Sāgarasaṅgama is a virtual confluence point for large online Sanskrit e-text collections. Each component project gets its own Atlas interface, and Sāgarasaṅgama sums over them.
An "atlas" is a navigational layer over a collection someone else hosts. It charts the navigable waters with improved navigation, and it plumbs their depths with sizes, counts, and growth dynamics that the source collection doesn't surface on its own. No Atlas hosts text content of its own, instead linking back to the source.
These large collections are poorly understood in their entireties because their own web interfaces are at once too bespoke and too limited. By standardizing navigation UX and giving more attention to issues of scope — the added info about sizes, counts, and growth dynamics — users will appreciate these collections at a deeper level and make better use of them day to day.
In addition, an Atlas surfaces structural problems in the source collection so that they can be fixed upstream.
Any one Atlas can say how large its own collection is, but only the confluence can say how multiple compare, or answer the question a reader arrives with: who if anyone has this text? Sāgarasaṅgama searches titles and authors across all Atlases, charts their scope and growth against each other, and keeps a front door for collections most readers meet one at a time.
| Collection | Atlas | E-Texts | MB (IAST) | Avg KB | PDFs |
|---|---|---|---|---|---|
| sa.wikisource.org | Sanskrit Wikisource | ||||
| ebharatisampat.in | E-bhāratīsampat | ||||
| sanskritdocuments.org | Sanskrit Documents |
While E-bhāratīsampat and Sanskrit Wikisource PDFs counts represent scans tied to their own texts and hosted on the site, Sanskrit Documents links e-texts to scans of editions only loosely, and the great majority are hosted not on the site itself, but elsewhere (chiefly archive.org). These outbound links are checked for liveness, and dead ones are not counted. Each unique PDF is counted only once (e.g., forty stotra documents may link to a single scan).
Sanskrit Wikisource's text count suffers from some insignificant noise for nine months across 2014–2015. Those months are shown here corrected to the level the months on either side agree about; the collection's total today is unaffected.
As many drops of water as collections, united and focused by a bindu of purposeful attention.
Source at github.com/tylergneill/sagara-sangama, released under CC BY-NC-SA 4.0. Part of kalpataru-grove.info.
code: