Review 1 for Allfrey 2025

The metadata of the dataset describes the dataset in as much detail as possible. They contain a human- and machine-readable license information. Standard data is referenced in a meaningful way (e.g. GND, Wikidata, GeoNames):

Comment: The metadata is generally good, with clear descriptions and human-readable licensing (CC BY 4.0). However, the dataset uses OS Open Names identifiers which are now deprecated (as the author acknowledges). The OS Open Names download indicates that 87% of these towns have alternative identifiers (GeoNames, DBpedia), but these were not included as additional properties. Adding Wikidata QIDs or GeoNames identifiers would provide stable, community-maintained identifiers and enable seamless integration with other georeferenced historical datasets. Creating additional Wikidata items where necessary would allow their sole use in place of other identifiers.

Data and metadata follow a common, suitable data standard (in that specific discipline) and are available in a common file format. The data standards used (and, if applicable, their versions) are named:

Comment: The data are provided in standard formats (GeoJSON and CSV) with clear, logical structure. The CSV column ordering is sensible, with the reference column placed last as it contains the most text. Data types are reasonably inferrable from the data itself. Minor enhancement: explicitly stating the GeoJSON specification version and character encoding (though UTF-8 is standard) would aid long-term documentation.

The authors of the dataset (and any other contributors) are clearly named and can be persistently identified (e.g. via the GND number, ORCiD), their functions are described in the metadata (e.g. via CRediT, MARC Relator or TaDiRAH):

Comment: The author is clearly identified with GND number (1362140252), ORCiD (0000- 0002-8733-3937), and comprehensive CRediT roles covering all aspects of the work. This is exemplary practice.

The data set and its subsets are clearly and meaningfully named and well structured. The data can be accessed directly via the PID without specialized tools, free of charge and barrier-free. Any access or usage restrictions are clearly justified (e.g. for copyright or privacy reasons):

Comment: The dataset is well-structured, clearly named, and accessible via DOI through Knowledge Commons Works (10.17613/cwart-61790) and GitHub. Access is free, barrier-free, and requires no specialized tools—both GeoJSON and CSV can be opened with standard software. The dual format enhances accessibility for different user needs (GIS applications vs. spreadsheet analysis).

The data is available in a file format that is suitable for long-term storage or in several file formats in order to increase the probability of long-term availability. A strategy for keeping the data and metadata permanently up-to-date, available and usable is in place and documented in the data set; the long-term archiving of the data and metadata is ensured. Changes are made transparent and traceable through versioning:

Comment: The data is available in stable, widely-used formats (GeoJSON, CSV). GitHub provides excellent version control and dissemination.

The data publication is an innovative tool that can be used by others and combined with other data sets. The data set contains information on completeness, context of origin, collection and processing methods and the quality of the data. Data gaps, uncertainties or difficulties are clearly stated:

Comment: The dataset is innovative and well-documented regarding methods, with transparent handling of uncertainties (77 locations flagged with location_uncertain=1). However, the context could be strengthened by engaging more explicitly with complementary datasets. The author's unique contribution—precise market square locations rather than parish-level geocoding—could be more clearly articulated relative to existing work like Bogart & Satchell (2017), which appears to exist only as an unpublished working paper (https://sites.socsci.uci.edu/~dbogart/blomemarkettowndatanote_may242017.pdf) but outlines an alternative algorithm for determining market locations within settlements.

Additionally, while Lawrenson (2023) is presented as comprehensive, his methodology relies substantially on archaeological evidence, which creates sampling bias toward areas where excavations have occurred. Documentary sources like Owen may actually provide more uniform geographic coverage, which is a strength worth highlighting. The discussion would benefit from acknowledging that archaeological and documentary evidence have different sampling characteristics and are complementary rather than hierarchical.

The data paper presents the context in which the data publication was created and the basic objective in a detailed and comprehensible manner:

Comment: The paper clearly presents the research context (taxation of armorial bearings, 1798–1944) and objective (locating markets to characterize urban clustering of taxpayers). However, some historical claims would benefit from citations—particularly the 6⅔ mile market separation regulation (paragraph 1).

The text also oversimplifies charter recipients: private individuals could hold market charters, and manorial lords often held multiple charters without residing at each location, so the assumption that rural market towns indicate gentry residences needs nuancing.

The relationship between market towns and gentry residence (paragraphs 1–2) is central to the research framework but presented without evidence. This assumption could be tested using Gadd & Litvine (2024) Index Villaris, which records both market existence and gentry/aristocracy residence circa 1680. While this predates the study period by 120+ years, it would provide historical context for whether this relationship exists.

For all published data, the data paper provides comprehensive information on the methodology of data collection and, where applicable, data cleansing and preparation. The choice of data schemes and file formats used is clearly explained:

Comment: The methodology is careful, systematic, and excellently documented with clear flowcharts (Figures 1, 2, 4). The three-pass approach with programmatic validation demonstrates rigor. Three suggestions for enhancement:

(1) Alternative validation approach: The NLS interface for the GB1900 dataset (https://maps.nls.uk/projects/os1900/) includes searchable text for "Market," "Market Place," "Market Hall," etc. Since these maps are georeferenced and date from ~1900 (close to 1813, before major 20th-century redevelopment), they could validate locations or help resolve some of the 77 uncertain cases. At minimum, acknowledging this resource would help future researchers undertaking similar work. The GB1900 dataset ought to be explicitly cited with reference to this and to your mention of Locolligo (see https://researchportal.port.ac.uk/en/projects/gb1900-crowd-sourced-gazetteer-of-britain).

(2) Silver Street variable (paragraph 17): This is an interesting exploratory variable, but needs brief explanation of the historical hypothesis—why might Silver Streets correlate with markets? (Silversmiths near markets? Revenue collection? Other reasons?) The same GB1900 interface enables identification of all ~76 Silver Streets in England (plus 1 each in Wales and Scotland), which could be compared against historic market locations to test whether the correlation is genuine or coincidental. The variable should be clearly flagged as exploratory/unvalidated for users.

(3) Reproducibility: The validation in Section 2.4 uses browser console transcripts, which demonstrate the process but aren't rerunnable. Providing executable scripts (Python/R/Node) as supplementary materials would enhance reproducibility.

Gaps in the data and weaknesses in the methodology are transparently stated and justified:

The paper is admirably transparent about limitations: 77 uncertain locations are clearly flagged with location_uncertain=1, potential gaps in Owen's coverage are discussed in Section 4, and the comparison with the 1822 Parliamentary return is honest about discrepancies (628 towns in common, 68 only in Owen, 102 only in the 1822 list). This transparency is a particular strength of the work.

The use of the data by the authors is described in a comprehensible manner and published research contributions published using the data are referred to. Reference is also made to other data publications that are similar or can be usefully combined with the data described:

Comment: The paper describes the author's intended use (analyzing armorial bearings taxpayer distribution) and engages with several relevant datasets (Letters 2013, Lawrenson 2023, Gadd's various works). However, several directly relevant publications deserve discussion:

(1) Bogart & Satchell (2017) "A Spatial Data Note on Market Towns in England and Wales, 1600–1850" geocoded 649 market towns from Blome (1673) at parish level for economic analysis. This provides important methodological precedent and demonstrates that geocoding historical market directories was already established practice. A brief discussion of how your precise market square locations build upon and enhance this parish-level approach would help readers understand your unique contribution. The apparent lack of formal publication for Bogart & Satchell (2017) makes this contextualization particularly important.

(2) Gadd & Litvine (2024) Index Villaris records gentry/aristocracy residence and market existence circa 1680, and is particularly relevant to your research question about gentry residence patterns and market towns.

(3) Relationship to my market gazetteer extension: You mention (paragraph 30) that I (Stephen Gadd) am extending Letters's gazetteer to the 19th century, covering your period of interest (1802–1830). A discussion of how your dataset complements or could be integrated with this work would be valuable. I would be happy to provide advance access to my extended market charter dataset to facilitate comparison or integration.

(4) Lawrenson (2023) methodology: While his thesis is presented as comprehensive, reliance on archaeological evidence creates systematic sampling bias toward areas where excavations have occurred. Documentary sources like Owen may provide more uniform geographic coverage, which is a strength worth highlighting.

Comprehensible potential usage scenarios are drafted and described:

Comment: The paper describes the author's intended use (analyzing armorial bearings taxpayer distribution) and mentions potential for combining with other datasets. However, since the armorial tax data is collected at parish level (paragraph 3), as is census data, and given that the precise locations of gentry residences are not known, it would be helpful to explicitly explain what analytical advantages precise market square locations provide over parish-level centroids for the stated research questions. The paper falls short on describing potential applications, which could include: (1) distance-based accessibility analysis from any point to nearest market; (2) integration with transport network data for analyzing market-road relationships; (3) longitudinal studies tracking whether markets moved within towns over time; (4) urban morphology studies relating market placement to street patterns; (5) network analysis for testing economic geography theories about market spacing. A brief discussion of such scenarios would help readers understand when this dataset's precision provides genuine analytical value versus when coarser geographic resolution would suffice.