[tei-council] open issues and planning

[Sebastian Rahtz]

Christian Wittern wrote:
>  While there is still some last minute surgery going on in some 
> quarters, I think we have to cordon off some areas now and start to do 
> the clean-up there, although we can't
> declare the whole building finished.
it makes initial sense to me to do this by module. Obviously namesdates
still has scaffolding on it,  while textstructure probably just needs
some last bits of painting. If we can declare some modules clean,
we can simultaneously check the prose and do any last minute
screen of the specs.

What we *must* have is a clear schedule of affected areas;
that's why finishing the SF work is so important. We don't
want any unexpected change to msdescription popping
up there. If there are 20 or 30 unresolved things, get them
listed, work on them, kill them one by one. Don't let them
get up again.

>   In fact, since there has been so much underground change in P5 it 
> seems likely that people will continue to discover changes, for 
> example in content models that where not intended but are a result of 
> the way we re-defined classes.
>
>
Sure, that'll always be true. We can assume a degree of risk
in anything we do.

If I was drawing up a list of risks which would stop the "Finish P5 by
October 1st" project from completing, unexpected bugs with wide impact
found by users would come in relatively low. A bigger risk is dependency
on volunteer or unscheduled labour; the greatest risk is, I would
say, that no-one uses P5 when we are done. If we were a business, that
would knock us dead, it would like no-one buying Windows Vista.

It might amuse you to know that I used the conversion TEI to XML
as my case study for a course on project management some years ago.
Building up these lists of risks and so on was quite illuminating.

Perhaps I should bore you all with a SWOT analysis?

-- 
Sebastian Rahtz      

Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

OSS Watch: JISC Open Source Advisory Service
http://www.oss-watch.ac.uk