Zookeeper, mail # dev - Re: Comment from Martin Fowler

Re: Comment from Martin Fowler

Ted Dunning 2011-11-02, 16:34

I have to say that the entire "my code is bigger than your code" episode in
the Hadoop community should just be considered embarrassing for all
concerned. That includes the ones who started it, Thomas K who posted a
holier-than-thou (but essentially correct) comment, and Thomas J who posted
a silly put-down in reply.

Let me also save anybody else the breath and say that I have said things at
least as rude as what Thomas J said and at least as naive as what Thomas K.
I plan to keep on saying both kinds of things, but in smaller quantities.

I think that Thomas Koch has good contributions to make and lots of energy
and lots of fascination with Zookeeper. That is great, but so far somewhat
less effective than it could be. If we can help him channel that energy a
bit more, then it will be much, much better.

I also think that Zookeeper has lots of technical debt, partly because we
didn't get to start with a codebase designed for testing. Paying off debt
is always extremely painful. We have a payment due.

On Wed, Nov 2, 2011 at 8:50 AM, Patrick Hunt <[EMAIL PROTECTED]> wrote:

> Perhaps we are being too nice when we should be direct. See how a
> similar thread was recently handled on hadoop:
> http://markmail.org/message/qswdx5yuvpsetwis
>
> Patrick
>
> On Wed, Nov 2, 2011 at 7:01 AM, Camille Fournier <[EMAIL PROTECTED]>
> wrote:
> > He also mentions the need for good judgement, which is a trait you appear
> > to be sorely lacking.
> >
> > From my phone
> > On Nov 2, 2011 5:58 AM, "Thomas Koch" <[EMAIL PROTECTED]> wrote:
> >
> >> Martin Fowler wrote a comment yesterday on our recent discussion about
> >> refactoring:
> >>
> >> http://martinfowler.com/bliki/OpportunisticRefactoring.html
> >>
> >> Regards,
> >>
> >> Thomas Koch, http://www.koch.ro
> >>
> >
>

Re: Comment from Martin Fowler

Patrick Hunt 2011-11-02, 17:10

On Wed, Nov 2, 2011 at 9:34 AM, Ted Dunning <[EMAIL PROTECTED]> wrote:
> I think that Thomas Koch has good contributions to make and lots of energy
> and lots of fascination with Zookeeper. That is great, but so far somewhat
> less effective than it could be. If we can help him channel that energy a
> bit more, then it will be much, much better.
>
> I also think that Zookeeper has lots of technical debt, partly because we
> didn't get to start with a codebase designed for testing. Paying off debt
> is always extremely painful. We have a payment due.

Ted, no disrespect but put your money where you mouth is, start
reviewing patches if you feel strongly. So far I've taken the brunt of
doing the reviews from Thomas and the rest of the community is just
getting pissed off by his attitude.

It doesn't matter what someones contributions might be, if they can't
work with the community.
http://communityovercode.com/over/

Patrick

Re: Comment from Martin Fowler

Ted Dunning 2011-11-02, 18:15

On Wed, Nov 2, 2011 at 10:10 AM, Patrick Hunt <[EMAIL PROTECTED]> wrote:

> On Wed, Nov 2, 2011 at 9:34 AM, Ted Dunning <[EMAIL PROTECTED]> wrote:
> > ...
>
> I also think that Zookeeper has lots of technical debt, partly because we
> > didn't get to start with a codebase designed for testing. Paying off
> debt
> > is always extremely painful. We have a payment due.
>
> Ted, no disrespect but put your money where you mouth is, start
> reviewing patches if you feel strongly.
Guilty (mostly) as charged.

I have only been able to review a few of the patches. My work schedule is
heinous right now.
> So far I've taken the brunt of
> doing the reviews from Thomas and the rest of the community is just
> getting pissed off by his attitude.
>

I have also tried to work on this issue. I have met with Thomas in person
and provided private coaching (with limited, but non-zero success). I also
have privately mediated some misunderstandings.
> It doesn't matter what someones contributions might be, if they can't
> work with the community.
> http://communityovercode.com/over/
Very true. That is why I have tried to work to help Thomas learn how to
work with this community. It isn't all about a snapshot in time; people
can develop new skills.

Re: Comment from Martin Fowler

Patrick Hunt 2011-11-02, 18:23

On Wed, Nov 2, 2011 at 11:15 AM, Ted Dunning <[EMAIL PROTECTED]> wrote:
> On Wed, Nov 2, 2011 at 10:10 AM, Patrick Hunt <[EMAIL PROTECTED]> wrote:
>
>> On Wed, Nov 2, 2011 at 9:34 AM, Ted Dunning <[EMAIL PROTECTED]> wrote:
>> > ...
>>
>> I also think that Zookeeper has lots of technical debt, partly because we
>> > didn't get to start with a codebase designed for testing. Paying off
>> debt
>> > is always extremely painful. We have a payment due.
>>
>> Ted, no disrespect but put your money where you mouth is, start
>> reviewing patches if you feel strongly.
>
>
> Guilty (mostly) as charged.
>
> I have only been able to review a few of the patches. My work schedule is
> heinous right now.

Totally understand (I'm in the same boat wrt being overloaded). Hope
you took it in the context I meant it. We can't do hugs through email
-- raincheck for next time we meet f2f. ;-)

>
>
>> So far I've taken the brunt of
>> doing the reviews from Thomas and the rest of the community is just
>> getting pissed off by his attitude.
>>
>
> I have also tried to work on this issue. I have met with Thomas in person
> and provided private coaching (with limited, but non-zero success). I also
> have privately mediated some misunderstandings.
>
>
>> It doesn't matter what someones contributions might be, if they can't
>> work with the community.
>> http://communityovercode.com/over/
>
>
> Very true. That is why I have tried to work to help Thomas learn how to
> work with this community. It isn't all about a snapshot in time; people
> can develop new skills.
>

That's the thing, so far everyone has been trying (incl Thomas) but we
are still seeing friction. Perhaps it's just the limited time
available, pressure to get 3.4.0 out, and non-alignment btw Thomas's
interests and our current goals?

Patrick

Re: Comment from Martin Fowler

Flavio Junqueira 2011-11-03, 18:25

I don't mean to interrupt the love story of this thread, but I'd like
to add a small comment.

It might not be easy to come up with such a list, but in general, it
sounds like a good idea to have a set of guidelines, accepted by the
community, that we could use to reject contributions. Otherwise, it
becomes a matter of taste, which is difficult to manage when we have
tens of people contributing. Even though it might be virtually
impossible to get rid of taste completely, we need a mechanism that
enables us to reject contributions without feeling guilty about doing
a disservice to the community. After all, I believe one of our key
goals is to maintain a community and to attract valuable
contributions, not to push people away without a reason. Any reason we
use to reject contributions should ideally be supported by the
community and should not be unilateral.

If I'm alone on this thought, then I'm happy to drop it, but otherwise
it would be great to hear some thoughts on how to get something like
this going. I think we already have some guidelines in the
documentation, but if I remember correctly, they are kind of weak with
respect to the perspective I'm laying above.

-Flavio

On Nov 2, 2011, at 7:23 PM, Patrick Hunt wrote:

> On Wed, Nov 2, 2011 at 11:15 AM, Ted Dunning <[EMAIL PROTECTED]>
> wrote:
>> On Wed, Nov 2, 2011 at 10:10 AM, Patrick Hunt <[EMAIL PROTECTED]>
>> wrote:
>>
>>> On Wed, Nov 2, 2011 at 9:34 AM, Ted Dunning
>>> <[EMAIL PROTECTED]> wrote:
>>>> ...
>>>
>>> I also think that Zookeeper has lots of technical debt, partly
>>> because we
>>>> didn't get to start with a codebase designed for testing. Paying
>>>> off
>>> debt
>>>> is always extremely painful. We have a payment due.
>>>
>>> Ted, no disrespect but put your money where you mouth is, start
>>> reviewing patches if you feel strongly.
>>
>>
>> Guilty (mostly) as charged.
>>
>> I have only been able to review a few of the patches. My work
>> schedule is
>> heinous right now.
>
> Totally understand (I'm in the same boat wrt being overloaded). Hope
> you took it in the context I meant it. We can't do hugs through email
> -- raincheck for next time we meet f2f. ;-)
>
>>
>>
>>> So far I've taken the brunt of
>>> doing the reviews from Thomas and the rest of the community is just
>>> getting pissed off by his attitude.
>>>
>>
>> I have also tried to work on this issue. I have met with Thomas in
>> person
>> and provided private coaching (with limited, but non-zero
>> success). I also
>> have privately mediated some misunderstandings.
>>
>>
>>> It doesn't matter what someones contributions might be, if they
>>> can't
>>> work with the community.
>>> http://communityovercode.com/over/
>>
>>
>> Very true. That is why I have tried to work to help Thomas learn
>> how to
>> work with this community. It isn't all about a snapshot in time;
>> people
>> can develop new skills.
>>
>
> That's the thing, so far everyone has been trying (incl Thomas) but we
> are still seeing friction. Perhaps it's just the limited time
> available, pressure to get 3.4.0 out, and non-alignment btw Thomas's
> interests and our current goals?
>
> Patrick

flavio
junqueira

research scientist

[EMAIL PROTECTED]
direct +34 93-183-8828

avinguda diagonal 177, 8th floor, barcelona, 08018, es
phone (408) 349 3300 fax (408) 349 3301

Re: Comment from Martin Fowler

Benjamin Reed 2011-11-03, 20:46

i agree. i proposed such a guideline on a separate thread. it wasn't
too well received. i think camille also made some good points. i would
love to establish a guideline of rejecting cleanup patches while fixes
and new functions are being added. it is just too distracting to
everyone involved.

ben

On Thu, Nov 3, 2011 at 11:25 AM, Flavio Junqueira <[EMAIL PROTECTED]> wrote:
> I don't mean to interrupt the love story of this thread, but I'd like to add
> a small comment.
>
> It might not be easy to come up with such a list, but in general, it sounds
> like a good idea to have a set of guidelines, accepted by the community,
> that we could use to reject contributions. Otherwise, it becomes a matter of
> taste, which is difficult to manage when we have tens of people
> contributing. Even though it might be virtually impossible to get rid of
> taste completely, we need a mechanism that enables us to reject
> contributions without feeling guilty about doing a disservice to the
> community. After all, I believe one of our key goals is to maintain a
> community and to attract valuable contributions, not to push people away
> without a reason. Any reason we use to reject contributions should ideally
> be supported by the community and should not be unilateral.
>
> If I'm alone on this thought, then I'm happy to drop it, but otherwise it
> would be great to hear some thoughts on how to get something like this
> going. I think we already have some guidelines in the documentation, but if
> I remember correctly, they are kind of weak with respect to the perspective
> I'm laying above.
>
> -Flavio
>
> On Nov 2, 2011, at 7:23 PM, Patrick Hunt wrote:
>
>> On Wed, Nov 2, 2011 at 11:15 AM, Ted Dunning <[EMAIL PROTECTED]>
>> wrote:
>>>
>>> On Wed, Nov 2, 2011 at 10:10 AM, Patrick Hunt <[EMAIL PROTECTED]> wrote:
>>>
>>>> On Wed, Nov 2, 2011 at 9:34 AM, Ted Dunning <[EMAIL PROTECTED]>
>>>> wrote:
>>>>>
>>>>> ...
>>>>
>>>> I also think that Zookeeper has lots of technical debt, partly because
>>>> we
>>>>>
>>>>> didn't get to start with a codebase designed for testing. Paying off
>>>>
>>>> debt
>>>>>
>>>>> is always extremely painful. We have a payment due.
>>>>
>>>> Ted, no disrespect but put your money where you mouth is, start
>>>> reviewing patches if you feel strongly.
>>>
>>>
>>> Guilty (mostly) as charged.
>>>
>>> I have only been able to review a few of the patches. My work schedule
>>> is
>>> heinous right now.
>>
>> Totally understand (I'm in the same boat wrt being overloaded). Hope
>> you took it in the context I meant it. We can't do hugs through email
>> -- raincheck for next time we meet f2f. ;-)
>>
>>>
>>>
>>>> So far I've taken the brunt of
>>>> doing the reviews from Thomas and the rest of the community is just
>>>> getting pissed off by his attitude.
>>>>
>>>
>>> I have also tried to work on this issue. I have met with Thomas in
>>> person
>>> and provided private coaching (with limited, but non-zero success). I
>>> also
>>> have privately mediated some misunderstandings.
>>>
>>>
>>>> It doesn't matter what someones contributions might be, if they can't
>>>> work with the community.
>>>> http://communityovercode.com/over/
>>>
>>>
>>> Very true. That is why I have tried to work to help Thomas learn how to
>>> work with this community. It isn't all about a snapshot in time; people
>>> can develop new skills.
>>>
>>
>> That's the thing, so far everyone has been trying (incl Thomas) but we
>> are still seeing friction. Perhaps it's just the limited time
>> available, pressure to get 3.4.0 out, and non-alignment btw Thomas's
>> interests and our current goals?
>>
>> Patrick
>
> flavio
> junqueira
>
> research scientist
>
> [EMAIL PROTECTED]
> direct +34 93-183-8828
>
> avinguda diagonal 177, 8th floor, barcelona, 08018, es
> phone (408) 349 3300 fax (408) 349 3301
>
>