Survey of programming languages for sound and music

2021/03/19 SIGPX #8

Transcript

1. 2021-03-19 SIGPX #8 Tomoya Matsuura/দӜ஌໵ 

   https://matsuuratomoya.com me@matsuuratomoya.com
   ԻͱԻָͷͨΊͷ

   ϓϩάϥϛϯάݴޠͷ

   αʔϕΠ
   1
   

   View Slide

2. • ਆಸ઒ݝכϲ࡚ࢢग़਎(1994)
   
   • ਆಸ઒૯߹ߴߍଔ(2013)
   
   • ౦ژܳज़େֶԻָ؀ڥ૑଄Պଔ(2017)
   
   • ۝भେֶେֶӃܳज़޻ֶ෎ 

   म࢜/ത࢜ޙظ՝ఔ(2017,2019~)
   
   • ࿇পࣥଠϑϧϑΟϧ(2018~)
   
   • YCAM(Intern,2015),teamlab(2015~2017),

   School for Poetic Computation(2018)
   ࣗݾ঺հ
   2
   photo: Takehiro Goto
   

   View Slide

3. 3
   minimal musical medium / a programming language as an Infrastructure for sound & music.
   https://github.com/mimium-org/mimium
   3
   

   View Slide

4. • Իָϓϩάϥϛϯάݴޠͷ֓೦త੔ཧɺൺֱΛͨ͠ࢿྉ͕গͳ͍
   
   • ಛʹ࣮૷ऀ͔Βݟͨڞ௨ͷޠኮɺඞཁͳ஌͕ࣝڞ༗͞Ε͓ͯΒ
   ͣɺԻָͷͨΊͷݴޠΛ࡞Δϋʔυϧ͕ߴ͍ɺධՁ͢Δͷ͕೉͠
   ͍
   
   • ۩ମతࣄྫͷࡉ͔͍঺հ͸աڈͷαʔϕΠʹͳΔ΂͘೚ͤͯཧ࿦
   త੔ཧΛ͍ͨ͠
   
   • ࠓճ͸ಛʹϦΞϧλΠϜ৴߸ॲཧͰ͖ΔݴޠΛத৺ʹ࿩͠·͢
   ൃදࣗମͷ໨త
   4
   

   View Slide

5. • ԻָϓϩάϥϛϯάݴޠΛ࡞Δ໨త
   
   • ԻָϓϩάϥϛϯάݴޠͷաڈͷαʔϕΠ
   
   • ୅දతͳ؀ڥͷ঺հ
   
   • 2000೥Ҏޙͷݴޠઃܭʹ͓͚Δओཁͳ՝୊
   • ݴޠͷ࣮૷ύλʔϯͱΞʔΩςΫνϟ
   • ݴޠΛಛ௃͚ͮΔཁૉͱಛ௃ͷ෼ྨ
   • ·ͱΊɺະདྷͷԻָݴޠͷ͋Γํ
   ߏ੒
   5
   

   View Slide

6. ԻָʹίϯϐϡʔλʔΛ༻͍ΔϞνϕʔγϣϯ
   ԻָϓϩάϥϛϯάݴޠΛ࡞ΔϞνϕʔγϣϯ
   • ϋʔυ΢ΣΞૢ࡞ʢϝϞϦɺεϨουɺεέδϡʔϦϯάʣ

   ͷӅṭʢԻָՈ͸ϝϞϦͷ͜ͱΛؾʹͨ͘͠ͳ͍ʣ
   
   • ԋ૗/࡞ۂΠϯλʔϑΣʔεͱͯ͠ͷϓϩάϥϛϯάʢLive Codingʣ
   
   • ϝσΟ΢Ϝͱͯ͠ͷιʔείʔυʢ#SC140ɺMML΍ςΩετԻָαΫϥͳͲʣ
   
   • ϓϩάϥϛϯάڭҭͷπʔϧͱͯ͠(EarSketch,Sonic Pi)
   Իָϓϩάϥϛϯάݴޠͷ໨త
   6
   • ίϯϐϡʔλʔͰͳ͚Ε͹ෆՄೳͳදݱɺ৽͍͠දݱʢmicrosoundͳͲʣ
   
   • ݱ࣮ͷָثɺ؀ڥͷγϛϡϨʔτ(VSTɺ෺ཧϞσϦϯά߹੒)
   

   View Slide

7. • "Languages for Computer Music”,Frontiers in Digital Humanities(2018), Roger
   B. Dannenberg
   
   • ԻָݴޠΛಛ௃͚ͮΔ6ཁૉͷఏҊɺԻָݴޠҰൠʹڞ௨͢Δ՝୊ͷ੔ཧ
   
   • “Computer Music Languages and Systems: The Synergy Between Technology
   and Creativity”,pp 1-49 in Handbook of Digital Games and Entertainment
   Technologies(2016),Hiroki Nishino and Ryohei Nakatsu
   
   • “The Development of Computer Music Programming Systems”, Journal of New
   Music research,42(1), pp97-100(2013) Victor Lazzrini
   
   • ↑2ͭ͸ओʹྺ࢙త؍఺ͷ੔ཧ
   
   • “Idiomatic Patterns and Aesthetic Influence in Computer Music Languages”,
   Organized Sound,25(1),pp53-63(2020),Andrew Mcpherson and Koray Tahlroǧlu
   
   • Իָϓϩάϥϛϯά؀ڥ͕ͲͷΑ͏ʹͦͷݴޠݻ༗ͷදݱΛੜΈग़͢ͷ͔ΛΠϯλϏϡʔ͔Β
   ԻָϓϩάϥϛϯάݴޠͷαʔϕΠ
   7
   

   View Slide

8. ԻָϓϩάϥϛϯάݴޠҰཡ
   8
   l-BOHVBHFTGPS$PNQVUFS.VTJD 3PHFS#%BOOFOCFSHz 'SPOUJFSTJO%JHJUBM)VNBOJUJFT /PW Ճච͸দӜ
   
   

   View Slide

9. ԻָϓϩάϥϛϯάݴޠҰཡ
   9
   l-BOHVBHFTGPS$PNQVUFS.VTJD 3PHFS#%BOOFOCFSHz 'SPOUJFSTJO%JHJUBM)VNBOJUJFT /PW Ճච͸দӜ
   
   Overtone
   Sam Aaron
   
   (2011?)
   Sonic PI
   Sam Aaron
   
   (2012)
   TidalCycles
   Alex Mclean
   
   (2006~)
   Chronic
   Eli Blandt
   
   (2002)
   Kronos
   Versa Norilo
   
   (2015)
   LC
   Hiroki Nishino
   
   (2013)
   ixi lang
   Thor Magnusson
   
   (2012)
   Gibber
   Charlie Roberts
   
   (2012)
   FoxDot
   Jean-François Parent
   
   (2015)
   Extempore
   Andrew Sorensen
   
   (2012)
   impromptu
   Andrew Sorensen
   
   (2010)
   Common Lisp Music
   Bill Schottstaedt
   
   (1989)
   SuperCollider-
   Clients
   SOUL
   ROLI
   
   (2018)
   Vult
   Leonardo
   Laguna Ruiz
   
   (2016)
   mimium
   Tomoya Matsuura
   
   (2020)
   Gwion
   Jérémie Astor
   
   (2017)
   ଞɺmidiϨϕϧΛϝΠϯͷର৅ʹͨ͠؀ڥʹAlda(2012)΍Takt(2014)ɺORCΛ(2017)ͳͲ
   

   View Slide

10. 10
    ΦʔσΟΦϓϩάϥϛϯάݴޠAdvent Calender 2020 by aike͞Μ
    ͋ΒΏΔݴޠͰαΠϯ೾+σΟϨΠΛ࡞ͬͯΈΔ
    https://qiita.com/advent-calendar/2020/audiolang
    

    View Slide

11. ྺུ࢙֓
    11
    50s~80s DSPϋʔυ΢ΣΞʹର͢Δ໋ྩ
    90s~ All CPUɺ൚༻OS্ͰͷϦΞϧλΠϜॲཧ(Max,Pd,SC,CSound)
    00s~ ৴߸ॲཧͷJITίϯύΠϧ/τϥϯεύΠϧ(Faust,Soul,Vult)
    ಛఆͷදݱʹಛԽͨ͠DSL in DSL(Sonic Pi,TidalCycles)
    Πϕϯτॲཧͷਖ਼֬ͳεέδϡʔϦϯά(ChucK,Extempore)
    

    View Slide

12. Pd,SuperColliderҎޙͷओཁ՝୊
    12
    2.௿ϨΠϠͷ֦ுੑ
    ৴߸ॲཧͷϓϦϛςΟϒ(UGen)Λ૿΍ͨ͢ΊʹC΍C++Λॻ͔ͳ͖Ό͍͚ͳ͍
    ৴߸ॲཧͱΠϕϯτॲཧͰผͷܕͷಉ͡ॲཧ(+ͱ+~ͱ͔)Λ࢖͍෼͚ͳ͖Ό͍͚ͳ͍
    3.ߴϨΠϠͷ֦ுੑ

    ݴޠͦͷ΋ͷͷ֦ுੑɺࣗݾ൓өੑʹݶ౓͕͋Δ
    ಛఆͷԻָྖҬʹ͞ΒʹಛԽͨ͠ݴޠ/ϥΠϒϥϦΛ࡞Γ͍ͨ
    1.ਖ਼֬ͳΠϕϯτεέδϡʔϦϯά

    ϓϦΤϯϓςΟϒϚϧνλεΫʢϢʔβϓϩάϥϜ͕OSͷλεΫεέδϡʔϦϯάʹ
    ؔ༩Ͱ͖ͳ͍ʣ؀ڥԼͰͲ͏΍ͬͯαϯϓϧϨϕϧͰਖ਼֬ͳΠϕϯτॲཧΛ͢Δ͔ʁ
    

    View Slide

13. 1.ΠϕϯτεέδϡʔϦϯά໰୊
    13
    ΦʔσΟΦόοϑΝʢྫ:256αϯϓϧʣຖʹΠϕϯτΩϡʔΛফԽ
    PuredataɺSuperColliderͳͲ
    ಉظ(Synchronous)
    ࿦ཧ࣌ؒʢLogical TimeʣͰॱংΛอূ
    ChucK,Gwion,mimiumͳͲ
    w ࣮૷͸ΘΓͱ؆୯ɺಈ࡞ܰྔ
    w .*%*΍6%1ͳͲ*0བྷΈͷ͕࣌ؒਖ਼֬Ͱ͸ͳ͍
    w $PODVSSFOUͳॲཧʹ࠷దԽ͠Α͏ͱࢥ͏ͱେม
    w *0ؔ࿈΋ൺֱతਖ਼֬ʹॲཧͰ͖Δ
    w εϨου҆શͳ࣮૷ͭΒ͍
    w ཁݸผ04ຖͷରԠɺ૊ΈࠐΈͱ͔ʹ͸Ҡ২ͮ͠Β͍
    ඇಉظ(Asynchronous)
    εϨου෼͚ͯϦΞϧλΠϜεέδϡʔϧ
    Max※,ExtemporeͳͲ
    ※ਖ਼֬ʹ͸Max͸ΦʔσΟΦͱεϨουΛ෼͚͍ͯΔ͕ΦʔσΟΦͱผϨʔτ(ඪ४
    20msຖ)ͰΠϕϯτΩϡʔΛpoll͍ͯ͠Δɻ

    Extempore͸λεΫ଴ͪ࣌ؒΛܭࢉͯ͠λεΫ͕͋Δ࣌ͷΈىಈ
    ϋΠϒϦου
    Ϣʔβʔ͕໌ࣔͯ͠ಉظ/ඇಉظΛ੾Γସ͑
    LC(Nishino2013)ͳͲ
    ྫɿͽͬͨΓαϯϓϧຖʹม਺Λมߋ͍ͨ͠ʜ
    ͚Ͳຖαϯϓϧॲཧ͢Δͷ͸ա৒
    ྫɿͽͬͨΓαϯϓϧຖʹม਺Λมߋ͍ͨ͠ʜ
    ͚Ͳຖαϯϓϧॲཧ͢Δͷ͸ա৒
    

    View Slide

14. • Pd΍SCͰ͸༻ҙ͞ΕͨجຊϞδϡʔϧɿUnit GeneratorʢUGenɺΦγϨʔλ΍
    ϑΟϧλͷ͜ͱʣΛ૊Έ߹ΘͤΔ͜ͱͰ৴߸ॲཧΛهड़Ͱ͖Δ
    
    • ৽͍͠جຊϞδϡʔϧΛ࡞Γͨ͘ͳͬͨΒͲ͏͢Δʁ

    →C or C++ͰϓϥάΠϯ࡞ͬͯͶʢͭΒ͍ʣ
    
    • JITίϯύΠϧ͢Ε͹ύϑΥʔϚϯεΛଛͳΘͣ௚઀৴߸ॲཧΛهड़Ͱ͖ΔͷͰ͸ʁ
    
    • DSPܥݴޠɿFaust,Soul,Vult,Kronos
    
    • JITͷόοΫΤϯυ͸͍͍ͩͨLLVMɺC΍C++ίʔυͱͯ͠τϥϯεύΠϧͰ͖Δ
    ΋ͷ΋ଟ͍ɻWebAssemblyհͯ͠ϒϥ΢βͰಈ࡞͢Δ΋ͷ΋
    
    • ໰୊ɿDSPݴޠࣗମ͸εέδϡʔϦϯάͷSemanticsΛ࣋ͨͳ͍ͷ͕΄ͱΜͲ
    2.௿ϨΠϠͷ֦ுੑ
    14
    

    View Slide

15. • ϦζϜύλʔϯ/Ի֊/࿨Իͱ͍ͬͨߴ࣍ͷύϥϝʔλΛϓϩάϥϜ͢Δʹ
    ͸ݴޠͷந৅Խೳྗʢಛʹܕ·ΘΓʣ͕଍Γͳ͍
    
    • ྫ͑͹PdͩͱϓϦϛςΟϒͱϓϦϛςΟϒ͔Β੒ΔϦετ͔͠ѻ͑ͳ͍
    
    • ϙϐϡϥʔͳղܾࡦɿSuperColliderΛωοτϫʔΫίϚϯυͰૢ࡞Ͱ͖Δ
    γϯηαΠβʔͱͯ͠࢖͍ɺ৽͍͠DSLΛ࡞Δ
    
    • SC͸v3ΑΓૄ݁߹ͳαʔόʔɾΫϥΠΞϯτϞσϧΛ࠾༻ɺ৽͍͠γϯηͷఆٛ(SynthDef)Ҏ֎
    ͸΄΅ͳΜͰ΋OSC(UDP্ͷ؆қϓϩτίϧ)Ͱૢ࡞Մೳ
    
    • Sonic Pi(Ruby+Erlang)ɺixi(Ruby)ɺFoxDot(Python)ɺTidalCycles(Haskell)ɺOvertone(Clojure)ͳͲ
    
    • ໰୊ɿωοτϫʔΫίϚϯυͷਖ਼֬ͳεέδϡʔϦϯάΛ…(σδϟϰ)
    3.ߴϨΠϠͷ֦ுੑ
    15
    DSL on DSLͷϗετͷબ୒ࢶ͸࣮࣭SuperColliderҰ୒ͳͷ͕ݱঢ়
    

    View Slide

16. • ࣈ۟ղੳ
    
    • ݴޠ֦ு
    
    • ݴޠಛघԽ
    
    • ύΠϓϥΠϯ
    
    • ιʔεtoιʔεม׵
    SpinellisʹΑΔDSL࣮૷ͷσβΠϯύλʔϯΑΓൈਮ
    Ͳ͏΍࣮ͬͯ૷͢Δʁ
    16
    Diomidis Spinellis. Notable design patterns for domain specific languages. Journal of Systems and Software,
    56(1):91–99, February 2001. (doi:10.1016/S0164-1212(00)00089-3)
    Ұ൪ετϨʔτͳΞϓϩʔνɺࣗ༝౓ߴ͍͕࣮૷େม
    ԋࢉࢠΦʔόʔϩʔυ΍ϚΫϩͳͲɺࣗ༝౓͸ϗετݴޠ࣍ୈ
    ϘΠϥʔϓϨʔτͱ૊Έ߹ΘͤͨΓɺҰ෦ͷػೳ͚ͩ࢖͏
    Cͷιʔεʹม׵ͨ͠Γ͢Δ͜ͱͰόοΫΤϯυ؆ུԽ
    SuperColliderʹ৴߸ॲཧ෦෼͸͓೚ͤ
    

    View Slide

17. SuperColliderʹΑΔه๏
    17
    {
    var noise = PinkNoise.ar(0.2);
    noise = Mix.new(Array.fill(10, {arg i;
    Resonz.ar(noise, i * 800 + 200, 0.05)}));
    noise = (noise * 0.2 + noise * Decay.kr(Dust.kr(0.5),10));
    noise = RHPF.ar(noise, LFNoise0.kr(0.5).range(220, 880), rq:0.001);
    CombL.ar(noise, delaytime:0.5, decaytime:25).dup() * 0.5
    }.play;
    (->
    noise = PinkNoise.ar(0.2)
    noise = Mix Array.fill 10, (i)->
    Resonz.ar(noise, i * 800 + 200, 0.05)
    noise = (noise * 0.2 + noise * Decay.kr(Dust.kr(0.5), 10))
    noise = RHPF.ar(noise, LFNoise0.kr(0.5).range(220, 880), rq:0.001)
    CombL.ar(noise, delaytime:0.5, decaytime:25).dup() * 0.5
    ).play()
    CoffeColliderʹΑΔه๏
    CoffeeScriptͷԋࢉࢠΦʔόʔϩʔυ͚ͩͰՄೳͳݶΓه๏Λࣅ͍ͤͯΔ
    http://mohayonao.github.io/CoffeeCollider/#noise.coffee
    

    View Slide

18. • $΍#͸ϗετݴޠͰ͋ΔHaskellʹ͓͚Δதஔԋࢉࢠ(ݴޠ֦ு)
    
    • sound΍gainɺeveryͳͲ͸Haskell্ͰϥΠϒϥϦͱͯ͠ఆٛ͞Εͨؔ਺
    
    • “”Ͱғ·Εͨ෦෼͸Haskell্Ͱ͸ͨͩͷจࣈྻ
    
    • Parsecͱ͍͏Haskellͷࣈ۟ղੳϥΠϒϥϦͰ಺෦తʹύʔε͍ͯ͠Δ
    
    • Ϣʔβʔ͸Haskellͷؔ਺ఆٛ΍ܕએݴͳͲߴ౓ͳจ๏Λ஌Δඞཁ͸ͳ͍
    ʢݴޠಛघԽʣ
    
    • TidalCycles͸SuperColliderʹରͯ͠OSCΛૹΓԻΛग़͢(ύΠϓϥΠϯ)
    HybridͳྫɿTidalCycles
    18
    d1 $ sound "bd*4" # gain (every 3 (rev) $ "1 0.8 0.5 0.7")
    ಉ༷ͳྫͱͯ͠͸FoxDot(ϗετݴޠ:Python)ͳͲ
    

    View Slide

19. ԻָݴޠͷΞʔΩςΫνϟ
    19
    ίϯύΠϥ/
    ΠϯλϓϦλ
    ϥϯλΠϜ
    ιʔείʔυ
    (ςΩετσʔλ)
    தؒදݱ
    UGenάϥϑߏ଄
    
    VM໋ྩྻ
    
    Ϛγϯίʔυ
    ϨΠϠʔ
    ߴ
    ௿
    ϥΠϒϥϦ
    ҙຯ֦ு
    • தؒදݱͷཻ౓ʹΑͬͯදݱͷ෯΍ಈతมߋͷ͠΍͕͢͞มΘͬͯ͘Δ
    50~80sɿ DSPϋʔυ΢ΣΞ
    
    00s: UGen
    

    View Slide

20. ϏδϡΞϧݴޠͷ৔߹
    20
    ίϯύΠϥ/
    ΠϯλϓϦλ
    ϥϯλΠϜ
    ιʔείʔυ
    (ςΩετ/όΠφϦσʔλ)
    தؒදݱ
    UGenάϥϑߏ଄
    
    VM໋ྩྻ
    
    Ϛγϯίʔυ
    ϨΠϠʔ
    ߴ
    ௿
    ϏδϡΞϥΠβ
    ʢIDEʣ
    ஌֮/ฤू
    Մೳͳදݱ
    ϥΠϒϥϦ
    ҙຯ֦ு
    • ͋Δҙຯී௨ͷݴޠ΋UTF-8ͳͲܾ·ͬͨόΠφϦσʔλΛɺςΩετͱ͍͏ਓؒʹೝࣝՄ
    ೳͳදݱʹม׵͍ͯ͠ΔͷͰ͜ͷਤʹ౰ͯ͸ΊΔ͜ͱ΋Ͱ͖Δ
    

    View Slide

21. தؒදݱ͸ͻͱͭͱ͸ݶΒͳ͍
    21
    ϥϯλΠϜ
    ιʔείʔυ
    (ςΩετ/όΠφϦσʔλ)
    ϏδϡΞϥΠβ
    ʢIDEʣ
    ஌֮/ฤू
    Մೳͳදݱ
    ίϯύΠϥ/
    ΠϯλϓϦλ
    தؒදݱ
    ϥΠϒϥϦ ίϯύΠϥ/
    ΠϯλϓϦλ
    தؒදݱ
    ϥΠϒϥϦ
    ίϯύΠϥ/
    ΠϯλϓϦλ
    தؒදݱ
    ϥΠϒϥϦ
    

    View Slide

22. TidalCyclesͷྫ
    22
    HaskellΠϯλϓϦλ(GHCI)
    TidalCycles
    BP ύʔα
    ύλʔϯ

    ϥΠϒϥϦ
    ιʔείʔυ
    (ςΩετσʔλ)
    ҙຯ֦ு
    SC
    ΫϥΠΞϯτ
    OSC
    ϝοηʔδ SCSynth
    Pattern
    σʔλ
    • ʢ͍·͍ͪ͏·͘ਤࣔͰ͖ͳ͍ʣ
    

    View Slide

23. • Syntax(౷ޠ࿦)
    
    • Semantics(ҙຯ࿦)
    
    • Runtime
    
    • Libraries
    
    • Programming Environments(IDE)
    
    • Community&Resources
    Dannenbergʢ2018ʣʹΑΔ
    ԻָݴޠΛಛ௃͚ͮΔཁૉ
    23
    

    View Slide

24. • ͜͜ͰͷExpressivenessͱ͸“ಡΈ΍͘͢ॻ͖΍͍͢”ɺ

    Generalͱ͸“Մೳͳදݱͷൣғ͕޿͍”ͱ͍͏ҙຯ
    Expressive vs General
    24
    E. Brandt, “Temporal type constructors for computer music programming,” School of Computer Science, Carnegie Mellon University, 2002.
    ࣮ࡍʹ͸࣮ߦ࣌ύϑΥʔϚϯε͕
    ѱ͘ͳΔͱ͍͏৽ͨͳτϨʔυΦ
    ϑ͕ൃੜ͍ͯ͠Δ
    

    View Slide

25. Իָϓϩάϥϛϯάʹ͓͍ͯଟݴޠΞϓϩʔνΛऔΔͱɺ
    Ϣʔβ͸൚༻ੑ (FOFSBMJUZ
    ͱޮ཰ੑ &GpDJFODZ
    ͷόϥϯεΛऔΓ΍͘͢ͳΔɻ
    λεΫʹԠͯ͡ϓϩάϥϚ͸ҟͳΔෳࡶ౓ͷΤϯτϦʔϙΠϯτΛબ୒͢Δ͜ͱ
    ͕ՄೳʹͳΔɻΑΓ௿͘൚༻తͳϨϕϧͰ͸ΑΓෳࡶͳίʔυͷઃܭ͕ඞཁ͕ͩ
    ෯޿͍݁ՌΛಘΒΕΔɻҰํɺΑΓߴ͍ɺಛघԽɺ۩ମԽ͞ΕͨϨϕϧͰ͸ϓϩ
    άϥϛϯάʹඞཁͳ࿑ྗͱ͍͏؍఺͔Βϓϩηε͸ΑΓޮ཰తʹͳΔͩΖ͏ɻ
    Lazzarini(2013)ΑΓ
    Efficient vs General
    25
    

    View Slide

26. 3ͭͷہॴԽͱ3ͭͷޮ཰ੑ
    ཁૉ͕༩͑Δಛ௃ͷ෼ྨʢͷҊʣ
    26
    6ͭΛશ෦৳͹͢ͷ͸ݫ͘͠ɺτϨʔυΦϑ͕͋Δ
    Expression Generality
    දݱͷہॴੑ
    Platform Independency
    ؀ڥͷہॴੑ
    Dynamic Modification
    ܧଓత࣮ߦͷہॴੑ
    Runtime Efficiency
    ࣮ߦ࣌ͷܭࢉίετ
    Learning Efficiency
    Ϣʔβʔͷֶशίετ
    Development Efficiency
    ։ൃίετ
    গͳ͘ͱ΋ɺ“͋ΒΏΔ؀ڥͰ͋ΒΏΔදݱ͕Ͱ͖ɺಈతมߋ΋͠·͘Εͯ௚ײతʹ࢖͍΍͢
    ࣮͘ߦ଎౓΋ωΠςΟϒฒΈ”ͷݴޠΛ࡞Ζ͏ͱ͢Δͱ։ൃίετ͕ͱΜͰ΋ͳ͍͜ͱʹͳΔ
    

    View Slide

27. 27
    Syntax
    Semantics
    Runtime
    Libraries
    Programming
    Environment
    Community
    Resources
    Development
    Efficiency
    Runtime
    Efficiency
    Learning
    Efficiency
    Expression
    Generality
    Dynamic
    Modification
    ֤ཁૉͱಛ௃ͷؔ܎
    Platform
    Independency
    …ຊ౰ʹ͜ΕΒͷཁૉ͚ͩͰݴޠΛධՁͰ͖ΔͷͩΖ͏͔ʁ
    

    View Slide

28. Idiomaticity
    ྔతධՁͰ͖ͳ͍ಛ௃
    28
    • ཁ໿ɿݴޠʹ͓͚Δཧ࿦తʹՄೳͳදݱͰ͖Δൣғ͕ಉͩͬͨ͡
    ͱͯ͠΋ɺͦΕͧΕͷݴޠ͕༠ൃ͢Δදݱͷ࣭ɿIdiomacticity͸
    ҟͳΔ
    
    • ݴΘΕͯΈΔͱ͋ͨΓ·͑
    l*EJPNBUJD1BUUFSOTBOE"FTUIFUJD*OqVFODFJO$PNQVUFS.VTJD-BOHVBHFTz 0SHBOJ[FE
    4PVOE
    QQ
    "OESFX.DQIFSTPOBOE,PSBZ5BIMSPǧMV
    

    View Slide

29. ϝλΫϦΤʔγϣϯͱͯ͠ͷݴޠઃܭ
    29
    զʑͷݚڀ݁Ռ͸ʢ༁஫ɿݴޠ͕༩͑ΔʣಛఆͷඒతӨڹͷূ໌Λҙਤͨ͠
    ΋ͷͰ͸ͳ͍͠ɺݸʑͷ࣮ફऀͷ૑଄ੑ͕ৗʹୈҰͰ͋Δ͜ͱʹมΘΓ͸ͳ
    ͍΋ͷͷɺσδλϧָثͷσβΠϯ͸ɺݴޠ࡞੒ऀͱָثσβΠφʔŠ࠷ऴ
    తʹͦͷݴޠͷϦιʔεʢͦͯͦ͠Εͱͱ΋ʹɺӅ͞ΕͨεΫϦϓτʣΛࣗ
    ਎ͷ૑଄త໨ඪʹ߹Θͤͯঌש͢ΔऀŠͱͷɺ࣌ؒͱڑ཭Λ௒͑ͨର࿩Ͱ͋
    Δͱݟͳ͢͜ͱ͕Ͱ͖ΔɺͱఏҊͰ͖Α͏ɻͦ͏ͯ͠ɺݴޠઃܭ͸Ұछͷϝ
    λΫϦΤʔγϣϯߦҝͱͳΓɺ࠷ऴతʹ͸Կे೥ʹ΋ΘͨͬͯඒతͳӨڹΛ
    ༩͑Δ͜ͱʹͳΔɻ
    l*EJPNBUJD1BUUFSOTBOE"FTUIFUJD*OqVFODFJO$PNQVUFS.VTJD-BOHVBHFTz 0SHBOJ[FE
    4PVOE
    QQ
    "OESFX.DQIFSTPOBOE,PSBZ5BIMSPǧMV චऀ༁ɺڧௐ͸චऀ
    

    View Slide

30. • 2000೥Ҏޙͷओཁͳ3՝୊
    
    • ࣮૷ํ๏͔ΒݟΔԻָݴޠ
    
    • ཉ͍͠จ๏ͷࣗ༝౓ʹΑͬͯݴޠ֦ுͰରԠͨ͠Γߏจղੳͨ͠Γ
    
    • தؒදݱͷཻ౓ͱ࣮ߦ࣌ύϑΥʔϚϯεɺಈతมߋͷ͠΍͕͢͞มΘΔ
    
    • ԻָݴޠΛಛ௃͚ͮΔ6ཁૉʹରͯ͠ɺ࣮ࡍʹݱΕΔ3ͭͷہॴ
    ੑɺ3ͭͷޮ཰ੑͱ͍͏ࢦඪΛఏҊ
    ·ͱΊ
    30
    ௿ϨΠϠͷ֦ு(JIT DSP) ߴϨΠϠͷ֦ு (DSL in DSL)
    ਖ਼֬ͳεέδϡʔϦϯά
    

    View Slide

31. • ԻϓϩάϥϛϯάʹඞཁͳυϝΠϯ஌ࣝͱݴޠॲཧܥʹඞཁͳυ
    ϝΠϯ஌͕ࣝ྆ํඞཁ͚ͩͲ྆ํϔϏʔ
    
    • DSPΛJITॲཧͰ͖ΔΑ͏ʹ͠Α͏ͱࢥ͏ͱ࣮૷ͷ95ˋ͸Իͱ͔
    ؔ܎ͳ͘ͳͬͯ͘Δ
    
    • ݴޠઃܭ͕Α͘Ͱ͖ͯΔ≠࢖ͬͯ΋Β͍΍͍͢ʢύοέʔδϯ
    άɺCIɺγϯλοΫεϋΠϥΠτɺυΩϡϝϯςʔγϣϯ…ʣ
    ࣮૷ऀͱͯ͠ײ͡Δࠔ೉
    31
    

    View Slide

32. • ΠϕϯτεέδϡʔϧͱJIT DSP͕1ͭͷݴޠମܥͰ׬݁͢Δݴޠ
    
    • ࣗݾ൓өੑΛڧΊͨݴޠ಺DSLΛ࡞Γ΍͍͢ݴޠ
    
    • ςΩετɺඇςΩετΛ૊Έ߹Θͤ/૒ํ޲ධՁͰ͖Δݴޠ
    
    • ෳ਺ͷݴޠʹ·͕ͨͬͯ࢖͑ΔLanguage Server/IDEج൫
    ະདྷͷԻָϓϩάϥϛϯά
    32
    Language Oriented Programming(Racket)ͷΑ͏ͳੈք؍ʁ
    ↓↑DAWʢԻ੍ָ࡞ιϑτ΢ΣΞʣͱͷΑΓࠜຊతͳ࿈ܞʁ
    →mimiumͰϝΠϯͰ࣮ݱ͍ͨ͜͠ͱ
    ໺ྑϓϩάϥϛϯάݴޠͷϦϙδτϦ͕਺ඦݸGithubʹస͕ͬͯΔΑ͏ʹɺ਺ඦݸͷԻָݴޠ͔Β޷͖
    ͳͷΛϑϨΩγϒϧʹ૊Έ߹ΘͤΒΕΔΑ͏ͳੈք͕དྷͯཉ͍͠ʢເʣ
    

    View Slide