Ted Husted wrote:
On 3/21/06, [EMAIL PROTECTED] <[EMAIL PROTECTED]>
I, for one, would never recommend to any business enterprise that they
use Struts for important applications if the source was not vetted and
controlled by a small, trusted committee. Your needs may not have such
requirements for trustworthiness.
In the case of the Apache Software Foundation, we do take intellectual
property very seriously. Before receiving an account, each committer
must file with the ASF a "Contributor's License Agreement". In this
way, when we make a commit, we legally donate the code to the ASF,
which is a not-for-profit US corporation. It is the ASF's intention to
have clear title to all the code in our repository, both for its
benefit and for the benefit of the people who make use of ASF code. As
the sole owner of the code, the ASF can also afford the individuals on
the PMC some legal protection, since we act as agents of the ASF.
Well, yeah... blah blah.
Let's examine what this means in plain English. Correct if I'm wrong but
I think the above means that to become a committer on an ASF project,
you have to print off some document and you sign it and send it in by fax.
Well, okay, fine. I previously suggested that the requirements for
commiting code culd be that someone (a) has a name and (b) has expressed
interest in working on the project. I append to those conditions that
they (c) print off this thing and fax it in to ASF.
Does this substantially change anything? Does it bolster or undermine
any arguments made so far on this topic? Frankly, it seems like a big
red herring.
We do encourage non-committers to submit patches, and we take care to
credit each person's contribution in the repository log when we make
the commit. Depending on the nature of the contribution, we may ask
someone to file a CLA, even if he or she is not a committer.
Yeah, okay, so other people fax the thing in too. Fine.
Now, in general, in this message, you're just repeating the theory,
aren't you? But the problem is that many people in this conversation
seem to believe that the theory isn't really working in practice.
Moreover, the fact that Struts was unable to stay competitive with
Webwork even given the huge advantages you should have in terms of
attracting collaborators, this seems to suggest that your model did not
work very well.
Many of the best features in Struts came from people who, at the time,
were not committers. The Validator, for example, as well as Tiles.
Features like the DispatchAction, roles-based authentification,
declarative exception handling, among many others were contributed to
the project by non-committers.
Most recently, opt-in cancel handling came in as a patch from a
non-committer, after a lengthy discussion of the best way to solve the
problem. Many ideas went into the patch, contributed by committers and
non-committer alike.
* http://issues.apache.org/bugzilla/show_bug.cgi?id=38374
For more on contributiing to the project, see
* http://struts.apache.org/helping.html
Sadly, there are occasions when we cannot offter committer status to
an individual. Usually, it is not because there is a problem with hsi
or her code,
I don't get it. If somebody can donate worthwhile code, why shouldn't
they be allowed to commit it?
but because all committers participate in the
decision-making process. We don't have any peon-committers.
I don't really understand what you're talking about here, Ted. I can't
refute it because I just don't understand it. I think I understood your
first paragraph. I read that 3 times or so and my analysis of it boiled
down to the fact that commmitters have to sign some legal boilerplate
and send it in. AFAICS, that's practically a non sequitir; it's just a
legal/procedural detail that isn't very relevant to the real issues in
running a project, but I processed what you were saying, I think. This
stuff about peon-committers, I just don't understand.
Every
committer is considered on track to become a PMC member, with a
binding vote on releases and other matters. In turn, some committers
and PMC members also become ASF members. The ASF members are the
"stakeholders" of the corporation and elect the board of directors.
So you are saying that the above means that somebody wants to hack the
code, is able to hack the code, and can contribute, but because of the
above, they can't become a committer.
I really think you should expound the various logical steps of your
argument more clearly. You know, it may seem clear in your own mind, but
of course, your goal must be to convey your message to others. I can't
understand your argument, and I'd say it's safe to say that if I don't
understand it, other people probably don't either. Of course, only very
few people who don't understand it have the self-confidence to say so
forthrightly as I just did. You know, it's like the emperor's new
clothes fable.
While ASF projects have a reputation for voting, most decisions are
made through informal discussions on the dev mailing list. Someone
commits some code, and the rest of us peer-review the change (by
following the commit list). Usually, that's the end of it, but any PMC
member can veto a product change if need be. It's rare that a PMC
member will abuse his or her veto power, but it does happen. On one
occasion, the board did have to strip an individual's commit
privileges. But, given that there are almost two thousand (2,000) ASF
committers now, working on more than thirty top-level projects, that
seems like a pretty good batting average :)
Oh, so there are actually more committers around than you can shake a
stick at, eh? All working away. Then why IYO did progress on Struts 1.x
completely stagnate?
We also take project management seriously.
Yeah, imagine how things would be if you didn't take things seriously....
(There maybe a problem of taking yourselves a tad too seriously, but
that is another matter, of course... ;-))
Every project has a
designated "Chair" who is a vice-president of the foundation. The
Chair/VP must tender a status report to the board on at least a
quarterly basis, to be sure the project remains vital and
collaborative.
For more about how the ASF (and by extension Apache Struts) works, see
* http://www.apache.org/foundation/how-it-works.html
You keep pointing people to this the way a devout muslim would point
somebody to the holy Koran. I can't read your mind. Which particular
verse(s) of the scripture you cite are you referring to?
Regards,
Jonathan Revusky
--
lead developer, FreeMarker project, http://freemarker.org/
FreeMarker group blog, http://freemarker.blogspot.com/
-Ted.
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
