couchdb-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Benoit Chesneau <>
Subject Re: [PROPOSAL] Use Influitive to help with marketing
Date Sun, 06 Oct 2013 16:25:42 GMT
On Sun, Oct 6, 2013 at 5:54 PM, Noah Slater <> wrote:

> > This system is quite odd and artificial.
> I don't understand what is "artificial" about it. Is JIRA also artificial?
> We're defining a list of things we would like help with, and then inviting
> people to help us with them. There's nothing artificial about that. We're
> not asking people to say or do things they don't believe in. We won't be
> asking people to write fake reviews. We will be asking people to review us
> honestly. (For example.)
I can see the syndrome of the "best employee of the month" behind the
system of reward. And I am definitely uncomfortable with that.

There is no reward system in Jira. Jira items are neutral. People can vote
(see the vote link on the right) on the tickets. What would be good is to
report regularly about waiting tickets in an automated way though.

> Plus, who will choose these items?
> Us! :)
> Well, initially me. But if you wanna help out, just help out. We will have
> full control.
> > Why not simply having a TODO list, aka roadmap in which people can simply
> pick an item on which  they think they can can help?
> That's exactly what I'm proposing! :) That's what Influitive helps us with.
> JIRA is a good TODO list for the code, and Influtive will make a good TODO
> list for marketing activities. Different tools to suit different parts of
> the project.
> > There is something really odd an artificial in giving a reward to do such
> thing
> Unusual for an open source project, maybe. But not unusual for a company.
> So it's an experiment, I guess. Perhaps it works, perhaps it doesn't.
> There's no worry here. If it doesn't work, we just stop using it. And I
> will find another tool that does.
> There is an ulterior motive here for me wanting to reward our passionate
> users with t-shirts, and swag. People wearing CouchDB t-shirts are like
> walking billboards. Free advertising!
> But for the RT stuff, or even just #FF, I think it's nice to recognise
> people who contribute to the project. I mean, really, we should be doing
> that with patches, and code. and stuff too!

You already recognised them when you take their patch in consideration and
discuss items with them. And to be honest we are not quite good at that
right now. Some patches land for a long time in the tracker, some
discussions about features aren't followed, ... If we want to recognise
better the people, we should first improve that.

About the "free advertising" I am all to have a "store" link like the one
on ubuntu:

It helps to create an identity.

And I don't see why sone couldn't be offered to some peoples from time to
time. But why not making it spontaneous instead of having it organised in a
"reward" system ? How would you make the distinction of the guy simply
retweeting a thing  and the other one giving a patch. Both helps a project
but in a different manner. How would you recognise the guy that do a lot of
support on the ml vs the other one. What about the guy busy, working on the
things but not participating to such system. How will you thank him?

Such limits are really hard to fix and I am not sure I want to fix them nor
I want to see someone fix them. But I am OK to discuss it anyway.

I understand that you want to experiment and I'm OK with that.  But I would
like to define it a little more the limit of it. In particularly to answer
to the following questions:

- How to ensure this reward system is optional?
- How to ensure that people that not participate do not feel unconsidered
by other members of the community because not rewarded by the Influtive
- What will define the success or failure of the experimentation?  What is
the deadline of the experimentation?

If we can find consensus around the responses to theses questions I am
definitely +1 to start it.

- benoit

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