> Well, Indiana is intended to be a product - an OpenSolaris distribution with a
> regular release schedule.
>
>
So, I'd expect to see some documents from you that describe this
product. Have you a 1-pager that describes the basics?

Project Description
Risks and Assumptions
Business Summary
    Problem Area
    Market/Requester
    Business Justification
    Competitive Analysis
    Opportunity Window/Exposure
    How will you know when you are done?
Technical Description
    Details
    Bug/RFE Number(s)
    In Scope
    Out of Scope
    Doc Impact
    Admin/Config impact
    HA Impact
    I18N/L10N impact
    Packaging & Delivery
    Security Impact
    Dependencies
Resources & Schedule
Prototype Availability

Following review and approval of the 1-pager, I would expect to see a
Requirements Document that describes all the features/functionality that
we will include in the product, so we can agree on expectations and
begin work. What is the deadline for producing drafts of these two basic
documents for Indiana?

Ok, I think I see where the vision is going to hit a wall. There is
definitely a process disconnect between agile development and Solaris
development.

In agile development, you begin with a list of simple requirements
without a detailed description of how you are going to get there. You
then rush straight to implementation. After implementation, you pass
back to customer for feedback at the soonest possible instance. You
then rush again for a period of time, (say 2 weeks) and kick out what
you have to the customer. Keep repeating indefinitely.

I really feel that if Indiana is to be what I think it should be, that
it needs to be implemented in an agile fashion. Let's just get
together a few like minded people, put together a short proposal for
autonomy and run with the ball. (No ARC, no OGB after the initial
project is OKed to fit under the OpenSOlaris.org umbrella).

To a head start, I propose that we take Nexenta and make it the
standard base for the Indiana/Linuxy Solaris. (OpenSolaris Community
Edition)

For the more classic OpenSolaris reference build (fully open source
Nevada), I think we should start with Benenix. This will replace SXCE
as the standard OpenSolaris development platform. (Solaris
customers/testers will still use SX) This will be called OpenSolaris
Enterprise Edition. This will be not be implemented using agile
methodology.

-brian
_______________________________________________
opensolaris-discuss mailing list
opensolaris-discuss@opensolaris.org

Reply via email to