Songs to pave the seasons

The data come as a large JSON file exported from Spotify. Parsing it is straightforward thanks to bqn-libs:

⟨P⇐Parse⟩ ← •Import "../../bqn-libs/json.bqn"
@ ⊣ spd ← P •file.Chars "../../supp/spdat/data.json"

For later queries I isolate the relevant fields (song index, artist, elapsed time) and set the length of the desired result:

s‿a‿m‿l ← ⟨7, 8, 3, 1+↕20⟩

Finally, I define a compact (50 chars) dyadic block function to perform ranked lookups and return the top entries:

Q ← {l≍˘∾(⍷𝕨⊸⊏˘)¨ l⊏ ((⍒(+´m⊸⊏˘)¨)⊸⊏ 𝕨⊸⊏˘⊐⊸⊔⊢) >1⊏¨ 𝕩}

The result is somewhat eclectic, yet a clear pattern emerges. Most pieces belong to a world of intricate harmonies and long forms: during these years, it's been all about progressive rock and progressive metal. A few simpler intrusions distort the spectrum: traces of a desert island I once inhabited.

s Q spd	

┌─                                                     
╵ 1  "Countless Skies"                                 
  2  "Divertimento I, K.136: Allegro"                  
  3  "The Numbers"                                     
  4  "Autre temps"                                     
  5  "Ghost of Perdition"                              
  6  "Crossing the Road Material"                      
  7  "Hoppípolla"                                      
  8  "Ether"                                           
  9  "Colossus"                                        
  10 "River"                                           
  11 "El Tete"                                         
  12 "Will o the Wisp"                                 
  13 "Pakumba"                                         
  14 "Damned Rope"                                     
  15 "Eternal Rains Will Come"                         
  16 "La femme d'argent"                               
  17 "Bajanda"                                         
  18 "Nimrodel - Medley"                               
  19 "Breathe (In The Air) - 2011 Remastered Version"  
  20 "In The Shadow Of Our Pale Companion"             
                                                      ┘

The same tendency appears at the level of artists. I gravitate toward music that builds worlds rather than merely expresses moods. Yet sometimes I descend to a simpler plane, to enjoy the breeze down there, or to feel the melancholia of my past existence³.

a Q spd

┌─                              
╵ 1  "Opeth"                    
  2  "Wolfgang Amadeus Mozart"  
  3  "Pink Floyd"               
  4  "Be'lakor"                 
  5  "Sigur Rós"                
  6  "Coldplay"                 
  7  "Mogwai"                   
  8  "Chocolate Mc"             
  9  "Radiohead"                
  10 "Joaquín Sabina"           
  11 "Iron & Wine"              
  12 "Rammstein"                
  13 "Alcest"                   
  14 "Buena Fe"                 
  15 "Silvio Rodríguez"         
  16 "Amon Amarth"              
  17 "In Mourning"              
  18 "Lamb of God"              
  19 "Camel"                    
  20 "Omnium Gatherum"          
                               ┘

This is the Opeth album I would recommend to anyone. The query function needs to be modified a bit for generating it. But before that, let's look at average album length in the official discography:

lo ← (↕∘⌈+´÷≠) 7‿5‿9‿7‿8‿6‿8‿8‿7‿10‿7‿11‿10

We can then filter all Opeth entries among the top songs, and select the first ≠lo:

O ← {lo (1⊸+∘⊣≍˘⊏) (⊢˝˘⊸∊⟜𝕨/⊏˘) ∾(⍷s‿a⊸⊏˘)¨ ((⍒(+´m⊸⊏˘)¨)⊸⊏ s⊸⊏˘⊐⊸⊔⊢) >1⊏¨ 𝕩}

Enjoy this whisper within a sigh⁴:

spd O˜⋈ "Opeth" 

┌─                             
╵ 1 "Ghost of Perdition"       
  2 "River"                    
  3 "Will o the Wisp"          
  4 "Eternal Rains Will Come"  
  5 "The Drapery Falls"        
  6 "Elysian Woes"             
  7 "Burden"                   
  8 "Harvest"                  
                              ┘