maven-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Paul Benedict <pbened...@apache.org>
Subject Re: Maven plugin naming pattern
Date Fri, 10 Oct 2014 14:58:13 GMT
Agreed, Stephen. A warning should be emitted. A build should not break.


Cheers,
Paul

On Fri, Oct 10, 2014 at 9:57 AM, Stephen Connolly <
stephen.alan.connolly@gmail.com> wrote:

> I agree that maven core should issue warnings on the plugin names... but we
> cannot break builds for people upgrading maven with a fully locked down pom
> (otherwise we'll never persuade them to upgrade, so IMHO core should warn
> only and never break... maven-plugin-plugin however should just break the
> build)
>
> On 10 October 2014 15:00, Karl Heinz Marbaise <khmarbaise@gmx.de> wrote:
>
> > Hi,
> >
> > On 10/10/14 3:41 PM, Paul Benedict wrote:
> >
> >> I would prefer this should be part of Maven Core's warning system. If
> the
> >> plugin starts with maven- and it's not an org.apache.maven.plugins
> group,
> >> then we should spit out the error. I am not sure enforcer is the right
> >> place for this rule; this is more of a global problem than a suggestion
> >> for
> >> good practice.
> >>
> >
> > In my opinion this should be part of Maven Core (Maven itself within the
> > next version 3.2.4 ?) otherwise we can't be sure that those warnings
> > (possible breaks) will ever happen...
> >
> > If you you use enforcer you can of course create such a rule which is
> > really simple but if you don't use that rule ......
> >
> > I would suggest to create for Maven 3.2.4 warning and for 3.3.X it should
> > create an error and fail the build...
> >
> >
> >
> >>
> >> Cheers,
> >> Paul
> >>
> >> On Fri, Oct 10, 2014 at 8:30 AM, Michael Osipov <1983-01-06@gmx.net>
> >> wrote:
> >>
> >>  Fine, I'd like to note that first:
> >>>
> >>> 1. Shouldn't we announce this publically on the users mailing list?
> >>> 2. I think that this deserves a major bump in plugin version. WDYT?
> >>>
> >>> Michael
> >>>
> >>>  Gesendet: Freitag, 10. Oktober 2014 um 15:23 Uhr
> >>>> Von: "Stephen Connolly" <stephen.alan.connolly@gmail.com>
> >>>> An: "Maven Developers List" <dev@maven.apache.org>
> >>>> Betreff: Re: Re: Re: Re: Maven plugin naming pattern
> >>>>
> >>>> That was the plan.... 3 years ago we decided to warn first and then
> >>>>
> >>> switch
> >>>
> >>>> on failing after a while... now is a good time, perhaps you could
> commit
> >>>> the change to fail the build?
> >>>>
> >>>> On 10 October 2014 13:48, Michael Osipov <1983-01-06@gmx.net>
wrote:
> >>>>
> >>>>  Yes, resposibility isn't always good.
> >>>>>
> >>>>> Shouldn't simply make the build fail instead of log when such a
> >>>>>
> >>>> collision
> >>>
> >>>> happens?
> >>>>>
> >>>>> Michael
> >>>>>
> >>>>>  Thankfully for you, you are not on the PMC... if you were on the
PMC
> >>>>>>
> >>>>> and
> >>>
> >>>> you did such a search you would then have to go and send C&Ds.
> >>>>>>
> >>>>>> /me runs away from this thread in case I happen to be made aware
of
> >>>>>>
> >>>>> any
> >>>
> >>>> trademark misuse ;-)
> >>>>>>
> >>>>>> On 10 October 2014 13:39, Stephen Connolly <
> >>>>>>
> >>>>> stephen.alan.connolly@gmail.com>
> >>>>>
> >>>>>> wrote:
> >>>>>>
> >>>>>>  Yes
> >>>>>>>
> >>>>>>> On 10 October 2014 13:12, Michael Osipov <1983-01-06@gmx.net>
> >>>>>>>
> >>>>>> wrote:
> >>>
> >>>>
> >>>>>>>  If you do a quick search on Central, you'll that even other
Apache
> >>>>>>>>
> >>>>>>> project
> >>>>>
> >>>>>> don't adhere to this convention. Should they receive a C&D
too?
> >>>>>>>>
> >>>>>>>> Michael
> >>>>>>>>
> >>>>>>>>  We just need to show best effort to defend our trademark...
if
> >>>>>>>>>
> >>>>>>>> we
> >>>
> >>>> *see*
> >>>>>
> >>>>>> anyone doing that then we have to send them C&D letters...
> >>>>>>>>>
> >>>>>>>>> Note: my understanding is that we only have to send
C&D letters
> >>>>>>>>>
> >>>>>>>> when we
> >>>>>
> >>>>>> know somebody is abusing our mark... we don't necessarily have
> >>>>>>>>>
> >>>>>>>> to go
> >>>
> >>>> actively looking for people abusing our mark... just if we went
> >>>>>>>>>
> >>>>>>>> looking
> >>>>>
> >>>>>> and
> >>>>>>>>
> >>>>>>>>> found any then we have to send them C&Ds quite
quickly
> >>>>>>>>>
> >>>>>>>>> On 10 October 2014 12:45, Benson Margulies <
> >>>>>>>>>
> >>>>>>>> bimargulies@gmail.com>
> >>>
> >>>> wrote:
> >>>>>>>>
> >>>>>>>>>
> >>>>>>>>>  On Fri, Oct 10, 2014 at 7:42 AM, Stephen Connolly
> >>>>>>>>>> <stephen.alan.connolly@gmail.com> wrote:
> >>>>>>>>>>
> >>>>>>>>>>> We, the PMC, agreed to allow permitted usage
of the form
> >>>>>>>>>>>
> >>>>>>>>>> ___-maven-plugin
> >>>>>>>>
> >>>>>>>>> as that clarified that the plugin was a plugin for
maven
> >>>>>>>>>>>
> >>>>>>>>>> not one
> >>>
> >>>> produced
> >>>>>>>>
> >>>>>>>>> by maven
> >>>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>> Yea, I know, and I'm not opposed to making the
tooling more
> >>>>>>>>>> obstreperous. I'm just warning people not to
have high hopes
> >>>>>>>>>>
> >>>>>>>>> of
> >>>
> >>>> enforcement for anyone who chooses to hack the tooling and not
> >>>>>>>>>> cooperate.
> >>>>>>>>>>
> >>>>>>>>>>
> >>>>>>>>>>> On 10 October 2014 12:40, Benson Margulies
<
> >>>>>>>>>>>
> >>>>>>>>>> bimargulies@gmail.com>
> >>>>>
> >>>>>> wrote:
> >>>>>>>>>>
> >>>>>>>>>>>
> >>>>>>>>>>>  Keep in mind that what we have here is
almost certainly a
> >>>>>>>>>>>> _convention_, not a point of trademark
law. As I
> >>>>>>>>>>>>
> >>>>>>>>>>> understand it,
> >>>
> >>>> we'd
> >>>>>>>>
> >>>>>>>>> as likely be laughed at for the suggestion that
reversing
> >>>>>>>>>>>>
> >>>>>>>>>>> the
> >>>
> >>>> order of
> >>>>>>>>
> >>>>>>>>> the components of a name leads to 'marketplace confusion'
> >>>>>>>>>>>>
> >>>>>>>>>>> at
> >>>
> >>>> the
> >>>>>
> >>>>>> level
> >>>>>>>>
> >>>>>>>>> at which trademarks can be enforced.
> >>>>>>>>>>>>
> >>>>>>>>>>>> On Fri, Oct 10, 2014 at 7:25 AM, Michael
Osipov <
> >>>>>>>>>>>>
> >>>>>>>>>>> 1983-01-06@gmx.net>
> >>>>>>>>
> >>>>>>>>> wrote:
> >>>>>>>>>>>>
> >>>>>>>>>>>>> They should rename going forward.
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>> At some point (probably we could
do so now) we will
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>> turn on
> >>>
> >>>> enforcement
> >>>>>>>>>>
> >>>>>>>>>>> in
> >>>>>>>>>>>>
> >>>>>>>>>>>>> the maven-plugin-plugin.
> >>>>>>>>>>>>>>
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> This will of course piss of a lot
of people. Wouldn't it?
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> There are, of course, several reasons
why people can't:
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> 1. Popularity of the old name
> >>>>>>>>>>>>> 2. Technical reasons
> >>>>>>>>>>>>> 3. Name collisions
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> etc.
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> Even if we enforce this, this should
not happen before
> >>>>>>>>>>>>>
> >>>>>>>>>>>> Maven
> >>>
> >>>> 4
> >>>>>
> >>>>>> and
> >>>>>>>>
> >>>>>>>>> should be added to the plugin dev center.
> >>>>>>>>>>>>
> >>>>>>>>>>>>>
> >>>>>>>>>>>>> Michael
> >>>>>>>>>>>>>
> >>>>>>>>>>>>>
> > Kind regards
> > Karl Heinz Marbaise
> >
> >
> >
> > ---------------------------------------------------------------------
> > To unsubscribe, e-mail: dev-unsubscribe@maven.apache.org
> > For additional commands, e-mail: dev-help@maven.apache.org
> >
> >
>

Mime
  • Unnamed multipart/alternative (inline, None, 0 bytes)
View raw message