Metadata Encoding in MEI for Spain's FMT Project

[martha-thomae]

The Fondo de Musica Tradicional (FMT) for folk Spanish songs would like to consult how to encode the metadata stored in their database in MEI in the best way. Here I provide the metadata fiels in their database and then the proposal for its encoding.

The metadata is given in by the following fields:

• Title [Song title]

• Text incipit [Song’s first words]

• Music incipit [A numeric incipit of first eight/ten intervals by semitones: + (ascending) – (descending), = (repeated note); C-D is +2; F-E is -1]

• Starting pitch [Do, Re, etc.]

• Final pitch [Do, Re, etc.]

• Ambitus/range [1, 2, 3, 4, 5 etc.]

• Instrumental / Vocal / Vocal and instrumental

• Genre (Can have a taxonomy, see example below with: Canto > Cancion religiosa > Cuancion de Cuaresma > Semana Santa > Jueves Santo)

  

• Informant [person who sang the piece]

• Location [Autonomous community—Province—Location]
  (Also can have a taxonomy as seen below: Other countries > Americas y Caribe > Cuba > La Habana)

  

• Source [Mission / Concurso; Example: MISIÓN M16]

• Piece ID [Mission number + piece number; Example: M16-046]

• Year [Year in which the piece was collected]

• Researcher [Name of person who collected the piece; last names, first name]

• Language

• Musical instruments

• Collaborating person [all the people who collaborate in any way, students, etc.]

• Collaborating institution [institutions of collaborating people]

• Transcriber [person who transcribes/codifies the piece]

• Submission author [person/s who introduced data]

• URL of the piece in Fondo de Música Tradicional IMF-CSIC

Encoding proposal:

I included some quewstions for the Metadata IG in the encoding. These can be found by doing a search for ASK METADATA IG:

<?xml version="1.0" encoding="UTF-8"?>
<?xml-model href="https://music-encoding.org/schema/5.0/mei-all.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="https://music-encoding.org/schema/5.0/mei-all.rng" type="application/xml" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<mei xmlns="http://www.music-encoding.org/ns/mei" meiversion="5.0">
    <meiHead>
        <fileDesc>
            <titleStmt>
                <title>[Song title]</title>
                <author> <!--to be converted into "creator" | Documentation reads: The name of the creator of the intellectual content of a non-musical, literary work-->
                    <!--ASK METADATA IG: (A) What is it meant by creator of the "intellectual content of a non-musical, literary work"?-->
                    <corpName>Fondo de Música Tradicional IMF-CSIC (FMT)</corpName>
                </author>
                <editor><!--Documentation reads: The name of the individual(s), institution(s) or organization(s) acting in an editorial capacity-->
                    <persName>Emilio Ros-Fábregas</persName>
                </editor>
                <respStmt>
                    <!--Responsibility Statement: Transcription of text that names one or more individuals, 
                        groups, or in rare cases, mechanical processes, responsible for creation, realization, 
                        production, funding, or distribution of the intellectual or artistic content-->
                    <persName role="informant">[PERSON WHO SANG THE PIECE]</persName>
                    <persName role="researcher">[PERSON WHO COLLECTED THE PIECE]</persName>
                    <persName role="transcriber">[PERSON WHO TRANSCRIBES THE PIECE]</persName>
                    <persName role="submission_author">[PERSON(S) WHO INTRODUCED THE (meta)DATA]</persName>
                </respStmt>
                <contributor><!--Documentation reads: Names of individuals, institutions, or organizations 
                    responsible for contributions to the intellectual content of a work, where the specialized 
                    elements for authors, editors, etc. do not suffice or do not apply-->
                    ["COLLABORATING PERSON" and "COLLABORATING INSTITUTION"]
                    <persName>[COLLABORATING PERSON]</persName>
                    <corpName>[ITS COLLABORATING INSTITUTION]</corpName>
                </contributor>
                <sponsor><!--Documentation reads: Names of sponsoring individuals, organizations or institutions. 
                    Sponsors give their intellectual authority to a project; they are to be distinguished from 
                    funders, who provide the funding but do not necessarily take intellectual responsibility-->
                    Institución Milá y Fontanals de Investigación en Humanidades—Consejo Superior de Investigaciones Científicas IMF-CSIC, Barcelona
                </sponsor>
            </titleStmt>
            <pubStmt/>
        </fileDesc>
        <workList>
            <work target="https://musicatradicional.eu/piece/12299" targettype="url"><!--ASK METADATA IG: (B) URL here is ok??? -->

                <!-- 3.6.1 WORK IDENTIFICATION: -->
                <identifier>M16-046</identifier> <!--Mission number + piece number-->
                <title>[Title of a bibliographic entity]</title> 
                <!--ASK METADATA IG: (C) SHOULD EVERYTHING IN TITLESTMT BE HERE OR IN TITLESTMT? and
                    (D) WHAT IS THE IDIFFERENCE BETWEEN THIS TITLE AND THE ONE IN TITLESTMT?-->

                <!-- 3.6.2 INCIPITS: -->
                <incip>
                    <incipText>
                        <p>[Only text incipit]</p>
                    </incipText>
                    <!--ASK METADATA IG: (E) We would like to keep the "music incipit" as in the database, as a sequence of semitones. Can this be done ???-->
                    <!--ASK METADATA IG: (F) And What is the best format from the three options: parsons, plaineAndEasie, and humdrumKern ???-->
                    <incipCode form="humdrumKern">
                        [Music incipit (with or without text)]
                    </incipCode><!--(There is also <score> for MEI incipit.)-->
                </incip>

                <!-- 3.6.5 WORK'S CREATION HISTORY -->
                <creation>
                    <date>[year]</date>
                    <!--ASK METADATA IG: (G) Where to add the information of the mission that recopilated this piece? Will <corpName> within <creation> work?-->
                    <corpName><!--Identifies an organization or group of people that acts as a single entity--></corpName> 

                    <!--ASK METADATA IG: (H) Would this work for the taxonomy of location?-->
                    <bloc>
                        <bloc type="big_category">Other countries</bloc>
                        <bloc type="small_category">Américas y Caribe</bloc>
                        <country>Cuba</country>
                        <region type="autonomous_community">[Autonomous community]</region>
                        <region type="province">[Province]</region>
                        <settlement>La Habana</settlement>
                    </bloc>
                    <!--bloc: Contains the name of a geopolitical unit consisting of two or more nation states or countries | 
                        region: Administrative unit, smaller than country but larger than settlement | 
                        settlement: city / town / village-->

                    <!--ASK METADATA IG: (I) Would this work for the taxonomy of 'genre'?-->
                    <styleName>
                        <styleName n="1">Canto</styleName>
                        <styleName n="2">Canción religiosa</styleName>
                        <styleName n="3">Canción de Cuaresma</styleName>
                        <styleName n="4">Semana Santa</styleName>
                        <styleName n="5">Jueves Santo</styleName>
                    </styleName>
                </creation>

                <!-- 3.6.6 LANGUAGE USAGE -->
                <langUsage>
                    <language>Catalán</language>
                </langUsage> 

                <!-- 3.6.7 PERFORMANCE MEDIUM -->
                <perfMedium>
                    <perfResList>
                        <head>[Vocal and Instrumental / Vocal / Instrumental]</head>
                        <perfRes>Soprano</perfRes>
                        <perfRes>Guitar</perfRes>
                    </perfResList>
                </perfMedium>

                <!--ASK METADATA IG: (J) Is the following ok?-->
                <notesStmt><!--Missing information from database: starting pitch, final pitch, ambitus-->
                    <!--Went for these to keep solfedge:-->
                    <annot type="starting_pitch">[Do, Re, etc.]</annot>
                    <annot type="final_pitch">[Do, Re, etc.]</annot>
                    <!--Went for this instead of <ambitus> in <staffDef> because ambitus is defined in the 
                        FMT database as a number of semitones rather than the lowest and heights <ambNote>:-->
                    <annot type="ambitus">[Number of semitones]</annot>
                </notesStmt>

            </work>
        </workList>
    </meiHead>
    <music/>
</mei>

[gucl-mu]

Thanks Martha for the encoding proposal, we have a lot to discuss tomorrow! Just a quick question first, maybe I missed it in Slack. Is the plan to continue processing/editing the data in MEI or is it more about archiving the data?

[ahankinson]

Are you considering using MerMEId for this project @martha-thomae ?

[martha-thomae]

Hi Andrew! Regarding your first question, it is just for archiving.

[ahankinson]

That wasn't me. 😁

[martha-thomae]

Sorry!

Hi @gucl-mu! Thank you for your feedback. Regarding your question, it is just for archiving.

[martha-thomae]

Now @ahankinson, they were thinking on using a script to parse the data from the database and store it in the MEI files they will generate from the transcriptions they have already in MuseScore.

I haven’t used MerMEId before. Would that be better?

They already have all this data in the database, so they just need to retrieve it and put it in the appropriate elements of the MEI files.

[ahankinson]

I was just thinking that a good place to start would be to look through the other work catalogues that exist, like the Nielsen catalogue, to see how they do similar things.

It was a short leap then to wondering if MerMEId was on the radar, since quite a few of those catalogues used MerMEId, but if you already have the data I'm not sure you would need that.

[pe-ro]

@martha-thomae, Here's my version of your example. Like the Nielsen and other catalogs created with MerMEId, it takes a FRBR approach.

<?xml-model href="https://music-encoding.org/schema/5.1/mei-all_anyStart.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="https://music-encoding.org/schema/5.1/mei-all_anyStart.rng" type="application/xml" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<mei xmlns="http://www.music-encoding.org/ns/mei" meiversion="5.1+anyStart">
  <meiHead>
    <fileDesc>
      <titleStmt>
        <title>[Song title] : an electronic transcription</title>

        <!-- <author> will be renamed to <creator> -->
        <author>
          <corpName>Fondo de Música Tradicional IMF-CSIC (FMT)</corpName>
        </author>
        <editor>
          <persName>Emilio Ros-Fábregas</persName>
        </editor>
        <respStmt>
          <!-- Informant can go here and/or with the expression below and/or in
            expression/creation. -->
          <persName role="researcher">[PERSON WHO COLLECTED THE PIECE]</persName>
          <persName role="transcriber">[PERSON WHO TRANSCRIBES THE PIECE]</persName>
          <persName role="submission_author">[PERSON(S) WHO INTRODUCED THE
            (meta)DATA]</persName>
        </respStmt>
        <contributor>
          <persName>[COLLABORATING PERSON]</persName>
          <corpName>[ITS COLLABORATING INSTITUTION]</corpName>
        </contributor>
        <sponsor>Institución Milá y Fontanals de Investigación en Humanidades—Consejo
          Superior de Investigaciones Científicas IMF-CSIC, Barcelona</sponsor>
      </titleStmt>
      <pubStmt>
        <!-- Publication info goes here. -->
      </pubStmt>
    </fileDesc>

    <!-- Taxonomies go in encodingDesc/classDecls. -->
    <encodingDesc>
      <classDecls>
        <!-- Expression taxonomy -->
        <taxonomy>
          <category xml:id="Canto">
            <label>Canto</label>
          </category>
          <category xml:id="CanciónReligiosa">
            <label>Canción religiosa</label>
          </category>
          <category xml:id="CanciónDeCuaresma">
            <label>Canción de Cuaresma</label>
          </category>
          <category xml:id="SemanaSanta">
            <label>Semana Santa</label>
          </category>
          <category xml:id="JuevesSanto">
            <label>Jueves Santo</label>
          </category>
        </taxonomy>
        <!-- Performance medium taxonomy -->
        <taxonomy>
          <category xml:id="Vocal">
            <label>Vocal</label>
          </category>
          <category xml:id="Instrumental">
            <label>Instrumental</label>
          </category>
          <category xml:id="VocalAndInstrumental">
            <label>Vocal and Instrumental</label>
          </category>
        </taxonomy>
      </classDecls>
    </encodingDesc>
    <workList>

      <!-- In FRBR, a work is an abstract entity. Not much can be said about it,
        especially if it's a folk song. In this case, only the title is recorded. -->
      <work target="https://musicatradicional.eu/piece/12299">
        <!-- What is the purpose of the URL in @target? -->
        <title>[Song title]</title>

        <expressionList>
          <!-- In FRBR, a particular recording/version of the work is an expression.
            There's a lot more information here since the focus is on the
            recording/performance. -->
          <expression>
            <identifier>M16-046</identifier>
            <title>[Song title]</title>
            <!-- <author> will be renamed to <creator> -->
            <author>
              <persName role="informant">[PERSON WHO SANG THE PIECE]</persName>
            </author>
            <incip>
              <incipText>
                <p>[Only text incipit]</p>
              </incipText>
              <!-- The values "humdrumKern", "parsons", and "plaineAndEasie" are
                suggestions. Other values are allowed. -->
              <!-- See https://www.humdrum.org/Humdrum/commands/mint.html for 
              documentation of mint representation. -->
              <incipCode form="humdrumMint">[Music incipit without text]</incipCode>
            </incip>

            <!-- Creation in this context captures information about the creation of 
              the performance, not the work. Again, we probably don't know anything
              about the creation of a folk song. -->
            <creation>
              <!-- The children of <creation> should each have @label attributes that 
                explain their function in this context. -->
              <persName label="Performer">[PERSON WHO SANG THE PIECE]</persName>
              <date label="Date recorded">[year]</date>
              <!--ASK METADATA IG: (G) Where to add the information of the mission that 
              recopilated this piece? Will <corpName> within <creation> work?-->
              <!-- What does "mission" mean? Same thing as archaeological "campaign"? -->
              <!-- What does "recopilated" mean? "recorded", "collected"? -->
              <geogName label="Place recorded">
                <!-- Is "Other countries" necessary? Is this a synonym for "World"? -->
                <!--<bloc type="big_category">Other countries</bloc>-->
                <!-- Use @type="continent" instead of "small_category"? -->
                <bloc type="small_category">Américas y Caribe</bloc>
                <country>Cuba</country>
                <region type="autonomous_community">[Autonomous community]</region>
                <region type="province">[Province]</region>
                <settlement>La Habana</settlement>
              </geogName>
            </creation>
            <langUsage>
              <language>Catalán</language>
            </langUsage>
            <perfMedium>
              <perfResList>
                <!-- IMO the values of <head> are another kind of classification, and
                  so should have terms assigned in <classification> below. The terms
                  "vocal and instrumental", "vocal", and "instrumental" can also be
                  defined/described in <classDecls>. -->
                <!--<head>[Vocal and Instrumental / Vocal / Instrumental]</head>-->
                <perfRes>Soprano</perfRes>
                <perfRes>Guitar</perfRes>
              </perfResList>
            </perfMedium>

            <!-- Use only the terms here that apply to this song/expression. -->
            <!-- The full taxonomy is recorded in encodingDesc/classDecls above. -->
            <classification>
              <termList>
                <!-- When @class is present, the term value can be omitted. -->
                <term class="#CanciónReligiosa"/>
                <term class="#SemanaSanta">Semana Santa</term>
                <term class="#JuevesSanto">Jueves Santo</term>
                <term>vocal and instrumental</term>
              </termList>
            </classification>

            <!-- Use <extMeta> instead of <noteStmt> for non-MEI metadata. -->
            <extMeta>
              <!-- The top-level element here can be anything you want, but it
                must have the @xmlns attribute, which should be one that the
                institution controls. -->
              <foo xmlns="https://musicatradicional.eu/">
                <startPitch>[Do, Re, etc.]</startPitch>
                <finalPitch>[Do, Re, etc.]</finalPitch>
                <ambitus>[1, 2, 3, 4, ...]</ambitus>
              </foo>
            </extMeta>

          </expression>
        </expressionList>
      </work>
    </workList>
  </meiHead>
  <music>
    <!-- A link to a digital recording can go in performance/recording. -->
    <performance>
      <recording>
        <avFile target="URL_to_digital_recording"/>
      </recording>
    </performance>
  </music>
</mei>

I look forward to a lively discussion tomorrow.

[pe-ro]

And here's a non-FRBR version that uses several modifications not yet available in MEI 5.1.

<?xml-model href="../../Downloads/mei-all_anyStart_DEV_physDesc_imprint_monogr_biblStruct_repository.rng" type="application/xml" schematypens="http://relaxng.org/ns/structure/1.0"?>
<?xml-model href="../../Downloads/mei-all_anyStart_DEV_physDesc_imprint_monogr_biblStruct_repository.rng" type="application/xml" schematypens="http://purl.oclc.org/dsdl/schematron"?>
<mei xmlns="http://www.music-encoding.org/ns/mei" meiversion="5.1+anyStart">
  <meiHead>
    <fileDesc>
      <titleStmt>
        <title>[Song title] : an electronic transcription</title>

        <!-- <author> will be renamed to <creator> -->
        <author>
          <corpName>Fondo de Música Tradicional IMF-CSIC (FMT)</corpName>
        </author>
        <editor>
          <persName>Emilio Ros-Fábregas</persName>
        </editor>
        <respStmt>
          <!-- Informant can go here and/or with the expression below and/or in
            expression/creation. -->
          <persName role="researcher">[PERSON WHO COLLECTED THE PIECE]</persName>
          <persName role="transcriber">[PERSON WHO TRANSCRIBES THE PIECE]</persName>
          <persName role="submission_author">[PERSON(S) WHO INTRODUCED THE
            (meta)DATA]</persName>
        </respStmt>
        <contributor>
          <persName>[COLLABORATING PERSON]</persName>
          <corpName>[ITS COLLABORATING INSTITUTION]</corpName>
        </contributor>
        <sponsor>Institución Milá y Fontanals de Investigación en Humanidades—Consejo
          Superior de Investigaciones Científicas IMF-CSIC, Barcelona</sponsor>
      </titleStmt>
      <pubStmt>
        <!-- Publication info goes here. -->
      </pubStmt>

      <sourceDesc>
        <source type="recording">
          <biblDesc>
            <identifier>M16-046</identifier>
            <titleStmt>
              <title>[Song title]</title>
              <!-- <author> will be renamed to <creator> -->
              <author>
                <persName role="informant">[PERSON WHO SANG THE PIECE]</persName>
              </author>
            </titleStmt>
            <physDesc>
              <incip>
                <incipText>
                  <p>[Only text incipit]</p>
                </incipText>
                <!-- The values "humdrumKern", "parsons", and "plaineAndEasie" are
                suggestions. Other values are allowed. -->
                <!-- See https://www.humdrum.org/Humdrum/commands/mint.html for 
              documentation of mint representation. -->
                <incipCode form="humdrumMint">[Music incipit without text]</incipCode>
              </incip>
            </physDesc>

            <!-- Creation in this context captures information about the creation of 
              the performance, not the work. Again, we probably don't know anything
              about the creation of a folk song. -->
            <creation>
              <!-- The children of <creation> should each have @label attributes that 
                explain their function in this context. -->
              <persName label="Performer">[PERSON WHO SANG THE PIECE]</persName>
              <date label="Date recorded">[year]</date>
              <!--ASK METADATA IG: (G) Where to add the information of the mission that 
              recopilated this piece? Will <corpName> within <creation> work?-->
              <!-- What does "mission" mean? Same thing as archaeological "campaign"? -->
              <!-- What does "recopilated" mean? "recorded", "collected"? -->
              <geogName label="Place recorded">
                <!-- Is "Other countries" necessary? Is this a synonym for "World"? -->
                <!--<bloc type="big_category">Other countries</bloc>-->
                <!-- Use @type="continent" instead of "small_category"? -->
                <bloc type="small_category">Américas y Caribe</bloc>
                <country>Cuba</country>
                <region type="autonomous_community">[Autonomous community]</region>
                <region type="province">[Province]</region>
                <settlement>La Habana</settlement>
              </geogName>
            </creation>
            <langUsage>
              <language>Catalán</language>
            </langUsage>
            <perfMedium>
              <perfResList>
                <!-- IMO the values of <head> are another kind of classification, and
                  so should have terms assigned in <classification> below. The terms
                  "vocal and instrumental", "vocal", and "instrumental" can also be
                  defined/described in <classDecls>. -->
                <!--<head>[Vocal and Instrumental / Vocal / Instrumental]</head>-->
                <perfRes>Soprano</perfRes>
                <perfRes>Guitar</perfRes>
              </perfResList>
            </perfMedium>

            <!-- Use only the terms here that apply to this song/expression. -->
            <!-- The full taxonomy is recorded in encodingDesc/classDecls above. -->
            <classification>
              <termList>
                <!-- When @class is present, the term value can be omitted. -->
                <term class="#CanciónReligiosa"/>
                <term class="#SemanaSanta">Semana Santa</term>
                <term class="#JuevesSanto">Jueves Santo</term>
                <term>vocal and instrumental</term>
              </termList>
            </classification>
          </biblDesc>
        </source>
      </sourceDesc>

    </fileDesc>

    <!-- Taxonomies go in encodingDesc/classDecls. -->
    <encodingDesc>
      <classDecls>
        <!-- Expression taxonomy -->
        <taxonomy>
          <category xml:id="Canto">
            <label>Canto</label>
          </category>
          <category xml:id="CanciónReligiosa">
            <label>Canción religiosa</label>
          </category>
          <category xml:id="CanciónDeCuaresma">
            <label>Canción de Cuaresma</label>
          </category>
          <category xml:id="SemanaSanta">
            <label>Semana Santa</label>
          </category>
          <category xml:id="JuevesSanto">
            <label>Jueves Santo</label>
          </category>
        </taxonomy>
        <!-- Performance medium taxonomy -->
        <taxonomy>
          <category xml:id="Vocal">
            <label>Vocal</label>
          </category>
          <category xml:id="Instrumental">
            <label>Instrumental</label>
          </category>
          <category xml:id="VocalAndInstrumental">
            <label>Vocal and Instrumental</label>
          </category>
        </taxonomy>
      </classDecls>
    </encodingDesc>
  </meiHead>
  <music>
    <!-- A link to a digital recording can go in performance/recording. -->
    <performance>
      <recording>
        <avFile target="URL_to_digital_recording"/>
      </recording>
    </performance>
  </music>
</mei>

[martha-thomae]

Thanks, @pe-ro !

Am I correct on guessing that it is best to always follow FRBR? I am not a metadata person, so I would prefer to follow your advice in this regard.

I am answering a few of the questions in the comments here:

• Yes, a "misión" is like an archelogical campaign
• And "recopilated" was supposed to be "collected" (sorry, my Spanish-speaking brain thought that "recopilated" was an actual word in English)
• Yes, "Other countries" would be like "World." They mostly assume the country is Spain so they usually don't have to give any bigger categories than country, except when the country isn't Spain. In the case of the latter, then they have this "Other countries" category and subcategories like the "Americas and the Caribbean" (Americas y Caribe).

And have a few questions of my own about the taxonomies (I am not sure I am getting the idea behind them):

1. The three categories defined in the "Performance medium taxonomy" (vocal, instrumental, and vocal and instrumental) are supposed to be mutally exclusive. While the categories defined in the "Expression taxonomy" (Canto, Cancion religiosa, Cancion de Cuaresma, Semana Santa, and Jueves Santo) are supposed to be subsets of one another—with Canto being the largest class in this case and Jueves Santo the most specific one. So, I think I am not fully understanding how to use taxonomies, arent they supposed to define an object in a similar way as in biology (with kingdom, phylum, class, order, family, genus, species) being more and more specific? More like in the way the "Expression taxonomy" is working (rather than the "Performance medium taxonomy")?

2. So, based on the comments, in the encodingDesc/classDecls we need to include all the existing classes while in the classification/termList we only include the terms for this particular song, right? If this is the case the classDecls would be soooo big just for Expression alone. There are way more categories than the Canto, Cancion religiosa, Cancion de Cuaresma, Semana Santa, and Jueves Santo, these are just related to this particular expression. So am I understanding correctly that all the existing categories in the database for the type of expression should be collected in the encodingDesc/classDesc/taxonomy?

What I understand, based on the complete file you sent with the comments, is that in encodingDesc/classDesc/taxonomy one should include the full classes for each "thing" that has classes (regardless of them being mutually exclussive classes or subsets of one another). And that all classes should be there (even if they are hundreds of them), and then in the classification/termList part one has only the classes that apply to the given expression. But that would make the header that accompanies each file (encoding just one song) incredibly big as all classes of the database would be included in them inside the encodingDesc/classDesc. Am I understanding this correctly?

[martha-thomae]

Oh, and in:

<classification>
    <termList>
        <!-- When @class is present, the term value can be omitted. -->
        <term class="#CanciónReligiosa"/>
        <term class="#SemanaSanta">Semana Santa</term>
        <term class="#JuevesSanto">Jueves Santo</term>
        <term>vocal and instrumental</term>
    </termList>
</classification>

Is there a way to indicate that each of these terms are being more and more specific in the classification of the song?

[martha-thomae]

And here:

<!-- Use <extMeta> instead of <noteStmt> for non-MEI metadata. -->
<extMeta>
    <!-- The top-level element here can be anything you want, but it
must have the @xmlns attribute, which should be one that the
institution controls. -->
    <foo xmlns="https://musicatradicional.eu/">
        <startPitch>[Do, Re, etc.]</startPitch>
        <finalPitch>[Do, Re, etc.]</finalPitch>
        <ambitus>[1, 2, 3, 4, ...]</ambitus>
    </foo>
</extMeta>

The @xmlns points to a place where these elements are defined or something? What is its use? Just want to understand this part. I think I have never fully gotten this part, like why does the @xmlns for MEI points to https://music-encoding.org/ns/mei? What is in here? (it gives a 404 in the browser) sorry if I am asking something very dumb, I just have never gotten this part.

[ahankinson]

<term> can also contain <term>...

<classification>
    <termList>
        <!-- When @class is present, the term value can be omitted. -->
        <term class="#CanciónReligiosa">
             <term class="#SemanaSanta">Semana Santa
                   <term class="#JuevesSanto">Jueves Santo</term>
             </term>
       </term>
    </termList>
</classification>

But this is awkward because of the mixed (text + XML tag) content. You could maybe fix this with:

<term class="#SemanaSanta">
    <name>Semana Santa</name>
    <term class="#JuevesSanto">
         <name>Jueves Santo</name>
     </term>
</term>