You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@lucene.apache.org by Erick Erickson <er...@gmail.com> on 2011/12/17 20:46:11 UTC
My role
I've been wondering what to do with this committer status thing. When
one has reached "my age", the prospect of understanding a code base as
complex and one that has been worked on by such awesome people is
daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
chapters titled "The Notebooks of Lazarus Long") "It's amazing how
much mature wisdom resembles being too tired". I've never been able to
figure out whether "how much" means "the degree to which" or "the
quantity of". I suppose they're both applicable.
Anyway, one of the things that's bothered me a bit is the patches
people submit that languish. Believe me, I understand that most of the
people who are actively involved as committers are doing some really
great and fundamental improvements to Lucene/Solr and I'm in awe of
their efforts; it's not like other committers have lots of time to
spare and are just lazy.
But it occurred to me that there's a role for someone to shepherd
patches submitted by people who are *not* committers through the
process. Or close them as "interesting but not something we should
add". So, the long and short of it is that I'm volunteering for that
role. I expect to lean heavily on the othes to give
thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
patches too, but don't be surprised by comments on JIRAs like "what do
y'all think? Kill or commit?".
I have a day job too, so this won't happen all at once. But I've been
amazed by how much gets accomplished by steady effort. Besides, it's
winter in Michigan. So anybody who has favorite JIRAs that they think
are particularly worthwhile, please let me know and I'll at least put
them on my list.
Let me know your thoughts
Erick
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: My role
Posted by Erick Erickson <er...@gmail.com>.
Oh, this thread will do. That way, people can see the process. I have
a little list of JIRAs that I'm tracking, I think if I can make it a
goal to nudge one along a day that'll at least help.
The problem is that I'm pretty time constrained. I'll try to look at a
patch a day, but that will often involve throwing up my hands and
asking for help from people who *really* know.
But at least we can make *some* progress, I'm still feeling my way
into this process, since I, too, have a "day job".
Best
Erick
On Tue, Jan 3, 2012 at 1:38 PM, Mikhail Khludnev
<mk...@griddynamics.com> wrote:
> Erick,
>
> 1. Here is one more patch which is worth to
> apply https://issues.apache.org/jira/browse/SOLR-2804.
> It's about multi-thread indexing, but impacts SolrJ and DIH users in 4.0 and
> 3.x.But there is an easy workaround by suppressing logger. Patches for both
> branches including a test are attached.
> Please have a look, I'm ready that my coding style can be considered as too
> heavyweight.
>
> 2. Is there any plans for SOLR-2933 and SOLR-2947?
>
> 3. What is an appropriate way for such communications beside of this thread?
>
> Regards
>
>
> On Thu, Dec 29, 2011 at 6:52 AM, Erick Erickson <er...@gmail.com>
> wrote:
>>
>> Mikhail:
>>
>> Fair warning. I know nothing, zero, zilch about DIH so as far as code
>> reviews etc, I can do some
>> superficial stuff, but a deep understanding of DIH is not something I
>> can provide.
>>
>> What I can do is help nudge it along, test, do the gruntwork that goes
>> with actually checking
>> something in. Here's what I propose (and I confess I haven't looked at
>> the bugs yet tonight,
>> I won't have time until tomorrow).
>>
>> Let's get on those bugs, and nudge them along. We can gently prompt
>> the people who *do*
>> know to provide some guidance and then we can see if we can get these
>> committed. It'll
>> take some persistence on your part, what I'm hoping most to provide is
>> closure when
>> a consensus is reached that they're ready to commit.
>>
>> And thanks for being willing to work on the *hard* parts. Threading is
>> notorious for
>> having ugly, hard-to-find bugs....
>>
>> Best
>> Erick
>>
>> On Wed, Dec 28, 2011 at 5:23 PM, Mikhail Khludnev
>> <mk...@griddynamics.com> wrote:
>> > Erick,
>> >
>> > I want to fix good old DIH multi-threading issue. It doesn't work at
>> > all.
>> >
>> > I've decided to start after SOLR-2382 is committed.
>> > Firstly I've found two minor issues introduced by SOLR-2382:
>> > - SOLR-2933 - It has patch polished by James D;
>> > - SOLR-2947 - I attached fix for the minor subject issue. It was
>> > reviewed by
>> > James. And I just attached patch which cover multi-threading _too_.
>> > Perhaps
>> > it's better to attach it to the separate issue. Let me know.
>> >
>> > I suppose it's worth to move these two forward, and multi-threading fix
>> > too. And btw, then I'll be able to cover one more minor SOLR-2961.
>> >
>> > Regards
>> >
>> > On Sat, Dec 17, 2011 at 11:46 PM, Erick Erickson
>> > <er...@gmail.com>
>> > wrote:
>> >>
>> >> I've been wondering what to do with this committer status thing. When
>> >> one has reached "my age", the prospect of understanding a code base as
>> >> complex and one that has been worked on by such awesome people is
>> >> daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
>> >> chapters titled "The Notebooks of Lazarus Long") "It's amazing how
>> >> much mature wisdom resembles being too tired". I've never been able to
>> >> figure out whether "how much" means "the degree to which" or "the
>> >> quantity of". I suppose they're both applicable.
>> >>
>> >> Anyway, one of the things that's bothered me a bit is the patches
>> >> people submit that languish. Believe me, I understand that most of the
>> >> people who are actively involved as committers are doing some really
>> >> great and fundamental improvements to Lucene/Solr and I'm in awe of
>> >> their efforts; it's not like other committers have lots of time to
>> >> spare and are just lazy.
>> >>
>> >> But it occurred to me that there's a role for someone to shepherd
>> >> patches submitted by people who are *not* committers through the
>> >> process. Or close them as "interesting but not something we should
>> >> add". So, the long and short of it is that I'm volunteering for that
>> >> role. I expect to lean heavily on the othes to give
>> >> thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
>> >> patches too, but don't be surprised by comments on JIRAs like "what do
>> >> y'all think? Kill or commit?".
>> >>
>> >> I have a day job too, so this won't happen all at once. But I've been
>> >> amazed by how much gets accomplished by steady effort. Besides, it's
>> >> winter in Michigan. So anybody who has favorite JIRAs that they think
>> >> are particularly worthwhile, please let me know and I'll at least put
>> >> them on my list.
>> >>
>> >> Let me know your thoughts
>> >> Erick
>> >>
>> >> ---------------------------------------------------------------------
>> >> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
>> >> For additional commands, e-mail: dev-help@lucene.apache.org
>> >>
>> >
>> >
>> >
>> > --
>> > Sincerely yours
>> > Mikhail Khludnev
>> > Lucid Certified
>> > Apache Lucene/Solr Developer
>> > Grid Dynamics
>> >
>> >
>> >
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: dev-help@lucene.apache.org
>>
>
>
>
> --
> Sincerely yours
> Mikhail Khludnev
> Lucid Certified
> Apache Lucene/Solr Developer
> Grid Dynamics
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: My role
Posted by Mikhail Khludnev <mk...@griddynamics.com>.
Erick,
1. Here is one more patch which is worth to apply
https://issues.apache.org/jira/browse/SOLR-2804.
It's about multi-thread indexing, but impacts SolrJ and DIH users in 4.0
and 3.x.But there is an easy workaround by suppressing logger. Patches for
both branches including a test are attached.
Please have a look, I'm ready that my coding style can be considered as
too heavyweight.
2. Is there any plans for SOLR-2933 and SOLR-2947?
3. What is an appropriate way for such communications beside of this thread?
Regards
On Thu, Dec 29, 2011 at 6:52 AM, Erick Erickson <er...@gmail.com>wrote:
> Mikhail:
>
> Fair warning. I know nothing, zero, zilch about DIH so as far as code
> reviews etc, I can do some
> superficial stuff, but a deep understanding of DIH is not something I
> can provide.
>
> What I can do is help nudge it along, test, do the gruntwork that goes
> with actually checking
> something in. Here's what I propose (and I confess I haven't looked at
> the bugs yet tonight,
> I won't have time until tomorrow).
>
> Let's get on those bugs, and nudge them along. We can gently prompt
> the people who *do*
> know to provide some guidance and then we can see if we can get these
> committed. It'll
> take some persistence on your part, what I'm hoping most to provide is
> closure when
> a consensus is reached that they're ready to commit.
>
> And thanks for being willing to work on the *hard* parts. Threading is
> notorious for
> having ugly, hard-to-find bugs....
>
> Best
> Erick
>
> On Wed, Dec 28, 2011 at 5:23 PM, Mikhail Khludnev
> <mk...@griddynamics.com> wrote:
> > Erick,
> >
> > I want to fix good old DIH multi-threading issue. It doesn't work at all.
> >
> > I've decided to start after SOLR-2382 is committed.
> > Firstly I've found two minor issues introduced by SOLR-2382:
> > - SOLR-2933 - It has patch polished by James D;
> > - SOLR-2947 - I attached fix for the minor subject issue. It was
> reviewed by
> > James. And I just attached patch which cover multi-threading _too_.
> Perhaps
> > it's better to attach it to the separate issue. Let me know.
> >
> > I suppose it's worth to move these two forward, and multi-threading fix
> > too. And btw, then I'll be able to cover one more minor SOLR-2961.
> >
> > Regards
> >
> > On Sat, Dec 17, 2011 at 11:46 PM, Erick Erickson <
> erickerickson@gmail.com>
> > wrote:
> >>
> >> I've been wondering what to do with this committer status thing. When
> >> one has reached "my age", the prospect of understanding a code base as
> >> complex and one that has been worked on by such awesome people is
> >> daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
> >> chapters titled "The Notebooks of Lazarus Long") "It's amazing how
> >> much mature wisdom resembles being too tired". I've never been able to
> >> figure out whether "how much" means "the degree to which" or "the
> >> quantity of". I suppose they're both applicable.
> >>
> >> Anyway, one of the things that's bothered me a bit is the patches
> >> people submit that languish. Believe me, I understand that most of the
> >> people who are actively involved as committers are doing some really
> >> great and fundamental improvements to Lucene/Solr and I'm in awe of
> >> their efforts; it's not like other committers have lots of time to
> >> spare and are just lazy.
> >>
> >> But it occurred to me that there's a role for someone to shepherd
> >> patches submitted by people who are *not* committers through the
> >> process. Or close them as "interesting but not something we should
> >> add". So, the long and short of it is that I'm volunteering for that
> >> role. I expect to lean heavily on the othes to give
> >> thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
> >> patches too, but don't be surprised by comments on JIRAs like "what do
> >> y'all think? Kill or commit?".
> >>
> >> I have a day job too, so this won't happen all at once. But I've been
> >> amazed by how much gets accomplished by steady effort. Besides, it's
> >> winter in Michigan. So anybody who has favorite JIRAs that they think
> >> are particularly worthwhile, please let me know and I'll at least put
> >> them on my list.
> >>
> >> Let me know your thoughts
> >> Erick
> >>
> >> ---------------------------------------------------------------------
> >> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> >> For additional commands, e-mail: dev-help@lucene.apache.org
> >>
> >
> >
> >
> > --
> > Sincerely yours
> > Mikhail Khludnev
> > Lucid Certified
> > Apache Lucene/Solr Developer
> > Grid Dynamics
> >
> >
> >
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: dev-help@lucene.apache.org
>
>
--
Sincerely yours
Mikhail Khludnev
Lucid Certified
Apache Lucene/Solr Developer
Grid Dynamics
<http://www.griddynamics.com>
<mk...@griddynamics.com>
Re: My role
Posted by Erick Erickson <er...@gmail.com>.
Mikhail:
Fair warning. I know nothing, zero, zilch about DIH so as far as code
reviews etc, I can do some
superficial stuff, but a deep understanding of DIH is not something I
can provide.
What I can do is help nudge it along, test, do the gruntwork that goes
with actually checking
something in. Here's what I propose (and I confess I haven't looked at
the bugs yet tonight,
I won't have time until tomorrow).
Let's get on those bugs, and nudge them along. We can gently prompt
the people who *do*
know to provide some guidance and then we can see if we can get these
committed. It'll
take some persistence on your part, what I'm hoping most to provide is
closure when
a consensus is reached that they're ready to commit.
And thanks for being willing to work on the *hard* parts. Threading is
notorious for
having ugly, hard-to-find bugs....
Best
Erick
On Wed, Dec 28, 2011 at 5:23 PM, Mikhail Khludnev
<mk...@griddynamics.com> wrote:
> Erick,
>
> I want to fix good old DIH multi-threading issue. It doesn't work at all.
>
> I've decided to start after SOLR-2382 is committed.
> Firstly I've found two minor issues introduced by SOLR-2382:
> - SOLR-2933 - It has patch polished by James D;
> - SOLR-2947 - I attached fix for the minor subject issue. It was reviewed by
> James. And I just attached patch which cover multi-threading _too_. Perhaps
> it's better to attach it to the separate issue. Let me know.
>
> I suppose it's worth to move these two forward, and multi-threading fix
> too. And btw, then I'll be able to cover one more minor SOLR-2961.
>
> Regards
>
> On Sat, Dec 17, 2011 at 11:46 PM, Erick Erickson <er...@gmail.com>
> wrote:
>>
>> I've been wondering what to do with this committer status thing. When
>> one has reached "my age", the prospect of understanding a code base as
>> complex and one that has been worked on by such awesome people is
>> daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
>> chapters titled "The Notebooks of Lazarus Long") "It's amazing how
>> much mature wisdom resembles being too tired". I've never been able to
>> figure out whether "how much" means "the degree to which" or "the
>> quantity of". I suppose they're both applicable.
>>
>> Anyway, one of the things that's bothered me a bit is the patches
>> people submit that languish. Believe me, I understand that most of the
>> people who are actively involved as committers are doing some really
>> great and fundamental improvements to Lucene/Solr and I'm in awe of
>> their efforts; it's not like other committers have lots of time to
>> spare and are just lazy.
>>
>> But it occurred to me that there's a role for someone to shepherd
>> patches submitted by people who are *not* committers through the
>> process. Or close them as "interesting but not something we should
>> add". So, the long and short of it is that I'm volunteering for that
>> role. I expect to lean heavily on the othes to give
>> thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
>> patches too, but don't be surprised by comments on JIRAs like "what do
>> y'all think? Kill or commit?".
>>
>> I have a day job too, so this won't happen all at once. But I've been
>> amazed by how much gets accomplished by steady effort. Besides, it's
>> winter in Michigan. So anybody who has favorite JIRAs that they think
>> are particularly worthwhile, please let me know and I'll at least put
>> them on my list.
>>
>> Let me know your thoughts
>> Erick
>>
>> ---------------------------------------------------------------------
>> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
>> For additional commands, e-mail: dev-help@lucene.apache.org
>>
>
>
>
> --
> Sincerely yours
> Mikhail Khludnev
> Lucid Certified
> Apache Lucene/Solr Developer
> Grid Dynamics
>
>
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: My role
Posted by Mikhail Khludnev <mk...@griddynamics.com>.
Erick,
I want to fix good old DIH multi-threading issue. It doesn't work at all.
I've decided to start after SOLR-2382 is committed.
Firstly I've found two minor issues introduced by SOLR-2382:
- SOLR-2933 - It has patch polished by James D;
- SOLR-2947 - I attached fix for the minor subject issue. It was reviewed
by James. And I just attached patch which cover multi-threading _too_.
Perhaps it's better to attach it to the separate issue. Let me know.
I suppose it's worth to move these two forward, and multi-threading fix
too. And btw, then I'll be able to cover one more minor SOLR-2961.
Regards
On Sat, Dec 17, 2011 at 11:46 PM, Erick Erickson <er...@gmail.com>wrote:
> I've been wondering what to do with this committer status thing. When
> one has reached "my age", the prospect of understanding a code base as
> complex and one that has been worked on by such awesome people is
> daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
> chapters titled "The Notebooks of Lazarus Long") "It's amazing how
> much mature wisdom resembles being too tired". I've never been able to
> figure out whether "how much" means "the degree to which" or "the
> quantity of". I suppose they're both applicable.
>
> Anyway, one of the things that's bothered me a bit is the patches
> people submit that languish. Believe me, I understand that most of the
> people who are actively involved as committers are doing some really
> great and fundamental improvements to Lucene/Solr and I'm in awe of
> their efforts; it's not like other committers have lots of time to
> spare and are just lazy.
>
> But it occurred to me that there's a role for someone to shepherd
> patches submitted by people who are *not* committers through the
> process. Or close them as "interesting but not something we should
> add". So, the long and short of it is that I'm volunteering for that
> role. I expect to lean heavily on the othes to give
> thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
> patches too, but don't be surprised by comments on JIRAs like "what do
> y'all think? Kill or commit?".
>
> I have a day job too, so this won't happen all at once. But I've been
> amazed by how much gets accomplished by steady effort. Besides, it's
> winter in Michigan. So anybody who has favorite JIRAs that they think
> are particularly worthwhile, please let me know and I'll at least put
> them on my list.
>
> Let me know your thoughts
> Erick
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: dev-help@lucene.apache.org
>
>
--
Sincerely yours
Mikhail Khludnev
Lucid Certified
Apache Lucene/Solr Developer
Grid Dynamics
<http://www.griddynamics.com>
<mk...@griddynamics.com>
Re: My role
Posted by Yonik Seeley <yo...@lucidimagination.com>.
On Sat, Dec 17, 2011 at 2:46 PM, Erick Erickson <er...@gmail.com> wrote:
> But it occurred to me that there's a role for someone to shepherd
> patches submitted by people who are *not* committers through the
> process.
It should also be noted that anyone can help move a patch forward by
reviewing, proposing changes, trying to reproduce (if it's a bug report), etc.
For those interested in becoming a committer, the best way is to act
like a committer.
-Yonik
http://www.lucidimagination.com
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org
Re: My role
Posted by Michael McCandless <lu...@mikemccandless.com>.
+1, this (pushing patches forwards or backwards, but not letting them
just languish) is very important and so rarely done!
Mike McCandless
http://blog.mikemccandless.com
On Sat, Dec 17, 2011 at 2:46 PM, Erick Erickson <er...@gmail.com> wrote:
> I've been wondering what to do with this committer status thing. When
> one has reached "my age", the prospect of understanding a code base as
> complex and one that has been worked on by such awesome people is
> daunting. To quote Robert Heinlein ("Time Enough for Love", one of the
> chapters titled "The Notebooks of Lazarus Long") "It's amazing how
> much mature wisdom resembles being too tired". I've never been able to
> figure out whether "how much" means "the degree to which" or "the
> quantity of". I suppose they're both applicable.
>
> Anyway, one of the things that's bothered me a bit is the patches
> people submit that languish. Believe me, I understand that most of the
> people who are actively involved as committers are doing some really
> great and fundamental improvements to Lucene/Solr and I'm in awe of
> their efforts; it's not like other committers have lots of time to
> spare and are just lazy.
>
> But it occurred to me that there's a role for someone to shepherd
> patches submitted by people who are *not* committers through the
> process. Or close them as "interesting but not something we should
> add". So, the long and short of it is that I'm volunteering for that
> role. I expect to lean heavily on the othes to give
> thumbs-up/thumbs-down here. Don't get me wrong, I'll look over the
> patches too, but don't be surprised by comments on JIRAs like "what do
> y'all think? Kill or commit?".
>
> I have a day job too, so this won't happen all at once. But I've been
> amazed by how much gets accomplished by steady effort. Besides, it's
> winter in Michigan. So anybody who has favorite JIRAs that they think
> are particularly worthwhile, please let me know and I'll at least put
> them on my list.
>
> Let me know your thoughts
> Erick
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
> For additional commands, e-mail: dev-help@lucene.apache.org
>
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@lucene.apache.org
For additional commands, e-mail: dev-help@lucene.apache.org