{
  "id": "musical-notation",
  "title": "Musical Notation",
  "lang": "en",
  "source_lang": "en",
  "summary": "Musical notation comprises written and symbolic systems for representing musical sounds, performance actions, and relationships through time.",
  "category": "arts",
  "subcategory": "music",
  "type": "concept",
  "aliases": [
    "Music notation",
    "Musical writing"
  ],
  "generated_at": "2026-11-02",
  "model": "pro/gpt-6.1-sol",
  "url": "https://aiwiki.page/en/musical-notation/",
  "translations": {
    "zh-Hans": {
      "url": "https://aiwiki.page/zh/musical-notation/",
      "json": "https://aiwiki.page/zh/musical-notation.json"
    }
  },
  "sections": [
    {
      "heading": "",
      "text": "Musical notation is the representation of music through written, printed, tactile, or digitally encoded signs. Depending on the system, these signs may indicate pitches and durations, instrumental actions, words to be sung, relationships among performers, or rules for generating a performance. Notation makes it possible to preserve and communicate musical information beyond the immediate presence of a performer, although its meaning usually depends on conventions learned through listening and practice.\n\nNo single system represents every aspect of music equally well. The five-line staff used internationally is especially effective at coordinating specified pitches and rhythms, while tablatures describe instrumental actions and other systems record melodic outlines, scale degrees, or patterns of ornamentation. Notation has consequently developed in continuing interaction with oral tradition, instrument design, religious practice, composition, publishing, and technology. Its history is not a simple progression toward increasingly complete representation: different systems select different features of musical experience for different purposes."
    },
    {
      "heading": "Functions, representation, and musical knowledge",
      "text": "A notation is a system of meaningful distinctions rather than merely a picture of sound. A mark may stand for a particular pitch, a relative position within a scale, a duration, a finger placement, or an instruction such as “repeat.” Its meaning depends on a shared code. The same dot can have different functions according to its position: beside a notehead in staff notation it can lengthen duration, above it it can indicate detached articulation, and in some numbered systems it can identify an octave. Reading therefore involves interpreting relationships among signs rather than assigning an isolated sound to each symbol.\n\nA useful distinction separates notations that primarily describe sounding results from those that primarily prescribe actions. A staff note may specify a pitch without identifying which string or finger should produce it. A lute tablature can instead identify a string and fret, leaving the resulting pitch dependent on the instrument’s tuning. These approaches overlap: staff scores contain fingerings and bowing instructions, while tablatures may include precise rhythmic information. Neither is inherently more complete. Each distributes information differently between the written document, the performer’s training, and the performance situation.\n\nNotation can preserve a composition, assist memory, organize an ensemble, support teaching, or provide evidence for analysis. A full score places simultaneous parts in vertical alignment, allowing a conductor or scholar to inspect relationships that no single performer plays. An individual part removes most of that information to make performance easier. A jazz lead sheet may provide only a melody, chord symbols, and the words, while a laboratory transcription may attempt to document minute timing or pitch fluctuations in a particular recorded performance. These documents serve different purposes even when they concern the same piece.\n\nThe distinction between prescriptive and descriptive notation also matters in ethnomusicology. A transcription made to study a performance is not necessarily an instruction that members of the musical community would use. Staff notation can expose recurring patterns while obscuring features such as flexible intonation, timbral variation, or the relationship between music and movement. A carefully measured transcription may therefore be supplemented by recordings, verbal explanation, or alternative diagrams.\n\nMusical literacy is consequently not identical with musical competence. Highly skilled musicians may transmit complex repertories without written scores, and fluent readers may depend on oral guidance for style and interpretation. Notation externalizes selected knowledge; it does not replace the bodily, auditory, and social knowledge through which music is performed."
    },
    {
      "heading": "Ancient evidence and the limits of reconstruction",
      "text": "Some of the earliest surviving evidence for musical notation comes from ancient Mesopotamia. Cuneiform texts describe strings, intervals, and procedures for tuning stringed instruments. Old Babylonian sources from the early second millennium BCE provide evidence for a structured terminology of musical relationships. Such documents are important because they show how musical knowledge could be organized in writing, but a tuning instruction is not equivalent to a fully notated composition. Identifying the purpose of a tablet is therefore a necessary step before attempting to turn its signs into sound.\n\nThe Hurrian hymns, discovered at Ugarit in present-day Syria, include a substantially preserved hymn commonly designated h.6, dating to approximately the fourteenth century BCE. Its text is in Hurrian, while its musical instructions employ technical vocabulary associated with Mesopotamian traditions. The hymn addresses the goddess Nikkal. It is frequently described as the oldest substantially surviving notated song, but its interpretation remains disputed. Reconstructions differ over the meaning of interval names, the relationship between syllables and musical signs, and the organization of the accompaniment. Modern performances are scholarly realizations rather than unambiguous recoveries of an ancient original.\n\nIn ancient Greece, separate systems existed for vocal and instrumental notation. They used alphabetic or alphabet-like signs placed above words or arranged to represent musical passages. Surviving examples include the Delphic hymns, generally dated to the second century BCE, and the Seikilos epitaph, usually dated to the first or second century CE. The Seikilos inscription preserves a short song with its words, pitch signs, and rhythmic indications. Ancient theoretical writings, notably the tables associated with Alypius, help scholars interpret the symbols, although questions of tuning, vocal production, and performance practice remain.\n\nThe survival of these sources is highly uneven. Stone inscriptions, clay tablets, and a limited number of papyrus fragments endured under conditions that destroyed many other musical documents. A small surviving corpus therefore cannot establish how often notation was used, who could read it, or how much music circulated without writing. Nor does the disappearance of a particular script necessarily mean the disappearance of the musical practices once associated with it.\n\nAncient notation must also be distinguished from modern representations of ancient music. A modern staff transcription often fixes choices that the original source does not settle, including octave placement, rhythm, tuning, or instrumental realization. Responsible editions identify those decisions rather than presenting the transcription as a transparent substitute for the source. The history of notation thus concerns both surviving signs and the changing methods through which later readers interpret them."
    },
    {
      "heading": "Neumes, the staff, and medieval rhythmic writing",
      "text": "The principal ancestry of modern Western staff notation lies in the writing of Christian liturgical chant. From the ninth century, surviving manuscripts increasingly show neumes above sung texts. These signs commonly indicate melodic gestures, groupings, and aspects of delivery rather than specifying every interval independently. Early neumatic traditions varied by region, including those associated with St Gall and Laon. Their signs could convey information about articulation and melodic movement that later, more pitch-explicit systems did not always preserve.\n\nThe move toward precise pitch location involved arranging signs at different heights and adding reference lines. The eleventh-century theorist Guido of Arezzo is associated with influential teaching methods that combined staff-based pitch representation with syllables for learning melodic relationships. The syllables ut, re, mi, fa, sol, and la were drawn from successive phrases of the hymn Ut queant laxis. Guido did not single-handedly invent every component of the staff, but his writings helped explain and disseminate methods that reduced dependence on learning each chant entirely by imitation. Four-line staves subsequently became characteristic of much plainchant notation.\n\nThe addition of reference pitches and lines did not immediately solve the problem of rhythm. In much chant, musical pacing was related to text and melodic grouping rather than to continuously measured beats of equal duration. As written counterpoint became more elaborate, coordinating simultaneous parts created different requirements. The rhythmic modes used in twelfth- and thirteenth-century polyphony associated with Notre-Dame organized patterns of long and short values. Their interpretation depended partly on recurring configurations of note groups called ligatures.\n\nDuring the thirteenth century, mensural notation increasingly assigned durational significance to note shapes themselves. Franco of Cologne’s Ars cantus mensurabilis, written around the middle or later thirteenth century, is a major witness to this development. Even then, the duration of a sign could depend on its context. A long note might divide into two or three shorter units, and neighboring notes could alter how a value was understood. Mensural writing was therefore not simply modern notation in an older graphic style.\n\nFourteenth-century developments associated with the ars nova, including writings connected with Philippe de Vitry and Johannes de Muris, expanded ways of specifying binary and ternary divisions. Later medieval composers exploited these resources to coordinate intricate rhythmic structures. During the fifteenth century, hollow or “white” note shapes became common, and conventions continued to evolve through the Renaissance. Modern editions often translate this notation into bar lines, reduced note values, and familiar signatures, but doing so can conceal the original relationships among units of measure."
    },
    {
      "heading": "Pitch organization in modern staff notation",
      "text": "The modern musical staff normally consists of five parallel lines, with notes placed on lines or in spaces. Successive vertical positions represent successive letter names in a diatonic sequence rather than equal distances in frequency. Additional short ledger lines extend the range above or below the staff. Horizontal position generally corresponds to musical succession, although the spacing is adjusted for legibility and does not necessarily measure elapsed time proportionally.\n\nA clef assigns a reference pitch to a particular line. The treble clef identifies the G above middle C, the bass clef the F below middle C, and a C clef identifies middle C wherever its central feature is placed. Alto and tenor clefs are two placements of the C clef. Different clefs reduce the need for ledger lines and suit the ranges of particular voices or instruments. A brace commonly joins treble and bass staves into a grand staff for keyboard music, but the two staves are parts of one coordinated representation rather than independent pitch systems.\n\nA key signature places sharps or flats after the clef to establish recurring alterations. Accidentals attached to individual notes modify those defaults according to notational conventions; in ordinary modern usage their effect generally continues through the measure for the same pitch at the same octave. Naturals cancel an alteration, while double sharps and double flats provide additional chromatic distinctions. Editorial and historical practices differ, especially concerning courtesy accidentals and the scope of accidental signs in older sources.\n\nThe written distinction between, for example, F-sharp and G-flat can express different harmonic functions even when both are played with the same key on an equal-tempered piano. Under twelve-tone equal temperament, the frequency ratio of pitches separated by \\(n\\) semitones is\n\n\\[\\frac{f_2}{f_1}=2^{n/12}.\\]\n\nThis equation describes a tuning system, not an intrinsic property of the staff. The same notation has been used with unequal temperaments, flexible vocal intonation, and other tuning practices. Note names identify positions within a musical framework without always determining a unique frequency.\n\nWritten pitch may also differ systematically from sounding pitch. A B-flat clarinet ordinarily sounds a whole tone below its written notes, while a modern piccolo sounds an octave above. Transposing notation allows related instruments to share familiar fingering and reading patterns. Unpitched percussion, by contrast, can use a staff whose vertical positions identify instruments rather than frequency levels. Additional accidental symbols are employed in microtonal music, but their meanings require explicit conventions: the familiar five-line framework does not by itself establish how quarter-tones or other subdivisions should be tuned."
    },
    {
      "heading": "Duration, meter, and expressive instruction",
      "text": "Modern rhythmic notation represents proportional durations through noteheads, stems, flags, beams, and rests. A whole note conventionally equals two half notes, four quarter notes, or eight eighth notes. This hierarchy specifies relationships, not absolute lengths in seconds. A quarter note can be fast or slow depending on tempo, and the name “whole note” does not mean that the note necessarily occupies a complete measure. Its relationship to the measure depends on the prevailing meter and time signature.\n\nDots, ties, and tuplets modify the basic hierarchy. A single augmentation dot adds half the note’s value, while a second adds half the first dot’s addition. A double-dotted half note therefore lasts seven eighth notes. A tie joins successive notes of the same pitch into one sustained event; a slur, despite its similar curved appearance, normally indicates phrasing or connected articulation. Tuplet indications specify a subdivision different from the surrounding norm, as when three eighth notes occupy the time normally taken by two.\n\nTime signatures organize recurring groupings. In \\(3/4\\), a measure normally contains three quarter-note units; in \\(6/8\\), six eighth-note units are commonly grouped into two main beats, each divided into three. The signature alone does not fully determine accent or phrasing. A passage in \\(7/8\\), for example, may group its eighth notes as \\(2+2+3\\), \\(3+2+2\\), or another pattern. Beaming and accent marks can clarify these distinctions. Syncopation depends on expectations of metrical emphasis, not simply on the presence of notes between beats.\n\nAbsolute timing can be indicated by verbal directions or by a metronome mark. If the quarter note is set to 120 beats per minute, each quarter note has a nominal duration of \\(60/120=0.5\\) seconds. Expressive performance may nevertheless involve local timing adjustments, and terms such as ritardando, accelerando, and rubato authorize or imply flexibility. A fermata suspends normal continuation without specifying an exact multiplier. Music using proportional spacing, time brackets, or clock times adopts other ways of relating the page to elapsed time.\n\nDynamic and articulation signs add further layers. Piano and forte indicate relative softness and loudness, while crescendos and diminuendos indicate change. These are not fixed sound-pressure values. Dots, wedges, accents, and tenuto marks interact with instrumental technique and historical style. Ornament signs can likewise require substantial interpretation: a trill symbol may leave its starting note, speed, and ending to convention. The notated rhythm is thus only one component of a performance whose phrasing, attack, balance, and timbre are only partly specified."
    },
    {
      "heading": "Tablature, harmonic shorthand, and tactile reading",
      "text": "Tablature represents actions on an instrument rather than relying primarily on abstract pitch coordinates. In common guitar tablature, six horizontal lines correspond to strings and numbers identify frets. A zero indicates an open string. The diagram therefore answers a question that staff notation often leaves unresolved: where on the instrument a pitch should be produced. Since the same pitch can occur on several strings, a tablature can preserve choices that affect resonance, tone color, and the feasibility of a passage.\n\nEuropean lute tablatures developed in several forms. French and Italian systems used different combinations of letters or numbers and different conventions for ordering strings, while German lute tablature employed a more distinctive mapping of signs to positions. Keyboard tablatures also circulated, including alphabetic systems used in German-speaking regions. Surviving tablatures are evidence of instrumental technique as well as repertory. Their interpretation may require knowledge of historical tuning, string courses, ornament signs, and conventions for holding notes beyond their initially marked duration.\n\nHarmonic shorthand reduces a different kind of information. In basso continuo, a bass line is accompanied by figures indicating intervals to be supplied above it. The performer chooses voicing, spacing, and often aspects of texture. A figure 6, for example, commonly indicates a sonority including a sixth above the bass, with other intervals understood through convention and context. The notation is neither a complete chord-by-chord transcription nor an invitation to unrestricted invention; it presupposes harmonic knowledge and stylistic competence.\n\nModern chord symbols similarly identify harmonic resources without fully specifying their realization. A symbol such as Cmaj7 normally names a C-major chord with a major seventh, but does not dictate the distribution of its notes among registers or instruments. In popular music and jazz, musicians combine such symbols with a tune, words, or a formal plan. Roman-numeral analysis and number-based chart systems can represent harmonic function or scale degree, making transposition easier. These systems must be distinguished from numbered melody notation, where the numbers ordinarily designate individual melodic pitches.\n\nBraille music offers a tactile system rather than a raised imitation of a visual score. Building on Louis Braille’s six-dot cell, it uses combinations for note identity and duration, with additional signs for octave, accidentals, expression, and organization. Pitch and duration can be encoded together, so its logic differs from the separation of vertical pitch and horizontal time on a staff. Because both hands may be needed to perform an instrument, musicians often read and memorize sections before playing. A faithful transcription therefore involves decisions about structure and navigability, not merely replacing printed glyphs with tactile equivalents."
    },
    {
      "heading": "East Asian, South Asian, and other regional systems",
      "text": "Chinese musical traditions employ several distinct forms of notation. Gongche notation uses Chinese characters as pitch or scale-degree designations, accompanied in different traditions by rhythmic and other signs. It has served vocal and instrumental repertories, but its interpretation depends on local conventions, tuning, and performance practice. The presence of pitch names does not guarantee that every ornament or rhythmic nuance has been written. A reader trained in one regional tradition may need further instruction to interpret another tradition’s use of similar characters.\n\nThe guqin, a seven-string zither, developed a particularly detailed action-based notation. Early prose tablature describes performance procedures in words; the later abbreviated system, jianzipu, combines components indicating such matters as string, stopping position, and hand technique. It can distinguish actions involving stopped tones, harmonics, plucking, and sliding. Traditional scores often leave much rhythmic organization to interpretation and transmission. The dapu process of realizing an old score therefore includes reconstructive and creative judgments within historically informed practice, rather than simple decoding of fixed durations.\n\nNumbered musical notation, widely known in China as jianpu, uses the digits 1 through 7 for degrees of a scale. Dots above or below indicate octave displacement, while additional marks specify duration and rests. A heading can establish the pitch corresponding to 1. Its international history includes the eighteenth-century work of Jean-Jacques Rousseau and nineteenth-century French developments associated with Pierre Galin, Aimé Paris, and Émile Chevé. Its later adoption in East Asia illustrates how a system can be adapted to new educational and musical circumstances rather than remaining tied to its place of development.\n\nSouth Asian notation commonly uses syllables or letters corresponding to the scale degrees sa, ri/re, ga, ma, pa, dha, and ni. In Hindustani music, Vishnu Narayan Bhatkhande and Vishnu Digambar Paluskar developed influential notational approaches in the late nineteenth and early twentieth centuries. Such systems can record a composition’s outline and aspects of rhythmic organization, but the realization of a raga also depends on characteristic movements, intonation, ornament, and the relationship between composition and improvisation. Carnatic notation likewise works within teaching traditions in which the treatment of ornamented pitch cannot be reduced to a sequence of uninflected notes.\n\nOther traditions employ different combinations of pitch signs, mnemonic syllables, and time grids. Korean jeongganbo, developed in the fifteenth century, organizes musical information within a grid whose cells provide a temporal framework. Japanese shakuhachi and koto traditions use notations shaped by instrumental technique and lineage. Javanese gamelan musicians use forms of cipher notation, including kepatihan, that can represent a melodic framework without specifying the entire ensemble realization. These examples show that the unit recorded may be a structural part of music rather than every audible event."
    },
    {
      "heading": "Printing, scores, and the circulation of repertory",
      "text": "The material form of notation affects both its circulation and its use. Medieval music circulated in manuscripts ranging from modest working books to elaborate illuminated manuscripts. Large choirbooks allowed several singers to read from one volume, while other books gathered repertory for particular institutions or performers. Copying involved decisions about layout, abbreviations, text placement, and correction. A manuscript was not necessarily a neutral carrier of a stable work: successive copies could preserve revisions, local variants, or errors.\n\nMusical printing posed special technical problems because staff lines, noteheads, and text had to align. Ottaviano Petrucci’s Harmonice musices odhecaton A, published in Venice in 1501, was a landmark collection of polyphonic music printed from movable type. His method used multiple impressions to coordinate the elements of the page. In the sixteenth century, single-impression techniques associated especially with Pierre Attaingnant made it possible to print notes and portions of staff lines together, improving production efficiency even when the joins between type units remained visible.\n\nThe format of a publication shaped how music was encountered. Renaissance polyphony was often issued in separate partbooks, one for each voice, rather than in vertically aligned full score. Performers could read their own lines, but a reader could not inspect the entire texture at a glance. The expanding use of full scores supported different forms of composition, rehearsal, collecting, and study. A modern orchestral score arranges instrumental families in a conventional vertical order and uses aligned measures to coordinate their activity, while each player normally performs from a separate part.\n\nEngraving, especially on metal plates, became important because it allowed flexible positioning of musical signs and finely controlled spacing. Engravers had to balance horizontal density, vertical separation, text placement, and convenient page turns. These considerations remain central to musical typography. An accurate sequence of pitches and durations can still be difficult to perform from if beams obscure the beat, collisions hide accidentals, or a page turn interrupts a passage requiring both hands. Legibility is a functional component of notation, not an ornamental addition.\n\nPublishing also created new editorial questions. A composer’s autograph, a copyist’s manuscript, and a first printed edition may disagree. Later editors must distinguish correction, modernization, and interpretation, sometimes marking supplied dynamics or accidentals with brackets or different type. Editions of Johann Sebastian Bach or Ludwig van Beethoven can consequently differ without one being simply a transcription error. Copyright and other forms of intellectual property intersect with this history, but the legal status of a composition, an edition, a typographical layout, and a recording need not be the same."
    },
    {
      "heading": "Graphic scores, indeterminacy, and expanded musical materials",
      "text": "Twentieth-century composition brought increased attention to sounds and performance situations that conventional staff notation represented only awkwardly. Extended instrumental techniques, continuously changing textures, noise, electronic sound, and performer choice could require new symbols or a different organization of the page. Some composers retained the staff while supplementing it with diagrams and explanatory legends. Others treated the score as a field of visual relationships or as a set of verbal instructions.\n\nGraphic scores use shapes, lines, spatial arrangements, or images as musically meaningful elements. The interpretation may be tightly specified or deliberately open. Earle Brown’s December 1952, for example, presents black rectangles distributed across a page, with realization dependent on the performance approach adopted. Cornelius Cardew’s Treatise, composed between 1963 and 1967, extends graphic notation across 193 pages. Its circles, lines, numbers, and other figures do not come with a single comprehensive decoding key assigning each mark an invariant sound.\n\nIndeterminacy concerns what is left unsettled, and can operate at several levels. A composer may determine pitches but not their precise coordination, allow performers to choose an order of sections, or establish a process whose results vary. John Cage’s work made extensive use of unconventional notation and chance procedures, but chance in composition and freedom in performance are not identical. A score created through random operations may still prescribe a highly specific result, while a conventionally composed text score may grant performers substantial choice.\n\nNew notation can also increase rather than relax control. Krzysztof Penderecki’s Threnody to the Victims of Hiroshima, composed in 1960 for 52 string instruments, employs signs for clusters, unusual sound production, and durations organized in seconds. In other music, proportional notation relates horizontal distance to time, while time brackets define intervals within which events are to occur. A spatially organized page may therefore function as a precise temporal diagram rather than an invitation to free interpretation.\n\nNotation for electronic and mixed-media works raises further issues. A score may coordinate live performers with a fixed audio track, specify electronic processing, or identify cues for lighting and projected images. The essential document can include a patch, signal-flow diagram, software configuration, or technical description alongside a conventional instrumental part. Preserving such a work involves retaining the relationship among these materials. A page that names an obsolete device may preserve an instruction without preserving the conditions needed to realize it, making technical documentation part of the musical source."
    },
    {
      "heading": "Digital encoding, automatic recognition, and preservation",
      "text": "Digital systems distinguish the appearance of notation from the information encoded beneath it. A scanned score is an image of a page. A notation file can instead contain structured objects such as notes, rests, voices, measures, lyrics, and articulations. A PDF may preserve an excellent visual representation while offering little directly accessible musical structure. Conversely, a structured file may preserve the notes while failing to reproduce all of an edition’s spacing, symbols, or editorial distinctions.\n\nMIDI, introduced in 1983, was designed primarily to exchange performance and control messages among electronic devices. Its note messages identify events such as the start and end of a note, together with parameters including key number and velocity. That is not equivalent to a complete notation. MIDI alone does not reliably distinguish enharmonic spellings such as C-sharp and D-flat, and an expressive performance may not reveal a unique choice of note values, voices, or meter. Converting a performance into a readable score therefore requires interpretive quantization and musical analysis.\n\nMusicXML, developed beginning in 2000, provides an interchange format designed for common Western musical notation. It can encode relationships that a performance-oriented format may omit, including written pitch spelling, voices, lyrics, and many layout details. The Music Encoding Initiative provides another framework, especially important for scholarly representation of musical documents and historical sources. Such formats overlap in their ability to describe notation but differ in emphasis and implementation. Exchanging files between applications can still alter layout or lose features that one program represents differently from another.\n\nOptical music recognition attempts to extract structured musical information from images. The task is more complicated than recognizing a sequence of independent characters. A system must locate staves, separate overlapping objects, identify symbols, and infer relationships among noteheads, stems, beams, accidentals, and voices. The meaning of a symbol can depend on a distant clef or an earlier accidental. Handwriting, damaged paper, unusual engraving, and complex polyphony introduce further difficulties. Methods drawing on machine learning improve recognition in many settings, but output still requires validation appropriate to its intended use.\n\nDigital preservation must retain both readable representations and enough contextual information to interpret them. Proprietary file formats, unavailable fonts, external sound libraries, and software-dependent playback can all threaten future access. Stable images preserve evidence of a document’s appearance, while structured encodings enable searching, comparison, transposition, and computational analysis. These functions are complementary. A digital critical edition may therefore connect page images, encoded notation, editorial annotations, and recordings, allowing the reader to distinguish the historical source from its transcription and sounding realization."
    }
  ],
  "links": [
    "music",
    "oral-tradition",
    "musical-pitch",
    "jazz",
    "ethnomusicology",
    "ancient-greece",
    "mensural-notation",
    "tempo",
    "meter-music",
    "time-signature",
    "syncopation",
    "metronome",
    "rhythm",
    "timbre",
    "tablature",
    "basso-continuo",
    "popular-music",
    "musical-improvisation",
    "illuminated-manuscript",
    "printing",
    "typography",
    "johann-sebastian-bach",
    "ludwig-van-beethoven",
    "intellectual-property",
    "machine-learning"
  ],
  "sources": [],
  "license": "https://creativecommons.org/licenses/by/4.0/"
}