db-torque-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From "Stephen Haberman" <steph...@chase3000.com>
Subject RE: [vote] plans for torque 3.1
Date Tue, 10 Dec 2002 00:54:23 GMT
> Do you think some of the template refactoring (proposed by Stephen
> Haberman I think) should take place in this release, or wait for 4.0?

I think template refactoring would be great; in doing so, I'll agree
with Jon that OJB is not the-end-all-killer-tool, however, I think they
have a much better (in terms of cross-database test cases, etc.)
implementation of persistence than Torque currently has.

Torque's strong point has always been code generation. I think it would
be great of Torque to focus on just that and offload all of the
persistence baggage to OJB where they have a dedicated team doing solely
that.

Assuming this happens, Torque could fall back on generating the nice om
layers it always has, except easier and with more reliable results. And
it also leads Torque towards doing more generation, e.g. of UI's (both
AWT/Swing/SWT-based and Turbine-based), as Martin was saying he'd really
like as a feature.

At that point it'd also be really awesome to explore the possibilities
of round trip engineering (e.g. of the om data layer and UI's). I'm a
bit naïve in knowing exactly how useful it would be, I just think it
sounds awesome (I ran a TogetherSoft trial a year or so ago and was
really impressed by what the round-trip engineering, if done correctly,
could do).

So, I guess in the long run I'd really like to see Torque mature into
pseudo-UML generation/reverse-generation framework. Kinda like ArgoUML,
but without the GUI, and ideally a real nice implementation and
integration between features like data layer and UI generation.

Dunno; with my current lurking status, my opinion is of little weight,
but I just thought I'd put in my two cents.

Thanks,
Stephen


Mime
View raw message