The Future
Noah Slater
As some of you may have already read, Damien Katz, Apache CouchDB’s
original developer, has publicly announced that he intends to focus his
time exclusively on developing other products for his company. Damien has
had very little involvement in the CouchDB project for a year or more now,
so, for many people, this is confirmation of what they already knew. We’re
sorry to see him go, and wish him all the best in his new direction. We all
owe him a huge debt of thanks for all the hard work he has put in over the
years, and for his original vision of the project.
Our biggest strength has always been the breadth and depth of our community
of developers and users. In the very near future, we’ll be voting in a new
committer, appointing a new PMC member, sprucing up the website, and making
a major new release. We’re happy to confirm that Cloudant has also publicly
made a commitment to help contribute BigCouch to the CouchDB project.
BigCouch, for those of you who have not had the pleasure of using it
already, is a fault-tolerant, horizontally scalable clustering framework
purpose-built for CouchDB.
Here’s to our future!
Relax,
Noah Slater
Mikeal Rogers
On Jan 6, 2012, at January 6, 20129:08 AM, Noah Slater wrote:
> Dear Community,
>
> As some of you may have already read, Damien Katz, Apache CouchDB’s
> original developer, has publicly announced that he intends to focus his
> time exclusively on developing other products for his company. Damien has
> had very little involvement in the CouchDB project for a year or more now,
> so, for many people, this is confirmation of what they already knew. We’re
> sorry to see him go, and wish him all the best in his new direction. We all
> owe him a huge debt of thanks for all the hard work he has put in over the
> years, and for his original vision of the project.
Yes, Damien has been pretty inactive for over a year, and has that gone unnoticed? How many releases in that time? How much new adoption?
For the last year CouchDB has suffered a vacuum of leadership. This is a thriving community, it can survive a loss like this, but only if people(s) step up and take responsibility. Having a lot of discussions and process around decisions leaves the flavor of meritocracy in everyones mouth but the viewable result of this has been a gigantic loss in momentum as the projects publicly viewable changes and advances have been close to zero.
>
> Our biggest strength has always been the breadth and depth of our community
> of developers and users. In the very near future, we’ll be voting in a new
> committer, appointing a new PMC member, sprucing up the website,
Sprucing up the website. *sigh* This is a great example, I've been hearing this for over a year with no noticeable change.
There have been countless false starts and arguments about direction ending in a stalemate. Maybe we all think this kind of discussion is just a healthy bi-product of a consensus based process but to the public it looks like it's output: nothing. Nothing piled upon nothing, and we've gotten comfortable with that.
> and making
> a major new release. We’re happy to confirm that Cloudant has also publicly
> made a commitment to help contribute BigCouch to the CouchDB project.
> BigCouch, for those of you who have not had the pleasure of using it
> already, is a fault-tolerant, horizontally scalable clustering framework
> purpose-built for CouchDB.
I challenge the idea that this is a positive development. Not because BigCouch isn't awesome, it definitely is, but the Couch community is much larger and more diverse than Apache CouchDB and many of the fringe projects have thrived without the Apache process to hold them back while CouchDB struggles to move forward in spite of it.
>
> Here’s to our future!
I'm sure I'll get lots of upset emails and some of those people are still clinging to the idea that this process is more important that being productive while at the same time in their other projects see frequent releases and contributions because they take responsibility for them.
I rely on Apache CouchDB tremendously and the reason I picked it a long time ago over alternatives was because the community was so great. Other projects, and even some of my own, have thrived after the loss of their creator because people were clamoring to step up and take responsibility. But what's happened here is that the leader has become fed up with the process, publicly pronounced it and moved on, and not only is everyone sticking with business as usual and not challenging the process, nobody is wiling to take responsibility for the projects future either.
>
> Relax,
I'm trying, I really am.
>
> Noah Slater
Adam Kocoloski
- Yes, we've talked about redoing the website before. It needs to be done, we're going to do it.
- Yes, the release schedule has slowed since 1.0; we're going to fix that by releasing 1.2. Releasing database software is serious business. We're constantly trying to streamline the process. One doesn't have to look too hard to see that the project has a constantly evolving codebase and a thriving developer community:
http://www.ohloh.net/p/couchdb/analyses/latest
- I'm personally very excited to be bringing BigCouch to Apache instead of splitting my time between two forks of a project. I'm taking a share of responsibility for the health and welfare of CouchDB and backing it up with working code.
It takes time to steer a ship like this. Noah's got us pointed in the right direction, let's not try to derail things before they've had a chance to develop. Cheers,
Adam
Noah Slater
At the core of all of them is one central point. This has nothing to do
with our consensus based approach. I find it frustrating that Damien
mentioned the Apache consensus based model as a concern, when to the best
of my knowledge he has not had any problems getting any feature he wants in
to CouchDB. I find it even more frustrating that you've used this as an
excuse to trot out your favorite hobby horse. Most people are aware of your
problems with Apache, but I don't think it's very helpful to bring them up
now, when they are tangental at best.
I do, however, agree that CouchDB could do with a bit more leadership. I
think we could do with being a bit more bold. Back in the old days, Jan
used to say it was easier to ask forgiveness than it was to ask permission.
I think that these days, we're too busy asking for permission most of the
time. I agree that there needs to be an attitude change within our ranks.
Be bold people! That's what I'll be doing with the website, shortly.
Jim Jagielski
Certainly the story being presented to the unwashed masses
is that Damien is still the driving force behind CouchDB,
that the nasty overhead of trying to build s/w via consensus-
based collaboration is damaging CouchDB (and in fact, has
damaged it so badly that Damien needs to seek more "energetic"
pastures elsewhere) and that CouchDB is heading towards the
bit bucket. It would be great to get out something that
refutes all those points.
Mikeal Rogers
Sure, my experience makes me believe, and I guess Damien's frustrations echo, that part of this problem is Apache process. You may not agree with that, but if you don't, and you admit that there *is* a problem, then what is it?
The first step in CouchDB's 12 step program to get get it's mojo back has to be admitting that it has a problem.
As much as it's a problem, I don't think that lack of leadership is the real problem, I think it's a symptom. I'm glad to see you say that we should get up and do more.
If you're stepping up to take on more leadership, say it, scream it! Own it!
-Mikeal
Jan Lehnardt
On Jan 6, 2012, at 22:21 , Jim Jagielski wrote:
> It might not be a bad idea to maybe blog something about this...
> Certainly the story being presented to the unwashed masses
> is that Damien is still the driving force behind CouchDB,
> that the nasty overhead of trying to build s/w via consensus-
> based collaboration is damaging CouchDB (and in fact, has
> damaged it so badly that Damien needs to seek more "energetic"
> pastures elsewhere) and that CouchDB is heading towards the
> bit bucket. It would be great to get out something that
> refutes all those points.
Jim, the committers discussed this (in a private email thread)
and opted to respond with positive action rather than clarifying
posts.
I just announced a new committer, a PMC addition is in the
works and the new major version 1.2.0 got dropped into the
release procedure. We hope that the news of all these events
(and many more to come in the near future) will create the
desired public perception.
Cheers
Jan
--
Adam Kocoloski
Adam
Adam Kocoloski
>
> On Jan 6, 2012, at 22:21 , Jim Jagielski wrote:
>
>> It might not be a bad idea to maybe blog something about this...
>> Certainly the story being presented to the unwashed masses
>> is that Damien is still the driving force behind CouchDB,
>> that the nasty overhead of trying to build s/w via consensus-
>> based collaboration is damaging CouchDB (and in fact, has
>> damaged it so badly that Damien needs to seek more "energetic"
>> pastures elsewhere) and that CouchDB is heading towards the
>> bit bucket. It would be great to get out something that
>> refutes all those points.
>
> Jim, the committers discussed this (in a private email thread)
> and opted to respond with positive action rather than clarifying
> posts.
>
> I just announced a new committer, a PMC addition is in the
> works and the new major version 1.2.0 got dropped into the
> release procedure. We hope that the news of all these events
> (and many more to come in the near future) will create the
> desired public perception.
>
> Cheers
> Jan
> --
Yes, that too. This statement and The actions and code to back it up are the way forward. Cheers,
Adam
Mikeal Rogers
About half noted Koco's response and Damiens inactivity over the last year and the other half were people flipping out and even asserting things that weren't in Damien's blog post (like that everyone at Couchbase is divesting in Apache which is clearly untrue).
I think you guys should print "Reports of my death are greatly exaggerated." in the welcome message for 1.2 :)
-Mikeal
Jim Jagielski
Jim Jagielski
the PRC is available to help if need be... Both Bad and Inaccurate
news always travel faster and deeper than Good and Accurate ;)
Mahesh Paolini-Subramanya
To put this in Marketing-Speak, in the Technology Adoption Lifecycle, we got to the 'Early Adopter' stage, but never managed to quite make it across the chasm to the 'Early Majority' stage.
To put this differently, we got to a place where the product was good enough to satisfy the Visionaries - who 'got it', and worked around any pesky inconveniences that popped up because they 'got it'. However, it was never good enough to satisfy the Pragmatists (who, face it, don't want to have to work around stuff)
(Where am I going with this? Hang on...)
So, to Mikeal's point about leadership - its not that there is any lack of leadership - there is plenty of it. Its just that leadership comes in many forms, and in the growth of any kind of disruptive product (which this was / is), the type of leadership necessary changes as it moves along the Technology Adoption Curve.
A quick perusal of users/dev shows that there tend to be many discussions about point-features, and most of these tend to get resolved fairly easily. So leadership in the technical/technological sense - Yes.
But thats pretty much where things tend to stop - i really don't see any person (or group of people) blasting through town with a banner saying Excelsior! In short, Vision Leadership does tend to be somewhat in short supply.
All that said, I don't think this is really any kind of major crisis. To paraphrase Benoit, its all noise, and not very interesting noise at that. There is clearly a group of people doing 'stuff', and quite happily doing it at that. Get some releases out, talk about whats in the releases, and just as importantly, what these features do and who is using them, and a lot of the vision will pretty much make itself self-evident.
Cheers
Mahesh Paolini-Subramanya
That Tall Bald Indian Guy...
Google+ | Blog | Twitter