Oracle FAQ | Your Portal to the Oracle Knowledge Grid |
Home -> Community -> Usenet -> c.d.o.server -> Re: Java to die in 2003
26 Dec 2002 09:14:09 -0600, Galen Boyer said (and I quote):
>
> Not sure what's so incredibly complex about J2EE. The complexity
Me neither. But here is an example.
Two years ago I was given the task of getting a database design going for
a pilot training system. Very similar to the classic "course-student-
results" application.
The J2EE mob spent 6 months discussing how they would "interface" to the database, to the servlets, if they would use Javascript on the front-end, etcetc. It seemed like even the most basic tasks were an insurmountable cacophony of "interfaces", "data-access-objects", instantiations, blah- -blah...
In the meantime, I came up with the design for the database, together with my design pal. We put together an API based on PL/SQL that basically provided object-relational mapping directly off Java objects in their DAOs. Objects, not classes.
Big bunfight with the J2EE morons, because we were "using PL/SQL and stored procedures and their book says that is *evil*"! All sorts of deranged, insane arguments ensued.
When it came time 6 months ago to design version 2 of the stuff, with much more functionality and extended code, we came up exactly with the same data design. Plus a series of ancillary tables/entities/relationships to provide the extra bits. Nothing major. Same API, extended of course.
Their solution? Re-code the ENTIRE J2EE code effort! The LOT!
Their explanation? They had not had access to the "big picture" of the
whole thing.
Never mind we didn't either...
Guess which methodology is now the standard for future development? Thank God someone with two brain cells listened...
-- Cheers Nuno Souto nsouto_at_optusnet.com.au.nospamReceived on Thu Dec 26 2002 - 23:01:28 CST