[tei-council] Attributes without examples

[Lou Burnard]

On 13/08/12 15:23, Paul F. Schaffner wrote:
> On Sun, 12 Aug 2012, Kevin Hawkins wrote:
>
>> Besides struggling to find times for the other things I've agreed to do,
>> my main problem is that I have no corpus of encoded literature to
>> consult to find real examples.  So I would be creating fake ones.  While
>> that's okay for some attributes, for many it's painfully obvious if it's
>> faked.
>

There's an obvious boot-strapping problem here! If we'd confined 
ourselves to pre-existing examples of TEI-encoded texts during the 
writing of the Guidelines we wouldn't have got very far. I wonder 
whether Kevin is interpreting "real examples" rather too literally -- 
for me it means that the text being encoded is a real one, not that the 
encoding already exists. So I think it's perfectly OK (as I just did in 
fact for the Lite tutorial) to grab an example from a favourite real 
text and encode it so as to demonstrate the feature concerned.

So Kevin, I happen to know from experience that just a few minutes walk 
from your office you have a pretty interesting collecion of books!