognition and no on the fly translation is necessary.

* (less radical idea) A third one is writing piano music vertically, like japanese. Why? Because the piano keys are vertical while the music is horizontal. If you turn the music 90 deg clockwise, from left to right the keys match up with the notes from low to high.

Of course if you're a naturally good sight reader, none of this stuff matters, but my gut feel says these will be improvements for new comers.

Very interesting. I've been trying to learn to read music so I gave it a shot with some of the example songs. I think I'm right in their target audience with where I'm at.

I liked the symbols as reminders (I still have to count All-cows-eat-grass or f-a-c-e sometimes), when I was unsure of a note it was quicker to think of the reminder, but still the position was primary, unlike other versions where the notes are labeled with their letter name (which makes it impossible for me to actually pay attention to the notes) it was just a hint.

But after I played through the song a few times, it felt a little "busy", it was more work to filter out the symbols. It also seemed a lot harder to count the rhythms with this notation (maybe this is just because I have more practice with traditional notation). Finally, I feel like this gives less of a sense of the overall flow of the song -- the ties and the phrasings.

As a sort-of-beginner, my reaction to this was that it didn't simplify things much, so I wouldn't want to invest the time in learning this and not being able to read the vast amount of preexisting music out there. On the other hand, I wish I had some of these cues when I was first starting.

I'd love to see this not as a replacement to traditional sheet music, but as a standard was to annotate traditional music for beginners. The symbols as hints in particular would have been great. I think this is a great idea but trying to do too much, just a small tweak to any language system is a huge undertaking (and can have huge results). But I'm glad someone is trying. Maybe it's because I'm older now, but learning to read music has felt much harder and more frustrating than learning to program computers.

Nobody has mentioned this yet, but this is just one of dozens of past attempts at creating a new music notation. Even if you think that Hummingbird is worse than traditional notation, don’t be so quick to dismiss the idea of alternative music notations in general. Each author’s attempt at a new notation changes different things about traditional notation. Read more about some other notations here:

http://musicnotation.org/systems/ – a list of various notations

http://twinnote.org/ – TwinNote, a notation. One trait it has in common with many other notations is a chromatic staff: the distance between notes on paper always exactly corresponds to their difference in pitch, so accidentals are not allowed or necessary. TwinNote comes with template files for the open-source music typesetting program LilyPond, so music written in LilyPond format can automatically be printed in TwinNote.

https://www.google.com/search?q=alternative+music+notation – more about music notations through Google

I like the idea a lot. Especially the embedding of note-name in the note itself as an additional measure to the spacial line-distance, as I'm not terribly great at dealing with interpreting the spacial gaps (especially when the note is more than 3 or so lines away).

But I find the symbols confusing in their own way, and not because of the language barrier others have mentioned (they're just a mnemonic, the words cease to matter after a certain point and other mnemonics can probably be devised for other languages). I find them confusing because I think there should be a progression to them. And the progression of A and B I find doubly confusing, because in music the notes go up, but in this the darkened part goes down. I look at a B and intuitively think it's an A and vice-versa.

I think I'd prefer a system where they look like clock hands, more or less. A is 10:30, B is 1:30, C is 4:30, D is 7:30, E is a slash going through 10:30 and 4:30, and F is a slash going through 1:30 and 7:30. I think I'd find that more intuitive. (note: all diagonal so they don't conflict with the spacial lines)

One thing normal notation is great for is hand writing score during composition. The filled in quavers become a diagonal line (running bottom left to top right). Whole notes a little circle. And the rhythms and slurs are easy to draw. In fact I think this is why classical notation is how it is. In medieval times the scores would have been hand written.

I would hate to handwrite with this during composition.

The whole aesthetic seems wrong. The perfect circles and even lines seriously clash with the naturally drawn clefs. In the example music, they adopt a new font for the time signature to blend in better, I guess, but the clefs are still out of place. There's an appealing naturalness to traditional notation that this system abandons and ends up looking like a schematic or alien code. Should really work the style into a more natural feel. There's potential here, but it's falling a bit flat (a pun and also literally the lines are too flat).

The sharps look like a guy flipping the bird. And the flats look a guy with a fist, ready to fight. Funny but a bit distracting.

The focus on pitch letters (abcdefg) and lack of key signatures reveals a weak music theory foundations. The key says what scale to use and then you think about the relative positions, you don't think about the letters.

It feels a bit like training wheels for reading music. Maybe thats the real purpose?

I completely support anyone who wants to invent new notations for things. It's fun. But I'd just like to note that if logical, regular notation was necessarily better, we'd all be speaking Lojban and programming in Scheme. Also, I suspect a conventional eighth note would be easier to make out in a dim concert hall...

Simplifying notation doesn't automatically make things better.

Theory: people prefer infix to prefix or suffix notation because it more closely mirrors the Subject-Verb-Object patterns of their native languages.

Corollary: lisp feels awkward because it doesn't map cleanly to native language thinking.

Lojban is mostly SVO as well.

If that theory was correct, one might expect Forth to be really popular among Japanese speakers. I don't see a lot of evidence for that.

I tend to think it's a more general effect where humans actually want a certain amount of irregularity in their languages/notations, to act as markers or error-detecting codes of some sort. e.g. "he", but "him" in accusative case. But who knows; English gets by with "you" being both singular and plural...

Most of the comments here seem to be about the qualities of the notation itself and whether it is good. I would like to ask why this notation of all possible notations?

Music notations - systems of analogies between visuals and sound - are tools for communication. If you have some new concept to communicate, inventing a suitable notation for it is a great way to gain recognizability for the concept. This holds with music notation as well as mathematics (ex: http://en.wikipedia.org/wiki/Graphic_notation). If you have no new concepts to communicate, inventing a new notation is like changing all the words in the python language to other words. Some things may be easier and others harder, but overall, the change isn't worth the trouble for most.

If there are no new concepts to communicate but there is a new medium to work with, that needs to be taken info account. Legacy notations were invented with the constraints of the media of the times they were developed in - in this case, paper and ink. There is no need to be bound by old media constraints when you're making something that has to be learned fresh anyway. Notice how hummingbird still does everything in black and white? We've had color displays for ages now (even on paper), so why not map note names to colors? Why map time to space when we've had dynamic displays for ages now and can just map time to time itself, or a mixture of time and space. Why should every instrument player see the same notation? Why can't we adapt the display to suit the instrument?

Guitar Hero, Dance Dance, the Japanese drumming games in arcades all illustrate what "notation" can be using an interactive medium. The would-be-musicians don't read paper notation. Learning the display is so easy that they just pretty much pick up the controller and begin playing. For other kinds of "notation" that can help the fresh ones, check out the iOS apps by Smule (Note: I don't have anything to do with Smule. I just like their work.)

The standard music notation and the criticism this new notation receives here look to me as if we were stuck for centuries with only one programming language, C++, and nobody could change it.

Sure, it has been working fine for many years, and a die-hard fan of C++ would come up with many criticisms of anything new: No pointers? It's for beginners only. Garbage collection? I can see the programmers that created it are not very proficient. Etc. etc. etc.

And you know what? There are really great programming languages that do many things differently and also work very well, or even better than C++. They are not perfect, but the traditional way of doing things (C++ or the standard music notation) was not perfect either to begin with.

So, I hope people experiment more and more with new notations, and maybe they will improve the standard notation or even replace it someday.

This looks pretty slick. Just the fact that its a bit easier to write is awesome. Even if the usefulness of this notation doesn't pan out, the hummingbird website is a great example of good marketing and messaging. It's the perfect name and the perfect website to promote something like this.

Is it really easier to write?

Every note has to have a line that extends its full rhythmic value past the notehead -- you can't just write 10 circles to say "this big chord fills the whole measure". Every accidental must be marked on the note -- so if you're writing a piece in C# major and it doesn't diverge from that key, you're going to have to notate the accidental on every single note.

EDIT: they do allow key signatures; they just don't mention it in the example or intro video.

And every notehead's shape varies based on the pitch class (i.e., C, D, E, etc.) -- so except for "E"s (empty) every note head has some decoration you'll have to do... you can't just make a spatter of dots/stems with a slash over them for eighth notes.

For some music, this seems like it would be a bit more work (mainly the varying noteheads); for other music it'd be a serious problem.

Mapping the pitch of notes to special symbols (to distinguish, eg, 'B' from 'C' or 'la' from 'ti') has been done before. See http://en.wikipedia.org/wiki/Shape_note and the accompanying images. Introduced in year 1801.

And I think it's a reasonable idea, especially for studying, practicing, or sight reading. I'm current struggling to map staff notation onto the guitar fretboard (I can play piano from staff notation, and play the guitar "by ear," so I'm trying to synthesize the two abilities), and I think this might make it slightly easier. For a lot of classical guitar transcriptions, or any guitar transcriptions that refuse to transpose (an octave), there is a lot of extreme ledger lines, and the shapes would help immensely.

If you're choosing between "rewrite this guitar music to transpose so there aren't crazy ledger lines" vs. "rewrite this guitar music into a radical new notation system", I think I'd go with the first choice.

Standard notation for guitar music also suffers from the "where on the fretboard do I play this" problem -- sometimes there are hints (like finger numbers), but often you just have to try a few options and see what's best... which is rough going if you're trying to sightread a piece you don't know yet.

TAB notation, of course, fills in the missing info on suggested finger placement, but omits other essential info (rhythm!), so by itself that's also broken....

OT: I'm curious. Does anyone besides me have a tendency to get off by one on the values when playing from a written score?

I have a tendency to play a note and then my eye moves to the next note--and I take the value of that note as how long to let the prior note play before playing the next note.

Love your thinking here, and your landing page is fantastic. Very inspiring. The 'before' and 'after' toggling is a very effective way to illustrate the value your system brings.

If you can help my mom put an Ipad on the music stand instead of a song book she has to manually flip through while playing you might just win another customer. Right now, when i do a jam session with her (I play guitar) she literally has to stop playing piano to flip the pages. She does this 3-4 times a song which is quite jarring. The humble paper song book and conventional notation system could use a 'shaking up', and you guys have started innovating this problem, so thanks!

I could easily see this becoming the tabulation (tabs) system for that piano and other instruments have not had to date.

Well done!

This is the musical equivalent of trying to improve the English language. Sure there's room for improvement and your ideas may be good ones but there's a massive established standard and you're not doing your students any favors by teaching them this over the standard.

I like it. It will no doubt be refined as more people learn. It will be interesting to see how it holds up.

On a side note, I think a majority of the arguments I've read against it aren't very good ones. I see most of them boiling down to, "it's different, I don't want to have to learn a new language". This is what I said when I was learning standard music notation. I can see the issue of its anglo-centricness, but even that is a minor detail. One doesn't need to understand the words "above" or "below" to be able to associate the symbols with the notes.

I don't think a fair critique can come out of a quick glance at the hummingbird website. I'd be more interested to hear what people say after a few months of working with it.

My first thought looking at the sample notation was that this would be hard to keep legible in hand-written music. It seems like it doesn't really do much to improve the legibility of notation while it does force the use of their software.

I was thinking the same thing - could be quite difficult to read under anything but good circumstances.

I've played in an orchestra from 8 to 18, performed live, etc. In my peak I was also a proficient trumpet, guitar and drum player.

My opinion:

* It is very nice to see someone trying to improve something that's hundreds of years old. This is an excellent teaching tool. Words cannot express how better it is!

* It is a poor tool for the live performer. When performing live, you already know the piece. What you need is visually clear, bold, guidelines. You also want to answer "I'm playing this now, what's next?" very fast; and to do that you need a clear visual relativity. I don't think Hummingbird does that well with the collection of small nuances and decorations.

Kudos to the team for coming up with a viable form of alternative notation! But personally (I've been playing classical music since I was 6 so my opinion is most likely biased) I find the new notation harder to read and comprehend but that's the consequence of my classical music education. So the new notation is meant to make it easier to read and learn even the "trickiest music"; I am just curious what this notation would look like with a genuinely complex piece of classical music (think Liszt, Rachmaninov or any other bits of classical music that one would think as technically challenging).

If you think hummingbird is strange, check out the notation system my former engineering prof came up with http://www.pianotheoryman.com - based on real-time systems engineering experience!

I'm a classically trained pianist and an engineer, and I get why tech-minded people are frequently tempted to hack western notation, but notation is that way for a reason, and it isn't what prevents people from becoming skilled musicians. It's talent.

Here's a suggestion: Add a sixth line to the staff. Now all clefs are identical two-octave portions, instead of getting shifted more and more as they retreat from middle C.

In my opinion, the greatest weakness of traditional music notation is that a same melody has to be written in 12 different ways, depending on in which key it is. After you've trained yourself to read in all 12 keys (or 5-6 most common ones), everything works nicely, but that's a lot of repetition.

A notation that would solve this problem, would be great progress.

But this Hummingbird notation does nothing to help with this problem.

This is an absolutely terrible idea; I hate to say it, because it looks very visually appealing. First mistake is to remove the key signature.

Seems counter-intuitive to me, but I'm interested to know who the target audience is for this? People who want to learn guitar for fun but think learning to read sheet music is too hard (I have never understood this), or do you expect that the big music institutions will adopt this? It doesn't really solve a problem for them (Although it might solve a problem for hobbyists).

A great idea and new way of looking at music, but with music and expressing it in a new visual manner, it is still missing a few things that may improve the piece stylistically. Crescendos and dynamics for example.

Although this might be an easier way to visualize and teach music for young musicians, it might be harder to translate to traditional music as one gets older.

Pitch is encoded by location on the staff, and also by the shape of the glyph. Isn't redundancy necessarily bad in any notation?

Redundancy can actually be useful in notation as a sort of error-detecting code. In this case, if the pitch-symbol and pitch-position don't match up, someone must have made a mistake in transcribing it.

Whether it helps one when reading music is another matter. I remember learning to play an electric organ as a youngster with a book of sheet music that came with the organ; being aimed at beginners, each note head had the note letter written inside it. Hummingbird is basically using the same idea, just replacing the letter with a symbol. It probably does make it easier to learn. Whether it would be of any use for an experienced musician, I kind of doubt. (I certainly outgrew expecting the note head to contain the note letter, myself.)

It depends on the purpose. Introductory materials sometimes use a large enough note that they can write the letter name inside the note head. (Works better for black note heads.)

The redundancy helps associate the name and position.

I don't think I would prefer it in this case. The extra note name information is visually noisy and detracts from the shape of the line, which is mostly what experienced musicians see. The shape of the line, the absence of sharp and flat marks, and remembering what key you are playing in, covers most of your reading.

Not necessarily. Note duration, for instance, is conventionally redundant (through note appearance and horizontal spacing).

not really, it can help in reading things faster&reduce errors and in this case the standard notation is just "wasting" some bits of information (the shape).

But, I, for one, am annoyed by the english language needing both grammatical structure _and_ a question mark to express questions ;)

I think the shapes would make more sense if they represented "do re me" instead of "ABC". The latter might be helpful to students, but the former is genuinely helpful to singers. And in fact, it's been done before:


Well, I don't have my glasses on hand so I can't really comment on what look like small smudgy bits but overall I like the idea... just can't see it. One advantage of current notation is that it is larger and easier to see plus familiar, so I know what it "should" be even when it's a tad blurry.

I'm a harmonica player. I don't play a chromatic instrument; I play diatonic instruments in various keys and tunings. The harmonicas I play most are in the Melody Maker tuning [1], which has a major diatonic scale in the "cross position" that harp players like to use.

When I read a score, the first thing I need to know is which harp to pick up, and whether it will be playable at all on this kind of harmonica.

Looking at the Hummingbird notation, how do I figure out which harp to play? I have to study the whole piece to see which sharps and flats it uses, and then translate that pattern back into a likely key for the piece. After I do that, I have to go over the sharps and flats a second time to understand which ones will be normal notes on that harp and which will be bends or overblows or impossible.

I guess I could look at the end to see what note it resolves on, but that wouldn't tell me whether it's in a major or minor key. And even if you told me what key the piece is in, I'd still have to study the entire thing to sort out which of the sharps and flats I can ignore and which I have to worry about.

Switching to the traditional notation, I can see from the key signature that the piece is either in Eb major or that key's relative minor, and I'd use the Eb Melody Maker for either. Also, I can see at a glance that all the notes in the treble clef follow the Eb major scale except for those E naturals in the middle. Those would be trouble, but the alto part looks easy enough.

It may not turn out to be a good harmonica piece anyway, but I immediately know which harp to play. And other than the accidentals, I know the notes will be the ones in my scale, so I can start noodling with it right away.

Traditional notation isn't great for a diatonic harmonica player. I don't really think in absolute scale notes at all, since I change harps to change keys. I think in terms of relative scale notes. Letter notes may be anywhere on a harp depending on what key the harp is, but the tonic is always going to be the draw 2, blow 6, and blow 9 regardless of the key.

So the ideal notation would be one I could always transpose to match the key of the instrument I'm playing. Maybe Hummingbird would be OK if it was only used on computers and always transposed on the fly. But give me a printed score and I'd be lost. At least with traditional notation I've got a chance.

[1] https://www.google.com/search?q=melody+maker+harmonica+tunin...

Shape-note systems for learners have a long history; this one has some clever touches, but ultimately that's it's niche -- it's not going to replace traditional notation, but it might help more non-musicians and non-reading musicians start reading, which is a good thing.

Might I suggest that barlines remain connected vertically on both the grand staff and on instruments in the same family? This is a very easy and intuitive way to make a score readable, especially when it involves large groups of instruments (such as orchestral scores).

I've spent the last year taking some serious time to compose some music. I always knew how to read sheet music and I think the current notation has survived cause its easy to read and write. This Hummingbird notation looks relatively much more complex to me.

Hey guys. My name is Blake West. I'm the co-inventor of Hummingbird. First, thanks for all the discussion. There really is no such thing as bad publicity. You've helped us crack 13k downloads in under 48 hrs. 2nd, I thought I'd just quickly respond here to some of the main points...

1.) "Traditional is fine. it's not broken": Neither were text-only command line interfaces. But GUI's are just easier to learn for most people.

2.) I am indeed a professional keyboardist, have been playing and reading traditional notation since age 7, and I also teach 25 students a week still. I know theory like the back of my hand, and can talk modes, b9 chords, and 12-tone rows all day long if you like. Jazz and pop are my thing and I play to lead sheets more often than not now a days. So I know this fro m both angles.

3.) We do have key signatures. They're at the start of each song in plain english. no need to be cryptic with symbols.

4.) Relative pitch notations seem like a good idea, but they really aren't. The function of a pitch is honestly pretty subjective and changes frequently in a song. Not to mention, they'd be much harder to learn, especially for young students. Relative pitch is an abstraction, and abstraction is the luxury of experts.

5.) Why not use a chromatic staff or other such layout? Because we actually wanted some adoption. Most other alternate systems have failed because they're SO different that they are completely alien. Ours is "backwards compatible", and also if you did want to switch over to traditional from Hummingbird, you could, and it's not that crazy.

6.) Why not use colors? Because music still gets printed and photo copied vey often, and will for at least another 5-10 yrs. And color printing is still 7x more expensive.

7.) "You can't hand-write the symbols". Yes, you can. It is slower, but our point is that most music is printed off of notation programs today, so hand-writing is usually reserved for small edits, or writing fragments from scratch. This is still completely fine even with an unsharpened pencil with Hummingbird. I have done it with my students many times.

8.) "Picking out lines and spaces isn't that hard" - If you spent time around kids you would be SHOCKED at how bad their spatial reasoning is before about 9-11 yrs old. It is really hard for them without a ton of frustration. That frustration often leads to them thinking they're "bad at music". That turns them away, and it shouldn't have to.

I know there's other stuff, but just not enough time...


Fixing the orthography of English language comes to mind. If that so-obviously-improvable notation is impossible to fix, this one probably is, too.

But good luck anyway finding something that helps students learn music more easily. It's clearly a goal worth fighting for!

I feel like a lot of semantics are lost here. Some of the elements are partially filled -- these elements seem more challenging to classify than the traditional approach. Also, it's unintuitive how "long" a "----" line should be held for.

If I'm not mistaken, the few bars from the homepage are Chopin's Revolutionary Etude, Op 10 No 12.

I think I prefer the way it looks in traditional notation, and I think music might be more intuitive to write as such – but heck. Always be innovating.

Whoa there. Too much too quickly.

This reminds me of EMACS -- it's going to take a lot of mental work for me even to begin to understand the strengths and weaknesses of this.

Even if it were better, beats me how you'd actually get people to give it a chance.

I've used traditional notation for more years than I spent minutes looking at this. I'm not sure anybody can say anything useful beyond "hey, this isn't what I'm used to".

It'd be nice if on the page they told us what kind of investment we would be looking at to actually make an informed decision -- "After only 12 hours of going through our tutorial, you'll be sight-reading better than ever!" or something.

Whether or not this is a good solution (I haven't examined it closely yet), there are certainly a lot of problems with traditional Western notation. For example: you're limited to 12 pitches, you can't precisely represent bends, you can't precisely represent note lengths, many of the symbols are cryptic at first glance, some of the words don't get translated (tempo markings), some of the symbols are imprecise (again, tempo markings), it's annoying to write music in non-traditional modes, it's annoying to write music with syncopation, etc. Sure, this has worked fine for musicians over the past few centuries, but music has changed a lot since Mozart. I think Western notation could use a little rethinking, so I'm happy that someone is at least looking into it.

Sure, these are generally valid points, but Hummingbird doesn't really solve them. I'd love to see a solution to those problems, though. >12-pitch solutions exist, but they're just cumbersome.

I'd disagree about the precision of note lengths and tempi being a problem, though, simply because humans aren't very good at keeping exact time. I mean, ask someone to count off exactly 60 beats evenly divided over the course of a minute. What about 80? 100? 133? The speed at which you count off is highly dependent on your heart rate (and possibly state of mind). So assuming we have an exact tempo, what's the likelihood a note length within it is going to be played with that same amount of precision? Well, it's all up to the musician's ability to interpret the composer's intentions, which is much the same as what we have now. In fact, there are ways even now to specify exact bpm (even if we can't exactly express slight tempo changes). The notes themselves are estimates, but fairly good ones at that, and they're effective enough to get the point across to the musician.

Well, with tempi, my main gripe isn't the exact BPM but rather the fact that you have to memorize all this obscure Italian (+ other?) terminology in order to understand what's going on. Why can't we have standard localized terms instead?

As for exact note lengths, I mostly agree with you, except for certain pieces of modern music (EDM, for instance) where the meter is fairly rigid but the note lengths are off from the defaults. It would be nice to be able to represent that accurately.

This notation is stupid as shit. In addition to the problems mentioned by others it is so hard to see from a reasonable distance while holding an instrument. Compare that with the traditional notation.

The flat symbols are tiny. I'm supposed to read this on a music stand?

I dunno, I already learned how to read sheet music, so maybe I'm biased, but my gut reaction is that it's not an improvement.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact