I'd *really* appreciate feedback on these thoughts. If you only answer one
mail from me, please make it this one.
Couldn't get my head to answer you till now, sorry for the delay. It's a pity that your Gump itch does not come at a time when I itch too ;-)
The topics are:...
1) Sanity check on what I'm doing w/ Gump (or hoping to) 2) Internal Community 3) External community
*Python Gump (as a future replacement for traditional Gump)
The way to go. +1
*Python Gump using Forrest
Not technically necessary but I applaud the effort +1
*Gump Statistics (as a tool for automatically communicating reuse/reliability/robustness, etc.)
Simply statistics, what they communicate is up to the reader ;-) +1
*Gump and Ruper (using Ruper to download packages)
+1 Important for easy installation. I'd add that Gump *must* install *hassle free*.
*Distributed Gump Trees (also using Ruper to download upstream Gump outputs)
Doh, you lost me here.
*Gump Documenter (as a tool for communicating the OSS map, and promoting)
Interesting.
*Promoting more gump projects in public Gump (do resources exist, will the 7+++ hours grate on Stefan?)
+1 as long as there are tools that help maintainers get the things fixed fast.
*Promoting more public/private/personal Gumps.
Private Gumps especially.
*Promoting more OSS Gumps (e.g. on SF.net. on Java.net, etc.)
COuld be interesting :-)
Questions: Anybody got any other itches they'd like to see scratched by Gump? I'm not volunteering, per se, just curious.
Integration with Eclipse.
Second, internal community:
... > Is there a way we can make a concerted effort to
promote Gump, and request developers? Thoughts?
The same thing I've told you for the Krysalis projects: release.
We gotta make something that all developers can use. Gump *is* incredible value, but we *must* release and make others use it.
I'll try to "install" it as a novice user and tell you where I get stuck, so we can work out the problems. As soon as I can install it without resorting to the manual more than once, we can so a milestone release, and from then on the ride starts.
Third, external community:
External community is key, and I see Gump as closed to those 300 or so modules it supports. Ought we propose a Gump on SF.net, or on java.net, or wherever? I think Gump's strengths are best felt with a large profile, so much as the GUI Gump is cool, I think big public Gumps are invaluable. Has anybody suggested a Gump on SF.net to SF.net, to others? Now that Gump is far easier to install (than when Andy Oliver tried) it really ought be possible for folks to have their own. Is this a good thing for future Gump?
I see Gump, I mean "The" Gump, as the Apache runs. Then I see developers using it on their machines to run the projects they have in the workspace.
Scenario: I have projects from CVS in a directory. I have a jar repository location. I download Gump, tell him where my workspace is, and it autoconfigures itself. Then I am able to use it fully to compile projects with dependencies resolved, and have it output reports.
When we get here, we have *the* cross-project buildtool.
What do folks think about Gump's future, about what it ought be trying to be, about who it ought be targeted at?
- Stefan and Sam (not joking, they are the power-integration users) - every developer
Thanks in advance for all responses.
Thank you for working on it. You really impressed me (again). :-)
--
Nicola Ken Barozzi [EMAIL PROTECTED]
- verba volant, scripta manent -
(discussions get forgotten, just code remains)
------------------------------------------------------------------------------------------------------------------------------------------ To unsubscribe, e-mail: [EMAIL PROTECTED] For additional commands, e-mail: [EMAIL PROTECTED]
