Corpora: Collaborative effort

[Jem Clear]

Phil

Indeed I am turning the Murray OED procedure on its head.
Re: quality control and sources: I do not believe that we should
try too hard to address these things directly for two reasons:

a) who would do the quality control? It would be a huge amount of
work for someone to read through all submitted examples and ensure
that they were "accurate". The key to my idea is that no-one has
to carry oiut any major work -- becuase if they do they won't do
it or they'll want funding to do it and once they've invested
a major amount of time and effort into it they will become
reluctant to hand the results over to the public domain. The
quality-control people will in effect become the guardians of
the growing database -- something I think we should avoid.

b) the idea of quality control implies that there are clear
criteria for acceptable and unacceptable input. I doubt this is
the case. Why don't we trust our colleagues and admit that if they
submit data that they believe is illustrative of the particular word/sense
then this is as valuable data as any other. My firm belief and hope is
that overall the central tendency will be towards a "correct" interpretation
despite some percentage of "wayward" submissions.

Cheers

Jem