[OpenLayers-Dev] Entering OSGeo Project Sponsorship

[Christopher Schmidt]

On Sun, Oct 12, 2008 at 08:47:51PM +0200, Kristian Thy wrote:
> On Sun, Oct 12, Christopher Schmidt wrote:
> > I hesitate to provide sponsors direct control over the usage of funds.
> > The reason for this is simply that I would prefer to not be 'beholden'
> > to the sponsors for how we use funds directly.
> 
> Organizations with specific needs can still pay developers consulting
> fees to fix their problems. 

Sure.

> Indeed, short of having money in surplus and no specific needs (i.e. for
> reasons of altruism) I fail to see why I, as a corporation, would give
> money through OSGeo instead of just paying someone to scratch my itch
> directly (you know how just rubbing your back through your t-shirt is
> never quite satisfactory?).

Scratching specific itches is beneficial for when you have a specific
itch. However, the success of OpenLayers as a project can achieve more
than that: things that you would have never thought that you'd need
scratched might get done. Things like better memory management, better
speed (fewer CPU cycles), improved documentation all allow for the
project as a whole to succeed in ways that few specific organizations
have a strong individual desire for, but a more general fund might build
up enough support for that they get priority.

Additionally, although there are a large number of OpenLayers
consultants of all shapes and sizes, the number of people who are core
developers in the project is relatively small. With funds being managed
by the project directly, those developers are likely to play a key role
in the design and implementation of plans that are paid for by
sponsorship funds -- and that kind of difference can make the difference
between a great patch for you and a great patch for the project. (I've
seen enough good designs come out of the OpenLayers development team
that I'd generally say that stuff which is worked on by the core
developers will always be of a relatively high quality compared to
individual contributors.) Since most of the core contributors to the
project work for organizations where their consulting time is unlikely
to be something you can get ahold of trivially, giving money to the
project directly might have a more direct correlation on improvement of
the project.

Another thing is that there are many organizations which based a fair
amount of their success upon the success of the OpenLayers project. For
example, the Ordnance Survey Open Space project depends critically on
OpenLayers, since a large part of their product is simply OpenLayers
with some additional configuration wrapped up into it. With that being
the case, though OSOS doesn't have any specific needs from the OL
project (or at least, we aren't aware of them), they still depend on
openLayers continuing to succeed as a project -- supporting new browsers
as they come out, improving upon things which might be bugs, etc.
Organizations like this may not have any specific itch to scratch, but
also don't have the resources to replace OpenLayers should the project
fail, so it may be in their best interests to support the project
monetarily to ensure continued community success.

By acting as a sponsor, organizations also get to use this fact in their
marketing materials. In the same way that purchasing a sponsorship slot
at a conference gets your name and logo in a prominent place in
conference materials, purchasing a sponsorship for OpenLayers gets your
name and logo in what may be a prominent place on our website/in our
marketing materials. It means that people can recognize that your
organization directly supports the OpenLayers project. to some extent,
some of MetaCarta's efforts in OpenLayers as a contributor to the
project can be seen as a marketing cost: a full one third of the search
results for MetaCarta on Google are tied directly to the OpenLayers
project. (32,100 out of 102,000) When I meet people at a conference, the
thing I usually hear isn't "MetaCarta... you guys are the search engine
company, right?" It's "MetaCarta: You guys do OpenLayers!" Sponsorship
allows for organizations which don't have the same level of developer
resources to get the same kind of participation in a project that they
support. 

Lastly, I expect that any sponsors will get more control over future
direction of the project via their feedback than people who aren't
directly suppoting the project. Certainly, my current answer to most
questions about "When will feature $x be done?" Are "When you write a
patch for it." However, given sponsorship, I think that there is a
chance that features that a large enough number of sponsors are in favor
of to be given priority -- and if developers still aren't interested, we
have cash that might be able to be thrown at the project. And In the
same way that I'm more likely to spend time helping someone who has
demonstrated the ability to help themselves -- by patching code,
offering documentation, or simply contributing to the mailing list --
I expect there is a certain level of credit you get for being a project
sponsor that will indirectly improve the communications you get from
developers when you have questions. 

In conclusion, (hm, 5th grade essay time!) Sponsorship offers a number of
benefits to the sponsor in a more indirect way than scratching an itch.
By collaborating with other sponsors, tasks which are too large for one
organization to support directly can be undertaken and core developers are
more likely to be involved in developments. Sponsoring helps ensure
the success of the project as a whole -- impotant for organizations
which depend critically on OpenLayers. Sponsoring has a certain
marketing appeal, and can help to popularize supporters of the
OpenLayers project even if they can't contribute developmetn resources
directly, and sponsorship helps to allow for the determination of future
direction by providing a direct pipeline to the project steering
committee for sponsors to offer project direction feedback.

All in all, for many organizations these benefits are probably worth the
$3k that they get out of it. In fact, many organizations sponsor OSGeo
with far fewer reasons. OSGeo sponsorship does not get you any direct
control/benefit from any OSGeo project -- just your name on a web page/
marketing materials. But there are a number of OSGeo sponsors:
http://www.osgeo.org/content/sponsorship/sponsors.html who do it
primarily for the encouragement/maintainance of the OSGeo community.
Some might well simply describe this as 'altruism', but I think that if
you were to ask Frank Warmardam, what he would say is that you are
ensuring the future stability of a community of developers who have
helped you in the past. Combined with a little bit of altruism, such
things can go a long way :)

> That said, Chris, I still think it's good idea, and I have probably
> overlooked something that makes it more attractive. Edumacate me :)

I don't know if I did so successfully, but i've just laid out the pitch
I plan to use on people who I want to be sponsors, if the PSC decides to
go this route: I'd love to hear whether you (or anyone else) thinks the
above reasons are convincing :)

Regards,
-- 
Christopher Schmidt
MetaCarta