Mailing-List: contact dev-help@struts.apache.org; run by ezmlm
Precedence: bulk
Reply-To: "Struts Developers List" <dev@struts.apache.org>
Received-SPF: pass (asf.osuosl.org: local policy)
Message-ID: 
 <15409.170.201.180.137.1145991189.squirrel@webmail.chiron.lunarpages.com>
In-Reply-To: <444E68F4.80908@progeeks.com>
References: <444D9319.1070702@omnytex.com>   
    <f8b39ace0604242136x1b5cf6b7k7bd409fd61ed0aa8@mail.gmail.com>
    <36159.170.201.180.137.1145977619.squirrel@webmail.chiron.lunarpages.com>
    <444E68F4.80908@progeeks.com>
Date: Tue, 25 Apr 2006 14:53:09 -0400 (EDT)
Subject: Re: Proposal for change
From: "Frank W. Zammetti" <fzlists@omnytex.com>
To: "Struts Developers List" <dev@struts.apache.org>
Cc: "Struts Developers List" <dev@struts.apache.org>
User-Agent: SquirrelMail/1.4.4
MIME-Version: 1.0
Content-Type: text/plain;charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Importance: Normal

On Tue, April 25, 2006 2:22 pm, Paul Speed said:
>
>
> Frank W. Zammetti wrote:
>
>>
>> You are of course right about this.  But, much like taking the ideas
>> about
>> inventory control and order processing and such from Dell and starting
>> your own business is possible, the likelihood that you would get
>> anything
>> but a small fraction of the attention and business that Dell gets is
>> slim
>> to none.
>
> Not to sidle in where I don't really belong but perhaps this last
> sentence exemplifies the disconnect with "getting it"?  If one wanted to
> take the code from an apache project and do something else with it then
> all they care about is the something else they want to do.  It isn't
> really a "business"... the code exists for the code's sake.

You aren't chiming in where you don't belong... if your interested, you
belong, at least as far as I'm concerned :)

I think there is definitely something to your point, and the analogy may
have been a bit flawed.  However...

I don't think it is accurate to think that ego doesn't play a part in just
about everything that just about everyone does.  We all want to see our
work benefit others.  For most of us I believe its because we genuinely
like the feeling we get when someone writes us and says "hey, your code
really helped me, thank you!".  I know speaking for myself, it makes my
day when I get those eMails!  Part of it is simply the ego stroke of
someone essentially saying your work is worth something, but I don't
believe that is the big factor for most people.  I know it isn't for me,
and I don't think it is for the Struts team.  I think the thank you note
means as much to them as it does me.

If you agree with that, then the idea of forking the code and doing it
with the belief that you aren't going to reach a wide audience because the
Apache version continues to be what people go to, is not appealing.  In
that regard, if we substitute ego for money in the analogy, I think it
still works (although just saying ego is dangerous because as I tried to
illustrate above, I think there is good ego and bad ego).

> I'm not a committer but I've been following this list and the tomcat dev
> list since the last millennium... I think before there even was a struts
> 1.0.  I can't speak in an official capacity, I can't even pretend, but
> here is my take on the "apache way".

Isn't kind of interesting that there can be more than one "take" on it
though?

> For an open source project to exist you need code.  All of apache
> projects seem to exist to benefit the code... and by extension the
> documentation.  Though, even without documentation you still have the
> code.  All of the other stuff is extraneous or the life support system
> depending on how you look at it.  I think most of the "apache way" is
> partially considering it to be extraneous... in a "if the code goes sour
> and you have nothing" sort of way.  It's definitely symbiotic but
> without the code, you have nothing.  You might as well be chatting on
> myspace.com.

Hehe, considering some of the recent threads around here, posting on
myspace.com might actually be safer! :-) LOL

> So, the only reason to be a committer is to contribute to the
> codebase... and all other committers have to live with each other.  The
> only reason to be able to cast a binding vote is if you have a stake in
> the code... ie: are a committer.

This is where I'm not sure I agree... why can you only have a stake in the
code, or in the community even, if you are a committer?  And certainly the
"community" is often touted as the most important part of any ASF
project... it's just that "community" in that context means the committers
only, which is where I disagree with the Apache Way I guess.

Simply putting code out there and sharing your work is great, but going
back to a point I made some weeks ago, I beleive there is a responsibility
that comes along with it when you do that.  Whether they should or not,
people become dependent on the project... not in a cocaine kind of way of
course, but they are "counting on you" basically.  That to me implies
taking into consideration their needs and wants.  Not above your own of
course, but to some degree.

> Bottom line: if a person isn't contributing to code and documentation in
> a way that the other committers are comfortable with then that person
> shouldn't be a committer on the project.  There is no other reason for
> being a committer.

This I absolutely agree with, and it was the reason my proposal didn't try
to change that.  I would NEVER propose that the PMC not have the final say
in who is invited.  It just to me seems right for that to be the case. 
But, I still see nothing wrong with being able to say "hey, PMC, we think
this guy or gal would be a good addition, please consider him".

> My personal (and probably unneeded) opinion on the original subject:
>
>  From my perspective, nominations don't matter so much... as I recall
> someone could nominate themselves.  If that person hasn't been
> contributing code then there is no reason to think they will become a
> committer.

That is correct.  I frankly was not aware that someone could do that, Ted
pointed it out to me.  As I replied previously, that indeed covers the
first principle of my proposal.  I always prefer things like that be more
concrete, i.e., rules layed out in document form, but even failing that I
think the principle is followed, so I'm happy.

> It would be nice if the process were a little more transparent as it
> would be interesting to know who was proposed, accepted, rejected, etc.
> even if we didn't know why.  (Though, even counter to that it was nice
> to know that someone who contributed to another apache project and
> stomped all over my contributed implementation because they didn't
> bother to patch to head was at least a controversial nomination.  But
> that's sort of personal and isolated reason for wanting to see the dirty
> laundry.)

I still have the concerns about people being embarassed by this.  However,
I think the idea of a nominee accepting the nomination first is a fair
idea.  Putting aside the original proposal, how would that simple change,
along with opening the vote process discussion for all to see, sit with
everyone?

> I guess I have trouble seeing how things could be improved much by your
> proposal... especially since I understood there to be nothing wrong with
> nominations coming from anywhere.  It was just explained to be easier
> with a committer's support.  I don't follow this list too closely, so
> maybe I missed someone who has been contributing lots of stuff and still
> was overlooked.

Agreed, once Ted explained that point to me, the proposal isn't quite as
strong as I thought at first.  I still think there is the issue of
transparency that could do with further discussion, but it seems the
nomination part of it is, more or less, already present.  Codifying it
would be nice, but I can live with it not being written anywhere.

Thanks for commenting, you are always welcome as far as I'm concerned :)

> -Paul

Frank

---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@struts.apache.org
For additional commands, e-mail: dev-help@struts.apache.org