Meteorologist, oceanographer, Python solvem probler, experimenter, Raspberry Pi tinkerer, geek dad, manager, space enthusiast, ham radio lover, science columnist, ADHDer, synthetic intellect curious, linux admin cosplayer.

Draft Corrections - Post 10: The Swedes Got There First

Post path: /_posts/2026-04-03-The-swedes-got-there-first.md Reviewed: 2026-05-17

Severity legend

  • HIGH: factually wrong; reader will get the wrong claim.
  • MEDIUM: imprecise; current wording slightly inaccurate.
  • LOW: stylistic / hedging issue, not a factual error.

No HIGH severity errors found

The post’s central claims (December 1954 first routine NWP, three times a week, North Atlantic, 45,000 troops September 1954 with nuclear protection equipment, Bolin’s 72-hour 0.85/0.82/0.70 correlations, IBM 701 JNWPU “very disappointing,” IBM 704 1958 catching up, all key people Rossby/Bolin/Phillips/Wiin-Nielsen/Crutzen/Riesel/Beurling/Karlqvist) are well supported by independent open-access primary and secondary sources.

MEDIUM corrections

M1. BARK acronym

Current (post line 58):

The first machine was BARK – the Binär Automatisk Relä-Kalkylator (Binary Automatic Relay Calculator).

Issue: Wikipedia and most Swedish-language sources expand BARK as Binär Aritmetisk Relä-Kalkylator (Binary Arithmetic Relay Calculator). The “Automatisk” form does appear in some sources but “Aritmetisk” is the canonical form documented by the BARK project.

Suggested fix:

The first machine was BARK – the Binär Aritmetisk Relä-Kalkylator (Binary Arithmetic Relay Calculator).

M2. BARK completion vs operational date

Current (post line 58):

It was already obsolete when completed in April 1950 – similar relay machines had existed a decade earlier.

Issue: BARK was completed in February 1950 and became operational on 28 April 1950 (Wikipedia BARK; Lundin 2006 oral history). “Completed in April 1950” conflates the two.

Suggested fix (minimal):

It was already obsolete when it became operational in April 1950 – similar relay machines had existed a decade earlier.

M3. BESK spec table “Operational | 1953”

Current (post line 97):

Operational 1953 1952 1952

Issue: BESK was first powered up in late 1953 (this is when it briefly held the “fastest computer in the world” claim, which the post correctly relies on later in the same section). However, the formal inauguration was 1 April 1954. The current table entry is defensible but ambiguous. Two options:

Option A (minimal, keep current): Leave as “1953” since first-run is 1953. The “for a few weeks in late 1953, BESK held the title of fastest computer” claim in the next paragraph already implies this.

Option B (clarify): Change to “late 1953” or “1953/1954” to avoid implying a full year of stable operation in 1953.

Recommendation: leave as 1953 if Option A retained, with no further wording change.

LOW corrections

L1. “Department of Meteorology (MISU)”

Current (post line 36):

He became the first professor at the newly established Department of Meteorology (MISU) and created the International Meteorological Institute (IMI) – a research centre funded by the Swedish Parliament

Note: Yang (2025) phrases this as “He was the first head of the Department of Meteorology” at Stockholm University, which matches. The acronym MISU (Meteorologiska Institutionen vid Stockholms Universitet) is the modern department abbreviation; in the 1940s-50s it was simply “Institute of Meteorology, University of Stockholm” (as the Tellus 6 (1954) paper cites). Both forms are used in the literature today. No change needed – the post correctly bridges old/new names.

L2. Norman Phillips “autumn 1953”

Current (post line 133):

In autumn 1953, Rossby brought another crucial visitor to Stockholm: Norman Phillips

Note: Open-access sources confirm Phillips visited Stockholm in 1953 to assist Rossby with a weather model (MIT News obituary 2019; Lewis 1998 BAMS quoting “6 months in Stockholm”). The specific “autumn 1953” timing is consistent with the Persson 2005 narrative the post cites, but not independently re-verifiable from open access. No change recommended; flag as relying on Persson 2005 (footnote [^1]).

L3. Paul Crutzen “1959”

Current (post line 197):

In 1959, a young man named Paul Crutzen arrived at MISU

Note: Wikipedia specifies July 1959. Minor; no change needed.

L4. Wiin-Nielsen “trained under Rossby in Stockholm and participated in the 1954 BESK forecasts”

Current (post line 191):

Aksel Wiin-Nielsen, a Danish meteorologist who trained under Rossby in Stockholm and participated in the 1954 BESK forecasts

Note: Confirmed Danish, ECMWF first Director, “Rossby’s Institute” connection. Whether he was personally hands-on with the December 1954 forecasts specifically (versus the broader 1954-55 Stockholm group) is sourced to Persson 2005. The post’s wording “participated in the 1954 BESK forecasts” is supportable by the Persson narrative; leave as is.

L5. BARK team “wanted to call it CONIAC”

Current (post line 60):

The BARK team had wanted to call it CONIAC – Conny Integrator And Calculator, after project leader Conny Palm. Officials blocked the name. Too close to “cognac.”

Note: This anecdote is cited via Lundin 2006 (footnote 3) and not in Wikipedia BARK. It is a well-known story in Swedish computing folklore. No correction required; relies on cited primary.

Summary

Post 10 is solid. Suggested edits:

  • Change BARK expansion to “Aritmetisk” (M1).
  • Reword BARK April 1950 to “became operational in April 1950” instead of “completed in April 1950” (M2).
  • Optionally clarify BESK operational date in spec table (M3) – low priority.

No HIGH-severity factual errors found.