As a relatively new adopter, I already feel like I face something like
50 configuration files scattered over a tree of uPortal core and
portlets, plus a mirrored tree of overlays, plus issues related to
desiring different overlay trees for each of 4 environments so that
configurations can propogate from Dev to Test to QA to Prod. Or, so
that a test environment portal can be configured to talk to a test
environment active directory service, while the production portal
talks to the production AD service. I've long since given up the
notion that I might memorize which features are in which files. (Grep
is my friend.)

I see both Drew and Cris's points - it is two views of the same data.
Both make sense depending on the task at hand. Either way, it's one or
three config files.

I can say early on I knew of SmartLdap being a feature. I only just
now heard of Quartz. My experience is just an anecdote. I don't know
if this perspective is helpful. For those who have worked with many
sites, which is the more likely feature to be encountered first? Is it
perhaps better to make the earliest features the easiest features? The
most commonly used features the easiest features?

Gatting back to "grep is my friend", if I search on a feature's name,
do I get hits on all of it's configuration items? Does each SmartLdap
item have "SmartLdap" in its name? Does each Quartz configuration item
have "Quartz" in its name?

-- 
Bruce Tong
Software Engineer
Office of Information Technology
Ohio University

-- 
You are currently subscribed to [email protected] as: 
[email protected]
To unsubscribe, change settings or access archives, see 
http://www.ja-sig.org/wiki/display/JSG/uportal-dev

Reply via email to