I would like to create a new web (desktop and mobile) application using the wonderful structured data that you’ve created, that would. help first with research and second with translating. I imagine some of the following features:
Ability to browse the Pali Canon at a high level using a snappy tree with descriptions for each sutta, like I created with Raindrop as bookmarks here. Very helpful to get an overview and navigate!
Possibility to cross reference terms from suttas. For example: you see a term in a sutta that is unclear, you click it (or search for it in the search bar, ideally in Pali) and it returns all the results where that term is used, ranked by relevance, first one being the one where the term is explained and defined. This would use Pali inflections from DPD to search for all variations of that term.
As for translations, I would like to propose some changes that would make it easier to create new ones:
Introduce variables into text. For example, instead of “rational application of mind”, put “<%=yoniso_manasikara%>” everywhere. Then, one can choose how they’d like that term to be expressed. This would allow us to quickly and easily translate and achieve consistency. Classical find & replace approaches will not work on all occasions because in many cases the same english word may be used for different Pali words and could affect meaning. Also, going one by one is tedious because every translator has to go through the entire Pali Canon. This way, we introduce the variable once, and we know for sure that they will fit in the right place, significantly reducing the chance for error.
Have a UI where these variables can be given values as the user (or translator) prefers.
What do people think? Would something like this be useful?
My main motivation is the fact that most Dhamma students who are not scholars, generally shy away from the suttas because they i) are cryptic, ii) hard to understand, and contain iii) foreign technical terms. Each translator has their own tradition which may or may not match the readers, take for example:
mindful (Sujato), aware (other)
aware (Sujato), clearly comprehending (other), clearly knowing (other), full comprehension of impermanence (Goenka)
rational application of mind (Sujato), wise attention (other)
placing the mind (Sujato), initial application of mind (comm.), thinking (dictionary)
Do you see the problem? This makes it very hard for people to read the suttas. My intention is to make the suttas alive and approachable using the latest technologies available today. Rather than being lost between thousands of scriptures, you can easily navigate between one another and clarify unclear terms within a few clicks.
Lastly, I am so grateful to SuttaCentral for solving the problem described here for me, because of introducing the line-by-line Pali view with the ability to click on words and see the dictionary entry. What a wonderful feature! This helped me approach the suttas and actually be able to make sense of them! Thank you so much.
Much mettā,
Gabriel
P.S. I am a very big fan of Bhante Sujato’s work and truly grateful for all these translations, which so far are the closest to contemporary human language. Wonderful work. Just want to make that clear so that my comparisons above aren’t somehow perceived as criticism. There is zero of that in them.
I think it’s fine to go ahead and use SC content to make your app. Just no AI! The content is freely available for all other uses.
If you are used to another translation then this is a place that can rub. I first read all the suttas in English with Venerable Bodhi’s translation. Adjusting to a different translation took time, but now I know what these phrases and words point to it isn’t a problem.
It’s like when you are learning any language, thinking in translation is the worst way to think. As soon as you can natively think in the new language then concepts flow smoothly. There is the caveat that concepts dont always have a one to one correlation. In English we have no real concept of Schadenfreude, so much so that we’ve adopted the German word.
I think that it’s fine to go ahead as you proposed. I have seen Bhante Sujato say, on his bluesky, that people just have their preferred translation for some things. I think that the difficulty will be in defining what terms to put markers around and how that will work with a case based language where there is no one to one word match.
DPD uses AI for development, so utilizing that in your workflow might also go against Sujato’s wishes. I am unclear where the line is that they’re drawing.
Well, the DPD is the default dictionary on SuttaCentral, so it’s clearly not a problem. My understanding is (at the moment) the only thing AI/LLM is used for in the DPD is coding, not the actual creation of the dictionary itself.
Even for coding assistance, Claude is reading the full codebase including the human articulated translations. And since Claude now runs on xAI’s Colossus (with Anthropic paying them over a billion dollars a month), that means DPD is indirectly supporting the very players Bhante Sujato has warned against so strongly. So I’m still not clear where the line is being drawn.
Let’s not derail this thread with that discussion. It doesn’t sound like @gbbr is planning on using AI, so it’s a non-issue. If you have concerns (which are, indeed, legitimate), you could start a new thread.
I don’t think this is really a major stumbling block for new readers. New readers should be sticking to a single translation whenever possible.
I think there is an idea out there that you can’t understand the Dhamma through translation, only reading in the original Pali. Therefore until you can read Pali it is important to read multiple translations to try and get to the core meaning. This is probably one of the worst things for a new reader to do.
Buddhadust.net has a resource that compares how different translators render technical terms, e.g. yoniso-manasikārā. But honestly, it has a lot of translators people are unlikely to encounter.
The issues you have raised about search ranking are very real and I think there is still the wish to improve the native search on SC. Ideally this would be fixed on the site so everyone has access to the improvements.
Note that we take a whole lot of data from here and copy it into sc-data:
sc-data contains a lot of stuff that is fairly insignificant (currency data for instance) or barely there (author bios has about three entries I think).
However, in there is the primary reason for SuttaCentral’s existance, the texts, the structure that contains them, the relationships between them. I’m yet to find it all myself.
Thank you so much for all the information and for your kind support. I will have to ponder whether a web app like this is worth the time investment and if it would actually be useful to others too, because I am not yet certain of that ! From what I gather from your responses (and elsewhere), there are already quite a few websites using this data. Ideally, it would be nice to have a strong powerful research tool that simply allows doing all of the things. Perhaps SuttaCentral can become this, and perhaps I should get involved with supporting the development of it (I don’t know anything about the plans here or what is going on). I will have to spend more time with both the suttas and the SC team to learn before I can draw a conclusion.
Ultimately, I’d love to make it easier for people to find their way around and read the suttas, in a beautiful interface that is easy to navigate.
From what I understand, Bhante Sujato is reluctant to add any new major features to suttacentral.net. As you may know, every new feature, no matter how small, increases the surface for bugs to appear. So major new features are probably not going to happen. But who knows!
In lieu of making major changes to suttacentral.net, all of the Bilara/segmented translations are made available for people to use any way they like other than as LLM/AI fodder. And I find that Bhante is quite happy to provide assistance for these re-use projects. The APIs are even public, so it makes it easy to create front end only projects. For example, this Sutta Diff app uses the API. So does this Anthology Builder
The other route is that of browser extension. In fact the reason I made the browser extension was to quickly add niche features that were unlikely to be added to the main site.
Search is something we all know needs improvement. Help with that is most welcome.
Oh, not at all I just spend a lot of time studying, and as I’ve started reading the suttas, my mind comes up with all kinds of ideas for improving the experience of learning from them online (having been a programmer and UI developer for many years). By “research” I mean specifically making it easy to learn and navigate the suttas, and making it easy to navigate from a concept that is only briefly mentioned in one sutta (assuming that the audience knew it) to another sutta where it is explained at length. However, many of these issues are solved by Bhante Sujato’s comments on statements. But that only happens on some main suttas. It would be nice if one would be able to click a term and find definitions in other places.
I just love this community and am so glad I got involved. Yes, I knew this website, but haven’t looked very closely at the Index. Wow! So wonderful. It seems like all the tools I am looking for are available, granted a bit spread out.
Thank you for sharing this. I will study it carefully.
More broadly speaking, that every line of code needs long-term maintenance.
Similarly that tense, context, etc etc can change the meaning of words (connotation), so a different translation can be more appropriate or easier to understand in a different section. (Although Bhante Sujato’s translations do try hard to be consistent.)
So Bilara, the Computer-Aided Translation (CAT) software that SC uses & develops, has a “translation memory” and other features that help with this. (I’m not entirely sure how they’re implemented though, as I’m not a translator and haven’t contributed to Bilara much yet)
Having the user use their own translation mid-sentence is interesting.
Making sure it works properly in every context may be tough; that is one of the jobs of a translator, and changing their words may go against their intent in word choice too (i.e. they chose specific words to try to get across a specific meaning).
When I first started contributing, I too discovered many interesting tools, micro-sites, etc related to the EBTs and more that I hadn’t known about (and still find a new one occasionally!). Very cool stuff around!
Dear @agilgur5, thank you very much for your kind and detailed reply.
I apologise for my ignorance, but I wonder what could be so bad with maintenance given that there is only one data set that will never actually change (the Pali Canon) and some translations coming from time to time. I don’t see it as such a dynamic website that it is bound to ask for bug fixes regularly. But I want to once more emphasize my ignorance because I am not aware of all the features. I will start watching the repositories right now and try to get up to date, and if that is the case, I may even try to help out from time to time, but no promises, as I’m already having a hard time staying off screen and it’s easy to find reasons not to
Very interesting. I will have to look at this project more closely.
I guess this can be both good and bad (or dangerous). Because you may think you know what you’re doing and may end up just confusing yourself more. But, if the Pali is there and you use the dictionary, the chances are low. I practice in the Goenka tradition, and it would be truly wonderful to read the suttas in the terminology used there. After all, any translation is fine, it all depends on which lineage you usually rely on in your practice, what each word means to you, and which ones you resonate most with. I’ll be honest that I have downloaded all of the JSON files locally and modify them myself, using a TamperMonkey JS script to inject them back into SuttaCentral when viewing. Quite hacky, but it is extremely helpful for me and am wondering if it would be for others too…
I truly like the Index on ReadingFaithfully a lot, and I’ll have to look more closely at the Bilara project.
I just spend a lot of time studying, and as I’ve started reading the suttas, my mind comes up with all kinds of ideas for improving the experience of learning from them online (having been a programmer and UI developer for many years). By “research” I mean specifically making it easy to learn and navigate the suttas, and making it easy to navigate from a concept that is only briefly mentioned in one sutta (assuming that the audience knew it) to another sutta where it is explained at length. However, many of these issues are solved by Bhante Sujato’s comments on statements. But that only happens on some main suttas. It would be nice if one would be able to click a term and find definitions in other places.