struts-user mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Steve Jovanovic" <ste...@noumenaut.com>
Subject RE: Struts Developers Needed
Date Thu, 10 Jul 2003 20:56:47 GMT
Hi Joe,

Here's some more information that I hope will give you a better idea of
what we're up to (excerpted from a post).

Thanks!

Steve

Maybe you might find this helpful:

http://dev.noumenaut.com/project/Skribe/Documents/BluePrint.jpg?title=Ar
chitectural+overv%2E%2E%2E&comments=+2+comments+%3A

This is a very high-level diagram of our 5-tier architecture. Our
development approach has been "from the bottom up." We began by defining
a comprehensive DB schema and then went on to write our Data Access
Objects. We still have a few more to do there. But next up is the
business tier-pure business logic that honors well-defined business
interfaces that are documented in our extensive architectural documents.
And above that is the presentation tier, where Struts sits. Finally,
Matt (our webmaster) is doing the client tier (the rest of us, who are
graphically un-gifted :) , will come up with rudimentary pages, and he's
going to Mattify them and add all the good stuff like Flash, JavaScript,
and killer graphics and sounds). 

Do you see the different components in the business tier? Each component
has a team lead, and each team consists of one or two people, although
we're spread a little thin in some areas. Our approach to development is
to do the things that we know how to do, contain the things that we
don't know how to do with well-defined interfaces, and then iterate on
those things repeatedly until we nail them. As a development team, very
few of us have any appreciable Struts experience, but what I expect will
happen is very similar to what happened with our DAO work.

One of our developers named Andy, who had a great deal of experience
with DAO's, kicked us off with a gold standard implementation that
everyone else closely followed in implementing the other DAO's.
Similarly, I expect that Bill or another of our senior developers will
kick us off with Struts, and everyone else will plough through.
 
Another nice thing is that what we're building is pretty well-defined.
For example, here's one of our use case documents:

http://dev.noumenaut.com/project/Skribe/Documents/BluePrint.jpg?title=Ar
chitectural+overv%2E%2E%2E&comments=+2+comments+%3A

It's pretty elaborate. We invest a lot of up-front design work to do our
best to get things right early on, so that coding becomes much easier.
Our goal is to do such a thorough job at architecture that coding
becomes a manageable and even fun activity rather than an overwhelming
and complex chore. We have a dedicated architect named Ping Wang and she
is amazing! She's single-handedly written dozens of pages of design
documents to get us to where we are now. Apart from Skribe, itself, one
of the neatest things about Noumenaut is that we're always trying
creative experiments and we try to adopt the best practices that we can
find out there: from RUP to patterns and everything else we can wrap our
minds around. A lot of times we'll try something and it'll fail, but our
successes-like architecture-have been big ones. It's a lot of fun. :)

There are areas of weakness, too. For example, we've traditionally had a
difficult time with project management. It's difficult to coordinate a
complex project like Skribe among developers located as far away as
Ireland, Serbia, and Brazil. :) We've also had a very hard time with our
app server's stability and sometimes we just throw every person we've
got on a problem and hope to make a breakthrough, stopping everything
else in the meantime. And then there's the occasional need to shuffle
people around as they have other obligations that sap up their time. But
overall, we've kept moving forward steadily and I'm optimistic that the
Struts work in the presentation tier won't be as hard as it might seem,
even though there is quite a bit to do.

We have a lot of other design documents that I could point you to, but I
hope that all this will give you a better idea of how we're going about
our work. What we need isn't developers with an expert command of Struts
(although that certainly wouldn't hurt! :) ), but enthusiastic people
passionate about Java and the Skribe vision. 

If I can speak personally, for a moment, Skribe arose from my efforts to
replace a PHP-based journaling system that I'd written for my private
use, but it's grown into something much larger and exciting than I'd
ever imagined. I think anyone who puts www.livejournal.com through its
paces can see the opportunities that exist for a high-end portal for
empowering people to find and connect with each other, and share the
creative productions of their personalities. We're building something
that we really want to use, and we hope that a great number of other
people will feel the same way. The blogging market is showing explosive
growth, and we really see a tremendous opportunity on many levels, but I
think the most important one is to allow people to find each other and
connect in meaningful ways that are, at present, difficult and limited.
When you connect enough "nodes" together, all kinds of exciting things
happen. And when those nodes are human minds and hearts, interacting
through our software, it's no wonder why we're excited! :)


-----Original Message-----
From: Joe Germuska [mailto:Joe@Germuska.com] 
Sent: Thursday, July 10, 2003 3:45 PM
To: Struts Users Mailing List
Subject: Re: Struts Developers Needed

It seems like a kind of elaborate application process without even a 
chance to see the code or the  design.  The site is pretty and the 
project sounds interesting, but this doesn't strike me as a way to 
encourage a lot of developers to participate...

Joe




---------------------------------------------------------------------
To unsubscribe, e-mail: struts-user-unsubscribe@jakarta.apache.org
For additional commands, e-mail: struts-user-help@jakarta.apache.org


Mime
View raw message