Re: People, roles, and traits
- --- In email@example.com, "Andrew Lawrence"
> --- In firstname.lastname@example.org, "Hugh R. Beyer"<beyer@i...>
> > But how do you prevent what I've seen many times--which is where
> > the "foundation" group decides to build the best and most
> > bullet-proof architecture there ever was, and spends months
> > on it, overrunning deadline after deadline, producing baselevelsnew
> > which are unusable for one reason or another (often performance)
> > until finally some other group does something quick to get their
> > own project out the door--at which point they are declared the
> > "tactical" foundation and the architecure group is cancelled?agreed -- this is the place where finacial or other resource
> Very interesting... come to think of it, the "other group" you
> mention is the exact situation I have been placed in, not once,
> but twice within the same company!
> I guess I should clarify. The projects I have been involved with
> have never started out with a separation of application and content
> teams. Initially they are always the same group, and usually only
> a few people. The group has to build something to meet a deadline.
> But I have always tried, while within that group, to be able to
> wear the two different hats. Sometimes you can separate the
> application and content as you are going, other times you just have
> to keep it in mind and then come back to it later. This is where
> agile methods come into play. Iteration planning and a good view
> of priorities (what has to be done, and what can drop off) are
> The result, is that after the initial deadline has passed, you are
> left with code that is very flexible for the next round -- which
> there always seems to be another round. Alistair Cockburn's book
> on Agile Software Development talks about software as a game -
> where a project usually has two goals: 1) deliver the software,
> and 2) create an advantageous position for the next game.
> So in essence, I don't think an intial stab at "let's take 3-6
> months to build a good foundation" ever works. You have to build
> an application front to back to meet a deadline. But you have to
> keep in mind the separation of "application" and "content" since
> it will make future life that much easier.
> -- Andrew
constraints usually drive what is and is not a good practice. Cecilia