Access vs. Visual Basic

[Thomas Kroljic]

WOW!!!

Per Thomas Kroljic:

Two separate issues/two separate judgments: front end and back end.
---------------------------------------------------------------------------
Back End: whatever floats your boat - especially in the light of mission
criticality and the judgment of IT.


Front End: If you have a record of the man hours you put into the current
version, I'd multiply them by a minimum of 3 to estimate a version using
VB6.

That's my experience from doing two apps that way in VB6. Others have
opined a
factor of five.

For one fairly large app that I wrote, and which IT converted to an Oracle
back
end and a DotNet front end - while adding maybe 40% new functionality, the
factor was a little over 100. ($225,000 vs 23 Mil)

 

[David W. Fenton]

<<host of things that are completely broken in A2K7>>

What major items are you talking about?

Allen Browne is very thoughtfully keeping an article on the subject
up-to-date on his website:

http://allenbrowne.com/Access2007.html

He doesn't mention the issue with replication (i.e., DAO direct
synchs that work just fine and dandy in A2K3 and all previous
versions fail in A2K7 and MS acknowledges it and says you have to
switch to JRO, even though they've been deprecating ADO and its
stepchildren in favor of DAO for Jet data for a couple of years now;
a hotfix is being contemplated), though that's only relevant to
those who are using replication in their apps already, I guess.

 

[(PeteCresswell)]

Per Thomas Kroljic:

WOW!!!

If you're reacting to the $225,000 vs 23 Mil; it may not be quite as bad as it
sounds.

I developed the original app as a sort of one-man gang: no red tape, nobody
telling me what to do, no rules, no regulations - just gallons of coffee and
some really smart and demanding-but-reasonable users. A purely tactical effort.

The replacement was a big IT project in a company that gone far, far, far down
the strategic path: hugely-detailed specs, thousands upon thousands of pages of
requirements, test plans, documentation, formal acceptance tests, signoffs, and
so-forth.

At the times I saw numbers, the team was around 50 people - but over the 4 or so
years the project ran people kept coming and going so there was substantial
slippage in getting newbies up to speed and it seemed to me like the team was
top-heavy: too many coordinators, project managers, and supervisors for my own
taste.

Also, now that I think of it, who am I to say that the new system only had 40%
more functionality? I'm probably biased...

But I keep coming back to two observations:
--------------------------------------------------------------------------------
1) The new system definitely doesn't have 100 times the functionality, nor does
it have fifty... or twenty-five.... In fact I can't see *anybody* claiming
more than ten.

2) The users didn't want the new system. They were happy with the old one right
up to cutover day and they're still not what I'd call ecstatic over the new
one.
--------------------------------------------------------------------------------

Some people in some big IT departments tend to knock guys like me - and I have
to say that in my case they have some legitimate complaints when users say "Why
can't you deliver as fast as that guy?"

But all-in-all, that particular experience made me feel pretty good about my
economic worth as a freelance tactical developer.