[sword-devel] Greek texts

Paul Gear sword-devel@crosswire.org
Sun, 16 Jul 2000 08:05:58 +1000

Trevor Jenkins wrote:
> On Saturday, 15 July, 2000 01:58:34, Paul Gear <paulgear@bigfoot.com> wrote:
> > Jeffrey Hoyt wrote:
> >>
> >> On Fri, 14 Jul 2000, you wrote:
> >> > Guys,
> >> >
> >> > Is anyone aware of any effort to spruce up our Greek texts a bit?
> >> >
> >> > I would like to see:
> >> >     - standardised encoding (so that all the texts look the same in
> >> > BibleTime)
> >> >     - breathings, punctuation, capitalisation, and maybe accenting
> >> >     - paragraph formatting
> >> >     - morphology codes
> >>
> ...
> How accurate are these parsed texts? Whilst not Greek the available
> morphologically parsed version of Biblia Hebraic Stuttgartensia (BHS) is
> known to be flawed---some say seriously flawed. Gramcord invested huge
> amount of effort in correcting these mistakes and errors for the version of
> BHS that appears in Accordance. (One of the reason that professional
> theologians etc recommend Accordance over all other products.)

Any professional who recommends a MacOS program needs their head read! 

> If there are morphologically parsed versions of the Greek then whose going
> to verify that the contents are accurate? Can't be me; I have limited Greek
> skills.

I was planning to do a large amount of the work myself as part of my
college research project.  If other people want to contribute, they can
do so.

> Having a plain Greek module is better than a seriously flawed super-duper
> one. In my book any way. Whilst I don't want to put anyone off this addition
> lets not get carried away with a because it's available let's do it
> mentality.

I'm going to have to disagree with you there Trevor.  We can't wait
until texts are perfect before publishing them.  If so, we'd never put
anything out.  The way that inaccuracies in these texts are found is by
people using them intensively.

Even if the codes have a 10% error rate, that's still 90% accurate.  As
long as there is an appropriate caveat provided with the text, and an
easy mechanism for reporting errors, there is no reason why we should
limit their publication.

Most of the work i am talking about would be able to be done
automatically by a text generation database once the appropriate raw
material is available.

"He must become greater; i must become less." - John 3:30