Re: Requirements?

Hi Pito,

Welcome to OSAF and Chandler!

You pose excellent questions. I'm the new product manager and am grappling
with exactly these issues. Given limited time and scarce resource, I think
your questions are best answered in a timeline fashion i.e. WHO do we serve
WHAT WHEN? I will be creating a product roadmap to start discussing and
addressing these issues.

Coming up to speed, I'm first pursuing things from a top down and bottoms up
approach so that we can hopefully arrive at the right framework.

From the top down, I see issues like:
- What are OSAF and Chandler's goals? What do we hope to achieve?
- Who are our target customers?
- What is our mass adoption strategy?

Being a non-profit and thus less bottom-line driven, I think we need to be
particularly clear, focused and aligned about the goals we want to achieve.
As a free (as in free beer) product, we lack price as a signal and feedback
mechanism. Hopefully, we can counterbalance this by active support from the
open source community through rapid releases and mass peer-review feedback.

Once we figure out a target customer, we need to figure out what the "whole
product" is for them. As Geoffrey Moore (www.chasmgroup.com) puts it, this
is a minimum set of requirements that will satisfy (for us, delight?) the
target customer. Bill Seitz has a nice section on customer scenarios and
positioning on his wiki (http://webseitz.fluxent.com/wiki/CrossingTheChasm)

Who we choose as target customers depends in part by our mass adoption
strategy (e.g. do we target power users first?) Andy also points out that
while we should focus on target customers, we should make sure we lay the
foundation for mainstream users and never detract from usability.

From the bottoms up, I'm trying to understand
- What are the major features and requirements from all users and how do we
classify them?
- What is our technical architecture? What are the constraints? What are
different architectures potentially good/bad at?
- What are the main components of the system and what are the dependencies
like?
- How is open source development different? How do they affect the timeline?

Do you think this is the right approach? Would appreciate your feedback.

You are definitely not late in the game. Hopefully, Andy has given a good
overview. The requirements are still in flux (e.g. Mitch has mentioned we
are rethinking about scalability for large organizations in
http://tinyurl.com/34nr).

chao


"Pito Salas" <[email protected]> wrote in message
news:<00c501c29614$51f26e70$c17ba8c0@RAVEN>...
> I've been following both this discussion and the dev one, and it seems
> to me that I'm missing a lot of context with which to make comments or
> recommendations or ask questions.
>
> For example:
>
> - Who and what do we expect a typical user to be? What level of computer
> experience? Working behind a firewall? Working in a company? A large
> one, a small one?
>
> - What volume of emails does this person work on daily? Hundreds,
> thousands, dozens of messages? Are they more personal or business? Who
> is he/she corresponding with? What platforms are they using?
>
> - How much information does she/he want to store in Chandler? Is it
> hundreds of meg, hundreds of gig? Any movies etc? How often is that
> information changed? Constantly? Or is it more write once, read often?
>
> - Does this person work on a small team of trusted colleagues, or is
> he/she part of numerous diverse teams, each of which they want to
> collaborate with? Or is this person mostly managing "personal"
> information with some exceptions?
>
> - What kind of connectivity do they have, primarily? Will it be typical
> or unusual that they are relying on a modem vs. a T1? What computer are
> they running on? Platform and capacity and capability?
>
> - How quickly do we want that person to be productive?  Is this to be a
> pick-up-and-use it kind of product or more something that you invest a
> little time to learn and get huge payback?
>
> - How important is interoperability with other email servers and
> clients? Do we expect Chandler to *replace* the user's current email
> client (e.g. Outlook) or supplement it?
>
> Those are some of the questions that are on my mind. It would seem to me
> that it is next to impossible to make design and architecture decisions
> without at least a first cut on questions like this. I have to assume
> that there is a thread somewhere and a document somewhere which addresss
> this.
>
> Thoughts?
>
> Pito
>
>
> _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
>
> Open Source Applications Foundation "Design" mailing list
> http://lists.osafoundation.org/mailman/listinfo/design
>

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _

Open Source Applications Foundation "Process" mailing list
http://lists.osafoundation.org/mailman/listinfo/process