Sourcing Vedanta texts

[Arun Prasad]

Our goal is to have all Sanskrit texts, but it is useful to start somewhere, and I think a nice place to start is Vedanta.

Major needs (please reply if you find a high-quality and public domain source for these):

- texts outside of the prasthanatrayi (ie outside upanishads, brahma sutras, & bhagavad gita). We have a good source for prakarana granthas by Adi Shankara but we need everything else.

- texts outside of advaita / vishishtadvaita. I know other sampradayas much less well.

- upanishads with svaras. We have the Isha upanishad but that's it.

Some of this work is duplicative but (1) I don't think duplication is inherently bad and (2) we can cross that bridge when we come to it. First, it's important to find good PDFs.

[विश्वासो वासुकिजः (Vishvas Vasuki)]

On Thu, 26 Mar 2026 at 11:25, Arun Prasad <aru...@gmail.com> wrote:

Some of this work is duplicative but (1) I don't think duplication is inherently bad and 

(2) we can cross that bridge when we come to it. First, it's important to find good PDFs.

This doesn't make sense to me. Only after you decide that duplication is worth it should you go about finding good pdfs, proofreaders etc..

Given paucity of resources (ie _good_ proofreaders mainly), it is best to scrupulously avoid duplication. And it's not like one has infinite time before literally the "gas runs out".

I generally find good (shAnkara, mAdhva, and to a lesser but good extant rAmAnujIya) vedAnta texts online on other portals already, and they can be scraped by those interested without much effort.

[Arun Prasad]

I thought the same but changed my mind.

- Digitizing all Sanskrit texts means a large and stable group of proofers that can train and recruit new members, simplify our workflows, and suggest improvements. Scale is important due to network effects, and proofreaders are not fungible due to differences in motivation and interest. So for me, the more the better. (That is, I am not choosing between 100% of capacity on new texts and 100% on old; I am choosing between 100% capacity on new and and additional 100% capacity on old.)

- The nature of proofing is changing, especially as OCR and automated tools become better. For example, a proofer could scan in a PDF, ask $LLM to reconcile errors against a specific scraped version, review differences, and publish the text. This is not costly.

- Ambuda's proofed texts are tied to a specific printed source version, down to specific pages and (maybe) specific lines. Scraped texts don't do this and often lack source information, so they can't be used as-is.

- Duplication also uncovers new errors. For example, I found several severe errors in various shAnkara stotras online. This is a kind of double keyed transcription and reduces errors for critical texts.

- I am willing to pay the cost of duplication at the margin if it means I can help people contribute to the vision in any capacity.

Arun 

[विश्वासो वासुकिजः (Vishvas Vasuki)]

On Thu, 26 Mar 2026 at 20:16, Arun Prasad <aru...@gmail.com> wrote:

I thought the same but changed my mind.

- Digitizing all Sanskrit texts means a large and stable group of proofers that can train and recruit new members, simplify our workflows, and suggest improvements. Scale is important due to network effects,

 

and proofreaders are not fungible due to differences in motivation and interest.

I see you're talking about volunteers (who might prefer to do some 15th edition of shankara bhAShya), while I was talking of hired proofreaders. 

 

- I am willing to pay the cost of duplication at the margin if it means I can help people contribute to the vision in any capacity.

But then this is confusing. If you're actually talking about paid proofreaders, what I said applies.

 

Arun 

On Thursday, March 26, 2026 at 12:51:44 AM UTC-7 Vishvas Vasuki wrote:

On Thu, 26 Mar 2026 at 11:25, Arun Prasad <aru...@gmail.com> wrote:

Some of this work is duplicative but (1) I don't think duplication is inherently bad and 

(2) we can cross that bridge when we come to it. First, it's important to find good PDFs.

This doesn't make sense to me. Only after you decide that duplication is worth it should you go about finding good pdfs, proofreaders etc..

Given paucity of resources (ie _good_ proofreaders mainly), it is best to scrupulously avoid duplication. And it's not like one has infinite time before literally the "gas runs out".

I generally find good (shAnkara, mAdhva, and to a lesser but good extant rAmAnujIya) vedAnta texts online on other portals already, and they can be scraped by those interested without much effort.

--
You received this message because you are subscribed to the Google Groups "ambuda-discuss" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ambuda-discus...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/ambuda-discuss/0cb04b40-af87-4203-b1c3-b5d42e65373en%40googlegroups.com.

--

--
Vishvas /विश्वासः

[Arun Prasad]

I mean volunteers, and "cost" in the general sense (time, energy, etc). Proofreaders we hire will not duplicate existing work.

Arun