Skip to content

Errata in Cox & D'Antonio (2017)

These are the 3 entries of the errata registry on Cox & D’Antonio (2017), in the order the registry lists them. The registry’s introduction says how each claim is checked against the printed page and what the status of an entry means.

Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Appendix D (two rows the table cannot tell apart)

Section titled “Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Appendix D (two rows the table cannot tell apart)”
  • Location: Appendix D, “Random incidence scattering coefficient table”, the group “Pyramids [6]”, on printed pages 499 and 500 (PDF pages 556 and 557). Non-normative source: a textbook.
  • The print: the group holds four rows. Printed page 499 carries ” cm, ” and then the same description again with the continuation line “One in four pyramid corners raised from baseplate” under it. Printed page 500 carries ” cm, ” and then, once more, ” cm, ” with the same continuation line under it.
  • The problem: the second and the fourth rows are printed with the same description, the same continuation line and different numbers: 0.38 against 0.44 at 1 kHz, 0.74 against 0.76 at 2 kHz, and 1.00 against an en dash at 5 kHz. Nothing printed beside either row distinguishes it from the other, so a reader who looks up “the 30.5 cm pyramids with one corner in four raised” finds two answers and no way to choose. A table of measured values has to identify its rows; this one does not.
  • Evidence: both pages read at six times magnification. The labels are character for character the same, the continuation lines are the same, and the values differ in thirteen of the eighteen bands, agreeing only at 100, 125, 160, 200 and 4000 Hz. The block’s own structure is what makes the reading plain: the table pairs a plain surface with a modified one, and printed page 499 pairs with its modified version. Printed page 500 opens with , so its second row is where the modified belongs, which would make the printed “2h” the defect. That is a reading of the pattern rather than something the page states, and the source the group is credited to, Sharma and Bradley, J. Acoust. Soc. Am. 134(5), 4095 (2013), is a one-page meeting abstract that this library has not read, so the errata records what the page does and not what it should have said. Verified on PDF page 556 (printed p. 499) and PDF page 557 (printed p. 500) of Cox and D’Antonio (2017), Acoustic Absorbers and Diffusers, third edition.
  • Library behaviour: both rows are kept, with the description and the continuation line as printed. Their keys carry the printed folio, which is the only thing that separates them, and test_two_pyramid_rows_are_told_apart_only_by_the_page_they_sit_on in tests/materials/diffusers/test_scattering_catalogue.py pins that they stay two rows with two spectra.
  • Status: not reported.

Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Appendix B (a width in centimetres that its own geometry makes metres)

Section titled “Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Appendix B (a width in centimetres that its own geometry makes metres)”
  • Location: Appendix B, “Normalized diffusion coefficient table”, section 1, the first surface of the width series, on printed page 482 (PDF page 539). Non-normative source: a textbook.
  • The print: the section is headed “Effect of changing diffuser periodicity and width. Semicylinder(s) non-absorbing surfaces, radius 0.3 m (1 cm flat section between each period)”, and its five surfaces are listed as “1 period, 0.61 cm wide”, “2 periods, 1.22 m wide”, “4 cylinders, 2.44 m wide”, “6 periods, 3.66 m wide” and “12 periods, 7.32 m wide”.
  • The problem: the first width is in centimetres and the other four are in metres, and the series doubles: 0.61, 1.22, 2.44, and then 3.66 and 7.32, which are six and twelve of the first. One period of the surface the heading describes is a semicylinder of radius m plus the cm flat section, so m. A width of 0.61 cm is six millimetres, a hundredth of what the heading’s own geometry gives and a hundredth of what the rest of the series requires.
  • Evidence: the same surface is listed in Table C.3 of the next appendix, under a heading with the same geometry, and there it reads “1 period, 0.61 m wide”. The book therefore prints both spellings of one surface twelve pages apart, and the metric one is the one its arithmetic supports. Verified on PDF page 539 (printed p. 482) and PDF page 549 (printed p. 492) of Cox and D’Antonio (2017), Acoustic Absorbers and Diffusers, third edition, with the unit re-read at six times magnification on both pages: the appendix really prints “cm” and Table C.3 really prints “m”.
  • Consequence for the book’s own tables: the description of one surface, and of the three rows that carry it. The numbers beside it are unaffected, which is what makes the defect easy to miss and worth recording: a reader comparing a measured semicylinder against this row would be comparing against a 0.61 m device whatever the label says.
  • Library behaviour: the row keeps the width as printed, because a catalogue that silently corrected it would be claiming a reading the page does not make. The three rows of the surface are keyed on it and test_every_section_heading_is_kept_whole in tests/materials/diffusers/test_diffusion_catalogue.py pins that the heading with the geometry travels with them, which is what lets a reader see the contradiction.
  • Status: not reported.

Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Table 6.7 (six porosities printed in per cent in a column of fractions)

Section titled “Cox & D’Antonio, Acoustic Absorbers and Diffusers 3e (2017), Table 6.7 (six porosities printed in per cent in a column of fractions)”
  • Location: Table 6.7, “Effective flow resistivity values for ground surfaces and other parameters”, the column “Porosity”, on printed pages 200 and 201 (PDF pages 257 and 258). Non-normative source: a textbook.
  • The print: the column is headed “Porosity” and states no unit, where the column beside it is headed “Water content (%)”. Every porosity it prints is a fraction between 0.15 and 1 (“0.5–0.9” for snow, “0.24” for the sports field, “0.44” for fine sand, “0.34–1” and the like for the fitted grasses), except on six rows: “Mineral layer beneath mixed deciduous forest” 36.5, “Humus on pine forest floor” 58.1, “Pine forest litter (6–7 cm thick)” 38.9, “Grass root layer in loamy sand” 48 ± 4, “Loamy sand” 37.5 and “Bare sandy plain” 26.9.
  • The problem: a porosity is the fraction of a volume that is open, so it cannot pass 1, and these six are percentages printed in a column that states no unit and holds fractions everywhere else. Read in the column’s own unit they are porosities of 26.9 to 58.1, which no material has; read as per cent they are 0.269 to 0.581, which is what soils like these are. The page leaves the reader to work out which, and a program that reads the column as printed takes the impossible value.
  • Evidence: Verified on PDF page 257 (printed p. 200) and PDF page 258 (printed p. 201) of Cox and D’Antonio (2017), Acoustic Absorbers and Diffusers, third edition: the heading prints no unit, and the six cells print the values quoted, the fourth as “48 ± 4”.
  • Library behaviour: in PUBLISHED_GROUND_SURFACES the six rows leave GroundSurface.porosity, which is a fraction from 0 to 1, empty, and hold the cell as misprinted: why_missing("porosity") and the refusal of printed("porosity") quote the figure the page prints, the ± 4 included, and say why it is not served. The library does not convert it to 0.365 and so on, because the page does not print the unit that conversion would assume. Every catalogue row checks when it is built that a porosity is a fraction. test_the_six_percent_porosities_of_cox_are_held_as_misprinted and test_why_a_per_cent_porosity_is_missing_quotes_the_page in tests/io/test_catalogue_row_contract.py pin the six.
  • Status: unreported.