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: Leaving DeviceMap
Date Mon, 20 Apr 2015 15:46:15 GMT
You forgot though a (not much bigger but better coordinated and yet more
democratic) team of committers lead by myself contributed ALL of
device-data in the first place;-)
Plus the seed of every example (the Spring and Servlet Filter stuff is
always just a minor variation of the /contrib codebase we brought in from
OpenDDR)

Some blurry visions about "let's rewrite everything in JSON because JSON is
cool now" etc. are not a real solution yet nor has the draft under a 2.x
branch ever happened as Bertrand and others suggested. The short, crisp but
clear 6 bullet points Volkan sent around his hackathon code seem to have
more actual substance than some of the theoretical piece on the Wiki.

No 2.x or 3.x branch would start from scratch, so the most important aspect
is how to get data into a new format without losing the precious
information we got about Ten Thousands of devices so far.

Werner

On Mon, Apr 20, 2015 at 5:28 PM, Reza Naghibi <rezan@apache.org> wrote:

> References to commits and/or other proof would be helpful here.
>
> I looked thru the SVN in regards to the examples and I see the same commit
> pattern, POM and syntax tweaks.
>
> https://svn.apache.org/viewvc/devicemap/trunk/examples/
>
> Im not trying to say that you need to be writing code to contribute here,
> there are lots of ways to contribute. But that fact that you have no
> substantial technical contributions but you are trying to direct the
> technical direction and architecture of this project goes a long way to
> explain why we are in the position we are in.
>
>
>
>
>
> On Mon, Apr 20, 2015 at 11:13 AM, Werner Keil <werner.keil@gmail.com>
> wrote:
>
> > I got (with Bertrand's help) the CI server to work after it was broken
> for
> > several weeks.
> > Similar with the VM instance. The whole examples (for every available
> > client) shows not a single trace of activity other than my own for over a
> > year.
> >
> > Data just look up the SVN, after all that's what makes the project, there
> > are other clients outside Apache.org, and aside from refactoring, and
> > occasional fixes for new devices, there's not a significant number of
> real
> > changes in the data files either.
> >
> > Packaging and deploying the archives, sure, that did happen and some of
> the
> > administrative stuff, but why cry on one side we don't have enough
> > resources (which in a smaller project can't be denied) and refusing to
> help
> > at least one or two of the guys who showed they want to help since at
> least
> > graduation?
> >
> > Asking others to help and delegating is not weakness, and if we had
> people
> > like Volkan on board since he offered to help (just check his first JIRA
> > tickets to see when that was) not some weeks from now, it would have been
> > beneficial to the whole project.
> >
> > Sure, there are other "never-ending stories" like the new web site we
> > talked about, but what people are more interested if and when they use
> > DeviceMap is "does it recognize my device properly?"
> >
> > And especially with several issues raised by multiple people about "wrong
> > OS version". ignoring the properly given UA value of say "Android 4.1"
> and
> > returning something else like "Android 2.2" is dumb and can't be
> justified
> > by arguments like "Our magic 2.0 Wand will fix this, in the meantime
> please
> > ignore the OS version".
> >
> > The information is there, it's just not used (at least in most clients)
> And
> > whether it's the OS or browser version, there will always be some element
> > of change, and even with 40 instead of 4 or 5 committers we would not be
> > able to cope with that just in the repository.
> >
> > Which is why a "healthy dose of builder" similar to what Volkan did
> during
> > his hackathon seems very beneficial. Like some of the patches he already
> > provided, it was just up to current committers to apply them. None of us
> > did, so maybe better accept help than refuse it or just tell people
> "it'll
> > come with 2.0 or 3.x" while someone else fixed the problem already for
> 1.x.
> >
> > Werner
> >
> > On Mon, Apr 20, 2015 at 4:33 PM, Reza Naghibi <rezan@apache.org> wrote:
> >
> > > >> Sorry but it was me who managed not only the graduation, but
> > everything
> > > else.
> > >
> > > I really need to chime in here otherwise people might believe these
> > > statements for truth. These are flat out lies. The only thing that you
> > did
> > > was point out that my the Java service was down and CI was broken. That
> > is
> > > a far stretch from actual doing things like fixing the Java service on
> > the
> > > VM, fixing CI, updating the website, contributing to releases, or
> > > graduating this project.
> > >
> > > I scanned the commit history for our website, I cannot find a single
> > update
> > > from you:
> > >
> > > https://svn.apache.org/viewvc/devicemap/site/trunk/
> > >
> > > You definitely had no role in the graduation process, I spent a good 2
> > > weeks doing all of the graduation tasks with much help from Bertrand.
> The
> > > fact that you take credit for graduating this project is sad.
> > >
> > > I look at your SVN commit history all the time, you have no releases
> and
> > > your commits amount to pretty much tweaking the legacy ODDR poms around
> > and
> > > changing the version number to compete with the 1.x releases.
> > >
> > > Are you aware that you constantly lie and manipulate the truth in your
> > > emails? To a newcomer, they probably read your emails and believe this
> > > stuff... which is scary.
> > >
> > > >> Records in this and other threads speak for themselves.
> > >
> > > Please please please start citing these records and threads. Otherwise
> > you
> > > are presenting nothing but a distorted reality.
> > >
> > >
> > > On Mon, Apr 20, 2015 at 7:44 AM, Werner Keil <werner.keil@gmail.com>
> > > wrote:
> > >
> > > > Sorry but it was me who managed not only the graduation, but
> everything
> > > > else.
> > > > CI server down for months. The website was out of date for many weeks
> > > after
> > > > the graduation.
> > > > The VM instance happened to be broken or stopped several times when
> > > people
> > > > wanted to test it during conferences. I had to ask Reza to start it.
> > > > At least one on this list who also contributed was willing to help,
> but
> > > > nobody bothered to accept his help till I took another management
> > > > responsibility in my own hands and started a ballot which we
> > fortunately.
> > > >
> > > > Records in this and other threads speak for themselves. The evidence
> is
> > > > right, no need to deny any of that.
> > > > The false impression that I was "co PMC Chair" by Reza himself says
> > > enough.
> > > > I WAS doing what he volunteered to do all the time;-)
> > > >
> > > > I don't see a problem to have him as committer/PMC member, should he
> > > decide
> > > > to stay, but e.g. Bertrand until someone really qualified should
> > > volunteer
> > > > again was clearly a better choice to act as PMC chair. I did most of
> > the
> > > > other managing like votes for new committers, etc. Happy to help with
> > > that
> > > > as well.
> > > >
> > > > Cheers,
> > > > Werner
> > > >
> > > > On Mon, Apr 20, 2015 at 12:38 PM, Bertrand Delacretaz <
> > > > bdelacretaz@apache.org> wrote:
> > > >
> > > > > On Sun, Apr 19, 2015 at 5:42 PM, Reza Naghibi <rezan@apache.org>
> > > wrote:
> > > > > > ...Instead of focusing on writing
> > > > > > and releasing software, all my time and energy is spent on how
to
> > > > manage
> > > > > > Werner....
> > > > >
> > > > > I think we now have abundant evidence that you and Werner cannot
> work
> > > > > together - evaluating who's right and who's wrong does not help
> much,
> > > > > but maybe we can find a creative solution that would cause you to
> > > > > reconsider, at least temporarily.
> > > > >
> > > > > How about defining a "perimeter" in our code and specs that Werner
> is
> > > > > not allowed to touch or discuss? It can be as simple as a subtree
> in
> > > > > our source code tree, and a wiki subtree.
> > > > >
> > > > > Combined with a ban (amicable or formal) of email arguments between
> > > > > you guys, this might be a workable solution, that we can try for
> say
> > 3
> > > > > months to see if things improve.
> > > > >
> > > > > That's unusual in an Apache project, and if this team was larger
> > > > > things would probably self-regulate, but with such a small team
> it's
> > > > > hard to find a balance without resorting to such artificial
> barriers.
> > > > >
> > > > > What do people think?
> > > > >
> > > > > -Bertrand
> > > > >
> > > >
> > >
> >
>

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