On 30-Apr-09, at 8:37 AM, nicolas de loof wrote:
Hi,(Hope my poor english will not create nonsense here)
Thanks for your reply, whatever can be said against my contribution
to Maven
core :p
I fully agree I was not a usefull contributor to maven core. The main
reasons are
- lack of time : I'm not a self-employee man. I get not a cents fom
maven -
maybe I'll with the incoming french book, but that lot's of efforts
for few
money. I like maven and want it to be as good as possible.
- hard to understand design. Sory to say that lack of typed
collections /
javadoc on APIs don't help to understand internals, even for plugin
developpers
Done in trunk, and Hervé has made the changes necessary in Modello for
the generated code.
- use of (maybe technologically excellent but) exotic and poorly
documented
tools. This don't make things easy for a newbee developer to bring
good code
in SVN.
Such is life but this barely impacts the code in core itself. Using
Guice, for example, is not going to help that problem. Though using
Guice would help when problems occurred. There's no avoiding the hard
work of digging in and spending hundreds of man hours to understand
the code base. I think this will be far simpler to understand in 3.x
and that's my primary goal with the first release.
I'd be pleased to contribute more to maven-core. To be honnest, I've
allready so much to do on plugins - maybe the visible part of the
iceberg,
but the one users request features
That's fine, comment where you have a right to comment.
For more and better contributions I'd need a cleaner view on maven
short
term roadmap. The maven 3.x plan is a very high level and I have not
spare
enough time to try understand it. The 2.2 release seems more
accessible to
me, and I plan to contribute more to this branch when I get time for
it.
Then contribute where you're comfortable but I a strongly disagree
with John and Brett and I'm not going to make a grand master roadmap
to have the vast majority of people do nothing. I have historical
precent on my side. If someone wants to be creative and helpful. I'd
be happy to chat with them and in 3 hours could probably record
something someone can work with.
No one wants to do the shitty work and it's not contributing to
features right now it's working on and fixing the integration tests so
that someone can work on the core and get immediate feedback on
problems. Completely working integration tests in Hudson running all
the time is more important to me then a roadmap. I'm refactoring and I
don't even know what future versions of Maven 3.x will look like I
just know that the code base is too hard to work with.
I really consider you make a technically great job on maven trunk,
and want
Sonatype to stay a main supporter or opensource maven. I just don't
like the
idea that people not working full time on maven have few chance to
get into
the project.
If you want to help us with the integration tests and refactoring then
that is the best way to enable others getting involved. Documentation
of code that changes on a daily basis is useless. Tests which describe
how the components work and the ITs where you see examples of behavior
is what will make it easier for people to try and make changes knowing
they haven't harmed the system.
I'd like to see a real roadmap for maven 2.x and 3.x, not just few
pages on
wiki. I'd like to know WHY we have replaced Foo with Bar not having to
search the mailing list history (@see the proposal wikis used on
apache
commons for sample). I'd like to ensure some major discution don't
occured
on IRC when I'm sleeping ;)
I'm not going to go into a huge long discussion, if people don't like
the direction then look at the code. At this point I'm telling you
that there is not a lot of interesting work going on insofar as
features. It's all refactoring and testing and it's full-time work. I
am merging to popular frameworks and I'm picking the ones I like
because I'm doing the work. If you don't like them write your own and
put them forward. We're never going to use Spring, so the choices for
IoC are XBR, Pico, and Guice and Guice is probably the best fit for
community, number of users, and compatibility with Plexus. For
transport who can do a better job then the Jetty people? Yes I just
try and preempt these low level discussions because do you really
care? If so say something.
You either help with all the crap work of integration testing, or wait
until the feature discussions start. As I try to solidify the base it
is hard to stop while I'm working full-time but the APIs are getting
smaller, the code base is getting smaller and more ITs are passing and
trunk is catching up to being compatible with 2.x.
If someone wants to chat and take the lead on writing the roadmap I
would be happy to work with them.
To make things shorter, I'd like to see Maven beeing documented on
what we
want it to be and how we plan to do so. Nowaday I can't find anything
concrete about this.
It's all where it's always been in the Wiki. You can ask the people
who have made proposals so you have to do a lot of work as well if you
want to understand what people have done. It's not my job to spend
half my time telling you what I'm doing so we can discuss and have me
do the work. If you want to get involved then ask questions. I'm happy
to tell anyone anything they want to know. At this point after 5 years
I know what the average person contributes and doesn't and that's just
the way it is. If someone wants to change that here then they have to
work with what's here and for the core on 3.x that's me. If someone
wants to make a completely detailed road map I'll point them to all
the source material which is plenty, if scattered, and I will be happy
to help someone put together a full plan. I want to code. I don't want
to be a project manager. My modus operandi is to focus in on and
select the couple people I know who can help in fundamental ways.
Benjamin is one of them for me. I would rather spend all my time with
him then write you a roadmap. If you want a roadmap I'll give you the
material to write a roadmap.
Maybe I'm not community-compliant with such expectation, maybe I can
get
blamed to publicly claim my point of view, maybe I missunderstood
Sonatype
implication on Maven development and maybe I'd better focus on
Archiva and
forget maven-core. If you thing so, please remove my write
permission on
maven-core SVN. I just would like to keep it on core-plugins if
possible, as
those one require people if we expect to get more frequent releases.
That's not what was voted so that's not what's done here. It's one of
the things I disagree with here but you take the parts you agree with,
along with the parts you disagree with.
Have a nice week-end,
Cheers,
Nicolas
2009/4/30 Jason van Zyl <jvan...@sonatype.com>
On 24-Apr-09, at 12:21 AM, nicolas de loof wrote:
Hi guys,
The previous translation sounds good, far better that any english
I could
write by myself, as you may notice in following lines ;).
First of all, this blog was not expected to be offensive. If you
consider
it
such please accept my apology, and feel free to attach any comment
to
expose
your point of view.
Really. An opening title of "The Hostile Takeover of Maven", siting
myself
in the first sentence and then listing a set of things for which
make little
sense to me so I'll address them as I believe by demonstration of
your own
behavior you have done exactly as I have. These sensationalist
titles and
blogs also immediately make me think of consultants abusing their
association with the Maven project for their own personal gain and
notoriety. It's certainly not helping the project. You can see the
Ivy guy
reacting exactly as you planned I'm sure. And, of course, by my
association
with the Maven project I am trying to further Sonatype. That is
nothing I'm
ashamed and I don't believe it's bad for the project either. I
think I have
just a tad more justification then you for doing so. This is above
all else
a meritocracy. People often forget here that it is a consensus
among those
who have done work, not just a consensus.
You, Nicolas, have 3 commits on the core of Maven, two of which we
rolled
back because you engaged in zero discussion with the list and made
changes
that were inappropriate. You actually made them to benefit Archiva
which is
a project you work on it while not thinking of the greater impact.
So the
exact thing what you are insinuating is _exactly_ what you did. I
know how
I work but I don't do one thing and say another. Especially about
someone
else on a public blog about a project I'm barely involved in -- and
I site
the single commit you have on the code base you're expounding
about. You
came into this project by association with Archiva and not by way
of your
sustained contribution to Maven. Which may explain the first
comment in your
blog.
- The creation of Nexus is just my personal project working on it
with a
bunch of my friends.
I don't feel that Archiva is something I wanted to be involved
with. I
preferred to take the oldest known codebase of a Maven repository
manager
(maven proxy I don't consider a manager) and support that effort.
There were
also people in Archiva that I frankly didn't want to work with.
These are
choices that we are all free to make and I really don't think that
the
existence of Nexus has harmed anyone in the Maven community. We
have very
aggressive schedules, we are very focused and we are very vocal. I
see none
of those things as a detriment to Maven users.
- The development of Plexus
Plexus is not a Maven project, it's not even an Apache project. But
if you
actually paid attention to the mailing lists of the code base you
would see
code is rapidly being culled and replaced in the hopes of being
able to swap
in Guice. You can get involved in Plexus just like you would any
external
project. Do you complain to the Spring developers when they change
the
internal dependency injection strategy?
- The overhaul of the transport
Have you even looked at the code to see how it works? It's not on by
default, and as I said to Brett a couple days ago and on the
mailing list a
couple months ago that there are no new features in 3.x. I don't
think you
understand. The only two people who really merit a decision in this
process
are really Brett, Brian and Mark because they have each made very
significant effort with the code related to that part of Maven and
it's an
incredible amount of work. Oleg has also sent many emails and there
are lots
of documents in the wiki. So there is amble place to make comments
but this
is not something you can just dabble in and yes it's hard to keep
up when
there are people that are working on it full time. If you're
interested ask,
or as Brett as for a summary (politely I might add) I will do that.
My blog is only for personnal opinions and uses a
caricatural style to get reader post comments. If I hurted you I
can post
an
eratum with my apologies, this was not expected to be read at first
degree.
For this reason I choosed this title in relation with current
finantial
activities around SUN & Java
All I can say is not very appropriate. I'll keep the rest of the
comments
to myself and trust me I'm biting my tongue.
During JUG sessions, many people ask me about maven roadmap, and
how the
project development is planed. Most of them are not used with
professional
opensource, and ask me how a company like Sonatype can get money
from
opensource.
Probably no different then a consultant trying to make money off open
source. Through involvement and experience you gain the trust of
people who
want to use the projects you work on. It's not that complicated.
You appear
to clearly mention that you're an "Apache Maven Committer" on your
LinkedIn
page so I'm sure what you are trying to do is not all together
different
then what Sonatype is trying to do.
What makes you assume we're making money off Maven? Sonatype's
business
model does have some training and revenue components but we're
primarily
product sales and we do almost no consulting unless it relates to our
products. So unlike consultants we have less reason to push in
specific
changes that benefit one client. I think we are actually in a
position where
we don't have to do questionable things. The massive changes I have
in Maven
that were client specific I have never committed.
This blog article has been written in reply to such discutions. It
tries
to
reflect how Sonatype guys work full-time on maven 3 and related
ecosystem
to
be "the maven company". As you may notice, I compare this business
model
to
Spring and JBoss way to opensource : open license, project open to
contribution, but project lead "hosted" by a company.
Except that your comparison is completely and entirely wrong.
Sonatype is
not trying to monetize Maven. We don't even remotely have the same
business
model. All the work we do on Maven is a community service. Sure we
want
recognition but we're not selling an enterprise version of Maven,
we don't
have publicly available forked versions of Maven that we promote
and I can
tell you our business model includes work on Maven that yields no
direct
return except for good will. If the Maven project has crappy
technical
underpinning it's not very useful for people. Yes, Sonatype's
business is
predicated on the widespread use of Maven-based technologies. Our
investment
is to increase the number of Maven users by increasing the quality
of Maven
itself and have it be used more. But that does not just directly
benefit us.
So how we make money is first making sure the ecosystem is healthy
because
if it isn't and people think we're doing devious things to then our
entire
model crumbles. Without a healthy Maven community we have no
business. We
invest in Maven without the expectation of a direct return.
We've invested in remaking Archetype, a couple full-time employees
working
on the 2.x codebase to make sure releases continue, work on
evolving the
artifact system, the transport system, Brian spends a ton of time
setting up
things like Nexus (and yes it's our product but yes it is useful to
Maven
users nonetheless), the Maven book, m2eclipse + book, Nexus + book
so I
think we're doing a lot for the community. We definitely hope people
recognize this and see because of this we understand the ecosystem
of tools
very well and puts us in a position to make great products that fit
well in
the ecosystem. I think Sonatype has done a pretty good job
balancing our
community and commercial interests.
As Daniel said, I really like Maven, I really like the maven 3
roadmap you
described (the post was also a way to link to the video). I also
don't
like
the current dev process : "release early, release often" - it
tooks 6
month
between two minor maven 2.0 release, we hardly got a maven 2.1
build.
So what are you doing about it? Aside from complaining? Seriously?
You
don't even ask for a summary of what you want, you just post a
diatribe
which doesn't do the community any good.
I
don't say people here are not activelly contributing : the dev
community
is
working hard and I myslef hardly find time to contribute
significantly. I
just thing we miss a good roadmap plan. Maven 2.x just have one
from few
time, maven 3 is still a "will be great, bu when" project.
Please post any comment you feel usefull for users to understand
your idea
of maven future.
Thanks all
Cheers,
Nicolas
2009/4/24 Jason van Zyl <jvan...@sonatype.com>
On 23-Apr-09, at 11:19 PM, Daniel Le Berre wrote:
Jason,
The summary is perfectly correct.
I would add that the author mentions too that your are friends,
and the
way the text is written is not offensive.
Good thing you translated. From the title and the Google
translated text
I
read it as offensive. We are not friends so I would guess an
attempt at
sarcasm.
(The author clearly does not agree with current maven development
process,
he would like it be more community driven,
but it looks like Maven still has some values on his eyes).
Daniel
Le 24 avr. 09 à 01:52, Jason van Zyl a écrit :
Thanks if that seems like a reasonable translation I will respond.
On 23-Apr-09, at 4:00 PM, Christian Edward Gruber wrote:
My french isn't perfect, but the article basically...
<summary>
...argues against you (Jason) personally having the habit of
imposing
dramatic changes by presenting them as a fait-acomplis.
There's a bit of a bill of rights: We have the right to choose
archiva
or nexus (which he labels as Jason & Friends' personal
project), the
right
to replace plexus with XBR?, the right not to have wagon
replaced by
the
jetty team without discussion (or something like that), and
something
about
you "offering" to integrate this person's GWT plugin and have it
hosted at
sonatype (with a sarcastic "what an honour" at the end).
Essentially it's an argument that you and your buddies have
created a
"putch" and taken over what should be an egalitarian, apache-
itarian
project. But it's now your little pet project (you and your
friends).
There's a bit about Eugene's being excluded because he doesn't
have
the
same high-profile as the rest of the team, etc.
He believes in the potential, even if M3 looks like far-off
promise,
but he's got a problem with the attitude of the core group.
He thinks
Maven's going the way of JBoss and Spring and becoming a
"closed"
opensource
project. The whole thing's nearly entirely under the control
of the
Sonatype people... how long until a "Maven Pro" comes out?
If things continue this way, he's going to have to revisit Ant
and
Ivy.
</summary>
Fun fun fun. My own 2c... I don't care if Sonatype, codehaus,
or an
anarchic comune develops it, as long as they're responsive to
bright
ideas
from the community and keep it open source. As far as I can
tell,
Jason,
Brett, John, Emmanuel, and a host of others have done awesome
things.
I
don't see Nexus as nefariously excluding Archiva, nor any other
maven-launched projects like Continuum, etc. I just see this
all as
churn
in the community as different ideas are tried. As long as
we're not
locked in, I have no problems with that. I know no one asked my
opinion,
but there it is.
cheers
Christian.
Could someone who speaks French please translate this for me
before I
respond:
http://blog.loof.fr/
I'm not a native French speaker so I won't speculate, but if
someone
would translate I would like to respond.
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
----------------------------------------------------------
What matters is not ideas, but the people who have them. Good
people
can fix bad ideas, but good ideas can't save bad people.
-- Paul Graham
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
Christian Edward Gruber
christianedwardgru...@gmail.com
http://www.geekinasuit.com/
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
----------------------------------------------------------
A man enjoys his work when he understands the whole and when he
is responsible for the quality of the whole
-- Christopher Alexander, A Pattern Language
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
----------------------------------------------------------
In short, man creates for himself a new religion of a rational
and technical order to justify his work and to be justified in it.
-- Jacques Ellul, The Technological Society
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
http://twitter.com/SonatypeNexus
http://twitter.com/SonatypeM2E
----------------------------------------------------------
the course of true love never did run smooth ...
-- Shakespeare
Thanks,
Jason
----------------------------------------------------------
Jason van Zyl
Founder, Apache Maven
http://twitter.com/jvanzyl
http://twitter.com/SonatypeNexus
http://twitter.com/SonatypeM2E
----------------------------------------------------------
First, the taking in of scattered particulars under one Idea,
so that everyone understands what is being talked about ... Second,
the separation of the Idea into parts, by dividing it at the joints,
as nature directs, not breaking any limb in half as a bad carver might.
-- Plato, Phaedrus (Notes on the Synthesis of Form by C. Alexander)
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscr...@maven.apache.org
For additional commands, e-mail: dev-h...@maven.apache.org
