[steering] Migration plan (was: Gonna start calling votes)

[Martin Burns]

On Sat, 11 Jan 2003, Marlene Bruce wrote:

> Here's my summary and take on the discussion:
>
> 1. Dump m.e.o. in favor of f2o.

Not that we can do more than recommend f2o... we can't do a simple move
people over there for a whole bunch of reasons, not least because we can't
make decisions on f2o's behalf.

> Give m.e.o. users notice, time to move (1-2 months?), and longer-term
> redirects (3-6 months?). We could zip unmoved files and archive them
> for sites that don't move themselves, and return the files if we get
> complaints.

As part of that, stop taking new subs *now*.

> It would be ideal to have evolters.org up first. Question: is it
> realistic to tie the two together?

Nah, separate them so people don't see evolters.org as 'a duff meo' but a
good something else.

> Will tieing them affect our ability to achieve the other goals listed
> below?

Yes as it'll take up our development effort needed for the below.

> 2. Move l.e.o. to:
> a) e.o.uk, or
> b) a sponsored (or inexpensive) service
> (How is CSS-D doing as a proving ground?)

Anyone who's a member want to comment on performance?

> 3. Create evolters.com to foster the community.
> What Isaac said:
> Some kinda blog space, member profiles, ability to network
> (professionally or socially -- find other people in your area).
> News/community articles, and recipes if people want to post them
> (taking care of FEO).

To make us financially stable, we need to do one or more of:
1) Charge money for this ($30 or $60/year)
2) Sell enough merchandise *cough*tagwear*cough* to cover hosting costs
3) Do well enough in cash donations

> 6. Create a new wiki and open it up to the community (specifics TBD).
> What Martin said:
> I'd also like a general wiki - wider in scope than the one we've used
> to date and explicitly open to all. Might even do some of the work done by
> the tips (plus allow community editing, getting round the 'this tip is
> wrong' issue)
>
> 7. Suspend the tip harvester if needed.

How many tips do we get these days?

7a. Maintaining the existing tips db so that we don't lose the resource

> (8. Redesign the sites.)

Implicit in the above.

> >Okay, if we start with that, what is the migration plan?  What issues
> >do we need to resolve?
> >
> >6) Exporting/importing users

css-d showed that that's easy.

> >8) Remove dependence on ColdFusion for l.e.o pages, unless hosting on CF box

That's, what, about 2 pages? Handleable in a weo FAQ article, surely.

> >9) Tracking sub/unsub in Oracle

Remind me why?

> >10) Aliases, virtusertable

Again, remind me why?

> >11) Mailman configuration

Initial list creation then per-list, done by the list admin in question.
An hour's work per list.

> >12) evolt.org patches to mailman

I'm still struggling to understand what these give us (other than more
work in the migration) above the current Mailman featureset.

Cheers
Martin
--
"Names, once they are in common use, quickly
 become mere sounds, their etymology being
 buried, like so many of the earth's marvels,
 beneath the dust of habit." - Salman Rushdie