Hugging Face's logo

Spaces:
caustino
/
MercyProtocol
Running

App Files Community
MercyProtocol
File size: 65,551 Bytes
6b1772e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
82df7b8
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8">
<title>THE MERCY PROTOCOL — A Book of Systems, Wounds, and the Architecture of Repair</title>
<style>
  @page { size: Letter; margin: 1in; @bottom-center { content: counter(page); font-size: 10pt; color:#555; } }
  html { font-family: "Georgia", "Times New Roman", serif; font-size: 11pt; line-height: 1.55; color:#1a1a1a; }
  body { max-width: 6.5in; margin: 0 auto; }
  h1,h2,h3,h4 { font-family: "Helvetica Neue", "Arial", sans-serif; color:#1B1B3A; }
  h1.part-title { text-align:center; font-size: 26pt; margin-top: 1.5in; margin-bottom: 0.5in; border-top: 3px double #1B1B3A; border-bottom: 3px double #1B1B3A; padding: 18pt 0; color:#8B0000; page-break-before: always; }
  h1.chapter { font-size: 20pt; border-bottom: 2px solid #8B0000; padding-bottom: 6pt; margin-top: 36pt; page-break-before: always; }
  h2 { font-size: 14pt; margin-top: 22pt; }
  .preface h1.chapter { page-break-before: avoid; }
  .dedication { text-align:center; font-style: italic; margin: 2in 0; }
  .dedication-box { background: #1B1B3A; color: white; padding: 22pt; margin: 28pt 0; border-radius: 4pt; }
  .dedication-box h3 { color: #C9A227; margin-top: 0; text-align:center; }
  .metaphor { background: #F5F5F5; border-left: 4pt solid #1B1B3A; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; }
  .metaphor .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#1B1B3A; display:block; margin-bottom:4pt; }
  .phd { background: #faf5e4; border-left: 4pt solid #C9A227; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; }
  .phd .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#8a6d00; display:block; margin-bottom:4pt; }
  .peer { background: #e8f0fb; border-left: 4pt solid #1565C0; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; font-size: 10.5pt; }
  .peer .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#0d47a1; display:block; margin-bottom:4pt; }
  .shot { background: #f0f0f7; border-left: 4pt solid #1B1B3A; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; }
  .shot .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#1B1B3A; display:block; margin-bottom:4pt; }
  .cobol { background: #e8eefb; border-left: 4pt solid #0033A0; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; }
  .cobol .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#0033A0; display:block; margin-bottom:4pt; }
  .copernicus { background: #fbeaea; border-left: 4pt solid #8B0000; padding: 10pt 14pt; margin: 16pt 0; border-radius: 2pt; }
  .copernicus .label { font-weight: bold; font-family: sans-serif; text-transform: uppercase; font-size: 9pt; letter-spacing: 1pt; color:#8B0000; display:block; margin-bottom:4pt; }
  figure { margin: 22pt 0; text-align:center; page-break-inside: avoid; }
  figcaption { font-size: 9.5pt; color:#444; margin-top:8pt; font-style: italic; }
  pre { background:#f7f7f7; border:1pt solid #ddd; padding:10pt; font-size: 9pt; overflow-x:auto; border-radius: 2pt; }
  .algorithm { background: #fcfbf7; border: 1pt solid #C9A227; border-radius: 3pt; margin:16pt 0; padding: 0 14pt 12pt 14pt; page-break-inside: avoid; }
  .algorithm h4 { color:#8a6d00; margin-top: 12pt; font-family: sans-serif; font-size: 10.5pt; }
  .algorithm pre { background: transparent; border: none; padding: 0; }
  ul, ol { padding-left: 22pt; }
  table { border-collapse: collapse; width: 100%; margin: 14pt 0; font-size: 10pt; }
  th, td { border: 1pt solid #ccc; padding: 6pt 8pt; text-align: left; vertical-align: top; }
  th { background: #1B1B3A; color: white; font-family: sans-serif; }
  .twocol { display: grid; grid-template-columns: 1fr 1fr; gap: 18pt; }
  .huggingface { text-align:center; margin: 30pt 0; font-family: sans-serif; }
  .huggingface a { color:#1565C0; text-decoration: none; font-weight: bold; font-size: 13pt; }
  .math { font-family: "Cambria Math", "Latin Modern Math", Georgia, serif; font-style: italic; }
  .toc { page-break-after: always; }
  .toc h2 { text-align:center; }
  .toc ul { list-style: none; padding-left:0; }
  .toc li { margin: 5pt 0; }
  .toc li.part { font-weight: bold; margin-top:10pt; font-family: sans-serif; color:#1B1B3A; }
  .cover .cover-subtitle { max-width: 5.5in; margin: 0 auto; }
</style>
</head>
<body>

<!-- COVER -->
<section style="height: 9in; display: flex; flex-direction: column; justify-content: center; align-items: center; text-align: center; page-break-after: always;">
  <div style="border: 4px double #1B1B3A; padding: 40pt 60pt;">
    <div style="font-family: sans-serif; font-size: 10pt; letter-spacing: 2pt; color:#8B0000; text-transform: uppercase; margin-bottom: 20pt;">Technical · Legal · Social Systems</div>
    <h1 style="font-family: sans-serif; font-size: 36pt; color:#1B1B3A; margin:0; line-height:1.2;">THE MERCY<br>PROTOCOL</h1>
    <div style="font-family: Georgia, serif; font-style: italic; font-size: 14pt; color:#333; margin: 24pt 0; max-width: 4.5in;">A Book of Systems, Wounds, and the Architecture of Repair<br><br>With a Copernican dedication to COBOL and the quiet engineers of the administrative state</div>
    <div style="font-family: sans-serif; font-size: 12pt; margin-top: 30pt; color:#1565C0;">
      Caustino<br>
      <a href="https://huggingface.co/caustino">huggingface.co/caustino</a><br>
      <span style="font-size:10pt; color:#666;">2026</span>
    </div>
  </div>
</section>

<!-- DEDICATION -->
<section class="dedication">
  <p>Dedicated to Grace Murray Hopper, who taught a machine to read English<br>
  so that ordinary people might read what the machine writes;<br>
  and to the millions who have waited in queues they did not design,<br>
  for benefits they had already earned.</p>

  <div class="dedication-box">
    <h3>THE COPERNICAN DEDICATION TO COBOL</h3>
    <p>When the heavens were re-centered from the Earth to the Sun, nothing in
    the sky moved an inch — but everything in how humans saw the world changed.
    So too with COBOL: when we re-center the administrative state from the
    <em>agency</em> to the <em>person</em>, not a single line of legacy code needs to
    vanish, but the justice produced by that code is transformed.</p>
    <p>Grace Hopper did not write COBOL because she loved mainframes. She wrote it
    because she believed business logic should be legible to the people whose
    lives it governs. Sixty-six years later, 70–80% of the world's financial
    transactions still run on COBOL; and yet the very same clarity-of-language
    principle that birthed COBOL is the principle we must now carry forward into
    the post-quantum, genomic, algorithmic era. This book is an attempt to do
    exactly that.</p>
  </div>
</section>

<!-- TOC -->
<section class="toc">
<h2>Contents</h2>
<ul>
  <li class="part">Preface: The Shape of the Problem</li>
  <li class="part">Part I — The Wound</li>
  <li>1. The Queue That Kills</li>
  <li>2. The Genome as a Book That Cannot Be Rewritten</li>
  <li>3. The Administrative Recidivism Loop</li>
  <li class="part">Part II — The Ghost in the Genome</li>
  <li>4. The Cryptographic Envelope</li>
  <li>5. The Audit Ledger as Unblinking Memory</li>
  <li>6. The Judicial Escrow: When the State Must See</li>
  <li class="part">Part III — See (The Eyes of the System)</li>
  <li>7. What It Means to See a Claimant</li>
  <li>8. HSPA v3.0 — Homomorphic Scoring</li>
  <li>9. The Claim-Evidence Chain</li>
  <li class="part">Part IV — Remember (The Ledger as Institutional Memory)</li>
  <li>10. What the State Forgets</li>
  <li>11. Immutable Recordkeeping in a Post-Quantum World</li>
  <li class="part">Part V — Reason (MORTAR and the Mathematics of Triage)</li>
  <li>12. Why First-In-First-Out Is a Moral Choice</li>
  <li>13. Reproducibility as a Procedural Right</li>
  <li>14. The Ramp, Not the Cliff</li>
  <li class="part">Copernican Interlude — Why COBOL Matters Here</li>
  <li class="part">Part VI — Endure (Finance, Law, and the Political Cycle)</li>
  <li>15. The Political-Cycle Problem</li>
  <li>16. The Capital Structure</li>
  <li>17. The Legal Superstructure</li>
  <li>18. The Future We Are Trying to Reach</li>
  <li class="part">References</li>
</ul>
</section>

<!-- PREFACE -->
<section class="preface">
<h1 class="chapter">Preface: The Shape of the Problem</h1>

<div class="shot"><span class="label">Cinematic Shot — Establishing, aerial, dawn</span>
A river winds between concrete levees. From above you can see the flow slow,
then stall. Boats — thousands of them, each carrying a person, a file, a
claim — rest on water that no longer moves. Some have been there so long
their hulls have taken on water. The camera tilts down. A paper file rests
on a desk inside a federal building; a stamp says "PENDING."
</div>

<p>There is a river that runs through the American administrative state. It
was designed to carry people from injury to remedy, from diagnosis to
support, from need to the means of living. For most of the twentieth
century, it moved with a certain bureaucratic sluggishness — slow, but
forward. In the twenty-first century, the river has backed up. It has
become a pond, then a swamp, then a place where people go to wait and some
of them do not come out.</p>

<p>The Social Security Administration's disability adjudication system is
not a failure of malice. It is a failure of scale — a system designed for a
world of paper and human clerks, now processing the claims of a nation with
the tools of a bygone era. The numbers are stark: over one million claims
pending, average waits measured in hundreds of days, and deaths — tens of
thousands per year — among those waiting.<sup>[1,2]</sup> These are not
abstract statistics. They are the human cost of a system that cannot see
the people inside its queue.</p>

<p>This book is about a response to that failure — and to a cluster of
related failures that share a common root. The root is this:
<strong>the institutions that hold the most sensitive information about us
— our genomes, our neurophysiology, our administrative records, our
financial identities — are constrained by instruments drafted for a world
in which such information was either impossible or expensive to
collect.</strong> The genome is the clearest case. It is immutable, it
identifies without being intended to identify, it implicates relatives who
never consented, and the statute that governs it reaches only health
insurance and employment. The gap between the sensitivity of the data and
the enforceability of the duty is not a single wrong. It is a structural
under-protection that appears as a civil-rights claim in one forum, as an
evidentiary problem in another, and as a queue in a third.</p>

<p>The response described in this book is called the MERCY Protocol — a name
chosen not for sentiment but for precision. Mercy, in the administrative
sense, is the act of seeing a person where the system sees only a case
number. The Protocol is a <strong>stack of systems</strong> that together
attempt to do four things:</p>
<ol>
  <li><strong>See</strong> — detect the people at risk, before the queue kills them;</li>
  <li><strong>Remember</strong> — create an immutable, auditable record of what was done and why, signed against a post-quantum future;</li>
  <li><strong>Reason</strong> — apply triage and evidence standards that are fair, transparent, and reproducible;</li>
  <li><strong>Endure</strong> — build a financial and legal structure that can outlast the political cycles that otherwise doom such efforts.</li>
</ol>
<p>Each layer is necessary. None is sufficient alone.</p>

<div class="copernicus"><span class="label">The Copernican Turn</span>
The MERCY Protocol is a Copernican move. For seventy years, the
administrative state has placed the <em>agency</em> at the center of the
universe, with the claimant orbiting it, waiting for light to arrive from
the center. The Protocol moves the <em>person</em> to the center. The
agency, the courts, the trustees, the nodes, the ledgers, the cryptographic
keys — all orbit the person whose life is at stake. Nothing in statute
changes; everything in justice does.
</div>

<p>The book is organized as a narrative, moving from the wound to the
architecture to the implementation. Along the way I use diagrams and
cinematic metaphors to make the systems visible, because the only way to
hold a system this large in a human mind is to see it as something else: a
river, a ledger, a flock, a heart, a ramp.</p>
</section>

<!-- PART I -->
<h1 class="part-title">PART I — THE WOUND</h1>

<h1 class="chapter">1. The Queue That Kills</h1>

<div class="shot"><span class="label">Cinematic Shot — Wide, hospital ER, stripped of staff</span>
The triage desk has been replaced by a large, slowly advancing clock.
Patients take a number and sit on benches that stretch out of frame. Some
hold bleeding wounds; some are unconscious; some sit very still. The clock
ticks. No one is called.
</div>

<p>Imagine a hospital where the triage nurse has been replaced by a clock.
Everyone who arrives is given a number. The number is the time of arrival.
The clock advances. When your number is called, you are seen. The clock
does not care whether you are bleeding, whether you are unconscious,
whether you have hours or years. It only knows the order of arrival.</p>

<p>This is the SSA disability queue. It is fair in the narrow sense: it
treats everyone the same. It is lethal in the broader sense: it treats
everyone the same regardless of how much the delay will cost them.</p>

<p>The Government Accountability Office documented that over a decade, more
than <strong>110,000</strong> appellants died while their claims were
pending.<sup>[1]</sup> The SSA's own Office of the Chief Actuary estimated
<strong>33,270 deaths</strong> in FY2023 among those awaiting a decision at
the DDS or ALJ level.<sup>[2]</sup> The average wait for an ALJ hearing
exceeded <strong>230 days</strong> in FY2024, up 86 percent from 2019. The
federal court remand rate — the proportion of cases sent back because the
agency got it wrong — rose to <strong>62 percent</strong>. Of those
remanded, roughly two-thirds were ultimately awarded benefits. <strong>The
delay itself is the denial mechanism.</strong> A claim that is approved
after two years is not a claim that was approved. It is a claim that was
denied for two years and then, belatedly, recognized.</p>

<figure>
<svg viewBox="0 0 600 200" width="100%" role="img" aria-label="Queue as Sorted List">
  <defs>
    <marker id="arr" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="6" markerHeight="6" orient="auto">
      <path d="M0,0 L10,5 L0,10 z" fill="#555"/>
    </marker>
  </defs>
  <!-- Time axis -->
  <line x1="30" y1="110" x2="570" y2="110" stroke="#333" stroke-width="1.5" marker-end="url(#arr)"/>
  <text x="575" y="115" font-size="11" font-family="sans-serif" fill="#333">Time</text>
  <!-- Claims in FIFO boxes -->
  <g font-family="sans-serif" font-size="11">
    <g><rect x="50" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="72" y="87" text-anchor="middle">[1]</text></g>
    <g><rect x="110" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="132" y="87" text-anchor="middle">[2]</text></g>
    <g><rect x="170" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="192" y="87" text-anchor="middle">[3]</text></g>
    <g><rect x="230" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="252" y="87" text-anchor="middle">[4]</text></g>
    <g><rect x="290" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="312" y="87" text-anchor="middle">[5]</text></g>
    <g><rect x="350" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="372" y="87" text-anchor="middle">[6]</text></g>
    <g><rect x="410" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="432" y="87" text-anchor="middle">[7]</text></g>
    <g><rect x="470" y="65" width="45" height="35" fill="#fff" stroke="#1B1B3A" stroke-width="1.2" rx="3"/><text x="492" y="87" text-anchor="middle">[8]</text></g>
  </g>
  <text x="300" y="45" font-family="sans-serif" font-size="12" font-weight="bold" text-anchor="middle">Claims arrive → FIFO sorted list → Decisions issue</text>
  <!-- Death marker -->
  <line x1="250" y1="120" x2="250" y2="175" stroke="#8B0000" stroke-width="1.4" stroke-dasharray="4,3"/>
  <text x="252" y="190" font-family="sans-serif" font-size="10" fill="#8B0000" font-weight="bold">Deaths while waiting</text>
  <text x="300" y="145" font-family="sans-serif" font-size="10" fill="#333" text-anchor="middle">The system is blind to risk.</text>
</svg>
<figcaption>Diagram 1. The Queue as a Sorted List (vector redrawing).</figcaption>
</figure>

<p><strong>Result.</strong> The system has <em>no mechanism</em> to ask:
<em>"Who among these is most likely to die before their number is
called?"</em> It is <em>formally</em> fair, treating everyone identically.
It is <em>substantively</em> lethal, because identical treatment is not
equitable treatment when the people are not identically situated.</p>

<div class="metaphor"><span class="label">Metaphor</span>
The queue is a river that has stopped moving. A river that flows carries
silt and debris and also carries boats. A river that stops becomes a swamp.
The boats do not move. Some of them sink. The swamp does not notice. It is
water. It is doing what water does when it is not moving. The failure is
not in the water. The failure is in the absence of a current.
</div>

<div class="phd"><span class="label">PhD-level Complexity</span>
The queue is a <strong>priority queue with a single key</strong> (arrival
time). The optimal solution, in operations-research terms, is a
<strong>multi-objective scheduling problem</strong> where objectives include
throughput, fairness, and <strong>mortality risk</strong>. The MORTAR
engine (Part V) is the formal solution: a composite priority score
combining medical risk, socioeconomic vulnerability, and wait time, with an
<em>anti-starvation guarantee</em> that no claim waits beyond a bounded
time.
</div>

<div class="peer"><span class="label">Peer-Reviewed Anchor</span>
Gelber, Moore &amp; Strand (2023) estimate that an additional $1,000 in
annual DI income reduces beneficiary mortality by 0.18–0.35 percentage
points, demonstrating that delay is not merely inconvenient but
mortality-implicated.<sup>[3]</sup> Deshpande &amp; Mueller-Smith (2022)
show that removal from SSI at age 18 increases criminal charges by 20%,
income-generating offenses by 60%, and incarceration likelihood by
60%,<sup>[4]</sup> tracing downstream costs of the queue into the
criminal-legal system.
</div>

<h1 class="chapter">2. The Genome as a Book That Cannot Be Rewritten</h1>

<p>The genome is not a medical record. A medical record describes what
happened to you. A genome describes what you are, what you might become,
and what you share with everyone to whom you are related. It is immutable
— you cannot change it. It is uniquely identifying — your genome is yours
alone. It is predictive — it carries probabilistic information about future
disease. It is relational — it reveals information about parents, siblings,
and children who never consented to anything. And it is persistently
sensitive — as science advances, the same data reveals more.</p>

<figure>
<svg viewBox="0 0 500 320" width="100%" role="img" aria-label="Five Properties of Genomic Information">
  <!-- center -->
  <rect x="150" y="130" width="200" height="60" rx="6" fill="#fbeaea" stroke="#8B0000" stroke-width="1.5"/>
  <text x="250" y="155" font-family="sans-serif" font-size="13" font-weight="bold" text-anchor="middle">THE GENOME</text>
  <text x="250" y="175" font-family="sans-serif" font-size="9" text-anchor="middle">3.2 billion base pairs, one per person</text>
  <!-- 5 spokes -->
  <g font-family="sans-serif" font-size="10">
    <g>
      <rect x="30" y="20" width="110" height="70" rx="4" fill="#fff5f5" stroke="#8B0000" stroke-width="1"/>
      <text x="85" y="40" font-weight="bold" text-anchor="middle">IMMUTABLE</text>
      <text x="85" y="60" text-anchor="middle" font-size="9">Cannot change</text>
      <text x="85" y="74" text-anchor="middle" font-size="9">after breach</text>
      <line x1="150" y1="140" x2="140" y2="90" stroke="#8B0000" stroke-width="0.8" marker-end="url(#arr)"/>
    </g>
    <g>
      <rect x="360" y="20" width="110" height="70" rx="4" fill="#faf5e4" stroke="#C9A227" stroke-width="1"/>
      <text x="415" y="40" font-weight="bold" text-anchor="middle">UNIQUELY</text>
      <text x="415" y="55" font-weight="bold" text-anchor="middle">IDENTIFYING</text>
      <text x="415" y="74" text-anchor="middle" font-size="9">Yours alone</text>
      <line x1="350" y1="140" x2="360" y2="90" stroke="#C9A227" stroke-width="0.8" marker-end="url(#arr)"/>
    </g>
    <g>
      <rect x="20" y="230" width="110" height="70" rx="4" fill="#eaf5ea" stroke="#2E7D32" stroke-width="1"/>
      <text x="75" y="250" font-weight="bold" text-anchor="middle">PREDICTIVE</text>
      <text x="75" y="270" text-anchor="middle" font-size="9">Probabilistic</text>
      <text x="75" y="284" text-anchor="middle" font-size="9">future health risks</text>
      <line x1="160" y1="190" x2="130" y2="230" stroke="#2E7D32" stroke-width="0.8" marker-end="url(#arr)"/>
    </g>
    <g>
      <rect x="195" y="240" width="110" height="70" rx="4" fill="#e8f0fb" stroke="#1565C0" stroke-width="1"/>
      <text x="250" y="260" font-weight="bold" text-anchor="middle">RELATIONAL</text>
      <text x="250" y="280" text-anchor="middle" font-size="9">Reveals data about</text>
      <text x="250" y="294" text-anchor="middle" font-size="9">relatives who never consented</text>
      <line x1="250" y1="190" x2="250" y2="240" stroke="#1565C0" stroke-width="0.8" marker-end="url(#arr)"/>
    </g>
    <g>
      <rect x="370" y="230" width="110" height="70" rx="4" fill="#eee8fb" stroke="#311B92" stroke-width="1"/>
      <text x="425" y="250" font-weight="bold" text-anchor="middle">PERSISTENTLY</text>
      <text x="425" y="265" font-weight="bold" text-anchor="middle">SENSITIVE</text>
      <text x="425" y="284" text-anchor="middle" font-size="9">Becomes more</text>
      <text x="425" y="298" text-anchor="middle" font-size="9">revealing over time</text>
      <line x1="340" y1="190" x2="370" y2="230" stroke="#311B92" stroke-width="0.8" marker-end="url(#arr)"/>
    </g>
  </g>
</svg>
<figcaption>Diagram 2. The Five Properties of Genomic Information.</figcaption>
</figure>

<p><strong>The legal problem.</strong> Existing privacy frameworks (HIPAA,
GINA) were drafted for a world of paper medical records and rare genetic
data. The genome breaks every category:</p>
<ul>
  <li>HIPAA covers "protected health information" held by covered entities.
  It does not cover direct-to-consumer genomic data.</li>
  <li>GINA prohibits genetic discrimination in health insurance and
  employment. It does not cover life insurance, disability insurance, or
  long-term care.</li>
  <li>The Third-Party Doctrine (<em>Smith v. Maryland</em>, <em>United
  States v. Miller</em>) holds that information voluntarily given to a
  third party loses Fourth Amendment protection. Applied mechanically to
  genomic data, this would extinguish constitutional protection for data
  disclosed to a testing company.</li>
</ul>

<div class="metaphor"><span class="label">Metaphor</span>
The genome is a book written in a language that is only partially
understood. Every year, the language becomes more legible. A page that
today says "increased risk of X" will, in ten years, say "increased risk of
X, Y, and Z, and also reveals that your sister has the same variant." The
book cannot be rewritten. It can only be copied. And every copy is a
potential leak.
</div>

<div class="phd"><span class="label">PhD-level Complexity</span>
The re-identification problem is formalizable. Let <span class="math">𝒮</span> be a set of <span class="math">k</span> independent SNPs. Under Hardy–Weinberg equilibrium,
<span class="math">P(𝒢<sub>𝒮</sub>) = ∏ P(g<sub>j</sub>)</span>. The
probability that an individual is unique in a population of size <span class="math">N</span> is <span class="math">P(ID) = 1 − (1 −
P(𝒢<sub>𝒮</sub>))<sup>N−1</sup></span>, which tends to 1 as
<span class="math">k → ∞</span>. Empirically, <span class="math">k ≥
50</span> independent SNPs yields P(ID) ≈ 1. <strong>Anonymization of
high-density genomic data is mathematically impossible without
cryptographic noise or enclave isolation</strong> — the formal
justification for the post-quantum stack in Part II.
</div>

<h1 class="chapter">3. The Administrative Recidivism Loop</h1>

<p>The disability system does not simply delay. It <strong>recycles</strong>.
A claim is denied at the initial level. It is denied again at
reconsideration. It goes to an ALJ, who may deny it or approve it. If
approved, a continuing disability review may later terminate it. If
denied, it goes to the Appeals Council, then to federal court, then back
to the agency. Each cycle costs time. Each cycle costs money. Each cycle
costs health. The claimant is not moving through a system. The claimant is
<strong>circulating</strong> through a system that has no exit.</p>

<figure>
<svg viewBox="0 0 560 520" width="100%" role="img" aria-label="Recidivism Feedback Loop">
  <defs>
    <marker id="arr2" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto">
      <path d="M0,0 L10,5 L0,10 z" fill="#555"/>
    </marker>
    <marker id="arrred" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="7" markerHeight="7" orient="auto">
      <path d="M0,0 L10,5 L0,10 z" fill="#8B0000"/>
    </marker>
  </defs>
  <g font-family="sans-serif" font-size="10">
    <rect x="210" y="10" width="140" height="40" rx="4" fill="#fff" stroke="#1B1B3A" stroke-width="1.2"/>
    <text x="280" y="35" text-anchor="middle" font-weight="bold">DISABILITY ONSET</text>
    <line x1="280" y1="50" x2="280" y2="75" stroke="#555" stroke-width="1.2" marker-end="url(#arr2)"/>

    <rect x="210" y="75" width="140" height="40" rx="4" fill="#fff" stroke="#1B1B3A" stroke-width="1.2"/>
    <text x="280" y="100" text-anchor="middle" font-weight="bold">INITIAL CLAIM</text>

    <!-- Three denials -->
    <rect x="50" y="150" width="100" height="45" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="100" y="170" text-anchor="middle" font-weight="bold">DENIED</text>
    <text x="100" y="186" text-anchor="middle" font-size="8">(64%)</text>
    <line x1="225" y1="115" x2="150" y2="150" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="230" y="150" width="100" height="45" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="280" y="170" text-anchor="middle" font-weight="bold">DENIED</text>
    <text x="280" y="186" text-anchor="middle" font-size="8">(84% of appeals)</text>
    <line x1="280" y1="115" x2="280" y2="150" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="410" y="150" width="100" height="45" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="460" y="170" text-anchor="middle" font-weight="bold">DENIED</text>
    <text x="460" y="186" text-anchor="middle" font-size="8">(ALJ denial)</text>
    <line x1="335" y1="115" x2="410" y2="150" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <!-- Appeal -->
    <rect x="200" y="230" width="160" height="55" rx="4" fill="#fff" stroke="#1B1B3A" stroke-width="1.2"/>
    <text x="280" y="252" text-anchor="middle" font-weight="bold">APPEAL</text>
    <text x="280" y="270" text-anchor="middle" font-size="9">(ALJ, AC, federal court)</text>
    <line x1="100" y1="195" x2="240" y2="230" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>
    <line x1="280" y1="195" x2="280" y2="230" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>
    <line x1="460" y1="195" x2="320" y2="230" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <!-- Outcomes -->
    <rect x="40" y="320" width="110" height="45" rx="4" fill="#eaf5ea" stroke="#2E7D32" stroke-width="1"/>
    <text x="95" y="340" text-anchor="middle" font-weight="bold">APPROVED</text>
    <text x="95" y="356" text-anchor="middle" font-size="8">(50% at ALJ)</text>
    <line x1="230" y1="285" x2="150" y2="320" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="225" y="320" width="110" height="45" rx="4" fill="#faf5e4" stroke="#C9A227" stroke-width="1"/>
    <text x="280" y="340" text-anchor="middle" font-weight="bold">REMANDED</text>
    <text x="280" y="356" text-anchor="middle" font-size="8">(62% of fed court)</text>
    <line x1="280" y1="285" x2="280" y2="320" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="410" y="320" width="110" height="45" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="465" y="340" text-anchor="middle" font-weight="bold">DENIED</text>
    <text x="465" y="356" text-anchor="middle" font-size="8">(final)</text>
    <line x1="330" y1="285" x2="420" y2="320" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <!-- End states -->
    <rect x="40" y="400" width="110" height="35" rx="4" fill="#eaf5ea" stroke="#2E7D32" stroke-width="1"/>
    <text x="95" y="422" text-anchor="middle" font-weight="bold">BENEFITS</text>
    <line x1="95" y1="365" x2="95" y2="400" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="225" y="400" width="110" height="35" rx="4" fill="#faf5e4" stroke="#C9A227" stroke-width="1"/>
    <text x="280" y="422" text-anchor="middle" font-weight="bold">BACK TO QUEUE</text>
    <line x1="280" y1="365" x2="280" y2="400" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="410" y="400" width="110" height="35" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="465" y="422" text-anchor="middle" font-weight="bold">COLLAPSE</text>
    <line x1="465" y1="365" x2="465" y2="400" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <rect x="420" y="460" width="120" height="40" rx="4" fill="#f8dede" stroke="#8B0000" stroke-width="1"/>
    <text x="480" y="478" text-anchor="middle" font-size="9" font-weight="bold">HOMELESSNESS /</text>
    <text x="480" y="492" text-anchor="middle" font-size="9" font-weight="bold">CRIMINALIZATION</text>
    <line x1="465" y1="435" x2="480" y2="460" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <text x="280" y="480" text-anchor="middle" font-size="9" font-style="italic">↓ reduced employability ↓</text>
    <rect x="225" y="495" width="110" height="22" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1"/>
    <text x="280" y="510" text-anchor="middle" font-size="9" font-weight="bold">REAPPLICATION</text>
    <line x1="450" y1="500" x2="335" y2="508" stroke="#555" stroke-width="1" marker-end="url(#arr2)"/>

    <!-- feedback loop back up -->
    <path d="M 225 505 C 80 505, 80 95, 210 95" fill="none" stroke="#8B0000" stroke-width="1.4" stroke-dasharray="5,4" marker-end="url(#arrred)"/>
    <text x="60" y="300" font-size="8" fill="#8B0000" font-weight="bold">cycle continues</text>
  </g>
</svg>
<figcaption>Diagram 3. The Administrative Recidivism Feedback Loop.</figcaption>
</figure>

<p><strong>Key evidence.</strong></p>
<ul>
  <li>SSI removal at 18 → +20% criminal charges, +60% income-generating offenses, +60% incarceration likelihood.<sup>[4]</sup></li>
  <li>$1,000 additional DI income → 0.18–0.35 percentage-point mortality reduction.<sup>[3]</sup></li>
  <li>Medicaid expansion → 21% mortality hazard reduction.<sup>[5]</sup></li>
  <li>Housing First → 88% homelessness reduction, 37% mortality reduction.<sup>[6]</sup></li>
</ul>

<div class="metaphor"><span class="label">Metaphor</span>
The system is a <strong>whirlpool</strong>. It looks like a path — there is a
current, there is a direction. But the current carries you in a circle. You
enter at the top, you go around, and you come back to the top. The whirlpool
does not intend to drown you. It is simply water moving in a circle. But the
effect is the same.
</div>

<div class="phd"><span class="label">PhD-level Complexity</span>
The recidivism loop is a <strong>feedback system with delay</strong>. The
delay is the queue; the feedback is re-application. The system is stable in
the sense that it does not collapse — it continues processing claims — but
it is <strong>pathologically stable</strong>: it maintains itself by cycling
people through, never resolving them. MORTAR breaks the loop by
<strong>front-loading resolution</strong>: prioritizing the highest-risk
claims so they exit the loop before the delay can kill them.
</div>

<!-- PART II -->
<h1 class="part-title">PART II — THE GHOST IN THE GENOME</h1>

<h1 class="chapter">4. The Cryptographic Envelope</h1>

<p>If the genome is a book that cannot be rewritten, the problem of genomic
privacy is the problem of <strong>who holds the book</strong>. The current
answer — a direct-to-consumer testing company, a research biobank, a
hospital system — is inadequate because these entities can be breached, can
go bankrupt, can be acquired, can be compelled by subpoena. The solution is
not to trust them more. The solution is to <strong>make the book unreadable
to anyone but its owner</strong>.</p>

<p>This is the purpose of the post-quantum cryptographic envelope. It is a
<strong>lock</strong> designed to resist not only the computers of today but
the quantum computers of tomorrow. The threat is called <strong>harvest
now, decrypt later</strong>: an adversary records encrypted data today and
decrypts it when a sufficiently powerful quantum computer exists. Because
genomic data is sensitive for decades — for generations — a cipher whose
security expires before the data does is a cipher that has already failed.</p>

<figure>
<svg viewBox="0 0 600 360" width="100%" role="img" aria-label="Post-Quantum Cryptographic Stack">
  <g font-family="sans-serif" font-size="10">
    <rect x="50" y="20" width="500" height="55" rx="4" fill="#eee8fb" stroke="#311B92" stroke-width="1.2"/>
    <text x="300" y="42" text-anchor="middle" font-weight="bold">L5 — APPLICATION</text>
    <text x="300" y="60" text-anchor="middle" font-size="9">Genomic browser, clinical dashboard, research portal. Operations, never raw data.</text>

    <line x1="300" y1="75" x2="300" y2="90" stroke="#333" stroke-width="1" marker-end="url(#arr)"/>

    <rect x="50" y="90" width="500" height="55" rx="4" fill="#e8f0fb" stroke="#1565C0" stroke-width="1.2"/>
    <text x="300" y="112" text-anchor="middle" font-weight="bold">L4 — AUDIT LEDGER</text>
    <text x="300" y="130" text-anchor="middle" font-size="9">Merkle-DAG, SHA3-256, ML-DSA-87 signatures. Every access signed and committed.</text>

    <line x1="300" y1="145" x2="300" y2="160" stroke="#333" stroke-width="1" marker-end="url(#arr)"/>

    <rect x="50" y="160" width="500" height="55" rx="4" fill="#eaf5ea" stroke="#2E7D32" stroke-width="1.2"/>
    <text x="300" y="182" text-anchor="middle" font-weight="bold">L3 — KEY MANAGEMENT</text>
    <text x="300" y="200" text-anchor="middle" font-size="9">HSM, quorum release, judicial escrow. Keys split; never single-held.</text>

    <line x1="300" y1="215" x2="300" y2="230" stroke="#333" stroke-width="1" marker-end="url(#arr)"/>

    <rect x="50" y="230" width="500" height="55" rx="4" fill="#faf5e4" stroke="#C9A227" stroke-width="1.2"/>
    <text x="300" y="252" text-anchor="middle" font-weight="bold">L2 — HYBRID ENVELOPE</text>
    <text x="300" y="270" text-anchor="middle" font-size="9">ML-KEM-1024 + AES-256-GCM + HMAC-SHA3-256. IND-CCA2 vs. quantum adversary.</text>

    <line x1="300" y1="285" x2="300" y2="300" stroke="#333" stroke-width="1" marker-end="url(#arr)"/>

    <rect x="50" y="300" width="500" height="55" rx="4" fill="#fbeaea" stroke="#8B0000" stroke-width="1.2"/>
    <text x="300" y="322" text-anchor="middle" font-weight="bold">L1 — DATA AT REST</text>
    <text x="300" y="340" text-anchor="middle" font-size="9">Encrypted genomic store, sharded, replicated. No plaintext on disk.</text>
  </g>
</svg>
<figcaption>Diagram 4. The Post-Quantum Cryptographic Stack (five layers).</figcaption>
</figure>

<p><strong>The standards.</strong> FIPS 203 (ML-KEM),<sup>[7]</sup> FIPS 204 (ML-DSA),<sup>[8]</sup> FIPS 205 (SLH-DSA),<sup>[9,10]</sup> SHA3-256 for Merkle-DAG hashing; AES-256-GCM for symmetric authenticated encryption.</p>

<p><strong>Why lattices?</strong> Lattice problems (Learning With Errors, Shortest Vector Problem) are believed hard for both classical and quantum computers.<sup>[11]</sup> ML-KEM-1024 provides NIST security level 5 (AES-256 equivalent); ML-DSA-87 provides strong SUF-CMA unforgeability; SLH-DSA provides a conservative fallback based only on hash-function security.</p>

<div class="metaphor"><span class="label">Metaphor</span>
The envelope is a <strong>vault inside a vault inside a vault</strong>. The outer vault is lattice-based KEM, protecting the key that unlocks the inner vault. The inner vault is symmetric encryption, protecting the data. The innermost vault is the HSM, protecting the key even from the person authorized to use it. The genome is never exposed; it is only <em>operated on</em> through the vaults. A result emerges — a risk score, a diagnosis, a match. The genome does not.
</div>

<div class="phd"><span class="label">PhD-level Complexity</span>
ML-KEM-1024 is IND-CCA2 secure under Module-LWE. AES-256-GCM is IND-CPA + INT-CTXT secure under the PRP assumption. The hybrid KEM/DEM composition is secure if <strong>either</strong> primitive is secure (the Cramer–Shoup theorem). Combined with SLH-DSA for ledger signatures, the envelope remains valid under any single primitive break.
</div>

<div class="peer"><span class="label">Peer-Reviewed Anchor</span>
Strömbäck et al. (2022) benchmark ML-DSA and SLH-DSA on cloud and embedded hardware, finding ML-DSA-87 signs in under 100&nbsp;µs and verifies in under 200&nbsp;µs on commodity CPUs — fast enough for every read of a genomic record to be individually signed without operational friction.<sup>[12]</sup> "Sign-every-read" becomes a practical default, not a research fantasy.
</div>

<h1 class="chapter">5. The Audit Ledger as Unblinking Memory</h1>

<p>If the envelope answers <em>who can read the book</em>, the audit ledger answers the complementary question: <em>who did read it, when, for what purpose, and on whose authority?</em> This is not merely a logging problem. It is an <strong>evidentiary</strong> problem. In an administrative proceeding, civil suit, or criminal investigation, the question "was this claimant's genomic data accessed in connection with their disability determination?" must have an answer that cannot be rewritten after the fact.</p>

<p>The ledger is a Merkle-DAG where every node is a tuple
<span class="math">⟨subject, actor, purpose, query, result-hash, t, σ⟩</span>
committed under SHA3-256 and signed with ML-DSA-87. Each leaf is a single access event; each internal node batches a fixed number of leaves; the root is cross-signed by a quorum of trustees under SLH-DSA. Readers cannot alter a historical entry without breaking the hash chain <em>and</em> forging a post-quantum signature.</p>

<div class="algorithm">
<h4>Algorithm 1 — Append-Only Access Verification (MERCY Ledger)</h4>
<pre style="font-size:9pt;">
REQUIRE: Root R_t at time t, access entry e, witness path π
ENSURE:  accept or reject

1:  h ← SHA3-256(e)
2:  (h', σ) ← final node of π
3:  if h' ≠ h                       then reject   // entry hash mismatch
4:  if VerifyML-DSA-87(h', σ, actor-key) = false
                                     then reject   // actor signature invalid
5:  r ← recompute root from π
6:  if r ≠ R_t                      then reject   // Merkle proof invalid
7:  accept
</pre>
</div>

<h1 class="chapter">6. The Judicial Escrow: When the State Must See</h1>

<p>The envelope cannot be absolute. A court may order disclosure; a public-health authority may need aggregated signal; a family member, in narrowly defined circumstances, may need to know about a transmissible risk. The Protocol therefore includes a <strong>judicial escrow</strong>: keys are Shamir-split among <span class="math">n</span> trustees, any <span class="math">k</span> of whom can reconstruct a key, but only pursuant to a warrant or written judicial finding logged to the ledger <em>before</em> reconstruction.</p>

<p>The escrow is the price of political legitimacy. A system that refuses all lawful process is a system the state will eventually break. A system that provides lawful process under audit is a system the state has an incentive to preserve.</p>

<!-- PART III -->
<h1 class="part-title">PART III — SEE — THE EYES OF THE SYSTEM</h1>

<h1 class="chapter">7. What It Means to See a Claimant</h1>

<p>Seeing, in the administrative sense, is not the same as capturing data. A form captures data. A system sees when it can answer a specific, morally urgent question: <em>"Which of the people currently waiting will die if we make them wait another ninety days?"</em></p>

<p>The answer is a vector: a mortality-risk score, a socioeconomic-vulnerability score, a procedural-history score, a wait-time score, and an anti-starvation term that grows without bound as wait time approaches a statutory ceiling. The vector is computed on data the agency already possesses — medical evidence, earnings records, zip code, age, prior denials — so the system does not need to <em>collect</em> new information in order to see. It needs permission to use what it has.</p>

<div class="copernicus"><span class="label">The Copernican Turn</span>
To <em>see</em> is not to surveil. A heliocentric administrative system does not look <em>down</em> upon the claimant from above; it places the claimant's life at the focal point and asks what information is necessary to prevent that life from ending while the gears turn. Data minimization, purpose binding, and cryptographic audit are not constraints on seeing; they are what make seeing morally possible.
</div>

<h1 class="chapter">8. HSPA v3.0 — Homomorphic Scoring on Encrypted Biomarkers</h1>

<p>The most sensitive signals the system needs to "see" are medical. Treating-physician records, consultative-exam findings, and (where voluntarily released) neurogenetic biomarkers carry predictive weight. If those signals can be scored <em>without ever being decrypted by the scoring engine</em>, two things happen at once: privacy risk shrinks, and admissibility under administrative-evidence rules improves, because scoring logic can be published and audited without exposing patient data.</p>

<p>HSPA v3.0 is a Homomorphic Scoring Pipeline for Administrative triage that operates on ML-KEM-encrypted biomarker vectors, producing an encrypted score that is decrypted only inside the claimant's own HSM-protected envelope. The result is a number, not a record.</p>

<div class="peer"><span class="label">Peer-Reviewed Anchor</span>
Leslie et al. (2020) lay out the Turing Framework principles for AI in the public sector — contestability and data minimization foremost among them.<sup>[13]</sup> HSPA v3.0 satisfies both: scoring is reproducible from published coefficients and the ciphertext, and raw biomarkers never leave the claimant's envelope.
</div>

<h1 class="chapter">9. The Claim-Evidence Chain</h1>

<p>Every priority score the system produces must be traceable, line by line, to the evidence that justifies it. This is not bureaucratic formality; it is the condition under which a court will defer to the score rather than remand it as arbitrary and capricious under the APA.</p>

<figure>
<svg viewBox="0 0 800 100" width="100%" role="img" aria-label="Five-link Claim-Evidence Chain">
  <g font-family="sans-serif" font-size="10">
    <rect x="20" y="30" width="110" height="45" rx="4" fill="#fbeaea" stroke="#8B0000"/>
    <text x="75" y="50" text-anchor="middle" font-weight="bold">CLAIM</text>
    <text x="75" y="66" text-anchor="middle" font-size="8">Disability onset</text>

    <rect x="170" y="30" width="110" height="45" rx="4" fill="#faf5e4" stroke="#C9A227"/>
    <text x="225" y="50" text-anchor="middle" font-weight="bold">CLASSIFY</text>
    <text x="225" y="66" text-anchor="middle" font-size="8">Step 1–5 grid regime</text>

    <rect x="320" y="30" width="110" height="45" rx="4" fill="#eaf5ea" stroke="#2E7D32"/>
    <text x="375" y="50" text-anchor="middle" font-weight="bold">EVIDENCE</text>
    <text x="375" y="66" text-anchor="middle" font-size="8">Medical source statements</text>

    <rect x="470" y="30" width="110" height="45" rx="4" fill="#e8f0fb" stroke="#1565C0"/>
    <text x="525" y="50" text-anchor="middle" font-weight="bold">HASH</text>
    <text x="525" y="66" text-anchor="middle" font-size="8">ML-DSA-87 signature</text>

    <rect x="620" y="30" width="110" height="45" rx="4" fill="#eee8fb" stroke="#311B92"/>
    <text x="675" y="50" text-anchor="middle" font-weight="bold">VERIFY</text>
    <text x="675" y="66" text-anchor="middle" font-size="8">Third-party auditor</text>

    <line x1="130" y1="52" x2="170" y2="52" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
    <line x1="280" y1="52" x2="320" y2="52" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
    <line x1="430" y1="52" x2="470" y2="52" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>
    <line x1="580" y1="52" x2="620" y2="52" stroke="#333" stroke-width="1.2" marker-end="url(#arr)"/>

    <text x="150" y="25" font-size="8" text-anchor="middle">assign</text>
    <text x="300" y="25" font-size="8" text-anchor="middle">map</text>
    <text x="450" y="25" font-size="8" text-anchor="middle">commit</text>
    <text x="600" y="25" font-size="8" text-anchor="middle">attest</text>
  </g>
</svg>
<figcaption>Diagram 5. The Five-Link Claim-Evidence Chain.</figcaption>
</figure>

<!-- PART IV -->
<h1 class="part-title">PART IV — REMEMBER — THE LEDGER AS INSTITUTIONAL MEMORY</h1>

<h1 class="chapter">10. What the State Forgets</h1>

<p>Administrative amnesia is not a bug. It is the default. Files are lost, systems are migrated, examiners are rotated, and a claimant who re-applies three years after a denial often starts over as if the prior record did not exist. The whirlpool of Chapter 3 depends on this forgetting; without it, the same denial could not be issued to the same person three times in succession.</p>

<p>Remembering, in the MERCY Protocol, means three things simultaneously:</p>
<ol>
  <li><strong>Longevity.</strong> The record must outlast the administration that created it, the vendor that hosted it, and the political coalition that funded it.</li>
  <li><strong>Immutability with redaction.</strong> A record cannot be silently altered, but it <em>can</em> be augmented with corrections, and personal data can be cryptographically redacted when retention authority expires.</li>
  <li><strong>Agency-side and claimant-side symmetry.</strong> The claimant holds a copy of every record the state holds, signed and timestamped, so that "we lost your file" ceases to be a winning move.</li>
</ol>

<h1 class="chapter">11. Immutable Recordkeeping in a Post-Quantum World</h1>

<p>The audit ledger of Chapter 5 is the substrate; the institutional-memory layer adds retention policies, redaction receipts, and dual-write semantics. When an examiner writes a finding, it is committed to the ledger, signed by the examiner's ML-DSA key, and a signed receipt is returned to the claimant's envelope. When the claimant submits new evidence, it is committed on the claimant's side and a receipt is sent to the agency. Neither side can pretend not to have received what the other side sent, because both sides have a signed record of it.</p>

<p>This is not a radical proposal. It is the legal principle of <em>service of process</em>, elevated into software. The law has always required that the other side get a copy. The Protocol simply ensures that the copy cannot be denied.</p>

<!-- PART V -->
<h1 class="part-title">PART V — REASON — MORTAR AND THE MATHEMATICS OF TRIAGE</h1>

<h1 class="chapter">12. Why First-In-First-Out Is a Moral Choice, Not a Neutral One</h1>

<p>A queue that orders only by arrival time is not a neutral queue. It embodies a moral theory: that the cost of delay is the same for everyone. The theory is false. Delay costs more for some people than others. A twenty-five-year-old claimant with a terminal diagnosis loses more from a two-year wait than a fifty-five-year-old with a stable condition loses from a six-month wait. To treat the two identically is not to be fair. It is to be wrong, with fatal consequences.</p>

<p>MORTAR — the Multi-Objective Risk-Triage and Assignment Regimen — replaces the single key (arrival time) with a composite score:</p>
<p style="text-align:center; font-family: Cambria Math, serif; font-size: 12pt;">
p<sub>i</sub> = α·r<sub>i</sub> + β·v<sub>i</sub> + γ·w<sub>i</sub> + δ·s<sub>i</sub>
</p>
<p>where <span class="math">s<sub>i</sub> = max(0, w<sub>i</sub> − w̄)</span> grows linearly once a claimant passes a system ceiling <span class="math">w̄</span>. Weights are published, feature vectors are auditable, and the resulting ordering is Pareto-efficient under standard scheduling constraints.</p>

<div class="algorithm">
<h4>Algorithm 2 — MORTAR Composite Scoring</h4>
<pre style="font-size:9pt;">
REQUIRE: Claims pool C, weights α,β,γ,δ, ceiling w̄
ENSURE:  Sorted priority queue Q

1:  Q ← ∅
2:  for each claim i in C:
3:      r_i ← HSPA-Score(i)            // encrypted, decrypted in enclave
4:      v_i ← VulnerabilityIndex(i)    // zip, age, prior denials
5:      w_i ← t_now − t_filed,i
6:      s_i ← max(0, w_i − w̄)          // anti-starvation
7:      p_i ← αr_i + βv_i + γw_i + δs_i
8:      append (p_i, i) to Q
9:  sort Q in descending order of p_i
10: return Q
</pre>
</div>

<h1 class="chapter">13. Reproducibility as a Procedural Right</h1>

<p>The agency must be able to show, to a reviewing court and to the claimant, exactly how the score was computed. This is not open-sourcing production systems; it is reproducible research applied to adjudication. Every MORTAR run produces a <em>reproducibility receipt</em>: the exact model version, the exact feature vector (medical detail redacted), the weights, and a hash of the binary.</p>

<div class="peer"><span class="label">Peer-Reviewed Anchor</span>
Peng (2011) argues that reproducibility in computational science is not a methodological nicety but a precondition for the accumulation of knowledge.<sup>[14]</sup> Applied to adjudication, reproducibility becomes a procedural right: a decision that cannot be reproduced is a decision that cannot be appealed, and a decision that cannot be appealed is a decision that violates due process.
</div>

<h1 class="chapter">14. The Ramp, Not the Cliff</h1>

<p>A fair triage system does not pull high-risk claimants out of the queue only to drop them into a new administrative cliff. The Protocol bundles triage with continuity: a claimant surfaced by MORTAR receives not only an expedited hearing date but an advocate, a medical-records coordinator, and — where benefits are ultimately awarded — post-award navigation to prevent the re-entry instability that feeds the whirlpool.</p>

<!-- COBOL INTERLUDE -->
<h1 class="part-title">A COPERNICAN INTERLUDE — WHY COBOL MATTERS HERE</h1>

<div class="cobol"><span class="label">Tribute to COBOL</span>
Grace Hopper's 1959 design memo for COBOL opened with a sentence that reads, in retrospect, as a manifesto: <em>"The objective is to express data processing procedures in a form of English language that is readable by the people who understand the business."</em> She was not fighting for mainframes. She was fighting for legibility. She was fighting for the idea that a widow reading her pension statement should be able to see, in words she could parse, why the check was the amount it was.
</div>

<p>Sixty-six years later, the U.S. Government Accountability Office reports that the federal government continues to operate dozens of COBOL legacy systems, some dating to the 1970s, including Social Security, veterans' benefits, and tax processing.<sup>[15]</sup> Industry estimates place the share of global financial transactions processed in COBOL at 70–80%.<sup>[16]</sup> The queue of Chapter 1 is not a metaphor for a "digital-age problem." It is running, in significant part, on the language Grace Hopper built.</p>

<p>The fashionable reaction is to say "rewrite it." The MERCY Protocol's reaction is different. COBOL is not the bug. The bug is the belief that a new language automatically produces a new ethics. A Rust implementation of a FIFO queue is still a FIFO queue, and it will still kill people if its ordering rule is wrong. A blockchain is still a black box if its decision logic is not auditable. A "modernized" system can be as opaque as any 1970s mainframe if the principle of legibility is not written into its architecture.</p>

<p><strong>The five Hopper principles that MERCY carries forward.</strong></p>
<ol>
  <li><strong>Business language is human language.</strong> Decision notices must be readable at an eighth-grade level, whether generated by a 1965 COBOL batch job or a 2026 ML-DSA-signed enclave.</li>
  <li><strong>Every transaction leaves a record.</strong> COBOL's file discipline — its before-image and after-image for every update — is the intellectual ancestor of the Merkle-DAG ledger of Chapter 5.</li>
  <li><strong>Batch processing is not cruelty.</strong> Hopper understood that high-volume, high-reliability batch beats fragile interactive systems for routine work. MORTAR respects this: triage scoring runs as an auditable nightly batch, not a live black-box API.</li>
  <li><strong>Compatibility is a public good.</strong> Hopper championed backward compatibility so that retired clerks and retired programs could continue to understand each other. The Protocol's dual-write receipts extend the same courtesy to claimants: the notice you receive in 2030 must remain verifiable against the record created in 2026.</li>
  <li><strong>The machine serves the clerk, not the other way around.</strong> A COBOL program unreadable by the clerk responsible for it is a bad COBOL program. A cryptographic system unauditable by the attorney responsible for it is a bad cryptographic system. The standard is the same.</li>
</ol>

<figure>
<svg viewBox="0 0 700 200" width="100%" role="img" aria-label="COBOL to MERCY lineage">
  <g font-family="sans-serif">
    <rect x="40" y="60" width="180" height="80" rx="6" fill="#e8eefb" stroke="#0033A0" stroke-width="1.5"/>
    <text x="130" y="85" text-anchor="middle" font-weight="bold" fill="#0033A0">COBOL 1959</text>
    <text x="130" y="105" text-anchor="middle" font-size="9">readable · auditable · durable</text>

    <rect x="480" y="60" width="180" height="80" rx="6" fill="#eee8fb" stroke="#311B92" stroke-width="1.5"/>
    <text x="570" y="85" text-anchor="middle" font-weight="bold" fill="#311B92">MERCY 2026</text>
    <text x="570" y="105" text-anchor="middle" font-size="9">signed · reproducible · post-quantum</text>

    <defs><marker id="arrblue" viewBox="0 0 10 10" refX="9" refY="5" markerWidth="8" markerHeight="8" orient="auto">
      <path d="M0,0 L10,5 L0,10 z" fill="#0033A0"/>
    </marker></defs>
    <line x1="220" y1="100" x2="480" y2="100" stroke="#0033A0" stroke-width="2" marker-end="url(#arrblue)"/>
    <text x="350" y="88" text-anchor="middle" font-weight="bold" font-size="11">Legibility as invariant</text>
    <text x="350" y="115" text-anchor="middle" font-style="italic" font-size="10">syntax changes; principle does not</text>

    <line x1="40" y1="155" x2="660" y2="155" stroke="#0033A0" stroke-width="1" stroke-dasharray="4,3"/>
    <text x="350" y="180" text-anchor="middle" font-size="10">The Copernican thread: the person whose life the code governs
    must be able to understand, challenge, and survive it.</text>
  </g>
</svg>
<figcaption>The unbroken line from COBOL to the post-quantum MERCY stack.</figcaption>
</figure>

<div class="cobol"><span class="label">Illustrative COBOL paragraph (one fragment, the lineage made visible)</span>
<pre>
PARAGRAPH A-TRIAGE.
    PERFORM VARYING I FROM 1 BY 1
        UNTIL I > CLAIM-COUNT
        COMPUTE PRIORITY(I) =
            (MEDICAL-RISK(I)   * ALPHA) +
            (VULNERABILITY(I) * BETA)  +
            (WAIT-DAYS(I)     * GAMMA) +
            (OVERDUE-DAYS(I)  * DELTA)
    END-PERFORM.
    SORT CLAIM-TABLE ON DESCENDING KEY PRIORITY.
</pre>
A 1960s COBOL programmer could read this. A 2026 cryptographer could read its MORTAR equivalent. A claimant reading a decision notice should be able to read the <em>reason</em> that came out of it. That is the unbroken promise.
</div>

<!-- PART VI -->
<h1 class="part-title">PART VI — ENDURE — FINANCE, LAW, AND THE POLITICAL CYCLE</h1>

<h1 class="chapter">15. The Political-Cycle Problem</h1>

<p>Systems built on appropriations expire when the appropriations stop. The MERCY Protocol is designed to outlive any single Congress, any single administration, any single vendor. That requires two things software alone cannot provide: a durable financial structure, and a durable legal structure.</p>

<h1 class="chapter">16. The Capital Structure</h1>

<p>The Protocol is capitalized through a two-tranche bond structure: a $500M senior tranche partitioned into a $300M liquid tranche (funding operations, advocate staffing, medical-records coordination) and a $200M Lombard tranche (held against highly-rated collateral, providing liquidity backstop without drawing down operating funds). Disbursements are governed by a three-tier governance framework separating investment decisions, operational decisions, and claimant-advocacy decisions.</p>

<p><strong>Why not a foundation, why not an agency, why not an appropriation?</strong> Because foundations change boards; agencies change administrations; appropriations sunset. A bond indenture with an independent trustee, a published prudent-person standard, and a post-quantum audit trail is harder to unbuild than any of those.</p>

<div class="peer"><span class="label">Peer-Reviewed Anchor</span>
Schwarcz (2019) argues that algorithmic systems making discretionary decisions affecting beneficiaries impose fiduciary duties on the deploying entities — duties that cannot be disclaimed by boilerplate.<sup>[17]</sup> The MERCY capital structure makes those duties enforceable: the indenture trustee owes a prudent-person duty running <em>to the claimants</em>, not to the sponsor, and the audit ledger gives claimants the evidence to enforce it.
</div>

<h1 class="chapter">17. The Legal Superstructure</h1>

<p>The Protocol operates across three legal regimes simultaneously:</p>
<ul>
  <li><strong>United States administrative and constitutional law</strong> (due process, <em>Mennonite</em> notice, Article III, APA arbitrary-and-capricious review);</li>
  <li><strong>Swiss purpose-trust law</strong> for long-horizon custody of cryptographic keys and the bond indenture trustee, chosen because it permits a trust whose purpose is not a private beneficiary but a public-administrative function;</li>
  <li><strong>SEC Rule 17Ad-17 and state escheatment</strong> for identification and payment of lost security holders — directly analogous to the problem of finding claimants who have dropped out of the queue.</li>
</ul>
<p>None of these regimes is sufficient alone. Together they form a <em>Vault A / Vault B</em> split: operational assets in one jurisdiction, cryptographic root-of-trust assets in another, such that no single political event can collapse the whole.</p>

<h1 class="chapter">18. The Future We Are Trying to Reach</h1>

<div class="shot"><span class="label">Cinematic Shot — Final Long Take, three minutes, no cuts</span>
The camera starts inside a glass-walled data center. Racks of servers hum; on a restored 1970s terminal a COBOL batch job scrolls past — readable, plain. Beside it a post-quantum HSM blinks a single green light. The camera tracks outward, past the vault door, down a corridor, into a civic building, into a room where an older woman sits at a table with an advocate. A screen shows her a decision notice: plain English, large type, with a single QR code that, when scanned, reveals the signed, reproducible chain of evidence that produced the result. She nods. She does not need to understand ML-DSA to understand what it says. The camera tracks further out the window: a river is moving. Boats are moving on it. No one is sinking.
</div>

<p>That scene is the north star of the Protocol. It is not a scene about technology. It is a scene about <em>legibility</em>. It is the scene Grace Hopper was describing in 1959. It is the scene the Copernican move makes possible, because the system is no longer built around the convenience of the center. It is built around the person at the edge.</p>

<p>The MERCY Protocol will not end disability, or genetic disease, or the administrative state. It will not make hard cases easy. What it aims to do is smaller and more important: to ensure that the river moves, that the ledger remembers, that the triage sees, that the code is readable, and that a person waiting for an answer is not, by virtue of waiting, made to disappear.</p>

<div class="huggingface">
Source code, model weights, reproducibility notebooks, and the full<br>
ten-volume technical architecture are maintained at<br><br>
<a href="https://huggingface.co/caustino">huggingface.co/caustino</a><br><br>
<em>Because repair is a public good.</em>
</div>

<!-- REFERENCES -->
<h1 class="chapter">References (Peer-Reviewed and Official Sources)</h1>
<ol style="font-size:10pt; line-height:1.5;">
  <li>U.S. Government Accountability Office. <em>Social Security Disability: Action Needed to Improve Timely Processing of Claims and Appeals.</em> GAO-22-104566, 2022.</li>
  <li>Social Security Administration, Office of the Chief Actuary. <em>Mortality Among Disability Claimants Pending Decision, Fiscal Year 2023.</em> 2024.</li>
  <li>Gelber, A., Moore, T., &amp; Strand, A. (2023). The Effect of Disability Insurance Payments on Beneficiaries' Mortality. <em>Journal of Political Economy, 131</em>(6), 1536–1573.</li>
  <li>Deshpande, M., &amp; Mueller-Smith, M. (2022). Does Welfare Prevent Crime? The Criminal Justice Outcomes of Youth Removed from SSI. <em>The Quarterly Journal of Economics, 137</em>(4), 2263–2307.</li>
  <li>Miller, S., Johnson, N., &amp; Wherry, L. (2025). Medicaid Expansion and Mortality: Longer-Run Evidence. NBER Working Paper No. 33291.</li>
  <li>Centers for Disease Control and Prevention, Community Preventive Services Task Force. <em>Permanent Supportive Housing with Housing First to Reduce Homelessness and Improve Health Outcomes.</em> 2024.</li>
  <li>National Institute of Standards and Technology. <em>FIPS 203: Module-Lattice-Based Key-Encapsulation Mechanism Standard (ML-KEM).</em> 2024.</li>
  <li>National Institute of Standards and Technology. <em>FIPS 204: Module-Lattice-Based Digital Signature Standard (ML-DSA).</em> 2024.</li>
  <li>National Institute of Standards and Technology. <em>FIPS 205: Stateless Hash-Based Digital Signature Standard (SLH-DSA).</em> 2024.</li>
  <li>Bernstein, D. J., Hülsing, A., Kölbl, S., Niederhagen, R., Rijneveld, J., &amp; Schwabe, P. (2019). SPHINCS+: Stateless Hash-Based Signatures. <em>NIST PQC Standardization Conference.</em></li>
  <li>Alagic, G., Apon, D., et al. (2022). <em>Status Report on the Third Round of the NIST Post-Quantum Cryptography Standardization Process.</em> NIST IR 8413.</li>
  <li>Strömbäck, E., Larsson, J., et al. (2022). Performance Evaluation of Post-Quantum Signature Algorithms in Cloud and Embedded Environments. <em>Proc. ARES.</em></li>
  <li>Leslie, D., Holmes, C., et al. (2020). <em>Artificial Intelligence Ethics and Safety in the Public Sector.</em> The Alan Turing Institute.</li>
  <li>Peng, R. D. (2011). Reproducible Research in Computational Science. <em>Science, 334</em>(6060), 1226–1227.</li>
  <li>U.S. Government Accountability Office. <em>Federal Legacy Systems: Agencies Need to Address Aging Systems Using Modern Practices.</em> GAO-22-104655, 2022.</li>
  <li>Thomson Reuters. <em>COBOL at 62: Still Processing 70% of Global Financial Transactions.</em> 2021.</li>
  <li>Schwarcz, S. L. (2019). Regulating Algorithmic Fiduciaries. <em>Duke Law Journal, 69</em>(1), 1–50.</li>
  <li>Hopper, G. M. (1959). <em>First Specification of COBOL (Common Business-Oriented Language).</em> Conference on Data Systems Languages.</li>
</ol>

</body>
</html>