Introduction
Goal
We've set out to build a database of subhAShita-s (= worthy quotes) which is:
- Universal
- Its goal is to contain within it every worthy subhAShita ever composed.
- In fact, the ambition encompasses all languages, verse and prose forms.
- Freely and easily available. Anyone should be able to
- Access it
- Export it in other formats
- Present it in any way users will find convenient.
- Eg. the upcoming [subhAShita-pratimAlA] app project.
- Growing constantly in number
- sarasvatI still suckles some amongst us at her bosom!
- Growing constantly in annotations/ ratings
- add annotations (rating, description, translations, metre, flaws, sources ...)
Expected extensions
- We hope that this will motivate other such long-sought-after universal databases for sanskrit connoiseurs, like: one for metres.
- The clients built for this database could serve as a model for other kAvya readers.
- Similarly, one can build a collaboratively annotated and rated collection of verses/ sentences within the context of long sequential works (rather than free floating subhAShita-s).
Motivation
- One of the greatest (and useful) pleasures I've had in tough times is retreat for a while into the world of beautiful subhAShita-s - and then burst back out like the vRtraghna armed with dadhIchi's bones.
- I especially like online collections curated by some friends and myself:
- since a book is not always available, and
- I want to collect + easily access choice ones for future enjoyment.
- But it is tedious (atleast for me) to sit in front of a computer to do the following:
- read them,
- or scour the internet for new ones
- or collect favorites in a spreadsheet
- or just annotate them with comments.
- So, it is desirable to:
- make the above as simple and easy as possible,
- and to share our collective labor so that we can benefit more easily from each others' work.
Database and API
- Periodic update announcements here.
Couchdb
API-s
- Querying quote entries: use the database and the index_headwords index
Database replicas
- You want to host a repilica and make things faster for folks in your geographical area? Just open an issue in this project and let us know.
- Ahmedabad, IN http://vedavaapi.org:5984/quote_db__sa/_all_docs
- Bay area, USA (dev machine, unstable) http://vvasuki.hopto.org:5984/quote_db__sa/_all_docs
Technical choices
- Use Nosql rather than a relational database as the primary canonical database.
- Reasons (from here ):
- The need to develop with agility
- Simplicity (less need to join n tables)
- Flexibility (schema can easily be changed)
- distributed nature: ability to operate at any scale.
- Database copies can sync with each other easily - no centralization is good.
- The need to develop with agility
- Comparison here .
- Alternatives for mobile offline use:
- Couchbase database, since it can be used in many mobile OS-s.
- Comparison with mongodb: on cb site.
- Realm.io - good for storing plain Java objects via JSON.
- Google Firebase for key-value storage - 1GB free.
- Couchbase database, since it can be used in many mobile OS-s.
- Reasons (from here ):
- Scala for the ingestion libraries (same justification as expressed here ).
Contribution
A note for subhAShita collectors
- Are you a collector of subhAShita-s? Would you like to contribute your collection to the database? You're very welcome!
- But you should make your collection available in such a way that a machine which does not know human languages can distinguish individual quotes, ratings and various other details.
- सङ्क्षेपेण - मानुषभाषाङ्कामप्य् अनवगच्छता यन्त्रेणापि "अस्यां सुभाषितावल्यां सन्तीमे सुभाषिताः। अमुकस्य सुभाषीतस्यैते विषयाः। अस्मिन् सुभाषीते कस्यचिदियं टिप्पणिः। सुभाषितमिदम् अस्मै ५/५ इति मानेन रोचते। अपरस्मै ३/५ इति मानेन। अस्य सुभाषितस्य लेखकोऽसौ।" इति ग्राह्यं स्यात्। तेन पाठकानुकूलं संलक्ष्य भवितुमर्हति सङ्ग्रहस्य यान्त्रिक-प्रस्तुतिः।
Code Contribution
Links to general comments
See indic-transliteration/README for the following info: