Ted Husted wrote:
On 4/18/06, Phil Zoio <[EMAIL PROTECTED]> wrote:
I'd be happy to pass Strecks on to Struts itself if the community really
wanted it, but I don't see that as essential to its existence in any way.
What we look for is a community forming around the codebase. We don't
just want code, we want people who are committed to maintaining the
code over the long term.
Yes, but if that is the case, doesn't that mean that you have failed
utterly and abysmally? After all, we have a situation in which the
Struts 1.x codebase is basically being abandoned and you are bringing in
a codebase that was, heretofore, that of a competing project.
So what is this about your "commitment to maintaining the code over the
long term"?
Not just one person, but a community of
"likely suspects" who will step and and volunteer if the creator of
the code loses interest.
Doesn't all this beg the question completely? If you and your
collaborators are not interested in developing the Struts 1.x codebase
any further, and Phil and maybe others are, why should you not just let
them come in and do what they will?
I mean, as regards Struts 1.x, you are not proposing to do anything with
it really, right? If other people have a plan for modernizing it and
making it better, why should you not just open the door and say: "Okay,
show us what you can do."
What is there to lose? AFAICS, the only alternative you are proposing
wrt Struts 1.x is simply to abandon it.
Many extensions that people now consider essential, like Tiles and
Validator, started as third-party extensions. Whether we want to add
an extension to the core usually depends on how many people recommend
the extension on the list -AND- whether the people developing the
extension follow the user list.
This is as true for committers as it is for non-committers. Hubert,
Niall, James Holmes, James Mitchell, Laurie Harper, Don Brown, myself,
and others, all have extensions to Struts that we have distributed
separately. Some of these extension were ultimately added to the
distribution, and others have not been. In the latter case, it's
usually because the committer chooses not to donate the extension. The
Struts Action distribution is already quite large, and we are all
sensitve to "code bloat
Well, whatever... but the set of policies you have followed wrt to the
ongoing development of Struts 1.x have to be considered an abject
failure. You have had to accept that this is something that is now
technically obsolete, and rather than trying to modernize/refactor it
forward, you have convinced the developers of a competing project to
donate their code, so that you have something more state of the art to
offer and to work on.
There is a recurring theme in your discourse, Ted, where you keep
saying: "We have been doing this, that, and the other thing..." and a
casual observer would think that you believe that everything you have
been doing has been wonderfully successful.
Do you really believe that?
It really makes me wonder whether you view certain things through
special rose-colored glasses.
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
".
In the case of Strecks, a similar case would the "EL" tags. At first,
it was a separate subproject, because EL was based on Java 1.3 when we
had set Java 1.2 as the baseline. For Struts Action 1.3, we moved the
baseline, and so EL is now integrated into the JSP Tags. The same
thing could happen here, if a community grows around the codebase, and
the community wanted to be an Apache Struts subproject. (Not everyone
does!)
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
