gump-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Leo Simons <>
Subject Re: Is Gump the obsticle?
Date Sun, 19 Feb 2006 10:59:35 GMT
On Sun, Feb 19, 2006 at 03:32:28PM +1030, Steve wrote:
> Incompatible version changes arise from:
>    (a) development errors
>    (b) planned version change typically flagged
>        by a major version identifier increment
> The first is something we want to trap while the second is something managed
> and intentional. Gump does a good job of identifying and reporting errors
> related to (a) but it is weak with respect to supporting managed change
> arising from (b).

Agreed. I think in general most of the software landscape is not so good at
this and most developers don't spend as much effort on it as would be a Good
Thing(tm) (probably because it isn't exactly easy).
> > It might be a nice feature to be able to create a 
> > packaged-foobar just from the metadata, without having to get 
> > Stefen involved with downloading the jars himself.  Maybe 
> > something for Gump 3.
> Yep - good suggestion! 

Heh. I would love to get back to working on this. Python is *so* much more
fun than java!

I wonder though whether the "true" place to work on this isn't "further down
the stack" (like in ant, make, scons, maven, ...). Gump is very focussed (or
the way we use it at least) on "latest of everything" and this focus helps
keep things somewhat understable (or so we hope). Its a useful focus. I'm
afraid of changing it (e.g. I think Stefan did the right thing by switching
to 1.5, we focus on whatever makes us build the largest part of the tree).


To unsubscribe, e-mail:
For additional commands, e-mail:

View raw message