deltaspike-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Mark Struberg <strub...@yahoo.de>
Subject Re: Sandbox for DeltaSpike
Date Thu, 28 Jun 2012 08:51:25 GMT
I know what you mean, but you worded it a bit too drastically :)

If a proposed feature is somehow related to CDI and sounds valuable, then we will for sure
add it.
But only after collecting additional ideas and doing a 'reality check' on the topic ;)

And if it helps I like to make this clear again: we will not even import stuff like the CODI
window handling 1:1 without a review. Actually I know quite a few parts which I like to do
radically different/easier. 


LieGrue,
strub




----- Original Message -----
> From: Gerhard Petracek <gerhard.petracek@gmail.com>
> To: deltaspike-dev@incubator.apache.org
> Cc: 
> Sent: Thursday, June 28, 2012 10:42 AM
> Subject: Re: Sandbox for DeltaSpike
> 
> for sure at least a vote can drop such parts - we did it already.
> i just mentioned the possibility because everybody has to be aware of it.
> (with an external sandbox it would be even worse.)
> 
> @ rest:
> agreed
> 
> regards,
> gerhard
> 
> 
> 
> 2012/6/28 Mark Struberg <struberg@yahoo.de>
> 
>>  I would not word it that drastically that we 'delete code if there is 
> no
>>  discussion upfront'.
>> 
>> 
>>  The discussion upfront is mainly important to raise visibility and
>>  attention. And to be able to get a response from many people about those
>>  new ideas. That way we can make good ideas even better and prevent easily
>>  overseen shortcomings. No one of us is perfect, but together we kick butt!
>> 
>>  Btw, the initial discussion is only a 'basic agreement' to kick off
>>  attention imo. If we see during implementation that other ways are
>>  superior, then there is no problem to amend the initially discussed topics.
>> 
>>  LieGrue,
>>  strub
>> 
>> 
>> 
>>  ----- Original Message -----
>>  > From: Gerhard Petracek <gerhard.petracek@gmail.com>
>>  > To: deltaspike-dev@incubator.apache.org
>>  > Cc:
>>  > Sent: Thursday, June 28, 2012 9:50 AM
>>  > Subject: Re: Sandbox for DeltaSpike
>>  >
>>  > i agree with mark.
>>  >
>>  > since we are talking about whole modules:
>>  > the idea of apache-labs [1] is a bit different but maybe it works for 
> us
>>  as
>>  > well.
>>  > (potential community members can clone it and follow our git
>>  > discussion-workflow.)
>>  >
>>  > in any case: there needs to be a discussion before moving such parts 
> to
>>  the
>>  > main repository -> that also means: if there is no agreement, we 
> have to
>>  > drop it again.
>>  >
>>  > regards,
>>  > gerhard
>>  >
>>  > [1] http://labs.apache.org
>>  >
>>  >
>>  >
>>  > 2012/6/28 Mark Struberg <struberg@yahoo.de>
>>  >
>>  >>  With 'public' I meant that the main communication tool is 
> the
>>  > mailing
>>  >>  list. There is a saying "if it's not on the list, it 
> didn't
>>  > happen".
>>  >>
>>  >>  IRC is fine as backing channel, but there are different time 
> zones etc.
>>  >>  It's also not logged (because freenode has a policy about not 
> logging
>>  >>  chats), thus other uses cannot simply search some archive to find 
> any
>>  old
>>  >>  information.
>>  >>
>>  >>
>>  >>  It's perfect if you drop a few lines of mail explaining what
>>  >>  problem/idea/feature you are working on and add a pointer to some
>>  github
>>  >>  repo.
>>  >>  But be aware that you must work alone on that gibhut repo or at 
> least
>>  must
>>  >>  _not_ accept patches/pull-requests from non-committers. Otherwise 
> you
>>  would
>>  >>  not be IP clean. And since goog vs orcl (Harmony,...) we _really_ 
> care
>>  >>  about that!
>>  >>
>>  >>  github is also a great tool, but it doesn't really strengthen 
> the team
>>  >>  collaboration spirit. It's more fore the lone fighter who 
> works on his
>>  >>  own...
>>  >>
>>  >>  Maybe I should explain it another way what could happen:
>>  >>
>>  >>
>>  >>  Imagine you get a cool new feature which has a decent complexity. 
> Say
>>  45
>>  >>  classes and 25000 lines of code. And all that in one big 
> merge-commit!
>>  >>  Compare that with work that evolves over a few weeks with 5 
> people
>>  working
>>  >>  on it and adding ideas. There would be much more understanding of 
> the
>>  topic
>>  >>  in the community and the quality would also be much better at the 
> end.
>>  >>  There will also be much less overlapping with other features in 
> the
>>  project
>>  >>  quite naturally...
>>  >>
>>  >>  LieGrue,
>>  >>  strub
>>  >>
>>  >>
>>  >>  ----- Original Message -----
>>  >>  > From: Jason Porter <lightguard.jp@gmail.com>
>>  >>  > To: "deltaspike-dev@incubator.apache.org" <
>>  >>  deltaspike-dev@incubator.apache.org>
>>  >>  > Cc: "deltaspike-dev@incubator.apache.org" <
>>  >>  deltaspike-dev@incubator.apache.org>; seam-dev@lists.jboss.org
>>  >>  > Sent: Wednesday, June 27, 2012 8:32 PM
>>  >>  > Subject: Re: Sandbox for DeltaSpike
>>  >>  >
>>  >>  > Why wouldn't this be in the public? The idea is to get 
> people to
>>  >>  contribute.
>>  >>  > If we need a separate Apache repo for a sandbox, okay fine 
> but then
>>  > we're
>>  >>  > back to the icla issue aren't we?
>>  >>  >
>>  >>  > Sent from my iPhone
>>  >>  >
>>  >>  > On Jun 27, 2012, at 14:10, Mark Struberg 
> <struberg@yahoo.de>
>>  > wrote:
>>  >>  >
>>  >>  >>  Btw, another thingy.
>>  >>  >>
>>  >>  >>  It is not the best community building approach to 
> develop
>>  > something 'in
>>  >>  > the dark' and then drop all that on all other community 
> members.
>>  >>  >>  Don't get me wrong, it's perfectly fine to 
> experiment
>>  > around if
>>  >>  > ideas are good at all. But doing this 'in public' is 
> much more
>>  >>  > appreciated. You can get lots or precious feedback that way.
>>  >>  >>
>>  >>  >>
>>  >>  >>  LieGrue,
>>  >>  >>  strub
>>  >>  >>
>>  >>  >>
>>  >>  >>
>>  >>  >>  ----- Original Message -----
>>  >>  >>>  From: Mark Struberg <struberg@yahoo.de>
>>  >>  >>>  To: "deltaspike-dev@incubator.apache.org"
>>  >>  > <deltaspike-dev@incubator.apache.org>
>>  >>  >>>  Cc:
>>  >>  >>>  Sent: Wednesday, June 27, 2012 7:33 PM
>>  >>  >>>  Subject: Re: Sandbox for DeltaSpike
>>  >>  >>>
>>  >>  >>>  basically +1
>>  >>  >>>
>>  >>  >>>
>>  >>  >>>  BUT we really have to be careful that we don't 
> do too
>>  > much at
>>  >>  > github!
>>  >>  >>>
>>  >>  >>>  All commits done on github must either be done by a
>>  > deltaspike
>>  >>  > committer or
>>  >>  >>>  someone who has at least an iCLA on file.
>>  >>  >>>
>>  >>  >>>  Commits from other people need to get added via an 
> attachment
>>  > in a
>>  >>  Jira
>>  >>  > ticket.
>>  >>  >>>  I know this sounds not really git-like, but 
> it's the only
>>  > way we
>>  >>  > can ensure
>>  >>  >>>  IP clearance.
>>  >>  >>>
>>  >>  >>>  LieGrue,
>>  >>  >>>  strub
>>  >>  >>>
>>  >>  >>>
>>  >>  >>>  ----- Original Message -----
>>  >>  >>>>  From: Mehdi Heidarzadeh 
> <heidarzadeh2@gmail.com>
>>  >>  >>>>  To: deltaspike-dev@incubator.apache.org
>>  >>  >>>>  Cc:
>>  >>  >>>>  Sent: Wednesday, June 27, 2012 7:28 PM
>>  >>  >>>>  Subject: Re: Sandbox for DeltaSpike
>>  >>  >>>>
>>  >>  >>>>  +1
>>  >>  >>>>  Great idea.
>>  >>  >>>>
>>  >>  >>>>  On Wed, Jun 27, 2012 at 4:52 AM, Shane Bryzak
>>  >>  > <sbryzak@redhat.com>
>>  >>  >>>  wrote:
>>  >>  >>>>
>>  >>  >>>>>    Fantastic idea, +1.
>>  >>  >>>>>
>>  >>  >>>>>
>>  >>  >>>>>    On 27/06/12 05:39, Jason Porter wrote:
>>  >>  >>>>>
>>  >>  >>>>>>    Hey everyone!
>>  >>  >>>>>>
>>  >>  >>>>>>    I wanted to bring up the idea of 
> having a
>>  > sandbox to add
>>  >>  > bits and
>>  >>  >>>  other
>>  >>  >>>>>>    non-core extensions. We have a great 
> bunch of
>>  > people from
>>  >>  > the Seam
>>  >>  >>>>>>    development group looking to add 
> their
>>  > extensions, but
>>  >>  > they're
>>  >>  >>>
>>  >>  >>>>  either not
>>  >>  >>>>>>    on the roadmap for DS, or are very 
> far down. I
>>  > suggest we
>>  >>  > setup a
>>  >>  >>>>  sandbox
>>  >>  >>>>>>    on github people can write to, or at 
> least do
>>  > pull
>>  >>  > requests to so
>>  >>  >>>  we
>>  >>  >>>>  can
>>  >>  >>>>>>    get some of these modules and other 
> ideas in
>>  > and pull
>>  >>  > them into
>>  >>  >>>  core as
>>  >>  >>>>  we
>>  >>  >>>>>>    get there. We can also use this as a 
> vetting
>>  > ground for
>>  >>  > new ideas
>>  >>  >>>  and
>>  >>  >>>>  other
>>  >>  >>>>>>    things which may not exactly fit into 
> core,
>>  > like the
>>  >>  > forge
>>  >>  >>>  extension.
>>  >>  >>>>>>
>>  >>  >>>>>>    To do this we need to
>>  >>  >>>>>>
>>  >>  >>>>>>    1. Setup the repo somewhere
>>  >>  >>>>>>    2. Seed it with a basic structure 
> (pom.xml,
>>  > contribution
>>  >>  >>>  instructions,
>>  >>  >>>>>>    etc)
>>  >>  >>>>>>    3. Get some CI setup somewhere (we 
> could
>>  > leverage
>>  >>  > OpenShift for
>>  >>  >>>  this if
>>  >>  >>>>>>    needed)
>>  >>  >>>>>>
>>  >>  >>>>>>    What does everyone else think? 
> I've
>>  > cc'd the Seam
>>  >>  >
>>  >>  >>>  Development
>>  >>  >>>>  list here
>>  >>  >>>>>>    hoping to get some feedback from them 
> as well
>>  > and
>>  >>  > hopefully
>>  >>  >>>  rekindle
>>  >>  >>>>  some
>>  >>  >>>>>>    of the fire we had there.
>>  >>  >>>>>>
>>  >>  >>>>>>    --
>>  >>  >>>>>>    Jason Porter
>>  >>  >>>>>>    http://lightguard-jp.blogspot.**com
>>  >>  >>>>  <http://lightguard-jp.blogspot.com>
>>  >>  >>>>>>    http://twitter.com/**lightguardjp
>>  >>  >>>>  <http://twitter.com/lightguardjp>
>>  >>  >>>>>>
>>  >>  >>>>>>    Software Engineer
>>  >>  >>>>>>    Open Source Advocate
>>  >>  >>>>>>    Author of Seam Catch - Next 
> Generation Java
>>  > Exception
>>  >>  > Handling
>>  >>  >>>>>>
>>  >>  >>>>>>    PGP key id: 926CCFF5
>>  >>  >>>>>>    PGP key available at: keyserver.net
>>  >>  > <http://keyserver.net>,
>>  >>  >>>>  pgp.mit.edu <
>>  >>  >>>>>>    http://pgp.mit.edu>
>>  >>  >>>>>>
>>  >>  >>>>>>
>>  >>  >>>>>>
>>  >>  >>>>>>
>>  > ______________________________**_________________
>>  >>  >>>>>>    seam-dev mailing list
>>  >>  >>>>>>  seam-dev@lists.jboss.org
>>  >>  >>>>>>
>>  >>  >>>>
>>  >>  >>>
>>  >>  > https://lists.jboss.org/**mailman/listinfo/seam-dev<
>>  >>  https://lists.jboss.org/mailman/listinfo/seam-dev>
>>  >>  >>>>>>
>>  >>  >>>>>
>>  >>  >>>>>
>>  >>  >>>>
>>  >>  >>>>
>>  >>  >>>>  --
>>  >>  >>>>  Mehdi Heidarzadeh Ardalani
>>  >>  >>>>  Independent JEE Consultant, Architect and 
> Developer.
>>  >>  >>>>  http://www.TheBigJavaBlog.com
>>  >>  >>>>
>>  >>  >>>
>>  >>  >
>>  >>
>>  >
>> 
> 

Mime
View raw message