Bounties for piecemeal work

13 vues
Accéder directement au premier message non lu

Avram Lyon

non lue,
4 févr. 2011, 01:44:4904/02/2011
à zotero-dev
Dear Zoterans,

Thinking about all the style and translator requests that hit the
forums, I often wonder whether it wouldn't be appropriate to design a
bounty system to encourage more development and perhaps attract more
developers who could contribute to these and other development
efforts. Translator and style development are pretty much orthogonal
to the core codebase development, and the core team is justifiably
devoted to improving the Zotero core.

There are philosophical and other issues to deal with, of course. We
wouldn't want users to feel obligated to pay for fixes, nor should
contributors feel that only paid work is worthwhile. And, of course,
there are lots of important areas for improvement that may be more
complex than what can be expressed in a bounty.

Still, I think that there is a pent-up demand for improvements in the
styles and translators definitely, and for other features as well, and
that some Zotero users are probably willing to pay for them. We can
safely assume that some research groups, labs, and universities are
saving thousands of dollars annually by using Zotero, and they may be
more comfortable contributing to concrete enhancements than simply
donating to the project in the abstract.

If bounties are implemented, I think that it would be appropriate for
the Corporation for Digital Scholarship (or something) to take a cut,
to support general operations and the project in general; indeed
administering and maintaining the contributions of bounty-motivated
developers would take some meaningful amount of staff time.

And I should say that I bring this up not because I want these
bounties, but rather because I want to see more people take
responsibility for the project, and more people learn to contribute to
it.

Maybe this whole idea is crazy or philosophically problematic, but I'd
be interested in what the community thinks of it.

Best wishes,

Avram

Bruce D'Arcus

non lue,
4 févr. 2011, 08:36:4004/02/2011
à zoter...@googlegroups.com

Well, I think you ask the right core question: how to "encourage more


development and perhaps attract more developers who could contribute

to these and other development efforts"? I've been asking this
question for years :-)

To me, the solution to the styles issue is obvious: make every user a
potential developer by investing in finishing and deploying this:

<https://bitbucket.org/csledit/csl-wysiwyg-editor>

If we do that right (and I'm a little worried about this, frankly), we
get the benefit of an even larger potential contribution pool, since
Mendeley and other projects have users that would contribute to it.

If a bounty might help do that; great. But I don't think we should
bother with bounties for writing the XML files by hand.

Not sure about translators. Obviously making them as easy as possible
to write (discussed here a fair bit, with a nice solution by Erik)
lowers the barrier to entry, and so potentially opens up more
contribution.

I'd also like to see Zotero move towards a more distributed
development model. Instead of putting everything into a single,
centralized, SVN repo, it'd be nice to start to spin off code
(citeproc-node, the server code, etc.) into separate projects hosted
at github, with separate wikis, issue trackers, etc., and to encourage
forking.

Bruce

Bruce D'Arcus

non lue,
5 févr. 2011, 10:19:3305/02/2011
à zoter...@googlegroups.com
A little addendum ...

On Fri, Feb 4, 2011 at 8:36 AM, Bruce D'Arcus <bda...@gmail.com> wrote:

...

> To me, the solution to the styles issue is obvious: make every user a
> potential developer by investing in finishing and deploying this:
>
> <https://bitbucket.org/csledit/csl-wysiwyg-editor>
>
> If we do that right (and I'm a little worried about this, frankly), we
> get the benefit of an even larger potential contribution pool, since
> Mendeley and other projects have users that would contribute to it.

Conversely, if we don't do this "right", one bad case scenario:

Feeling pressure from their users, Mendeley will deploy an incomplete
version of this editor client-side that will create styles that are
rather brittle (hard to debug, and often won't format correctly; that
effectively force different individual users to edit the style to get
it to work for their needs), and more-or-less not good enough to
share.

There will then be an explosion of styles of wildly varying quality,
and effectively two (or more) separate camps of CSL users, and Zotero
will still be burdened by a lack of a good, easy-to-use, style editor.

This is my worry.

I can certainly help with advice on finishing the editor, but I don't
have the time or skill to contribute actual code myself.

Bruce

Frank Bennett

non lue,
6 févr. 2011, 01:30:4706/02/2011
à zotero-dev
One useful project that I think might have some prospect of being kick-
started by a bounty arrangement is a plugin for Abiword. The Abiword
collaborative editing design is similar to Zotero, with the document
stored locally on each user's local machine, editable offline, but
capable of smart synchronization with copies held by multiple users
through a central sync service. With Zotero support, it could make a
really sweet collaborative editing platform.

Someone would have to undertake the initial work, though, and that
first step is a doozy. A user bounty pool on it just might help get
the ball rolling.

Frank

Avram Lyon

non lue,
6 févr. 2011, 06:33:4606/02/2011
à zoter...@googlegroups.com
2011/2/6 Frank Bennett <bierc...@gmail.com>:

> Someone would have to undertake the initial work, though, and that
> first step is a doozy. A user bounty pool on it just might help get
> the ball rolling.

I think that this is the way to go-- if anyone has experience with
specific bounty sites or systems, maybe we can gather advice and set
up a pilot site to see how it goes.

Avram

Bruce D'Arcus

non lue,
6 févr. 2011, 10:33:0806/02/2011
à zoter...@googlegroups.com
On Sun, Feb 6, 2011 at 1:30 AM, Frank Bennett <bierc...@gmail.com> wrote:

> Someone would have to undertake the initial work, though, and that
> first step is a doozy. A user bounty pool on it just might help get
> the ball rolling.

Is there room now to consider generalizing this sort of code so that
it's easier to support different editors?

Bruce

skornblith

non lue,
6 févr. 2011, 15:48:5606/02/2011
à zotero-dev
On Feb 6, 10:33 am, "Bruce D'Arcus" <bdar...@gmail.com> wrote:
> On Sun, Feb 6, 2011 at 1:30 AM, Frank Bennett <biercena...@gmail.com> wrote:
> > Someone would have to undertake the initial work, though, and that
> > first step is a doozy. A user bounty pool on it just might help get
> > the ball rolling.
>
> Is there room now to consider generalizing this sort of code so that
> it's easier to support different editors?
>
> Bruce
In what sense is it not general now? You should be able to write a
plug-in for Abiword without touching any Zotero internals as long as
you can implement these interfaces:

https://www.zotero.org/svn/extension/trunk/idl/zoteroIntegration.idl

Simon
Répondre à tous
Répondre à l'auteur
Transférer
0 nouveau message