{"slug": "adelaide", "name": "ADELAIDE", "meter_label": "5.4.5.4.D", "meter": {"pattern": [5, 4, 5, 4], "doubled": true, "refrain": false, "alleluia": false, "irregular": false}, "composer": "george-c-stebbins", "composer_date": "1907", "canonical_key": {"tonic": "ees", "mode": "major"}, "preferred_harmonization": "adelaide-tabernacle-1921", "provenance": {"layer": "tune", "source": "tabernacle-hymns-no-2-1921", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Tune by George C. Stebbins (1846-1945), written for Adelaide A. Pollard's words. The received account \u2014 NOT verified against a printing here \u2014 is that it first appeared in the Northfield Hymnal with Alexander's Supplement (1907). Read here from the dated printing in Tabernacle Hymns, No. 2 (Chicago: Tabernacle Publishing Co., 1921), hymn 21 \u2014 archive.org tabernaclehymnsn00chic leaf n29 \u2014 which credits \"Geo. C. Stebbins.\" on the staff and foots \"Copyright, 1907, by Geo. C. Stebbins. Used by permission. Hope Pub. Co., owner.\", and collated notehead by notehead against a second, independent scan of that same plate, archive.org tabernaclehymnsn02chic leaf n27. The 1921 book's own title page (n00chic leaf n7) reads \"Tabernacle Hymns No. 2 \u2026 Copyright 1921 \u2026 Published by Tabernacle Publishing Company, 29 South La Salle Street, Chicago, Illinois\". US public domain: the notice on the hymn is 1907 and the printing read is 1921, both before 1931. THE 1907 FIRST PRINTING ITSELF WAS NOT OPENED \u2014 the pd_basis rests on the 1921 plate and its own 1907 notice, not on a copy of the Northfield Hymnal. The tune name is not printed on the 1921 plate \u2014 gospel songbooks of this line rarely name their tunes \u2014 so ADELAIDE is the name the later hymnals and Hymnary carry for it, after the author's given name; no printed alt_name is recorded. The soprano's first fifteen note-groups as encoded, in scale degrees, are 3 2 3 4 3 1 7 1 2 2 1 2 3 2 2, which matches Hymnary tune authority adelaide_stebbins's incipit 323431712212322 exactly, and tools/tune_fingerprint.py finds no corpus melody closer than 0.557 (RUTHERFORD and HAMBURG, tied), so this is not a duplicate.", "pd_jurisdictions": {"US": true}, "hymnary_ids": {"tune_auth": "adelaide_stebbins", "total_instances": 337}}, "harmonization": {"slug": "adelaide-tabernacle-1921", "tune": "adelaide", "music_file": "adelaide-tabernacle-1921.ly", "canonical_key": {"tonic": "ees", "mode": "major"}, "time": "9/4", "pickup": "2.", "system_breaks": [10, 20, 30], "breaks_manual": true, "final_barline": "|.", "amen": false, "verse_variable_melisma": false, "refrain": null, "source_line": "Setting: George C. Stebbins, 1907, as printed in Tabernacle Hymns, No. 2, 1921. Public domain.", "provenance": {"layer": "harmonization", "source": "tabernacle-hymns-no-2-1921", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Four-part setting of Stebbins's ADELAIDE as printed in Tabernacle Hymns, No. 2 (Chicago: Tabernacle Publishing Co., 1921), hymn 21 \u2014 the leaf carries no printed page number \u2014 headed \"Have Thine Own Way, Lord.\" over \"A. A. P.\" (left) and \"Geo. C. Stebbins.\" (right), marked \"Slowly.\", in E-flat major and 9/4 with a three-quarter pickup, and footed \"Copyright, 1907, by Geo. C. Stebbins. Used by permission. Hope Pub. Co., owner.\" US public domain: the underlying copyright is 1907 and the printing read is 1921, both well before 1931, and no modern arranger stands anywhere in the chain. TWO DATED SCANS OF THAT PLATE WERE READ AND REQUIRED TO AGREE \u2014 archive.org tabernaclehymnsn00chic leaf n29 (2365 px across, staff space 29.28 px, i.e. 14.6 px per diatonic step) and archive.org tabernaclehymnsn02chic leaf n27 (1877 px across, staff space 23.35 px, i.e. 11.7 px per step); the two items are two scans of the one book \u2014 n00chic's own title page, leaf n7, reads \"Tabernacle Hymns No. 2 \u2026 Copyright 1921 \u2026 Published by Tabernacle Publishing Company, 29 South La Salle Street, Chicago, Illinois\" \u2014 and at this hymn n00chic runs two leaves later than n02chic. THE NOTEHEADS WERE MEASURED, NOT EYEBALLED: each scan was binarised against its own local paper level (ink darker than 82% of a 41x41 grey dilation smoothed by a 41x41 box filter) and closed 5x5 to seal the gaps the plate's ink leaves inside a head; holes smaller than 1.2 staff spaces squared were then filled, which turns each hollow dotted-half OUTLINE into a disc; staff lines were suppressed by vertical run length AFTER that fill, so a hollow head's thin arcs survive; heads were isolated by morphological opening (0.62 staff space horizontally, 0.45 vertically), and every core converted to a diatonic step against a FRESH five-line fit made in a window around that core's own x. Half and dotted-half heads on this plate are WIDER than the filled ones \u2014 44 px against 30 px on scan 1 \u2014 and were invisible to the detector until the width window was opened to 1.80 staff spaces. Counting steps from the bottom staff line, the readings inside the staff are tight: the 25 cores on the bottom line measured -0.139 to +0.110, the 21 on the second line 1.903 to 2.086, the 10 in the top space 6.974 to 7.081. Outside it the plate squeezes the heads back toward the staff: the 9 cores in the space above the bass staff measured 8.655 to 8.818, the 9 in the space below the treble staff -0.695 to -1.007, the 13 between the first and second ledger lines below -2.630 to -2.916. The classes never overlap. 144 of the 160 note positions on scan 1 were covered by a core within 0.5 step of the reading that ships (mean distance 0.094 step, largest 0.496) and 117 of 160 on scan 2 (mean 0.118, largest 0.484); the rest the automated pass did not resolve \u2014 hollow dotted-half heads it still drops, and the bass chord of soprano note 6, whose merged pair it split at 7.53/4.50 instead of 7/5 on both scans. SO EVERY ONE OF THE 160 WAS READ A THIRD WAY, off crops with the fitted five-line grid drawn over the page at 3.4x to 9x, ON BOTH SCANS, and it is that third reading and not any centroid that ships. It corrected two chords the eye had got wrong at low magnification \u2014 the alto of the first chord of system 2 (D4, not E-flat 4) and the bass under \"will,\" (A-flat 2 in the space, not G2 on the line) \u2014 and the automated figure agreed with the correction both times. The two grid readings agree with each other on all 40 chords in all four voices \u2014 160 notes, 0 disagreements. ACCIDENTALS WERE SWEPT SEPARATELY AND BY EYE, at high magnification, over every column of BOTH leaves; the automated figures above see staff positions only and never covered accidentals. THIS SWEEP FIRST GOT IT WRONG, and the error reached the notes: it reported exactly two printed accidentals, both naturals in the soprano, and said none appears in the alto, tenor or bass anywhere. There are THREE. The plate prints, on both leaves: a natural on the bottom line before soprano note 12 (\"art\", E natural); a natural in the second space before soprano note 27 (\"-er\" of \"Aft-er\", A natural); AND A FLAT before the ALTO notehead of the first chord of the third system (soprano note 26, \"Aft-\"), making that alto D-FLAT 4, with the measure rule carrying it to note 28. Both were first encoded as D natural, putting an E-flat MAJOR seventh where the plate writes an E-flat DOMINANT seventh resolving to the A-flat at \"will,\". Corrected. Rule 2b in one line: a notehead probe reads heads, never the flats in front of them. The three-flat signature stands at the head of every staff. NO SCAFFOLD WAS USED AND NOTHING WAS CHECKED AGAINST ONE: no MIDI, no .nwc, no Cyber Hymnal asset and no existing encoding was opened at any point. The only external check is Hymnary's melodic incipit for tune authority adelaide_stebbins, \"323431712212322\", which is the first fifteen soprano note-GROUPS of this reading exactly. DEPARTURES FROM THE PAGE, in full: (1) the page casts three systems and breaks after soprano notes 13 and 25, and this encoding breaks after 10, 20 and 30, so all four systems hold different content from the plate's three \u2014 the plate's first break falls inside poetic line 3 and the house rule puts breaks at poetic line ends; (2) the plate's \"Slowly.\" tempo word is not carried, this corpus holding no tempo marks. Nothing else departs: every pitch, every duration, both naturals, all four ties and the closing light-heavy double bar are the plate's, and the four voices restrike together at all 40 chords because that is what the plate prints.", "pd_jurisdictions": {"US": true}, "encoding_origin": "Hand-encoded from the page images of Tabernacle Hymns, No. 2 (1921), hymn 21 \u2014 archive.org tabernaclehymnsn00chic leaf n29, collated notehead by notehead against archive.org tabernaclehymnsn02chic leaf n27. No MIDI, no Cyber Hymnal asset and no existing encoding was used at any point.", "encoding_license": "PD-dedication"}, "music_ly": "%% ADELAIDE (5.4.5.4.D) \u2014 George C. Stebbins. Four SATB voice streams,\n%% canonical key E-flat major, 9/4, a three-quarter (dotted-half) pickup,\n%% eight full bars plus that pickup and a six-quarter close.\n%% RULES (CI-enforced): exactly the variables soprano/alto/tenor/bass/global;\n%% absolute octaves; each voice rhythmically complete; NO chords <>, NO <<>>,\n%% NO \\secondShift/\\unisonWholeShift/tightSpacing, NO \\break, NO \\bar, NO lyrics.\n%%\n%% SOURCE OF THE NOTES. Tabernacle Hymns, No. 2 (Chicago: Tabernacle Publishing\n%% Co., 1921), hymn 21 \u2014 the leaf carries no printed page number \u2014 headed\n%% \"Have Thine Own Way, Lord.\"\n%% over \"A. A. P.\" on the left and \"Geo. C. Stebbins.\" on the right, marked\n%% \"Slowly.\", and footed \"Copyright, 1907, by Geo. C. Stebbins. Used by\n%% permission. Hope Pub. Co., owner.\" Read off TWO independent scans of that\n%% same plate: archive.org tabernaclehymnsn00chic leaf n29 (2365 px across,\n%% staff space 29.28 px = 14.6 px per diatonic step) and archive.org\n%% tabernaclehymnsn02chic leaf n27 (1877 px across, staff space 23.35 px =\n%% 11.7 px per diatonic step). The two items are two scans of the one book \u2014\n%% n00chic's own title page (leaf n7) reads \"Tabernacle Hymns No. 2 \u2026 Copyright\n%% 1921 \u2026 Published by Tabernacle Publishing Company, 29 South La Salle Street,\n%% Chicago, Illinois\" \u2014 and at this hymn n00chic runs two leaves later than\n%% n02chic.\n%%\n%% HOW THE NOTEHEADS WERE READ (not by eye). Per scan, with its own\n%% binarisation and its own staff-line fits: ink is whatever is darker than 82%\n%% of the LOCAL paper level (41x41 grey dilation then a 41x41 box filter),\n%% followed by a 5x5 binary closing to seal the pale gaps the plate's ink\n%% leaves inside a head; then holes smaller than 1.2 staff spaces squared are\n%% filled, which turns each hollow (dotted-half) OUTLINE into a disc without\n%% flooding the boxes that staff lines and stems enclose; then staff lines are\n%% suppressed by vertical run length \u2014 AFTER the fill, so a hollow head's thin\n%% top and bottom arcs are already part of a tall run and survive; then a\n%% horizontal opening of 0.62 staff space and a vertical opening of 0.45 staff\n%% space delete stems, flags, barlines and ledger lines and leave a head core.\n%% Every core was converted to a diatonic step against a FRESH five-line fit\n%% made in a +-2.6-staff-space window around that core's own x, so the plate's\n%% skew cannot drag a reading. Half and dotted-half heads on this plate are\n%% WIDER than the filled ones (44 px against 30 px at scan 1) and were lost\n%% until the width window was opened to 1.80 staff spaces; merged\n%% soprano-over-alto blobs were split at the midpoint of their two clean outer\n%% edges with the separation snapped to a whole step.\n%%\n%% Counting steps from the bottom staff line (0 = bottom line, 8 = top line),\n%% the measurement inside the staff is tight: the 25 cores on the bottom line\n%% measured between -0.139 and +0.110, the 21 on the second line between 1.903\n%% and 2.086, the 10 in the top space between 6.974 and 7.081. OUTSIDE the\n%% staff the plate itself squeezes the heads back toward it: the 9 cores in the\n%% space above the bass staff measured 8.655 to 8.818, the 9 in the space below\n%% the treble staff -0.695 to -1.007, the 13 between the first and second\n%% ledger lines below -2.630 to -2.916. The classes never overlap, so no\n%% rounding is in doubt.\n%%\n%% 144 of the 160 note positions on scan 1 were covered by a head core within\n%% 0.5 step of the reading below (mean distance 0.094 step, largest 0.496); the\n%% other 16 the automated pass did not resolve \u2014 hollow dotted-half heads it\n%% still drops, plus the bass chord of soprano note 6, whose merged pair it\n%% split at 7.53/4.50 instead of 7/5. On scan 2, 117 of the 160 were covered\n%% (mean 0.118, largest 0.484). So EVERY one of the 160 was read a third way as\n%% well, off crops carrying the fitted five-line grid drawn over the page at\n%% 3.4x to 9x, on BOTH scans \u2014 and it is that third reading, not any centroid,\n%% that ships. It corrected two chords the eye had got wrong at low\n%% magnification: the alto of the first chord of system 2 (D4, not E-flat 4)\n%% and the bass under \"will,\" (A-flat 2 in the space, not G2 on the line). The\n%% two grid readings agree with each other at all 40 chords in all four\n%% voices \u2014 160 notes, 0 disagreements.\n%%\n%% ACCIDENTALS ARE NOT WHAT A NOTEHEAD PROBE SEES. Every column of both leaves\n%% was swept separately BY EYE at high magnification for printed sharps, flats\n%% and naturals; the automated figures above cover staff positions only and\n%% never covered accidentals. The sweep found exactly TWO in the body of the\n%% hymn, both naturals, both in the soprano, and both are on both leaves:\n%% a natural on the bottom line before soprano note 12 (the syllable \"art\",\n%% E natural) and a natural in the second space before soprano note 27 (the\n%% syllable \"-er\" of \"Af-ter\", A natural). Nothing else: no accidental in the\n%% alto, tenor or bass anywhere, and the three-flat signature stands at the\n%% head of every staff.\n%%\n%% NO SCAFFOLD WAS USED, and nothing here was checked against one. No MIDI, no\n%% .nwc, no Cyber Hymnal asset and no existing encoding was opened at any\n%% point. The one external check is Hymnary's melodic incipit for\n%% tuneAuthNumber adelaide_stebbins, \"323431712212322\", which is the first\n%% fifteen soprano note-GROUPS of this reading exactly.\n%%\n%% 40 soprano noteheads carry 36 syllables. The four two-note groups are TIES,\n%% not slurs \u2014 the two heads are the same pitch in all four voices \u2014 and they\n%% fall on the last syllable of poetic lines 2, 4, 6 and 8 (\"way!\", \"clay.\",\n%% \"will,\", \"still.\"): soprano notes 9-10, 19-20, 29-30 and 39-40.\n%%\n%% DEPARTURE FROM THE PLATE'S CASTING. The page casts three systems, breaking\n%% after soprano notes 13 and 25. The first of those falls in the MIDDLE of\n%% poetic line 3, which this corpus does not do, so the systems here break\n%% after notes 10, 20 and 30 \u2014 the ends of poetic lines 2, 4 and 6, which are\n%% also barlines \u2014 giving four systems of ten noteheads instead of the plate's\n%% 13/12/15.\n\\version \"2.24.0\"\n\nglobal  = { \\key ees \\major \\time 9/4 \\partial 2. }\n\nsoprano = {\n  g'4 f'4 g'4 |\n  aes'2. g'2. ees'4 d'4 ees'4 |\n  f'2.~ f'2. f'4 e'4 f'4 |\n  g'2. f'2. f'4 ees'4 d'4 |\n  ees'2.~ ees'2. g'4 f'4 g'4 |\n  aes'2. g'2. bes'4 a'4 bes'4 |\n  c''2.~ c''2. c''4 d''4 c''4 |\n  bes'2. g'2. f'4 ees'4 f'4 |\n  ees'2.~ ees'2.\n}\n\nalto = {\n  ees'4 ees'4 ees'4 |\n  ees'2. ees'2. bes4 bes4 bes4 |\n  d'2.~ d'2. d'4 d'4 d'4 |\n  d'2. d'2. bes4 bes4 bes4 |\n  bes2.~ bes2. ees'4 ees'4 ees'4 |\n  ees'2. ees'2. des'4 c'4 des'4 |\n  ees'2.~ ees'2. ees'4 ees'4 ees'4 |\n  ees'2. ees'2. bes4 bes4 bes4 |\n  bes2.~ bes2.\n}\n\ntenor = {\n  bes4 aes4 bes4 |\n  c'2. bes2. g4 f4 g4 |\n  aes2.~ aes2. aes4 g4 aes4 |\n  bes2. aes2. aes4 g4 f4 |\n  g2.~ g2. bes4 aes4 bes4 |\n  c'2. bes2. g4 g4 g4 |\n  aes2.~ aes2. aes4 bes4 aes4 |\n  g2. bes2. aes4 g4 aes4 |\n  g2.~ g2.\n}\n\nbass = {\n  ees4 ees4 ees4 |\n  ees2. ees2. ees4 ees4 ees4 |\n  bes,2.~ bes,2. bes,4 bes,4 bes,4 |\n  bes,2. bes,2. bes,4 bes,4 bes,4 |\n  ees2.~ ees2. ees4 ees4 ees4 |\n  ees2. ees2. ees4 ees4 ees4 |\n  aes,2.~ aes,2. aes,4 aes,4 aes,4 |\n  bes,2. bes,2. bes,4 bes,4 bes,4 |\n  ees2.~ ees2.\n}\n"}, "harmonizations": [{"slug": "adelaide-tabernacle-1921", "tune": "adelaide", "music_file": "adelaide-tabernacle-1921.ly", "canonical_key": {"tonic": "ees", "mode": "major"}, "time": "9/4", "pickup": "2.", "system_breaks": [10, 20, 30], "breaks_manual": true, "final_barline": "|.", "amen": false, "verse_variable_melisma": false, "refrain": null, "source_line": "Setting: George C. Stebbins, 1907, as printed in Tabernacle Hymns, No. 2, 1921. Public domain.", "provenance": {"layer": "harmonization", "source": "tabernacle-hymns-no-2-1921", "source_kind": "hand_encoded", "pd_basis": "pub_pre_1931", "pd_evidence": "Four-part setting of Stebbins's ADELAIDE as printed in Tabernacle Hymns, No. 2 (Chicago: Tabernacle Publishing Co., 1921), hymn 21 \u2014 the leaf carries no printed page number \u2014 headed \"Have Thine Own Way, Lord.\" over \"A. A. P.\" (left) and \"Geo. C. Stebbins.\" (right), marked \"Slowly.\", in E-flat major and 9/4 with a three-quarter pickup, and footed \"Copyright, 1907, by Geo. C. Stebbins. Used by permission. Hope Pub. Co., owner.\" US public domain: the underlying copyright is 1907 and the printing read is 1921, both well before 1931, and no modern arranger stands anywhere in the chain. TWO DATED SCANS OF THAT PLATE WERE READ AND REQUIRED TO AGREE \u2014 archive.org tabernaclehymnsn00chic leaf n29 (2365 px across, staff space 29.28 px, i.e. 14.6 px per diatonic step) and archive.org tabernaclehymnsn02chic leaf n27 (1877 px across, staff space 23.35 px, i.e. 11.7 px per step); the two items are two scans of the one book \u2014 n00chic's own title page, leaf n7, reads \"Tabernacle Hymns No. 2 \u2026 Copyright 1921 \u2026 Published by Tabernacle Publishing Company, 29 South La Salle Street, Chicago, Illinois\" \u2014 and at this hymn n00chic runs two leaves later than n02chic. THE NOTEHEADS WERE MEASURED, NOT EYEBALLED: each scan was binarised against its own local paper level (ink darker than 82% of a 41x41 grey dilation smoothed by a 41x41 box filter) and closed 5x5 to seal the gaps the plate's ink leaves inside a head; holes smaller than 1.2 staff spaces squared were then filled, which turns each hollow dotted-half OUTLINE into a disc; staff lines were suppressed by vertical run length AFTER that fill, so a hollow head's thin arcs survive; heads were isolated by morphological opening (0.62 staff space horizontally, 0.45 vertically), and every core converted to a diatonic step against a FRESH five-line fit made in a window around that core's own x. Half and dotted-half heads on this plate are WIDER than the filled ones \u2014 44 px against 30 px on scan 1 \u2014 and were invisible to the detector until the width window was opened to 1.80 staff spaces. Counting steps from the bottom staff line, the readings inside the staff are tight: the 25 cores on the bottom line measured -0.139 to +0.110, the 21 on the second line 1.903 to 2.086, the 10 in the top space 6.974 to 7.081. Outside it the plate squeezes the heads back toward the staff: the 9 cores in the space above the bass staff measured 8.655 to 8.818, the 9 in the space below the treble staff -0.695 to -1.007, the 13 between the first and second ledger lines below -2.630 to -2.916. The classes never overlap. 144 of the 160 note positions on scan 1 were covered by a core within 0.5 step of the reading that ships (mean distance 0.094 step, largest 0.496) and 117 of 160 on scan 2 (mean 0.118, largest 0.484); the rest the automated pass did not resolve \u2014 hollow dotted-half heads it still drops, and the bass chord of soprano note 6, whose merged pair it split at 7.53/4.50 instead of 7/5 on both scans. SO EVERY ONE OF THE 160 WAS READ A THIRD WAY, off crops with the fitted five-line grid drawn over the page at 3.4x to 9x, ON BOTH SCANS, and it is that third reading and not any centroid that ships. It corrected two chords the eye had got wrong at low magnification \u2014 the alto of the first chord of system 2 (D4, not E-flat 4) and the bass under \"will,\" (A-flat 2 in the space, not G2 on the line) \u2014 and the automated figure agreed with the correction both times. The two grid readings agree with each other on all 40 chords in all four voices \u2014 160 notes, 0 disagreements. ACCIDENTALS WERE SWEPT SEPARATELY AND BY EYE, at high magnification, over every column of BOTH leaves; the automated figures above see staff positions only and never covered accidentals. THIS SWEEP FIRST GOT IT WRONG, and the error reached the notes: it reported exactly two printed accidentals, both naturals in the soprano, and said none appears in the alto, tenor or bass anywhere. There are THREE. The plate prints, on both leaves: a natural on the bottom line before soprano note 12 (\"art\", E natural); a natural in the second space before soprano note 27 (\"-er\" of \"Aft-er\", A natural); AND A FLAT before the ALTO notehead of the first chord of the third system (soprano note 26, \"Aft-\"), making that alto D-FLAT 4, with the measure rule carrying it to note 28. Both were first encoded as D natural, putting an E-flat MAJOR seventh where the plate writes an E-flat DOMINANT seventh resolving to the A-flat at \"will,\". Corrected. Rule 2b in one line: a notehead probe reads heads, never the flats in front of them. The three-flat signature stands at the head of every staff. NO SCAFFOLD WAS USED AND NOTHING WAS CHECKED AGAINST ONE: no MIDI, no .nwc, no Cyber Hymnal asset and no existing encoding was opened at any point. The only external check is Hymnary's melodic incipit for tune authority adelaide_stebbins, \"323431712212322\", which is the first fifteen soprano note-GROUPS of this reading exactly. DEPARTURES FROM THE PAGE, in full: (1) the page casts three systems and breaks after soprano notes 13 and 25, and this encoding breaks after 10, 20 and 30, so all four systems hold different content from the plate's three \u2014 the plate's first break falls inside poetic line 3 and the house rule puts breaks at poetic line ends; (2) the plate's \"Slowly.\" tempo word is not carried, this corpus holding no tempo marks. Nothing else departs: every pitch, every duration, both naturals, all four ties and the closing light-heavy double bar are the plate's, and the four voices restrike together at all 40 chords because that is what the plate prints.", "pd_jurisdictions": {"US": true}, "encoding_origin": "Hand-encoded from the page images of Tabernacle Hymns, No. 2 (1921), hymn 21 \u2014 archive.org tabernaclehymnsn00chic leaf n29, collated notehead by notehead against archive.org tabernaclehymnsn02chic leaf n27. No MIDI, no Cyber Hymnal asset and no existing encoding was used at any point.", "encoding_license": "PD-dedication"}, "music_ly": "%% ADELAIDE (5.4.5.4.D) \u2014 George C. Stebbins. Four SATB voice streams,\n%% canonical key E-flat major, 9/4, a three-quarter (dotted-half) pickup,\n%% eight full bars plus that pickup and a six-quarter close.\n%% RULES (CI-enforced): exactly the variables soprano/alto/tenor/bass/global;\n%% absolute octaves; each voice rhythmically complete; NO chords <>, NO <<>>,\n%% NO \\secondShift/\\unisonWholeShift/tightSpacing, NO \\break, NO \\bar, NO lyrics.\n%%\n%% SOURCE OF THE NOTES. Tabernacle Hymns, No. 2 (Chicago: Tabernacle Publishing\n%% Co., 1921), hymn 21 \u2014 the leaf carries no printed page number \u2014 headed\n%% \"Have Thine Own Way, Lord.\"\n%% over \"A. A. P.\" on the left and \"Geo. C. Stebbins.\" on the right, marked\n%% \"Slowly.\", and footed \"Copyright, 1907, by Geo. C. Stebbins. Used by\n%% permission. Hope Pub. Co., owner.\" Read off TWO independent scans of that\n%% same plate: archive.org tabernaclehymnsn00chic leaf n29 (2365 px across,\n%% staff space 29.28 px = 14.6 px per diatonic step) and archive.org\n%% tabernaclehymnsn02chic leaf n27 (1877 px across, staff space 23.35 px =\n%% 11.7 px per diatonic step). The two items are two scans of the one book \u2014\n%% n00chic's own title page (leaf n7) reads \"Tabernacle Hymns No. 2 \u2026 Copyright\n%% 1921 \u2026 Published by Tabernacle Publishing Company, 29 South La Salle Street,\n%% Chicago, Illinois\" \u2014 and at this hymn n00chic runs two leaves later than\n%% n02chic.\n%%\n%% HOW THE NOTEHEADS WERE READ (not by eye). Per scan, with its own\n%% binarisation and its own staff-line fits: ink is whatever is darker than 82%\n%% of the LOCAL paper level (41x41 grey dilation then a 41x41 box filter),\n%% followed by a 5x5 binary closing to seal the pale gaps the plate's ink\n%% leaves inside a head; then holes smaller than 1.2 staff spaces squared are\n%% filled, which turns each hollow (dotted-half) OUTLINE into a disc without\n%% flooding the boxes that staff lines and stems enclose; then staff lines are\n%% suppressed by vertical run length \u2014 AFTER the fill, so a hollow head's thin\n%% top and bottom arcs are already part of a tall run and survive; then a\n%% horizontal opening of 0.62 staff space and a vertical opening of 0.45 staff\n%% space delete stems, flags, barlines and ledger lines and leave a head core.\n%% Every core was converted to a diatonic step against a FRESH five-line fit\n%% made in a +-2.6-staff-space window around that core's own x, so the plate's\n%% skew cannot drag a reading. Half and dotted-half heads on this plate are\n%% WIDER than the filled ones (44 px against 30 px at scan 1) and were lost\n%% until the width window was opened to 1.80 staff spaces; merged\n%% soprano-over-alto blobs were split at the midpoint of their two clean outer\n%% edges with the separation snapped to a whole step.\n%%\n%% Counting steps from the bottom staff line (0 = bottom line, 8 = top line),\n%% the measurement inside the staff is tight: the 25 cores on the bottom line\n%% measured between -0.139 and +0.110, the 21 on the second line between 1.903\n%% and 2.086, the 10 in the top space between 6.974 and 7.081. OUTSIDE the\n%% staff the plate itself squeezes the heads back toward it: the 9 cores in the\n%% space above the bass staff measured 8.655 to 8.818, the 9 in the space below\n%% the treble staff -0.695 to -1.007, the 13 between the first and second\n%% ledger lines below -2.630 to -2.916. The classes never overlap, so no\n%% rounding is in doubt.\n%%\n%% 144 of the 160 note positions on scan 1 were covered by a head core within\n%% 0.5 step of the reading below (mean distance 0.094 step, largest 0.496); the\n%% other 16 the automated pass did not resolve \u2014 hollow dotted-half heads it\n%% still drops, plus the bass chord of soprano note 6, whose merged pair it\n%% split at 7.53/4.50 instead of 7/5. On scan 2, 117 of the 160 were covered\n%% (mean 0.118, largest 0.484). So EVERY one of the 160 was read a third way as\n%% well, off crops carrying the fitted five-line grid drawn over the page at\n%% 3.4x to 9x, on BOTH scans \u2014 and it is that third reading, not any centroid,\n%% that ships. It corrected two chords the eye had got wrong at low\n%% magnification: the alto of the first chord of system 2 (D4, not E-flat 4)\n%% and the bass under \"will,\" (A-flat 2 in the space, not G2 on the line). The\n%% two grid readings agree with each other at all 40 chords in all four\n%% voices \u2014 160 notes, 0 disagreements.\n%%\n%% ACCIDENTALS ARE NOT WHAT A NOTEHEAD PROBE SEES. Every column of both leaves\n%% was swept separately BY EYE at high magnification for printed sharps, flats\n%% and naturals; the automated figures above cover staff positions only and\n%% never covered accidentals. The sweep found exactly TWO in the body of the\n%% hymn, both naturals, both in the soprano, and both are on both leaves:\n%% a natural on the bottom line before soprano note 12 (the syllable \"art\",\n%% E natural) and a natural in the second space before soprano note 27 (the\n%% syllable \"-er\" of \"Af-ter\", A natural). Nothing else: no accidental in the\n%% alto, tenor or bass anywhere, and the three-flat signature stands at the\n%% head of every staff.\n%%\n%% NO SCAFFOLD WAS USED, and nothing here was checked against one. No MIDI, no\n%% .nwc, no Cyber Hymnal asset and no existing encoding was opened at any\n%% point. The one external check is Hymnary's melodic incipit for\n%% tuneAuthNumber adelaide_stebbins, \"323431712212322\", which is the first\n%% fifteen soprano note-GROUPS of this reading exactly.\n%%\n%% 40 soprano noteheads carry 36 syllables. The four two-note groups are TIES,\n%% not slurs \u2014 the two heads are the same pitch in all four voices \u2014 and they\n%% fall on the last syllable of poetic lines 2, 4, 6 and 8 (\"way!\", \"clay.\",\n%% \"will,\", \"still.\"): soprano notes 9-10, 19-20, 29-30 and 39-40.\n%%\n%% DEPARTURE FROM THE PLATE'S CASTING. The page casts three systems, breaking\n%% after soprano notes 13 and 25. The first of those falls in the MIDDLE of\n%% poetic line 3, which this corpus does not do, so the systems here break\n%% after notes 10, 20 and 30 \u2014 the ends of poetic lines 2, 4 and 6, which are\n%% also barlines \u2014 giving four systems of ten noteheads instead of the plate's\n%% 13/12/15.\n\\version \"2.24.0\"\n\nglobal  = { \\key ees \\major \\time 9/4 \\partial 2. }\n\nsoprano = {\n  g'4 f'4 g'4 |\n  aes'2. g'2. ees'4 d'4 ees'4 |\n  f'2.~ f'2. f'4 e'4 f'4 |\n  g'2. f'2. f'4 ees'4 d'4 |\n  ees'2.~ ees'2. g'4 f'4 g'4 |\n  aes'2. g'2. bes'4 a'4 bes'4 |\n  c''2.~ c''2. c''4 d''4 c''4 |\n  bes'2. g'2. f'4 ees'4 f'4 |\n  ees'2.~ ees'2.\n}\n\nalto = {\n  ees'4 ees'4 ees'4 |\n  ees'2. ees'2. bes4 bes4 bes4 |\n  d'2.~ d'2. d'4 d'4 d'4 |\n  d'2. d'2. bes4 bes4 bes4 |\n  bes2.~ bes2. ees'4 ees'4 ees'4 |\n  ees'2. ees'2. des'4 c'4 des'4 |\n  ees'2.~ ees'2. ees'4 ees'4 ees'4 |\n  ees'2. ees'2. bes4 bes4 bes4 |\n  bes2.~ bes2.\n}\n\ntenor = {\n  bes4 aes4 bes4 |\n  c'2. bes2. g4 f4 g4 |\n  aes2.~ aes2. aes4 g4 aes4 |\n  bes2. aes2. aes4 g4 f4 |\n  g2.~ g2. bes4 aes4 bes4 |\n  c'2. bes2. g4 g4 g4 |\n  aes2.~ aes2. aes4 bes4 aes4 |\n  g2. bes2. aes4 g4 aes4 |\n  g2.~ g2.\n}\n\nbass = {\n  ees4 ees4 ees4 |\n  ees2. ees2. ees4 ees4 ees4 |\n  bes,2.~ bes,2. bes,4 bes,4 bes,4 |\n  bes,2. bes,2. bes,4 bes,4 bes,4 |\n  ees2.~ ees2. ees4 ees4 ees4 |\n  ees2. ees2. ees4 ees4 ees4 |\n  aes,2.~ aes,2. aes,4 aes,4 aes,4 |\n  bes,2. bes,2. bes,4 bes,4 bes,4 |\n  ees2.~ ees2.\n}\n", "preferred": true}], "license": "The hymn texts and tunes are in the public domain; per-layer provenance is recorded on each record."}