Comment by SamBam | Hacker News Reader

Comment by SamBam | Hacker News Reader

Parent Full thread

Was the text added using OCR? I ask because I see a number of text errors that look like that.

* "1" for "l" (http://nkanaev.github.io/zen101/en/009/)

* Unmatched quotation marks or differing symbols (‘ vs ') (http://nkanaev.github.io/zen101/en/002/)

* Several times I've seen "yon" for "you" (http://nkanaev.github.io/zen101/en/004/)

* "l" for "r" (loom/room) (http://nkanaev.github.io/zen101/en/099/)

If so, what's the original source for these stories? I don't see credits anywhere, though I am probably just missing them.

Yeah, definitely. E.g. mistaking "alone" for "done":

    he used to travel done as a wandering mendicant

As for the source, this[0] archive of a 1957 book has the same typo, so I'm guessing that's where it came from?

[0] https://archive.org/stream/ZenFleshZenBones/zen_djvu.txt

I recognize the stories from Zen Flesh, Zen Bones (I have an old paper copy).

And, although the stories may be ancient, the translation is presumably not. Are these out of copyright?

Hmm... while I don't know, it'd be interesting to write something that would automatically fix those things - which I find interesting because I would think that a good OCR service should use some model of word frequency or sentence structure to aid symbol resolving.

Perhaps sort of digital wabi-sabi meant to test the zen level of the reader? If so, I'm failing.

Reply on news.ycombinator.com