Just so people know, progress on WaylaidWriter is well under way. See here.

Comment

  1. Kyllorac on 11 February 2010, 19:46 said:

    Nice! Your coding is disgustingly neat, though. Disgustingly.

    Is that line-by-line critiquing I spy? If so, you might want to nip over to Critique Circle. They have a paragraph-by-paragraph critique feature that you might want to take a look at for ideas and stuff, though their method uses JavaScript.

    P.S. – You never did extol to me the virtues of Ruby. :P

  2. SlyShy on 11 February 2010, 21:02 said:

    Heh, as a matter of fact, it is sentence-by-sentence critiquing, which no one else has been able to manage thus far (because having a computer figure out where sentences end and begin is difficult).

    Critique Circle is crazy annoying to use, compared to what I have planned. You know how lovely critiquing with a pen on a manuscript is? It’ll be as close to that as I can replicate on a computer.

    And yes, I should extol the virtues of Ruby. I’ll get on that after my hellish week of simultaneous assignments ends.

  3. Kyllorac on 11 February 2010, 22:36 said:

    I eagerly await this sentence-by-sentence critiquing. :3

    I suppose it’s too much to hope that the ability to doodle diagrams would be included. Ah, hardcopies – even in this day and age of technological advancement, you still can’t be beat. XP

    As for the sentence bit, have you tried denoting it by (punctuation mark)(space/line or paragraph break)? There are only so many punctuation marks that end sentences in the English language, after all. :D

  4. SlyShy on 11 February 2010, 22:50 said:

    Periods on their own aren’t enough. The problem is you get all sorts of ambiguity with honorifics, abbreviations, etc.

    Stuff like “She needs her car by 5 p.m. Saturday evening. At 5 p.m. I had to go to the bank. She has an appointment at 5 p.m. Saturday afternoon. By 5 p.m. Sunday I have to be at home.” will wreck attempts to naively use periods and capitalization to act as sentence markers.

    Sentence tokenization is considered a pretty difficult natural language processing task. I’ve made some breakthroughs though… much lower error rate than most published algorithms. :-)

  5. Kyllorac on 11 February 2010, 22:53 said:

    Completely forgot about abbreviations. @_x

    But progress is good! I expect great things from you Sly. Great things. >:3

  6. RomanticVampireLover on 12 February 2010, 17:43 said:

    Well done, Sly. :D I’m looking forward to seeing your finished product.

  7. Steph the Sue on 14 February 2010, 07:53 said:

    Sly, this is fantastic :)

  8. sansafro187 on 14 February 2010, 23:43 said:

    You’re kind of a genius, man.

  9. Artimaeus on 17 February 2010, 21:26 said:

    I can’t wait to see how this works.