Score voting

• Jan 31, 2015 - 23:39

Hey! I'd like you to add some voting system to all sheet music published on website. It would be easier to find good scores and avoid these not very good. It's not directly MuseScore feature, but I think this is the place where such thread should be written.
Cheers!


Comments

In reply to by kujaw

I still don't know how you would define 'good'; it could mean one's opinion of the music, or it could reflect an assessment of how skillfully the score has been typeset. Or it might mean a combination of those two elements - but in what proportion? Or it could mean something else entirely.

Unless the criteria were objective (unlike the guesses I've made about what 'good' could mean and how it might be measured), a rating system doesn't seem meaningful or useful to me.

In reply to by [DELETED] 448831

IMHO the only kind of evaluation which would make sense in a MuseScore forum / community is about typesetting / engraving quality; the quality and the relevance of the music engraved should be completely ignored.

It would still be subjective, of course, but any evaluation would be. Even in this restricted area, some problems would still remain, though:

A) the correctness, skilfulness, even beauty of the engraving and the appropriateness to the intended purpose do not necessarily go hand in hand: a reference score with, say, all Beethoven piano concerts and the same pieces typeset for performance would require different layout approaches; a scholar in need of a reference work may dislike some aspects of a performing score, while a performer may have reserves on a more compact, reference score.

B) For specific engraving tasks, there are different schools and what looks correct to someone does not necessarily looks correct to other.
________________________

Something different, but perhaps related, could be some kind of 'mark' / 'mention' / '...' for scores which presented complex engraving problems and which the engraver had skilfully solved. No idea about who would be in charge of assigning these mentions, though.
_______________________

In summary, the matter is rather thorny.

NEVERTHELESS, some way of expressing appreciation and / or marking a well done job may help other users ("this is how these kind of things are / could be done"), increasing the average quality of scores engraved with MuseScore, tightening the community, raising the MuseScore status.

So, it MIGHT be worth pursuing; it is very important to aim it rightly since the beginning, though.

In reply to by Miwarre

I really object to the basic premise of this, although I believe this was proposed with a good heart. Among many other thoughts, if one wants to express an opinion it is much better to describe ones reaction rather than some easily clickable "like - "hate" relationship which communicates nothing substantive other than a chance for upset. If you like it, let 'em know!

In reply to by xavierjazz

As it probably appears from my previous post, I have mixed feelings myself.

My basic premises are that this community gather some rather specialized knowledge about music engraving and this knowledge could be put to some use. The goals could the ones I outlined above (rewarding good work, letting people know that MuseScore can be used for good work, highlighting ingenuous solution for complex problems, etc...) and / or other ones.

It is true, not everybody here is a specialist in music engraving and probably nobody has a good knowledge of all styles and use cases of music engraving. Then, for sure, a single "Like / don't like" button is not appropriate and I would object too.

Something more evolved would be needed. It may takes many different forms:

*) A specialized "Reviews" sub-forum, where users could rate and comments about posted scores. Its relationship with the current "Made with MuseScore" sub-forum should be thought carefully. As it probably should be a way to combine freedom of comment with some sort of aggregable evaluation.

*) A complex rating system like the ones used by Stack Overflow and other similar web sites. With its interactions between numbers of posts, number of replies, numbers of votes and so on determining who can vote for what, it is rather complex and, as such, it would probably be overkill; a simplified system along these lines could make sense, though.

*) A form-like rating system with a few (2 to 4) topics to be rated from 1 to 5 (say: readability, ease of use, appropriateness to stated goal, general engraving quality, ...); its wording would discourage hasty answers and encouraging a minimum of thought. It could include additional free-text comments / explanations.

*) ...

Of course, the whole concept could turn out to be simply impracticable but, who knows, a few thoughts and a some brainstorming may point to a solution worth pursuing.

Thanks,

M.

In reply to by Miwarre

I think it will be impossible to satysfy everyone, as there's always somebody's needs that have to be sacrificed. But I think that any, any evaluation type is better that none.

I need an answer on one question: Why do you write your pieces and put it on this website? What are main reasons for doing this.

I just wrote an idea, looking for community opinion on this. I didn't mean only simple "like-dislike" button. I actually think that there is need for complex rating system. We should have possibility to sort pieces by date, by number of downloads, by rating, by some combination of rating/date (something that can be called "hot" I think?)

Or as you wrote - readability, ease of use etc. but...
isn't readibility similar to ease of use?
what do you mean by appropriateness to stated goal? what goal?
general engraving quality - could you explain more? isn't it also similar to readability?

And what I think is a must-be - an optional comment during rating a piece.

In reply to by kujaw

I would keep distinct, at least logically, sorting by inherent parameters (date, no. of downloads, ...) from rating: the firsts can be computed from the archive data themselves, the seconds are the result of a human being's action. Both can be useful, of course, but need to be treated differently.

About the categories I listed, they are just proposals; there is surely some overlap, but I see them somewhat different.

Goal: each score has a goal; I mean the score itself, the physical (or potentially physical) object, not the music it contains. It may be a reference score for study or for following a performance, it may be a performing score, it may be an Urtext, it may be a critical edition, and so on. If the engraver didn't set his own goal while engraving the score, the result is likely to be unfocused.

A reference score may be of small size (or small scale) and still be useful (perhaps even more useful, as it is easier to carry around and / or to browse), it should allow to easily locate individual pieces or sections. A performance score cannot be too small and should pay attention, for instance, to page turnings; it will possibly need indications not needed in the former, fingerings for instance. And so on.

All these elements, as well as others, are important to rate the appropriateness of a score.

Readability is more concerned with the ease of identifying the nature and the purpose of each element: is it clear from which note to which note each slur goes? Is spacing (both vertical and horizontal) appropriate? too cramped? too loose? Or with too marked changes from one spot the another? For multi-instrument scores, at the end of a system, it is easy to jump to 'my staff' in the next system or there is the risk to jump to the wrong part? Can the various kinds of texts (tempo, dynamics, technique indications, ...) be easily distinguished and associated with the right element?

Ease of use has some overlap with the previous, but to me is more related with physical properties of the score: too many pages? too many page turnings, maybe in impractical spots? If a bunch of loose sheets fall on the ground, it is easy to reconstruct the sequence? Are separate parts clearly marked? And so on.

General engraving quality: lack of errors, proper usage of the available space, adherence to established practices and traditions, and so on...

I think it should be clear at this point that in my opinion, as we are in the context of a music notation software like MuseScore, the whole evaluation / rating system should be about the engraving, not the music itself.

Again, just my opinion...

In reply to by Miwarre

I like this - very clear and focused, and by it's nature it brings us to the basics. Thanks a lot Miwarre.

My impression is that the OP was suggesting rating "emotional connection" (good/bad) and using that to determine musicality, which again is subjective, and, as you point out, is not the main reason for this project.

Best regards,

In reply to by kujaw

The main purpose of this project has been stated many times: to provide a high quality, open source notation transcription program dedicated to providing the ability to notate music at a high, professional level. I have often expressed my support for this basic goal. To me, that's the only thing I want from Musescore. I have questioned, occasionally, threads that ask for and promote aspects other than this basic goal, but as long as the energy is mostly directed to that goal, I don't object to other's wants.

As for me, when I upload music to the Musescore sites, I upload .pdf notation as I am most interested in providing opportunities for players to learn and reproduce. If you don't want to do the work, that's okay. When I upload audio anywhere, it is from actual performance. My own preference is to capture the energy of live performance, so virtually all my audio is "Live off the Floor", with all the hazards that presents. I work with musicians that are capable of this.

I was heavily involved with MIDI here in Canada when it first came out - I was involved with major manufacturers, using their early equipment developing MIDI files for their use. Although there are undeniable uses for it, and I used it a lot then to examine my work, I rapidly became disenchanted with the mechanical aspects of it and moved on. I don't use the MIDI playback abilities in Musescore at all, but that's personal.

I have no issues with those who wish to use digital means to manipulate their sounds, and I am considering using it again to manipulate sounds that are not achievable with analogue instruments, but generally it just holds no interest to me.

As for my initial response to this thread, I feel that the culture of instant, effortless "like" and "dislike" criticism so popular on "social media" today, while it gives exposure to an individual's vague "feelings", it is basically harmful to a real examination and discussion of the worth of a particular piece of music and in a way destroys the contact it seems to espouse. I feel that it waters down discussion and exchange. Instead of a button to push (which really takes no effort and to me is a vapid kind of expression), I much prefer to contact the individual(s) involved with the music I want to interact with and have a bit deeper exchange. I suppose it's adequate for much pop music, which generally doesn't deal with development but rather on the momentary interest of a "hook", but that is not an area of direct interest to me.

I can go on but I'd rather not. I DO find it interesting that so many people want the ability to "humanize" their work without using humans, but that leads to philosophical discussions I would prefer to pursue elsewhere.

As I wrote earlier, I think you probably have what you feel is a good motive for wanting this "facility". I just strongly disagree with the whole premise.

:)

In reply to by Miwarre

I think that Readibility and Ease of Use you describe have too many similarities so they shold be one thing. Because if something is easy to read it surely is easy to use. If It's not clear "from which note to which note each slur goes" it is an issue either of Readibility and Ease of Use.
I think there should be only three of these conditions: Goal, Ease of use(Redability), General engraving quality.
And one thing about Goal - how am I supposed to know what was composer's goal when writing this or that piece of music?

Remember that I look at this problem as a performer, not composer. I wrote here my point of view.

I think we should make a possibility for performers to choose best (subjectively) scores to play - so sorting by downloads, date, general rating.
Also we (performers) should have possibilty to rate all these engraving quality things, this is for composers to learn and make better scores and perfomers to have a great time working with these good quality scores.

And as composer I think that I would like other composers to check my scores and write engraving quality tips and rate my engraving.

So maybe we can make two types of rating - for performers and for composers.

FWIW, I'm not as uncomfortable with the notion as some here seem to be. I mean sure, what does it really "mean" if a lot of people "like" a score. But "like" it or not, we live in an age of social media, social media generally uses this mechanism without a second thought, musescore.com is part of that world, so why not meet the expectations?

Of course, msuescore.com does have the notion of "Favorites" which accomplishes largely the same thing.

In reply to by Marc Sabatella

Present social media conventions are not necessarily etched in stone, nor necessarily a good way to approach thing. It seems to me that people simply press "like" and feel as if they have taken some concrete action. In important cases I think this removes the incentive to actually contribute thought and ideas and replaces action with some sort of fuzzy "feel good about myself so I can move on to other things" rather than any real contribution. If you have a strong feeling about something, it surely isn't too much to ask for a real comment rather than some vague "like". Also if you hate something. :)

(Looking for the "dislike" button for this thread).

In reply to by xavierjazz

I think it's a bit absurd to imagine we should bend to present conventions of social media, especially when those conventions - already clearly of a lowest-common-denominator variety - are in a race to the bottom. A case in point is the growing prevalence of 'stickers' on Facebook in lieu of any verbal comment whatsoever - despite the fact that (1) the meaning of the 'sticker' is generally unclear or, at least, ambiguous; and (2) it would take longer even to choose an 'appropriate' sticker for a given circumstance than simply to write a brief comment.

There's something passive-aggressive about that whole phenomenon, if you ask me, and we are rapidly dumbing down to a pre-verbal state in which, carried to a logical conclusion, grunts and gestures will replace actual words because so many people just cannot bother to communicate clearly or lack the ability to do so. It is not, in my opinion, something to give into because that's the world we live in. We can, and we should aspire to, do better than that.

I, too, would like a 'dislike' button for this entire thread. Consider that the original poster has not returned with an answer to what exactly would be meant by the concept of 'good' in this 'voting' context - never mind any further examination of how 'goodness' would be measured quantitatively or qualitatively.

In reply to by [DELETED] 448831

Whoa I didn't expect that this thread will emerge so big!
Firstly I want to point that I just need a lot of time to unerstand your posts, sorry, I'm not native english speaker.
I'll give you and example.
I'm looking for a piece to play on my exam, it can be solo or any conterporary music. I'm checking percussion instruments since it's my main instrument that I play on. I spend two hours on searching, then I realize that I think more than 50% of all these pieces are just crap. For player. Either they cannot be played properly according to the composers writing, some other have many mistakes (harmony etc.), some are just so so so basic that I'm actually not interested in playng them. Of course a beginner would like it, but it's another issue.
Let's say I found an interesting piece, I download/buy, practice, have a little conversation with composer about modification the notes, interpretation etc. Then I rate it. I give it 5/5. Why? Because I like it, simple. And of course there is a possibility to write a comment explaining the reason of my rating.

I think it works with android app store, people rate apps and they go up so that more & more can try them. I think it works on sites like Reddit and many more. I buy a product on amazon... and I rate it! I rate Amazon as very good online shop, I rate mu product. And there's everywhere a possibility to write few words why does one give a product this note.

I think the players are the best evaluators of compositions, because it's us who played tons of them during entire life and we've seen a lot of various approaches to composition, from typography, to harmony, ease of reading and many many more. I don't think that some random music listener would register on musescore.org, download music then give it 1/5 rate because he didn't like it.

In reply to by kujaw

kujaw, I sincerely appreciate that you have given a lengthy explanation of the motivations and goals behind your initial post in this topic. For what it's worth, the only statement you made with which I would disagree is: 'But I think that any, any evaluation type is better that none.' Otherwise, in my opinion, you've expressed valid concerns and made valuable points.

Also, please don't concern yourself with any perceived language barrier because English isn't your native language. I think you've expressed yourself very well and very clearly. I hope you'll continue to do so and give further input if this subject continues to be explored.

In reply to by kujaw

@kujaw thx for taking the time to explain your case.

"I think the players are the best evaluators of compositions". I concur with this. My hope is that we will be able to evaluate compositions based on the time spent/played/practised/performed on these scores via the MuseScore mobile apps. That would be the ultimate metric.

In reply to by Miwarre

Popularity, I guess. I still don't see that as a terrible thing to be able to measure, as long as people take it with an appropriate grain of salt.

As it is, we can already measure this some extent via number of views, and comments; also potentially plays and downloads. Seems the mobile app could definitely be reporting some more interesting statistic than any of these.

In reply to by Marc Sabatella

A mobile app wound be a miracle! But even on the website please. Because I will not spend amounts of hours digging to find some nice sheet music. It's worthless.

I think that we should make a direct answer on questions: Why do you, composers, compose and upload you sheets on this website? What is the main reason? For who do you compose? What are your goals, why are you doing this?

In reply to by kujaw

@kujaw: although your personal concern is that of a performer looking at composers' scores, we should keep in mind that not all who use MuseScore to create scores are composers. Some (perhaps many) of us use MuseScore instead to re-typeset historical scores or to create transcriptions and arrangements, and there are probably other uses for the software as well of which I'm not aware.

Any and all of us may be uploading scores to to musescore.com. I have not, but that's because I've been actively using Trunk builds leading toward 2.0 ever since I found them stable enough for my own purposes. I've been living dangerously, of course, but very little of my work is in 1.3.

My only point here is to clarify and remind that the MuseScore community can't be reduced to 'composers' and 'performers'. We're more diverse than that simple binary, and any proposed score rating system needs to take that into account (to the extent that parameters and metrics may vary accordingly - specifically 'Goals' and possibly others as well).

Many people seem to be against a voting system because of the difficulty distinguishing between negative views such as "I don't like this type of music" vs. "I don't like this particular piece" vs. "I don't like the way it is written". Similar problems occur all the time on shopping websites where people don't know whether to rate the product or the service.

Perhaps the solution is to go down the Facebook route of only allowing positive feedback (i.e. have a "like" button but not a "dislike" button), or by reporting some additional stats like "number of times favourited", "number of return visits" or "number of downloads". This would provide a more objective way of measuring the popularity of a piece than resorting to a 5 star rating system or similar.

If it is decided to go down the 5 star route, it would be great if some kind of versioning system could be implemented to take into account changes made in response to feedback. It would be frustrating for users to have negative reviews based on something that they have subsequently gone to the trouble of correcting and improving.

In reply to by shoogle

A single rating system (our of 5 stars), is used on various sites around the web, but is it fair.
For musical scoring, there is more to it that just a black or white like or dislike.
To cover all aspects of an uploaded piece, possibly an average of multiple ratings ranging from the way the piece was written (asthetically pleasing or not), how does it sound, ignoring any soundfont anomolies between the creators system to the online system.
Does there appear to be any effort put into the composition, too often, I have noticed a number of uploads that appear to be just raw midi imports that have not been altered to what most musicians would be expecting to see.
Then there is, how do you rate an original composition against a transcription? most would rate it based on their own likes and dislikes.
As has been mentioned before, because a piece of music is popular, does not mean it is good.
A French research team has "proven" that today's Pop music has been dumbed down over the last few decades, but it is still popular.
So, how will a hip hop artist rate a classical piece, a jazz artist rate a country composition, and vice versa.
The favourites button gives, IMO, adaquate ratings.

Do you still have an unanswered question? Please log in first to post your question.