[ale] [OT] AJC Article on State Welfare System

[Jeff Hubbs]

On Mon, 2002-08-26 at 13:43, Stuffed Crust wrote: 

> It is my understanding that the reason we have four failed attempts and
> $30 million down the tubes is because the state can't get the
> requirements sorted out long enough to actually design a system around
> it.

So what would make a fifth attempt any different?  If "sorting out the
requirements" is intractable, then no progress will be made if you
insist that requirements must be sorted out before actually creating
anything.   

Now, earlier, you curtly dismissed my Web-app/green-screen musings as
premature while appearing to want to focus on requirements.  But, the
State is unlikely to perform any better at cooperating with
requirements-gathering on the fifth go-round.  

When I was talking about the Web-app/green-screen issue, I was talking
about a lot more than just UI.  I was talking about what kind of
components (Web server vs. app session server), connectivity (Internet
vs. modem server), and what kind of expertise might be required (Web
coder or "Zopatista" vs. language-of-choice-plus-ncurses, for example). 

If the goal is to create what XPers call a "spike solution," then what
you did along the lines of inventing requirements might be all there is
to go on.  A spike solution that fully and completely addresses some
arbitrarily-drawn requirements is better than having no solution at all
because you're awaiting requirements that never come.   

So, again, by way of a spike solution, all I was trying to do is discuss
some very coarse characteristics based on some suppositions I was making
about how this software entity might be accessed.  I have not gone in
and studied the situation.  However, no real study is necessary to
arrive at the blind-man-encounters-an-elephant conclusion that you need
a database app.  I don't think it's premature to wonder about how the
app will present itself to users, which naturally leads to questions
like 1) by what means are the users *able* to access such an app?  2)
*Who* are the users anyway?  

Yes, you are right in that actually coding the UI will require some
requirements-type decision-making on someone's part.  But, that isn't at
all tightly coupled to the issue of what KIND of UI it is; I was talking
Web vs. green-screen but even at that, those aren't the only two choices
- it's just the two that are the least likely to drive client platform
dependence IMHO.  All kinds of platforms are likely to be able to run a
term emulator or a browser.  

- Jeff


---
This message has been sent through the ALE general discussion list.
See http://www.ale.org/mailing-lists.shtml for more info. Problems should be 
sent to listmaster at ale dot org.