xml-dev - Re: Gutenberg Project <longish>

[Len Bullard <cbullard@hiwaay.net>]

Frank Boumphrey wrote:
> 
> > I'd be interested in a detailed technical rebuttal of why the TEI does not
> > already provide exactly you need.
> 
> We tried using it. Unfortunatly our volunteers found it extreamly difficult
> to use. When I tested it on a group of 50(non-honors) students, I got
> exactly two texts marked up over a three month period (and they were being
> offered extra credit!).
> 
> At this rate it would take approximately. 1000 years to mark up the
> Gutenberg library!
> 
> OTH with more intuitive DTD's 30 volunteers have marked up 120 texts in
> under a month, and only two of these volunteers had formal training in XML.
> 
> Not a technical argument I know, but an argument from necessity.

All true.  Large complex DTDs are hard to use.  However, 
subsets based on these are easily associated to a larger 
corpus.  Building in aggregate namespaces requires us 
to use smaller sets of the aggregate yet get ease of 
association through the common names.  It is a DUH, 
but I think the TEI can easily be made serviceable 
through subsets.

All your test says is the volunteers found it hard to 
navigate and service a dense information set.  It 
is like learning to drive a car in heavy traffic. 
If markup mastery is part of that learning curve, 
it is even more difficult.  Or they aren't willing.

Carve up the TEI but keep the names insofar as possible.
You make it easier for the next team to use that way. 
DTDs and schemas work better when shared and best when 
shared on a network.  The TEI definition is well understood.

len


***************************************************************************
This is xml-dev, the mailing list for XML developers.
To unsubscribe, mailto:majordomo@xml.org&BODY=unsubscribe%20xml-dev
List archives are available at http://xml.org/archives/xml-dev/
***************************************************************************