devicemap-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Werner Keil <werner.k...@gmail.com>
Subject Re: Deleting the legacy ODDR client and related artifacts from SVN
Date Mon, 05 Jan 2015 11:50:15 GMT
See the other thread and reply to Bertrand's message.

"deprecated" is totally untrue, since the W3C DDR implementation is well
established, stable and mature.
The only reason a release was delayed was the 1.0 Data first not being
released (it would have been odd to release it and use OpenDDR data though
technically possible) and when it got released, Reza accidentally
disfigured the class references in the Builder.xml files (let's assume it
was not intentional;-) so data 1.0 was also unusable.
The data 1.0.1 out on even MavenCentral now offers a proper baseline for a
1.x W3C DDR implementation and every current 1.x data does, too.

Looking at major players like Facebook or large companies that either
directly use the OpenDDR or other compatible W3C DDR implementations this
is the only client offering them a seamless migration path easier than say
moving a Java Servlet web application from JBoss or WebLogic to Tomcat;-)

Technically a OpenDDR 2.x client running against the W3C compliant data
would also work, but the Builder references in the DDR data must be
preserved, so if the data files say "org.apache.devicemap.simpleddr.*" so
must be those immediate classes, even if the code was "forced away" to an
OpenDDR repository again.

It would require dual- or relicensing of the client code and more
importantly "disfranchise" Apache DeviceMap not only as W3C compliant
implementation (to some companies or commercial clients those things
matter;-) but by driving a certain user base off back into another
community. We offered the OpenDDR data and (W3C) client to try build a
single, stronger community out of at least 1 or 2, so that would mean a
certain fraction and proliferation of the community.

>From a process point, based on the board's input a single "-1 in such
structural cases is enough to prevent it, but the code only makes sense, if
it's also released at least once.
See the other thread, let's discuss there.

Werner


On Mon, Jan 5, 2015 at 9:31 AM, Bertrand Delacretaz <bdelacretaz@apache.org>
wrote:

> Hi,
>
> On Fri, Jan 2, 2015 at 6:58 PM, Reza Naghibi
> <reza.naghibi@yahoo.com.invalid> wrote:
> > ...Any objections to deleting the legacy ODDR java client and its
> related artifacts
> > from SVN?...
>
> It looks like there's strong objections from Werner in this thread,
> and I also think that even if some code was deprecated it would be
> better to just move it under a "contrib" or "deprecated" folder first.
>
> Even if with svn that doesn't matter much as it's easy to bring things
> back, but in terms of gradual evolution I would prefer that.
>
> Anyway, it looks like some discussion is needed first, I'll start that
> in a new thread.
>
> -Bertrand
>

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