DLM in uPortal 3

[Andrew Petro]

Peter, others,

Could you explain your understanding of what is required to port or
re-implement DLM as it exists in uPortal 2.5.2 into uPortal 3, thereby
bringing in the attractive increment of features beyond those available in
SLM?

Thanks,

Andrew

------------------------------------
Unicon, Inc.
Andrew William Petro
Software Developer
ape...@unicon.net
3401 N. Arizona Ave
Suite 113
Chandler, AZ 85225
mobile: 203.915.7333
------------------------------------

[Adam Rybicki]

Andrew,

After reading Shoji's note on JASIG-PORTAL I realized that we will have
to ask the same question about the upcoming Sungard contribution of i18n.

How can the uP3 development keep up if all the contributions are going
into 2.x? :-(

Adam

[Andrew Petro]

If the problem is getting uP3 to keep up, then we're doing it wrong. A
next-generation implementation doesn't have trouble keeping up. It's
revolutionarily far out ahead such that the previous version can't catch up
to it and it addresses the requirements and pain points so that people feel
naturally compelled to move forward. Attracted. Flock to it, even.

Tomcat 5.5.x doesn't have trouble keeping up with Tomcat 4.

So let's stop trying to get the development to keep up, and switch focus to
identifying and nailing the smallest set of requirements that will make
uPortal 3 deployable at all, confident that fundamental architectural
decisions (Spring, performance-on-steriods-via-caching, etc.) will make this
platform not be in the business of keeping up and instead in the business of
being far out ahead with incredible suppleness, performance, standards
support, etc.

We don't immediately need SunGard's i18n code contribution to uP3 (of
course, SunGard should feel completely welcome to be as involved as they can
and if they're looking to put that code here, let's have that conversation).
But immediately, we don't need or want that specific code. We do need and
want to identify what requirements are driving that code, and make very sure
that we've got an architecture here that will accommodate them.

What concretely do we expect ESUP-Portail to have to do to uP3 to make it
French? Bilingual English-French? Do these features actually work today?
I'd be very concerned about going after that and having an attractive story
for them.

And about getting this project to the point where it is deployable, such
that would-be deployers can plausibly contribute to it.

> How can the uP3 development keep up if all the contributions are going
into 2.x? :-(

How can we expect anyone to contribute to a product they cannot immediately
use in production?

Andrew

[jayshao]

> How can we expect anyone to contribute to a product they cannot immediately
> use in production?

+1