New blog post, "Parsing: a timeline"

[Jeffrey Kegler]

I've just pushed a new blog entry: "Parsing: a timeline".

[klca...@gmail.com]

Some discussion is occurring at Hacker News: https://news.ycombinator.com/item?id=8290681

[Ron Savage]

In my head I've written another, non-technical, article, about how animals+humans use DNA-based parsing to recognize sound (aural input) and text (visual input).

Then the text transitions to (very briefly) lex/yacc and flex/bison, and then to Marpa. It'd link neatly to Jeffrey's article, and to the link at Hacker News.

This is all cultural knowledge (background) by way of introduction to specifics about Marpa.

I've only committed about 10 lines to 'paper'. Since I want a number of tables to display relationships, I'll probably keep writing it directly in HTML, even though it's a bit laborious.

[Helmut Wollmersdorfer]

Am Montag, 22. September 2014 03:21:57 UTC+2 schrieb Ron Savage:

In my head I've written another, non-technical, article, about how animals+humans use DNA-based parsing to recognize sound (aural input) and text (visual input).

Ron, this sounds interesting.

Maybe I should post my approach and the status of my work on parsing old books froms scans, correct the OCR (with help of big data lexicons), then parse it (with the help of big data lexicons) etc.

Where big is up to 5 million entries per language, or feature.

Helmut Wollmersdorfer

[Ron Savage]

Hi Helmut

I don't remember ever seeing your name hereabouts.

More below.

On Thursday, 25 September 2014 02:12:42 UTC+10, Helmut Wollmersdorfer wrote:

Am Montag, 22. September 2014 03:21:57 UTC+2 schrieb Ron Savage:

In my head I've written another, non-technical, article, about how animals+humans use DNA-based parsing to recognize sound (aural input) and text (visual input).

Ron, this sounds interesting.

The way I started is to imagine myself giving a talk. Typing is much more labourious.

 

Maybe I should post my approach and the status of my work on parsing old books froms scans, correct the OCR (with help of big data lexicons), then parse it (with the help of big data lexicons) etc.

Where big is up to 5 million entries per language, or feature.

That's not big. It's shockingly big!

 

[Jeffrey Kegler]

Helmut was very helpful when I was wrestling with Unicode.

--
You received this message because you are subscribed to the Google Groups "marpa parser" group.
To unsubscribe from this group and stop receiving emails from it, send an email to marpa-parser...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.