You are viewing a plain text version of this content. The canonical link for it is here.
Posted to user@bigtop.apache.org by Konstantin Boudnik <co...@apache.org> on 2015/03/06 08:03:18 UTC

Annual feeback for project's Chair

Bigtop'ers

It's already been a year since I took over the role of the Bigtop PMC Chair. I
think it won't be an exaggeration to say It was an exciting and very eventful
year. I want to use this occasion and set up of what I hope might become a
tradition: a mix of annual report to the community and a feedback gathering
for the Chair and the PMC.

Here're the highlights of what was great and fun! 

- The number of the contributors to the project have increased significantly;
  there's 70+ contributors in the project
- The committer-ship base grew by more than 25% in the last year
- We have rolled a very important 0.8 release with a bunch of updates and
  fixes (190+ total)
- The next release is going to be pretty pivotal (no pun intended) as we have
  modernized both the development and user experience, and keep on working to
  improve how our users can enhance the stack, deploy and test it, and
  develop applications for it. So far 200+ commits made it into this version
- we are evidently leaning towards a better and faster ways of data
  processing, adding and supporting latest versions of Spark, Tachyon, Ignite
  (incubating), Kafka, and other interesting new technologies
- a few meetups and hackathons were organized
- there are a lot of signs that a significant number of commercial companies
  in the Big- and Fastdata space have high interest in the Apache Bigtop, as
  the ability to quickly and robustly deploy a standard fully open ASF
  data-processing stack becomes a critical requirement for many enterprises
- clearly, it becomes more fun to work with the stack as we adding support for
  Puppet 3.x and Hiera; latest Groovy runtime, and Gradle build system
- Bigtop presentations were accepted to both ApacheCon events in 2014;
- we had a super-successful 4 days appearance at SCALE13x including a full day
  workshop. It was incredibly gratifying to see that people who never had any
  experience with Hadoop, Docker, or Puppet can get from 'git clone' to a
  working custom-built cluster in about 2 hours!
- independent analysts estimate the user base of Apache data stack at over 51%
  of total users of any Hadoop derivatives. It might be bold to say, but I
  believe there are very few people out there who will try to build a cluster
  from scratch using just tarballs and shell scripts. I believe these 51%
  means us - Apache Bigtop.
  
  
All of above was possible because of you and your generous contributions of
the time, code, documentation, talent, knowledge and pizzas!
Thank you all very much!

And to the feedback part. For the benefit of this and all following PMC Chairs
I'd like to ask all to donate a few more minutes of your time and share your
thoughts on what could've been done differently, what and how the project Chair
can do better, what new things you would like to see happening in the project?

-- 
With best regards,
  Dr. Konstantin (Cos) Boudnik
  Apache Bigtop PMC Chair


Re: Annual feeback for project's Chair

Posted by Konstantin Boudnik <co...@apache.org>.
Thanks Roman - appreciate the feedback and will certainly be happy to get
Bigtop to be a foundation of ODP - seems to be a pretty exciting thing to do!

I guess The Chair could play a passive and/or just mediating role or be more
pro-active in the project development. And by that I don't mean only coding,
but also growing up the connections with the other projects, as you said.

Hopefully, in the next a couple of weeks recent dealing and wheeling in this
area will bring us some fruits, so stay tuned for new thread popping out on
dev@ ;)

Regards,
  Cos

On Tue, Mar 10, 2015 at 08:43PM, Roman Shaposhnik wrote:
> Hi!
> 
> As usual of these days -- I'm late to the thread. $DAYJOB
> is taking up all the cycles :-(
> 
> On Thu, Mar 5, 2015 at 11:03 PM, Konstantin Boudnik <co...@apache.org> wrote:
> > Bigtop'ers
> >
> > It's already been a year since I took over the role of the Bigtop PMC Chair. I
> > think it won't be an exaggeration to say It was an exciting and very eventful
> > year.
> 
> Indeed! And while you and I know have known each other for quite some time,
> I honestly mean the following: I was really happy to see how the project grew
> and progressed in the past year. There's a great deal of your personal sweat
> in there -- and I think it is something that needs to be recognized
> and appreciated!
> 
> > I want to use this occasion and set up of what I hope might become a
> > tradition: a mix of annual report to the community and a feedback gathering
> > for the Chair and the PMC.
> 
> +1 to that!
> 
> > Here're the highlights of what was great and fun!
> >
> > - The number of the contributors to the project have increased significantly;
> >   there's 70+ contributors in the project
> > - The committer-ship base grew by more than 25% in the last year
> > - We have rolled a very important 0.8 release with a bunch of updates and
> >   fixes (190+ total)
> > - The next release is going to be pretty pivotal (no pun intended) as we have
> >   modernized both the development and user experience, and keep on working to
> >   improve how our users can enhance the stack, deploy and test it, and
> >   develop applications for it. So far 200+ commits made it into this version
> > - we are evidently leaning towards a better and faster ways of data
> >   processing, adding and supporting latest versions of Spark, Tachyon, Ignite
> >   (incubating), Kafka, and other interesting new technologies
> > - a few meetups and hackathons were organized
> 
> And we need more. I'm guilty for not pushing for that hard enough on the Pivotal
> side of things. I promise to change that in the May-August time frame.
> 
> > - there are a lot of signs that a significant number of commercial companies
> >   in the Big- and Fastdata space have high interest in the Apache Bigtop, as
> >   the ability to quickly and robustly deploy a standard fully open ASF
> >   data-processing stack becomes a critical requirement for many enterprises
> > - clearly, it becomes more fun to work with the stack as we adding support for
> >   Puppet 3.x and Hiera; latest Groovy runtime, and Gradle build system
> 
> This is something that makes me personally very excited about the past year!
> 
> > - Bigtop presentations were accepted to both ApacheCon events in 2014;
> > - we had a super-successful 4 days appearance at SCALE13x including a full day
> >   workshop. It was incredibly gratifying to see that people who never had any
> >   experience with Hadoop, Docker, or Puppet can get from 'git clone' to a
> >   working custom-built cluster in about 2 hours!
> > - independent analysts estimate the user base of Apache data stack at over 51%
> >   of total users of any Hadoop derivatives. It might be bold to say, but I
> >   believe there are very few people out there who will try to build a cluster
> >   from scratch using just tarballs and shell scripts. I believe these 51%
> >   means us - Apache Bigtop.
> 
> A bit self-serving there was also ODP. My hope is that Bigtop is going to serve
> a fundamental role in there as well. Ok, may be it is something for
> *next* year ;-)
> 
> > All of above was possible because of you and your generous contributions of
> > the time, code, documentation, talent, knowledge and pizzas!
> > Thank you all very much!
> >
> > And to the feedback part. For the benefit of this and all following PMC Chairs
> > I'd like to ask all to donate a few more minutes of your time and share your
> > thoughts on what could've been done differently, what and how the project Chair
> > can do better, what new things you would like to see happening in the project?
> 
> Great question. The only thing that comes to mind is perhaps, somehow,
> more aggressively reaching out to sister projects in ASF (guys like
> Flink, etc.).
> 
> At the end of the day, a Chair of a PMC is more of a passive (but not
> passive-agressive! ;-))
> roles. If a Chair also leads by just coding away and providing
> feedback on JIRAs/MLs
> that's all I could ask for.
> 
> Thanks,
> Roman.

Re: Annual feeback for project's Chair

Posted by Konstantin Boudnik <co...@apache.org>.
Thanks Roman - appreciate the feedback and will certainly be happy to get
Bigtop to be a foundation of ODP - seems to be a pretty exciting thing to do!

I guess The Chair could play a passive and/or just mediating role or be more
pro-active in the project development. And by that I don't mean only coding,
but also growing up the connections with the other projects, as you said.

Hopefully, in the next a couple of weeks recent dealing and wheeling in this
area will bring us some fruits, so stay tuned for new thread popping out on
dev@ ;)

Regards,
  Cos

On Tue, Mar 10, 2015 at 08:43PM, Roman Shaposhnik wrote:
> Hi!
> 
> As usual of these days -- I'm late to the thread. $DAYJOB
> is taking up all the cycles :-(
> 
> On Thu, Mar 5, 2015 at 11:03 PM, Konstantin Boudnik <co...@apache.org> wrote:
> > Bigtop'ers
> >
> > It's already been a year since I took over the role of the Bigtop PMC Chair. I
> > think it won't be an exaggeration to say It was an exciting and very eventful
> > year.
> 
> Indeed! And while you and I know have known each other for quite some time,
> I honestly mean the following: I was really happy to see how the project grew
> and progressed in the past year. There's a great deal of your personal sweat
> in there -- and I think it is something that needs to be recognized
> and appreciated!
> 
> > I want to use this occasion and set up of what I hope might become a
> > tradition: a mix of annual report to the community and a feedback gathering
> > for the Chair and the PMC.
> 
> +1 to that!
> 
> > Here're the highlights of what was great and fun!
> >
> > - The number of the contributors to the project have increased significantly;
> >   there's 70+ contributors in the project
> > - The committer-ship base grew by more than 25% in the last year
> > - We have rolled a very important 0.8 release with a bunch of updates and
> >   fixes (190+ total)
> > - The next release is going to be pretty pivotal (no pun intended) as we have
> >   modernized both the development and user experience, and keep on working to
> >   improve how our users can enhance the stack, deploy and test it, and
> >   develop applications for it. So far 200+ commits made it into this version
> > - we are evidently leaning towards a better and faster ways of data
> >   processing, adding and supporting latest versions of Spark, Tachyon, Ignite
> >   (incubating), Kafka, and other interesting new technologies
> > - a few meetups and hackathons were organized
> 
> And we need more. I'm guilty for not pushing for that hard enough on the Pivotal
> side of things. I promise to change that in the May-August time frame.
> 
> > - there are a lot of signs that a significant number of commercial companies
> >   in the Big- and Fastdata space have high interest in the Apache Bigtop, as
> >   the ability to quickly and robustly deploy a standard fully open ASF
> >   data-processing stack becomes a critical requirement for many enterprises
> > - clearly, it becomes more fun to work with the stack as we adding support for
> >   Puppet 3.x and Hiera; latest Groovy runtime, and Gradle build system
> 
> This is something that makes me personally very excited about the past year!
> 
> > - Bigtop presentations were accepted to both ApacheCon events in 2014;
> > - we had a super-successful 4 days appearance at SCALE13x including a full day
> >   workshop. It was incredibly gratifying to see that people who never had any
> >   experience with Hadoop, Docker, or Puppet can get from 'git clone' to a
> >   working custom-built cluster in about 2 hours!
> > - independent analysts estimate the user base of Apache data stack at over 51%
> >   of total users of any Hadoop derivatives. It might be bold to say, but I
> >   believe there are very few people out there who will try to build a cluster
> >   from scratch using just tarballs and shell scripts. I believe these 51%
> >   means us - Apache Bigtop.
> 
> A bit self-serving there was also ODP. My hope is that Bigtop is going to serve
> a fundamental role in there as well. Ok, may be it is something for
> *next* year ;-)
> 
> > All of above was possible because of you and your generous contributions of
> > the time, code, documentation, talent, knowledge and pizzas!
> > Thank you all very much!
> >
> > And to the feedback part. For the benefit of this and all following PMC Chairs
> > I'd like to ask all to donate a few more minutes of your time and share your
> > thoughts on what could've been done differently, what and how the project Chair
> > can do better, what new things you would like to see happening in the project?
> 
> Great question. The only thing that comes to mind is perhaps, somehow,
> more aggressively reaching out to sister projects in ASF (guys like
> Flink, etc.).
> 
> At the end of the day, a Chair of a PMC is more of a passive (but not
> passive-agressive! ;-))
> roles. If a Chair also leads by just coding away and providing
> feedback on JIRAs/MLs
> that's all I could ask for.
> 
> Thanks,
> Roman.

Re: Annual feeback for project's Chair

Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
Hi!

As usual of these days -- I'm late to the thread. $DAYJOB
is taking up all the cycles :-(

On Thu, Mar 5, 2015 at 11:03 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Bigtop'ers
>
> It's already been a year since I took over the role of the Bigtop PMC Chair. I
> think it won't be an exaggeration to say It was an exciting and very eventful
> year.

Indeed! And while you and I know have known each other for quite some time,
I honestly mean the following: I was really happy to see how the project grew
and progressed in the past year. There's a great deal of your personal sweat
in there -- and I think it is something that needs to be recognized
and appreciated!

> I want to use this occasion and set up of what I hope might become a
> tradition: a mix of annual report to the community and a feedback gathering
> for the Chair and the PMC.

+1 to that!

> Here're the highlights of what was great and fun!
>
> - The number of the contributors to the project have increased significantly;
>   there's 70+ contributors in the project
> - The committer-ship base grew by more than 25% in the last year
> - We have rolled a very important 0.8 release with a bunch of updates and
>   fixes (190+ total)
> - The next release is going to be pretty pivotal (no pun intended) as we have
>   modernized both the development and user experience, and keep on working to
>   improve how our users can enhance the stack, deploy and test it, and
>   develop applications for it. So far 200+ commits made it into this version
> - we are evidently leaning towards a better and faster ways of data
>   processing, adding and supporting latest versions of Spark, Tachyon, Ignite
>   (incubating), Kafka, and other interesting new technologies
> - a few meetups and hackathons were organized

And we need more. I'm guilty for not pushing for that hard enough on the Pivotal
side of things. I promise to change that in the May-August time frame.

> - there are a lot of signs that a significant number of commercial companies
>   in the Big- and Fastdata space have high interest in the Apache Bigtop, as
>   the ability to quickly and robustly deploy a standard fully open ASF
>   data-processing stack becomes a critical requirement for many enterprises
> - clearly, it becomes more fun to work with the stack as we adding support for
>   Puppet 3.x and Hiera; latest Groovy runtime, and Gradle build system

This is something that makes me personally very excited about the past year!

> - Bigtop presentations were accepted to both ApacheCon events in 2014;
> - we had a super-successful 4 days appearance at SCALE13x including a full day
>   workshop. It was incredibly gratifying to see that people who never had any
>   experience with Hadoop, Docker, or Puppet can get from 'git clone' to a
>   working custom-built cluster in about 2 hours!
> - independent analysts estimate the user base of Apache data stack at over 51%
>   of total users of any Hadoop derivatives. It might be bold to say, but I
>   believe there are very few people out there who will try to build a cluster
>   from scratch using just tarballs and shell scripts. I believe these 51%
>   means us - Apache Bigtop.

A bit self-serving there was also ODP. My hope is that Bigtop is going to serve
a fundamental role in there as well. Ok, may be it is something for
*next* year ;-)

> All of above was possible because of you and your generous contributions of
> the time, code, documentation, talent, knowledge and pizzas!
> Thank you all very much!
>
> And to the feedback part. For the benefit of this and all following PMC Chairs
> I'd like to ask all to donate a few more minutes of your time and share your
> thoughts on what could've been done differently, what and how the project Chair
> can do better, what new things you would like to see happening in the project?

Great question. The only thing that comes to mind is perhaps, somehow,
more aggressively reaching out to sister projects in ASF (guys like
Flink, etc.).

At the end of the day, a Chair of a PMC is more of a passive (but not
passive-agressive! ;-))
roles. If a Chair also leads by just coding away and providing
feedback on JIRAs/MLs
that's all I could ask for.

Thanks,
Roman.

Re: Annual feeback for project's Chair

Posted by Roman Shaposhnik <ro...@shaposhnik.org>.
Hi!

As usual of these days -- I'm late to the thread. $DAYJOB
is taking up all the cycles :-(

On Thu, Mar 5, 2015 at 11:03 PM, Konstantin Boudnik <co...@apache.org> wrote:
> Bigtop'ers
>
> It's already been a year since I took over the role of the Bigtop PMC Chair. I
> think it won't be an exaggeration to say It was an exciting and very eventful
> year.

Indeed! And while you and I know have known each other for quite some time,
I honestly mean the following: I was really happy to see how the project grew
and progressed in the past year. There's a great deal of your personal sweat
in there -- and I think it is something that needs to be recognized
and appreciated!

> I want to use this occasion and set up of what I hope might become a
> tradition: a mix of annual report to the community and a feedback gathering
> for the Chair and the PMC.

+1 to that!

> Here're the highlights of what was great and fun!
>
> - The number of the contributors to the project have increased significantly;
>   there's 70+ contributors in the project
> - The committer-ship base grew by more than 25% in the last year
> - We have rolled a very important 0.8 release with a bunch of updates and
>   fixes (190+ total)
> - The next release is going to be pretty pivotal (no pun intended) as we have
>   modernized both the development and user experience, and keep on working to
>   improve how our users can enhance the stack, deploy and test it, and
>   develop applications for it. So far 200+ commits made it into this version
> - we are evidently leaning towards a better and faster ways of data
>   processing, adding and supporting latest versions of Spark, Tachyon, Ignite
>   (incubating), Kafka, and other interesting new technologies
> - a few meetups and hackathons were organized

And we need more. I'm guilty for not pushing for that hard enough on the Pivotal
side of things. I promise to change that in the May-August time frame.

> - there are a lot of signs that a significant number of commercial companies
>   in the Big- and Fastdata space have high interest in the Apache Bigtop, as
>   the ability to quickly and robustly deploy a standard fully open ASF
>   data-processing stack becomes a critical requirement for many enterprises
> - clearly, it becomes more fun to work with the stack as we adding support for
>   Puppet 3.x and Hiera; latest Groovy runtime, and Gradle build system

This is something that makes me personally very excited about the past year!

> - Bigtop presentations were accepted to both ApacheCon events in 2014;
> - we had a super-successful 4 days appearance at SCALE13x including a full day
>   workshop. It was incredibly gratifying to see that people who never had any
>   experience with Hadoop, Docker, or Puppet can get from 'git clone' to a
>   working custom-built cluster in about 2 hours!
> - independent analysts estimate the user base of Apache data stack at over 51%
>   of total users of any Hadoop derivatives. It might be bold to say, but I
>   believe there are very few people out there who will try to build a cluster
>   from scratch using just tarballs and shell scripts. I believe these 51%
>   means us - Apache Bigtop.

A bit self-serving there was also ODP. My hope is that Bigtop is going to serve
a fundamental role in there as well. Ok, may be it is something for
*next* year ;-)

> All of above was possible because of you and your generous contributions of
> the time, code, documentation, talent, knowledge and pizzas!
> Thank you all very much!
>
> And to the feedback part. For the benefit of this and all following PMC Chairs
> I'd like to ask all to donate a few more minutes of your time and share your
> thoughts on what could've been done differently, what and how the project Chair
> can do better, what new things you would like to see happening in the project?

Great question. The only thing that comes to mind is perhaps, somehow,
more aggressively reaching out to sister projects in ASF (guys like
Flink, etc.).

At the end of the day, a Chair of a PMC is more of a passive (but not
passive-agressive! ;-))
roles. If a Chair also leads by just coding away and providing
feedback on JIRAs/MLs
that's all I could ask for.

Thanks,
Roman.

Re: Annual feeback for project's Chair

Posted by Konstantin Boudnik <co...@apache.org>.
Thanks for the feedback Olaf! Let me add a bit of my own colors on the points
you've brought up:

2. I think we have decided to fix the issue with JIRA traffic via adding new
mailing list. In fact - thanks for reminding - I am doing just that right now!

3. I don't think this means that Bigtop failed in any sense. But I'd rather
not make any comments on the topic, considering how evidently sensitive this
issue is for a few members of the community.

The Twitter account - as any other media channels - will be used by all
project committers. And the policy to this effect is being hashed out as we
speak.

Thanks again!
  Cos

On Tue, Mar 17, 2015 at 10:30PM, Olaf Flebbe wrote:
> Hi Konstantin,
> 
> thank you very much about sharing background information about the PMC work
> and overall status.
> 
> Me personal 3 cents:
> 
> One of the most embarrassing thing about bigtop is its enormous pace
> forward, and it is a great pleasure to me to be part of it.
> 
> One of the most hindering things is the amount of noise the JIRA produces,
> with some important non-JIRA signals in between.
> 
> One of the irritating things is the ODP anouncement and its message: Bigtop has failed.
> 
> I was wondering about the @ASFBigtop twitter account, too. But the
> discussion following explained that to me, and I appreciate if we can reach
> a common understanding who can use it and how its to be used.
> 
> Greetings
>   Dr. Olaf Flebbe
> 



Re: Annual feeback for project's Chair

Posted by Olaf Flebbe <of...@oflebbe.de>.
Hi Konstantin,

thank you very much about sharing background information about the PMC work and overall status.

Me personal 3 cents:

One of the most embarrassing thing about bigtop is its enormous pace forward, and it is a great pleasure to me to be part of it.

One of the most hindering things is the amount of noise the JIRA produces, with some important non-JIRA signals in between.

One of the irritating things is the ODP anouncement and its message: Bigtop has failed.

I was wondering about the @ASFBigtop twitter account, too. But the discussion following explained that to me, and I appreciate if we can reach a common understanding who can use it and how its to be used.

Greetings
  Dr. Olaf Flebbe