RE: Re: Sticking point

Hi, Dale, J.B.,

> -----Original Message-----
> From: J. B. Rainsberger [mailto:[email protected]]
> Sent: Wednesday, December 11, 2002 3:42 PM
> To: [email protected]
> Subject: Re: [XC] Re: Sticking point
<snip>
>
> All right. Let's step back from the shippable CD, then. I'll take a few steps
> back and see where the sweet spot is. Go down the list until you find the
> first thing you think provides business value every iteration.
>
> 1. A physical CD with working features to date.
> 2. An install image of the working features to date, available on an internal network.
> 3. An install image of the working features to date sitting on someone's machine.
> 4. A version of the working features to date installed on someone's machine.
> 5. A built, but unpackaged version of the working features to date.
> 6. A "tagged version" of the source of the working features to date.

I've been thinking a lot about this, since it was posted.

Here's my reason for thinking #1 is the best solution.

In even the saner jobs that I've had, the one thing that really messes us developers up
is demos.  Frequently the demo is requested at a short time span, which would be *within*
an iteration. So they're requested the next day, or two days.

And invariably there's some cool feature that the Requestor (Marketing, Manager, whoever)
wants that they ask to be included.

So everything stops (which is frustrating) while the team does a quick-and-dirty
implementation of "Cool Feature."  In an XP shop, that wouldn't break previous
functionality.  In my experience with non-XP shops, it always breaks *something*, and
that something usually breaks at the demo.  So even if the cool functionality works, the
rest of the demo makes everybody look bad...

So, having a physical CD at the end of the iteration means:

1.  Anybody can do a demo with the last CD.

2.  Demos are always the latest *tested* and robust work.  Less chance for failure, less
opportunities for excuses.

3.  No "demo" interruptions to the Flow of the iteration.  XPers have a pretty good idea
of what they are going to do each iteration...  and knowing that they will not be
interrupted and sent in a different direction gives them more confidence...

4.  No quibbling over adding last minute "cool features."  Everything implemented and
tested is on the CD.  Next "cool feature" in the next iteration.

5.  Practice making the CD.  More people have hands on experience with it... so if the
person who does the CD isn't there, somebody else can do it.

6.  The CD, Install test, all that stuff is factored into the time estimates.  We always
have to do it, so it's always in the estimates.

7.  The up front time for preparing for a demo is reduced to the time it takes to pick up
and install the CD. So that guy who's only in town for today can request a demo at 4 and
get it at, say, 4:30.

Don't know about you guys, but that would totally blow away anyone I've ever done a demo
for...

X: Can we get a demo soon?
Me: Sure, 20 minutes... is that soon enough?
X:  How many weeks ago was this the latest version..
Me: Oh, last friday.

<grin>

Kay












------------------------ Yahoo! Groups Sponsor ---------------------~-->
Get 128 Bit SSL Encryption!
http://us.click.yahoo.com/CBxunD/vN2EAA/xGHJAA/NhFolB/TM
---------------------------------------------------------------------~->

To unsubscribe from this group, send an email to:
extremecustomering-unsubscribe-hHKSG33TihhbjbujkaE4pw@public.gmane.org

 

Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/