[steering] Moving Forward

[.jeff]

martin,

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> From: Martin Burns
>
> > I think it is generally accepted, without working out
> > the details of full functionality (thetips,
> > archiving), to move the lists to e.o.uk. b.e.o is
> > already there, or at least a mirror.
>
> Should be doable very quickly.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

john may disagree after his trials and tribulations with css-discuss.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > So, the big problem is w.e.o and or.e.o.  We're not
> > likely to get another host with Oracle so we need to
> > migrate the database.  With the database change there
> > are CMS changes.  We need to finalize the database
> > changes quickly and start work on the porting of code.
>
> Jeff - the best understanding I have is that it's "a
> lot" of work. How much is that in man/days, and how much
> time do you have to give it over the next 6 weeks?
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

it's too slow to check files over ftp to get a good grasp on the total
number of templates that need work, but with my existing knowledge of the
applications, i'd estimate about 3-4 man days to migrate the database to
another platform.

i have at least that much time between now and our deadline to put into this
task.

><><><><><><><><><><><><><><><><><><><><><><><><><><><><><
> > I think, we should try to make the changes so that
> > future porting will be database independent, i.e.,
> > don't use INDENTITY, AUTO-INCREMENT, SEQUENCES, etc.
> > if possible.
>
> Totally agree - it also lets us start work before the
> final platform decision.
><><><><><><><><><><><><><><><><><><><><><><><><><><><><><

well, database abstraction to the level suggested isn't easy.  right now we
simply don't have a large enough window to do it right.  so, rather than
futzing around with something that will be a moot point once we settle on db
platform for weo, i'd rather put a shorter amount of time into porting the
existing to work with a set db platform in the interim.

on top of the time estimate above to port the code to work with another db
platform, we're also going to have loads of fun porting the data from oracle
to the new database.  we'll want to make a couple of test runs before the
final one and then shut down weo and do the port one last time while we wait
for dns changes to propagate.

.jeff

http://evolt.org/
jeff@members.evolt.org
http://members.evolt.org/jeff/