
🔍How Japan Search's Image Similarity Search API Works
Investigating the undocumented API behind Japan Search's image upload similarity search
1091 articles in total

Investigating the undocumented API behind Japan Search's image upload similarity search

KotenOCR is a free iOS app that runs the NDL Koten OCR-Lite model entirely on-device, enabling offline recognition of kuzushiji (classical Japanese cursive script) from photos.

A comparison of five methods (SingleFile, Playwright PDF, ArchiveBox, WARC, yt-dlp) for locally archiving Yahoo News articles and videos, covering output format, file size, and use cases

How to build an end-to-end pipeline that captures simulator screenshots with XCUITest, generates marketing images with Python Pillow, and uploads them to App Store Connect via API -- all from a single shell script.

A hands-on tutorial on collecting bibliographic data from the National Diet Library Search API and fine-tuning a 1.8B Japanese LLM with LoRA to classify books by their NDC (Nippon Decimal Classification) category from title alone.

researchmap does not support linking KAKENHI research projects to achievements via API or CSV import, so I created a Playwright script to automate the browser-based workflow.

How cache configuration and parameter tuning of the Cantaloupe IIIF server achieved up to 7.6x faster tile delivery, with methods and results.

How to diagnose and fix a greyed-out Publish button in Contentful caused by locale misconfiguration in a multilingual setup.

A record of using Claude Code's worktree and agent features to fix 6 GitHub Issues in parallel for a Nuxt.js TEI viewer project.

Tackling 391 "Crawled - currently not indexed" pages in Google Search Console by implementing schema.org structured data (JSON-LD) on a digital humanities site.

An introduction to CATMA, the web-based text annotation and analysis platform developed by forTextLab at the University of Hamburg.

Datawrapper is a data visualization tool for easily creating charts, maps, and tables. It supports 20+ chart types, choropleth maps, and provides responsive and accessible visualizations.

Flourish is a platform for creating interactive data visualizations including race charts, animated maps, and interactive stories. With 30+ templates, it is adopted by BBC, Google, and more. Free plan available.

An overview of FromThePage, the crowdsourcing transcription platform for historical documents, and its applications in Digital Humanities.

An introduction to Gephi Lite, the browser version of the popular network visualization tool Gephi, and its practical applications in Digital Humanities research.

Hypothes.is is an open-source annotation tool compliant with the W3C Web Annotation standard. It enables highlights and comments on any web page, used in education, research, and journalism. BSD licensed.

An overview of Internet Archive, the world's largest digital archive, and how it can be used in Digital Humanities research.

An overview of Kepler.gl, Uber's large-scale geospatial data visualization tool, and how it can be used in Digital Humanities research.

An overview of Mirador, a IIIF image viewer developed by Stanford and Harvard, and its applications for comparative research in digital archives.

An introduction to Observable, the JavaScript-based data analysis and visualization notebook platform by D3.js creator Mike Bostock, and its applications in Digital Humanities.