STRATA
Skip to study
XLVII

SOURCE RECORD / 1987

Dynamic Markov modelling

DocumentedR01 · 2026-11-17
Scope of this record

Bibliographic existence, authorship, and date only. Algorithm correctness and present-day performance are not established by this record.

Statement

Cormack and Horspool’s compression paper is listed in The Computer Journal, December 1987. [source]

Equation

Unrecorded

No equation has been transcribed or verified for this record.

Observation

The author bibliography records a 1987 paper on data compression using dynamic Markov modelling. [source]

Progressive questions

Which assumptions and update rules survive a change of application?

Responses

The bibliography identifies the original work. It is not a substitute for inspecting the paper. [source]

Interpretation

The linked later study provides a separate application to examine, not retroactive proof of every original claim.

Conclusion

Retain the source record. Leave uninspected mechanisms unresolved.

Application

Acquire and inspect the original algorithm before implementing a reproduction. No implementation or benchmark is claimed here.

Work through a related example.

IV.14 / “Nigel compression” identity resolution and Dynamic Markov Compression

Resolve a name using independent identifiers.

Create separate records for the informal label and each candidate publication. Compare title, authors, year, venue, algorithm description, and any supplied code. A shared word such as Markov or compression is insufficient on its own; a matching citation or implementation history would provide a concrete link.

X.05 / Information theory, coding, and compression

Make a codebook part of the size calculation.

Suppose a short document uses four words repeatedly. Replacing each word with a short symbol can shrink the body, but the decoder also needs the symbol-to-word dictionary. Count both components and compare the total with the original. Repeat with a much longer document to see how shared overhead changes the result.

Follow earlier and later work.

Before

No linked record.

After

Follow the chain
  1. Compression → classification · 1 link away · KNOWN
  2. Compression, reopened · 2 links away · UNKNOWN
  3. Information is not meaning · 2 links away · UNKNOWN
  4. What survives a replication? · 3 links away · UNKNOWN
  5. Memory as a collective state · 3 links away · UNKNOWN
  6. Erasure has conditions · 3 links away · UNKNOWN
  7. Compute has a budget · 4 links away · UNKNOWN
  8. Prediction and residual error · 4 links away · UNKNOWN
  9. Reversibility and the record · 4 links away · UNKNOWN
  10. Attention without recurrence · 5 links away · UNKNOWN
  11. The chemical compass · 5 links away · UNKNOWN

Placement.

Technology

No evidence of abandonment or suppression is asserted.

Sources.

  1. Publications of G. V. Cormack

    University of Waterloo · author bibliography · Journal publications: December 1987 entry, Computer Journal 30:6, 541–550.

    Bibliographic listing read; the original 1987 full paper was not inspected in this pass. Checked 2026-11-17.

  2. Spam Filtering Using Statistical Data Compression Models

    Journal of Machine Learning Research · Volume 7(97), 2673–2698 (2006); abstract.

    Journal metadata and abstract read; no independent replication performed. Checked 2026-11-17.

Record history
  1. 2026-11-17 / R1 / CREATED

    Source record created.