How does one know if a score is notated well?

Question

I'm a programmer and in programming a code can be bad or good. It is easy to see by the way the structure is built and how blocks of code are arranged, what functions and procedures are used, and how everything is organised.

In music as a composer I'm finding it hard to understand whether a score is notated well (whether it corresponds to a professional music notation style guide or rules of notation) or badly. This is probably because I'm not so good at sight-reading music myself, since I have little practice and haven't seen a lot of professional scores (I can't make a comparison to my scores). Anyway, I compose in Guitar Pro 7 (notate using this software) and I always ask myself, "Is the score I am notating good? Is it correctly done? Does is correspond to rules or style guides? Won't a sight-reader have problems playing the score?"

What makes a score well notated, let's say for piano for instance?

Edit: Someone might have misunderstood the question so I'll add what I mean by bad or good:

Is this a piece of score that is badly notated? Well, I definitely think it's a very badly notated score (I did it this way on purpose, to describe what I personally think is a badly notated score).

My point of view is that here the notes are improperly durated and tied to each other. The rests aren't optimised for readability. The key signature is missing, although we can see that the piece is written in a key with sharps or flats. The accidentals are a mess, which makes reading harder. Articulation is missing whereupon we can't say how the piece is to be played and what dynamics are used.

A lot of scores I find on the Internet are really badly notated in my opinion. I think this is a very problematic topic. i think that making the score look nice, easily readable and non-problematic for conductors or musicians is a very difficult task in fact.

Another Edit: Is it required to include dynamics and articulation in a score? Is it necessary to include fingering?

Answer 1

I consider a score to be well notated if it efficiently communicates a musical idea to a performer. A great musical idea that is badly communicated is a bad score. So, how do you communicate effectively? Some thoughts:

• Choose the most conventional way of doing things. This pretty much encompasses every other point here. Unless there is a good reason, notate in the same way as other scores. There are standard ways to write certain rhythms and use accidentals. There are standard score order layouts. Use them. Why? Experience has conditioned me (and every other musician) to expect certain norms to be followed. If you follow them, then we're going to understand each other quicker and better. The less time I spend trying to find the bassoon part because you've put it next to the tuba, the better. Exception: convention is just convention. If there is a better way of expressing your idea, you should do it. The trick is knowing whether it is better. I'd suggest trying to understand why the convention exists, so that you understand when you can break it.
• Notate rhythms consistently, and make them as easy to read as possible. This is connected to convention. Can I look at the rhythm, and figure out how to count it? Would it be better if I split some notes? Combined them? Changed the time signature? Used a triplet? Is there a standard way of writing this? A good rule of thumb is asking whether you can easily see where the strong beats fall in the bar.
• Use key signatures and accidentals appropriately. For a start, use the right key signature. Use the accidentals that would be expected. In C# Major, don't write a Db. Why? Because I have to think about it more, which reduces the effectiveness of your communication. Don't use the more obscure accidentals (double flats and sharps) unless it makes the music clearer. Don't switch between Gb and F# in the same bar/phrase/piece. The more time I spend thinking about accidentals, the less time I think about expression and emotion.
• Make sure the notes you expect an instrument to play are actually playable. Real instruments have physical limitations. If you can't ask a player, there are many inexpensive or free resources that can tell you what you should do here. Real flutes will struggle to a low C at fortissimo. Real saxophones may not be able to hit a low Bb at pianissimo.
• Choose the right clef. For example, if you write tenor sax music in bass clef, your idea will not be well communicated, and the music will suffer. If you are writing a score for a conductor, you can use concert pitch, or transposed pitches. Either is usually ok, but if you can ask the conductor, do so. Don't use a mix of the two, though. And be aware that concert pitch music for instruments like bass clarinet may need to be written in bass clef, whilst transposed is written in treble. Here's more good points in another answer.
• Are dynamics required? Depends on what you are trying to achieve. If you have an idea that requires specific dynamics (and most do), then you should include them. But you should be aware that they are relative levels. A big band forte is different to a concert band forte. Even within the same piece, I'll interpret forte differently depending on the part I have, and the 'feel' of the piece. Lack of dynamic markings does not mean that we won't play dynamics. We'll just choose them ourselves. Even if you do mark them, they will often be adjusted via pencil markings anyway.
• You should use articulations where the player would not automatically do what you want. That's pretty vague. What I mean is that, for example, in a jazz big band, players are naturally going to articulate rhythms in a certain way that is idiomatic to that genre. If you want to change that, add articulations. But be a little cautious of unnecessarily articulating when a player would do that anyway. Point number two. Articulations mean different things in different genres (e.g., classical staccato vs big band staccato). Be aware of what you are communicating with each symbol. Same letters, different language. Point number three. Be careful of just adding articulations to make the computer playback sound right. Real musicians will get annoyed. If you must, add them, but hide them from the printed parts. Sibelius can do this easily, so I assume the rest probably can as well.
• Fingering depends on the audience. If you put trumpet fingering on a Grade 5 Concert Band chart, people will think you don't know what you are doing. Why? Because you're communicating too much information. They already know that. No need to repeat it. However, if it's a piece for a beginner, maybe you should print fingering on the harder notes. Fingering on every note should be reserved for technical exercises or when teaching a specific concept. Otherwise it's just redundant.
• Know your (musician) audience Simple example. I have a score of Silent Night notated in 6/4. It always causes problems with the junior band. If it was written in 3/4, it would fit their expectations more, and they would find it easier to play. Now, 3/4 is actually 'incorrect', because it's not a compound time signature. But I know from experience that it would communicate the musical idea better than 6/4 or 6/8 (my preferred option), or anything else.
• Score navigation is a big deal. Don't use a complicated sequence of codas and repeat lines if a few extra bars will suffice. Put page turns in logical places, if at all possible. You can almost never have too many rehearsal markings (those numbers/letters that you can use to refer to a specific section). I prefer bar numbers to letters, but either is much better than none. If you are using a computer program, why not add those little bar numbers at the start of every line? Again, Sibelius does this by default. Every second we spend hunting for a bar is a second we don't rehearse your music.
• The typography and design can be significant. Handwritten is ok as long as we don't end up having a fifteen minute discussion over whether that crotchet is a C or a B because we can't read it properly. There's a whole art form in typesetting a score, and I'm useless in that area, but other people have already answered on that topic. To be honest, I personally care less about this point that many of the others. Opinions may differ. My argument is that a score with bad spacing can still clearly communicate the idea adequately. Sure, it would be improved by proper typesetting, but you might not have the time or budget for that. Don't spend all your time on this, and get all the rest wrong.

This has become much longer than I intended. My main point is that score writing is about communicating a musical idea. A good score communicates clearly.

Answer 2

It's extremely obvious that your score is not optimally spaced. I would also add that the choices of rests and tied notes make the rhythms needlessly difficult to read.

These days, scores are usually typeset by computer rather than by hand. Computer programs, such as lilypond, have algorithms that evaluate the beauty and readability of various alternative layouts.

Here are some examples of decisions that the programmers have implemented:

• there is too much space before the time signature
• the stems of the beamed notes are too long
• the second and fourth measures are too narrow
• the slur is awkward-looking
• the trill marks are too big
• the stems are too thin

There are readability issues that the LilyPond developers have identified in other software:

Finale doesn’t adjust the positions of interlocking note heads, which makes the music extremely difficult to read when the upper and lower voices exchange positions temporarily:

They even criticize scores from professional publishers for looking too rigid and mechanical, since all the notes and bar lines are too regularly aligned.

In summary, yes, it is possible to evaluate scores for beauty and readability. For the details, you can read the source code of LilyPond. =)

Answer 3

Edit: OP's edit has made clear that I misunderstood the original question. I would edit my response, but I think endorph's answer is already terrific. Nevertheless, I'll leave this answer here, unless it's asked to be deleted.

---

I think it depends on whether or not these scores will be played by humans or be computer-generated. If it's computer-generated, I think this question is mostly opinion-based. Since problems of technique, acoustics, etc. can be rectified digitally, any opinion of the score seems to be mostly subjective.

But if the scores will be played by humans, there are definitely some things to consider:

Physical Considerations

In terms of the piano, there are some things that a single human simply cannot do. With (usually) only 5 fingers on each hand and a limited range within a single hand, a well-written score needs to be physically playable by a human body. Other instruments have other limitations, especially concerning range: you wouldn't write a violin below the bass clef, for instance.

Technical Considerations

Related to the above, even if something is physically possible, that doesn't mean it's practical. Some passages can be too technically demanding, and although they're possible in theory, they aren't possible in practice. Wacky intervals that bounce around the piano in 32nd notes for 6 minutes straight might sound cool, but pretty much every pianist will just scoff at it once they see the score. Similarly, brass players can't articulate leaping staccato sixteenth notes at 200 bpm, nor can they hold an uninterrupted drone for minutes on end like a string player could (circular breathing notwithstanding).

You also need to know who you're writing for. The technical limitations for an 8th-grade symphonic band are very different than those for the Tokyo Kosei Wind Orchestra.

Acoustical/Textural Considerations

Especially with the advent of notation software, inexperienced composers often don't realize just how differently their piece will sound in real life compared to the notation program. A well-written score will always account for nuances in acoustics and texture. You wouldn't put a clarinet solo in its throat-tone (ie, quiet) register while a quartet of trumpets plays forte in the same register. As I see it, this is the only limitation that computer-generated music can't always get out of, but it still usually can.

Answer 4

I'll fill this in with things that endorph misses.

Ledger Lines, Clefs, and Ottava Lines

All three of these are often used to notate notes that are not in the most-often-used range of the instrument. But some of them are more unreadable than others. For example, having five or more ledger lines for a single note is often a sign of badly written sheet music, as that many ledger lines are hard to read. They're usually a sign that you need to switch clefs or use an ottava line (whether octave above or octave below).

Regarding clefs, some clefs are not used for some instruments, so don't use them for those instruments. For example, alto clefs are virtually never used in piano music, so alto clefs in piano music are signs that the sheet music is badly written. Likewise, bass clarinet music ironically does not use bass clefs, so don't use them in bass clarinet sheet music if you don't want that music to be badly written.

Also, changing clefs more than twice in a single measure makes music hard to read and therefore is a sign that the sheet music is either badly written or has an awkwardly enormous range. Consider using ottava lines or ledger lines in those cases.

Don't overuse ottava lines, though. I recommend that if you can use 3 or fewer ledger lines instead of an ottava line, use the ledger lines. Otherwise, people may wonder why you're misusing those ottava lines.