re: [gutvol-d] Language free version of guiguts?
seems like you should welcome new tools... anyway, when are scans going to go online? until the general public can compare e-texts to the page-scans, you simply aren't using the best "checker" at your disposal -- all of their eyeballs. "debugging is parallelizable" (a.k.a. distributable), but hey, _only_ if you actually _set_it_up_ that way... of course, if you _want_ to do all the checking yourself... -bowerbird
On Mon, 16 Jan 2006 16:16:47 EST, Bowerbird@aol.com wrote: | |seems like you should welcome new tools... If they work on my projects |anyway, when are scans going to go online? When I get tools which work half way OK, and can get the submission process sorted to my satisfaction. -- Dave Fawthrop <dave hyphenologist co uk> 17,000 free e-books at Project Gutenberg! http://www.gutenberg.net For Yorkshire Dialect go to www.hyphenologist.co.uk/songs/
On Tue, 17 Jan 2006 14:13:09 +0000, Dave Fawthrop <hyphen@hyphenologist.co.uk> wrote: | |Finally guiguts is as it stands unusable on my texts. No doubt I will find |other equally drastic problems I have now played with gutcheck a bit more and worked out what some of the specifically *American* tests are which are not true in Yorkshire dialect are, or indeed *English* which translates to Queens English in American. Without the code or indeed a working knowledge of Perl, these are based purely on the gutcheck false errors found. My computer languages are C Fortran77, Basic, some Pascal, and a little Cobol. 1. Gutcheck objects to words consisting of one or more consonants without a vowel. These are common in Yorkshire dialect Th' T' etc.etc. the apostrophe indicates missing letters 2. Gutcheck objects to words of three vowels or more eea and similar words occur in Yorkshire dialect. 3. Gutcheck gets its knickers completely in the twist on single quotes ' It assumes that the single quote is speech, Whereas in English (Queens English in American) single quotes are uncommon and double quotes indicate speech. Hartley, especially uses double quotes for speech. It misinterprets apostrophes indicating missing unsounded letters as single quotes. 4. Gutcheck also gets its knickers in a twist about double quotes and objects to " " I have not worked out why, but no doubt that will give me more sleepless nights come to me at about 3 am :-( 5. Gutcheck also assumes that a line end is a paragraph end, which is not true in poetry, even American poetry. Speech commonly spans lines in all poetry, many lines in the poetry I work on, and may well span stanzas in Harleys work. No doubt I will find other systematic errors in the way Gutcheck works on my text. Sorry about the threading. Agent is sending things by email which I had meant to send to the list. :-( -- Dave Fawthrop <dave hyphenologist co uk> "Intelligent Design?" my knees say *not*. "Intelligent Design?" my back says *not*. More like "Incompetent design". Sig (C) Copyright Public Domain
On Wednesday 18 January 2006 09:55 pm, Dave Fawthrop wrote:
On Tue, 17 Jan 2006 14:13:09 +0000, Dave Fawthrop
<hyphen@hyphenologist.co.uk> wrote: |Finally guiguts is as it stands unusable on my texts. No doubt I will find |other equally drastic problems
I have now played with gutcheck a bit more and worked out what some of the specifically *American* tests are which are not true in Yorkshire dialect are, or indeed *English* which translates to Queens English in American. Without the code or indeed a working knowledge of Perl, these are based purely on the gutcheck false errors found. My computer languages are C Fortran77, Basic, some Pascal, and a little Cobol.
Dave, perhaps you're labouring under a misapprehension here. Your comments seem to indicate some confusion. Jim Tinsley's gutcheck is pretty much 100% plain vanilla C code. Steve (thundergnat's) guiguts is 100% perl, with the ability to act as a front end interface to external programs such as gutcheck. Source code to both is readily available and included in the downloads last I checked. I'm not sure what you expect either of those developers to do about your situation. It seems to me that you are trying to use a hammer when what you really need is a 4mm Torx driver. Obviously, both programs were developed with the intent of checking the most common texts submitted to PG--English. Others with specialised needs (such as decrufting poorly OCR'ed Fraktur, or old long-ess texts) have developed their own specialised tools for those specific purposes. They had the subject matter expertise and the technical skills to implement these. Quick Google searches will reveal other similar tools directed at their niches. I completely support the preservation of strongly localized texts such as those you are working with. Have you considered applying your skills in C and Yorkshire to create a customised version of gutcheck for your needs?
participants (3)
-
Bowerbird＠aol.com -
D Garcia -
Dave Fawthrop