I'm trying to help a scholar who'd like to create a website that can do two things:
1) Display a digitized songbook; and
2) Allow the user to see differences among multiple editions of this songbook.
As one might imagine, this songbook includes both musical notation and text. Currently, the images are (I believe) PDFs.
I'm thinking this sounds Juxta-ish, but can Juxta be deployed on the Web? What would you do?
Thanks!
The most complete way to represent common Western music notation digitally is with the MusicXML format:
http://www.recordare.com/musicxml
This format is supported by over 140 programs. If the songbook was created in the past 10 years, it most likely was created by a program like Finale or Sibelius that can export MusicXML files as well as create the PDF files.
There are a variety of web-based tools for displaying and playing MusicXML files within a web browser, including the free Myriad music plug-in, the Noteflight products, and the enterprise-strength Legato Sheet Music Viewer. Just being able to display and play music notation in the web is challenging enough. I don't know of any product at this point that can do differences in the way that musicians would expect to see them. Sure, you can use an XML differencing tool on the MusicXML files, but I don't think that would be very helpful for musicians.
Maybe some of the analysis toolkits using MusicXML could help, such as MIT's music21 software? Perhaps it could be hooked up with Juxta or other types of differencing tools?
You can find links to these different MusicXML software tools and lots more at:
http://www.recordare.com/musicxml/community/software
It would help to know more about the source material. What specific music are we talking about? What is the instrumentation? Are there digital source files available besides the PDF image files?
Good luck with your project!
Michael
Hi Michael -- and thanks! Unfortunately, I don't think this particular songbook is appropriate for MusicXML because the notation is so specialized. (It's a shape-note songbook.) However, I'm very glad to learn about MusicXML.
Having thought about it a bit, I guess a better way to frame my question might be: How would you go about displaying multiple sets of PDFs so that a user might most easily discern the differences among them? Perhaps the solution is just a page-turner plugin.
Thanks again.
The Juxta team received a Google DH grant to prepare a web version of the tool -- really, a set of services you could use for a project like this. I've pinged them and hope they can share a timeline or get in touch. Also, I'm not sure if MEI (the Music Encoding Initiative) markup supports shape notes, but you should know about that as well! I've dropped Perry Roland a line about your question. Cheers!
Arno Bosse
Member
I don't personally know of a online scholarly edition that offers structured musical comparisons in the same way a tool like Juxta allows one to visualize the differences between two texts (AFAIK the Online Chopin Variorum Edition http://www.ocve.org.uk/index.html probably comes closest to this). However you might want to refer to a useful list of online musical editions & resources put together by Joanna Swafford: http://www.nines.org/news/?p=1183 If nothing else, these should give you some further pointers to discover what's available in this area and how such editions are organized on a website.
On a related note, I've recently been researching text editions that offer comparison views of facsimiles/transcripts/translations. You might find some of these sites helpful as well. See for example the "Van Gogh Letters" http://vangoghletters.org/vg/ and "Die Fackel" (requires registration) http://corpus1.aac.ac.at/fackel/
Joanna Swafford
Member
Hi Miriam,
If you want to have any sort of markup, then I'd second Bethany's recommendation for MEI (Music Encoding Initiative). It's an XML music markup based on TEI standards that lets users track scholarly information as opposed to just formatting information (which is all that MusicXML can do). I think people can use it to track differences between editions, and it can also be used for more specialized notation. It was created by Perry Roland at UVa, and he's been very helpful answering all the questions I've had.
Here's the website for more information about MEI: http://www2.lib.virginia.edu/innovation/mei/
Here's a tutorial if you want to learn how to write MEI: http://music-encoding.org/documentation/tutorial
Obviously, using MEI would be more involved than putting some PDFs online and letting users spot the differences themselves, but it's a great option if you have the time.
Hope this helps, and good luck!
Replying to @Miriam Posner's post:
Thanks very much for your advice! It's enormously helpful. I've also had my eye on the Van Gogh letters as an example of how this might be done well, and I was interested to read about the project's software and components.
Thanks to your help, I feel I have a much clearer idea of the options I might present to this scholar. I'll let you know what we decide. I really appreciate it!
Miriam
pdr4h
Member
Hi Miriam,
MEI does indeed support shape-note notation.
It also allows the comparison of image regions without encoding the entire group of songs. In the following example --
<facsimile>
<surface>
<graphic xml:id="g1" xlink:href="A_1.jpg"/>
<zone xml:id="g1z175" ulx="120" uly="992" lrx="450" lry="290"/>
</surface>
<surface>
<graphic xml:id="g1" xlink:href="A_2.jpg"/>
<zone xml:id="g1z175" ulx="120" uly="992" lrx="450" lry="290"/>
</surface>
</facsimile>
surface elements are used for each page, graphic elements declare the image associated with the page, and zone elements define the regions of interest of the image. The attributes ulx, uly, lrx, and lry (upper-left-x, upper-left-y, lower-right-x, and lower-right-y, respectively) capture the image region's coordinates using the coordinate system of the image, similar to the way image maps are done. The @decls attribute is available on facsimile, surface, and graphic elements in order to associate these elements with metadata in the file header.
I don't believe this will work for PDFs so you'd have to convert them to some raster format.
However, having done this, features encoded in the markup (measures, notes, etc.) can be associated with one or more zones using the @facs attribute.
<measure xml:id="m2" n="2" facs="g1z175 someOtherRegion"/>
If the differences are all at the measure level, then this markup may be all you need. This provides enough information to locate, in this case, measure 2 in all the appropriate images and navigate between the various images and regions.
If you are interested in providing commentary about the markup features, then use the annot (annotation) element, like so --
<annot plist="m2">Measure 2 is really cool!</annot>
The @plist attribute contains the ID of the feature you want to comment on.
This is just a basic introduction. If you'd like more info, feel free to write to me at info (AT) music-encoding (DOT) org.
Good luck with your project!
Hey Miriam, Juxta is trying to roll out the web-service by the end of the summer. Right now in July there will be a JuxtaCamp where many folks interested in taking Juxta in this or that direction will get together to brainstorm and hack. The idea is to get a set of web-services that works for most. I don't think there is any talk yet of making it amenable to music comparison. I will bring it up, though! This would be a great use for it.
On an unrelated note, as soon as I read the question, my brain tricked me into believing somebody tweeted a diff for music project last week. For the life of me, I can't find that tweet. The sense that the tweet is tweal is too strong, though.
Douglas Knox
Member
People interested in this thread might want also to take a look at Aruspix, which is software for editors of early music prints. Aruspix does optical recognition of music and text, and has features to help with superimposition of images and collation of different editions. I have never used it and don't know the current state of its development, but saw an impressive presentation about it a year and a half ago. The optical music recognition (OMR) features at the time were specifically tuned to particular styles of early music printing, and it's more of a desktop tool for editors than a web presentation tool, so it's probably not directly suited to Miriam's question.
Hi all -- and wow! What great answers. I'm so interested to hear that MEI can handle shape-note notation, and I'm really pleased to hear that Juxta is continuing to expand. JuxtaCamp sounds awesome!
In case anyone's on the edge of her seat, I spoke to the scholar this afternoon and shared this information. He was very interested to learn about Juxta and the MEI. We decided, though, that the significance of the variances could be just as well explained by highlighting a few examples as by conducting a systematic, larger-scale analysis. That, combined with his timeline, led us to settle on a page-viewer plugin (either Archive.org's or Google's) plus WordPress or Omeka.
However, he's now thinking about ways he might work with others on developing Juxta for MEI. (Conveniently, he happens to have a tech background.) So I'll be sure he has your information in case you all might make a connection.
Thanks a million for your help!
Miriam
Hi Miriam,
MusicXML handles shape-note music as well, and has working software to exchange shape-note music between multiple programs that musicians use.
MEI has a lot of good ideas, but there is little or no software that uses it. This makes it difficult to use for practical projects such as web displays. It also restricts the usage of your markup to a small community that excludes performing musicians. Any performing musician who uses a notation program can use MusicXML files.
If you are on a tight timeline, image display of examples does seem to be your best bet. Good luck with your project!
Michael
You must log in to post.
A project of the Association for Computers and the Humanities and the Chronicle of Higher Education's ProfHacker
Icons by famfamfam
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 3.0 Unported License.
