Curatorial Statement

News Report Curation

The Wilde Trials International News Archive currently features European and North American news reports on Oscar Wilde's trials and imprisonment in 1895. Overall it includes representative news coverage from twenty-five countries in five different languages, amounting to nearly 1.5 million words. The entire list of multilingual news reports can be accessed on the All Reports page.

The majority of these reports are sourced from digitized newspaper collections. These include large open-access national collections, multi-title commercial databases, as well as smaller digitized collections devoted to individual titles. To ensure a diversity of content, the archive also includes reports from non-digitized newspapers. National collections generally prioritize the digitization of major titles, rather than specialist and foreign-language newspapers. As foreign-language newspapers are important sites of transnational cultural transfers, many of those that carried reports on the trials have been transcribed by hand at collection sites from print and microfilm sources.

Digital search within databases is the primary method for discovering news reports related to Wilde's trials and imprisonment in 1895, beginning in April and extending until the end of the year. Visual search via browsing is a complementary method used when sources are not digitized or only partially digitized. Various other methods are deployed to extract news reports, depending on the technologies and layouts of the different databases. To consult the source databases and institutions for the reports, see News Sources.

Transcriptions

The text for most news reports in the archive is computer generated using optical character recognition technologies (OCR). Non-digitized reports have been transcribed by hand as accurately as possible. Transcriptions of news reports are checked and corrected for high accuracy, but not at an editorial standard. The transcriptions aim to provide a good quality text for reading as well as for computational text comparison.

Determining what constitutes an individual news report presents some challenges. News reports differ widely in format across different newspapers, and sometimes even within a newspaper. Some reports appear as stand-alone items. Others add on additional late-arriving telegraphic dispatches. Yet others are mixed with other news items in longer columns. In each case, the archive counts these as single reports. Where reports originally appeared in longer columns containing unrelated news, the unrelated news is stripped from the text. Links to facsimile pages allow for the consultation of news in wider contexts.

The original formatting of the news reports is not strictly preserved, as the archive's primary interest is in the text of the reports rather than their appearance, and the text-comparison tools necessitate standardized encoding of line breaks and crossheads. Headlines are preserved, but are reconstructed to show the title of the column (if relevant) and the headline of the individual report within the column (if provided).

Many hands have been involved in the correction and transcription process, acknowledged in the Contributors page.

Translations

Machine translations of news reports are available for non-English reports into English, using the Google Cloud Translation API. These translations have not been corrected and are not polished. They are meant as a bridge for understanding and analysis.

Bibliographic & Matching Metadata

Each news report includes basic bibliographic metadata about the newspaper from which it was derived, as well as date of publication, word count, and source information. If sourced from open-access databases, a link to the facsimile is provided. In the case of certain databases behind paywalls, however, it is not possible to provide a direct link to the report.

News reports also include matching metadata when there is evidence that a small or large portion of the report reappeared in other reports collected in the archive. Next to each report are match buttons. These show the number of similar paragraphs from other reports in the archive. On the same page, there is a list of news reports that match the displayed report, at the document level, along with a match percentage.

To export and analyze the metadata collected from all the reports, see Data.

Comparisons

The match buttons quickly reveal whether a news report shares content with other news reports and enable comparative understanding of the reports. To allow for further analysis, links from these reports open side-by-side comparison pages of the selected reports, at the paragraph level as well as at the document level. Match metrics and text highlighting also facilitate close, detailed comparisons.