gump-general mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Leo Simons <>
Subject Re: Gump3 or Gump 2.1
Date Tue, 15 Mar 2005 10:57:38 GMT
Hi Adam!

Good to hear from you.

On 14-03-2005 15:01, "Adam  Jack" <> wrote:
> I've been pretty flaky these last months; not able to commit the time/effort
> I'd enjoy committing. Basically, of the three days a week I can work (the
> others I grow with my daughter) I've been working for a start-up that
> consumes me. It is fun to build something from scratch, but I need to let
> things mature/settle now that the live customers using it. So, maybe (this
> attempt) I can eke out a little time.
> So, I keep coming back to this ... what Gump can I make increment
> improvements on, given that small amount of time. Could I tweak Gump2 to a
> Gump 2.1 that is closer to what we want, or is Gump3 really the starting
> point. Just 'cos there is a bunch of code in Gump2 doesn't mean Gump3
> ougtn't be the primary move, it is just very daunting to start from scratch
> (again).

What would you enjoy most? :-D

/I/ would love to see you helping out with Gump3. I'm making some slow
progress. Doing things componentized with big barriers between components,
and unit tested (working my way up to test-first), is a lot of work. Going
on my own we're months away from something usable. I would love to engage in
discussion about how to componentize Gump2 into many small utilities and
plugins. I'm still a few weeks away from having most of the boring "kinks"
worked out so that I can lower the barrier for others to contribute, but I
doubt they'd get in your way.

The current live/trunk gumps really are suffering a little from less tender
love and care from you. There's a lot of work that really needs to be done
(like making sure we can use one CVS|SVN master tree and properly testing
all that) if at all possible. And Stefan could no doubt use some help
polishing the configure|make integration?

I wouldn't worry too much about duplication of effort. I find that with
python, writing the code isn't the problem, its figuring out how to write it
(like how to properly and portably run programs from the shell), and that
kind of stuff isn't duplicated.

> I know that a Gump 2.1 could be valuable, it really isn't very far from
> where we are to populating the database (as Gump3 wishes to do) and maybe it
> is a valuable contribution, for comparing Gump3 to. That said, I can't write
> the Cocoon-based DB presentation layer, so w/o some good will/interest from
> others that could be a pointless venture.

Even not talking about the actual programming, I don't know yet what the
presentation of our data /should/ look like, so I have some doubts about
spending a lot of time conforming to a data model of which we are currently
not quite sure it offers exactly what we need. I for one need some more
"hard" requirements in the form of, O, I dunno, GUI mockups.

> So, I'm looking to see what other's thoughts are, to see if I could tie in.
> Thanks in advance.

There's a whole lot of stuff you could do. Just make sure to have fun :-D



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

View raw message