In the vein of if one person has a question, a hundred are afraid to ask
it.....
I've been reading and coding and lurking on things CF and FB related for
awhile and reading over everything I can find. I've gone over the FuseDocs
and XFB etc from Hal and various other additions and while I've used my own
very watered down version of FB I don't have the sheer glut of projects to
get as proficient as many of you.
I'm almost always the lone coder/project manager/architect/janitor on my
projects and anything I can do to make it easier on myself and those that
come after me is a Good Thing *TM*. Being able to spec out a project, write
the fusedocs, and then magically generate the CF from the docs seems like
Xanadu - but I've yet to see a clear cut demonstration of this. I just hear
that's how you do it.
So after all that, the question. Does anyone have an actual project plan
with accompanying FuseDocs and maybe even code for review? I've seen the
examples on Hal's site, and read the FB book but nothing seems to give the
complete picture at least to my eyes.
I'm not even sure this is what I need but it seems the closest to expressing
it in my journey to groking all of this. Unless of course Hal is willing to
take one for the team and we can turn him into stew :)
Regards,
Scott Whittaker
Recnetly Allaire Certified Developer (was anybody elses test nothing but
custom tag questions?)
----- Original Message -----
From: "Steve Nelson" <[EMAIL PROTECTED]>
To: "Fusebox" <[EMAIL PROTECTED]>
Sent: Thursday, April 05, 2001 6:10 PM
Subject: Re: Fwd: extreme programming
> It will be iterative.
>
> The idea is that you spend X days between the architect, client, end
> users building the specification using this tool (no coding yet). Then
> the architect generates the fusedocs and the fusecoders spend only a
> couple days writing the code for all the fusedocs. While writing the
> code, the fusecoders will make notes to the architect as to where
> problems were encountered or where better solutions have appeared after
> having written the code. Then the completed Fuses would be sent back to
> the central server for testing and deploying, and the entire process
> would start over. It'll make sense when you see this thing.
>
> Whether you use 100 developers or your current team, my goal is to have
> a 10 fold improvement in software productivity. Forget 6 months to 2
> months, I want to go from 1 month to 3 days, possibly even less.
>
> It's going to be tough to make it work right, but I think we've built a
> community that could accomplish it.
>
> Steve
>
> Patrick McElhaney wrote:
> >
> > I don't know, Steve. Based on my understanding of XP, I don't think
> > it would really work that way. Because XP is so iterative, the
> > spec and the code are developed almost simultaneously.
> >
> > You don't want to spend an 1600 man-hours building something, test
> > it two days later, and realize it doesn't work. You also don't want
> > 100 different people coding 100 different ways, some of them
> > much more efficient than others. That's where patterns, refactoring,
> > and moving people around come in.
> >
> > I'm not saying it's a bad idea. I just think what you're talking
> > about sounds more like waterfall than XP.
> >
> > Patrick
> >
> > > -----Original Message-----
> > > From: Steve Nelson [mailto:[EMAIL PROTECTED]]
> > > Sent: Thursday, April 05, 2001 1:32 PM
> > > To: Fusebox
> > > Subject: Re: Fwd: extreme programming
> > >
> > >
> > > Well my thinking is that the the 2 days would not actually begin until
> > > the specification was complete.
> > >
> > > Just like when building a house, you don't hire the contractor to
build
> > > the house until the architect is finished drawing it.
> > >
> > > You'd be surprised at how fast you could build an application when it
is
> > > down to that level of detail.
> > >
> > > Steve
> > >
> > > Todd Ashworth wrote:
> > > >
> > > > I don't know .. looking at this project, I can't easily
> > > envision it being
> > > > done in 1 to 2 days (it's rather ambitious). With 100 programmers
and
> > > > database people .. maybe 2 weeks. I wish I had that many to
> > > try it. What a
> > > > trip that would be ;)
> > > >
> > > > Todd Ashworth --
> > > >
> > > > ----- Original Message -----
> > > > From: "Steve Nelson" <[EMAIL PROTECTED]>
> > > > To: "Fusebox" <[EMAIL PROTECTED]>
> > > > Sent: Wednesday, April 04, 2001 11:16 PM
> > > > Subject: Re: Fwd: extreme programming
> > > >
> > > > | i want to take your 2 month development time down to 1-2 days
> > > (after the
> > > > | specification is complete). This would be done by having anywhere
from
> > > > | 1-100 Fusecoders on the project. According to the mythical man
month,
> > > > | this isn't possible because adding manpower to a project
> > > won't speed up
> > > > | development. Common sense tells me that it's because no has
> > > figured out
> > > > | how to linearly manage people.
> > > > |
> > > > | Think about this... If an olympic sprinter could run their
sprinting
> > > > | speed for a mile the world record would be under 3 minutes.
> > > > |
> > > > | The reason this hasn't been done is that the human body can't pump
> > > > | enough blood and suck in enough air. BUT! If the human body was
> > > > | redesigned to solve those problems, the 3 minute mile could
> > > be possible.
> > > > |
> > > > | I'm about halfway through this book for the second time and I'm
> > > > | realizing that almost all of these problems Mr. Brooks has come
across
> > > > | could potentially be solved with a handful of very
> > > sophisticated project
> > > > | management tools and a standardized methodology which we have
already
> > > > | developed.
> > > > |
> > > > | Steve
> > > >
> > > >
> >
>
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Structure your ColdFusion code with Fusebox. Get the official book at
http://www.fusionauthority.com/bkinfo.cfm
Archives: http://www.mail-archive.com/[email protected]/
Unsubscribe: http://www.houseoffusion.com/index.cfm?sidebar=lists
