[OpenLayers-Dev] Entering OSGeo Project Sponsorship

[Frank Warmerdam]

Paul Spencer wrote:
> Frank,
> 
> In relation to OpenLayers pursuing sponsorship via OSGeo, could you 
> provide us with a brief summary of your experience with OSGeo 
> sponsorship program as it relates to GDAL/OGR?  I think this would help 
> us crystalize our thoughts on how OpenLayers would like to proceed.

Paul,

I'm not sure the experience will be all that similar for OpenLayers, but
I'll make a few observations with regard to GDAL, and perhaps editorialize
a bit.

First, please be aware that participating in the sponsorship program
involves some effort.  Beyond setting up initial documents, and soliciting
sponsors, it is expected that any project that participates in the sponsorship
program will produce an annual report to sponsors summarizing at least how
the sponsorship funds have been used over they year.   This isn't onerous,
but someone has to get this done.

Second, I would observe that Tyler takes care of the invoices for sponsorship,
and the collection of money.  So generally speaking once I have a verbal
(or email) agreement from an organization to sponsor GDAL I pass on the
contact details, the start date, and the sponsorship level to Tyler who
prepares the invoice, and takes care of contact with the sponsor.  This is
a huge help and without Tyler handling this professionally the whole effort
might well be too much for me pursue.

For GDAL it is considered allowed to sponsor for any amount of above
$500, but the invoicing and other hassles involved in small sponsorships have
resulted in me only actively pursuing sponsorships of $3000 or more (a
silver sponsorship).  I accept smaller sponsorships (gracefully I hope),
but I don't pursue smaller sponsors for annual renewals.

An items that has proven a bit challenging for GDAL is that sponsors are
permitted a logo and block of text on our sponsor page.

   http://www.gdal.org/credits.html

It has proven surprisingly hard to get an appropriately sized logo, and
appropriate text from sponsors.  Sometimes they don't respond.  Sometimes
they give me huge or otherwise inappropriate logos.  And they give me
highly variable text for inclusion.  So I've had to set fairly specific
rules for the maximum number of pixels in the logo (I take care of resizing
and confirm with the sponsor) and an upper limit on the descriptive text
(50 words at silver).  I still have some sponsors I've not been able to get
the logo and text from and they go uncredited.  In retrospect I wonder if I
would have been better off with a simplier scheme - at least perhaps no text.

As for actually soliciting sponsors, I have (I think) been responsible for
soliciting all the GDAL sponsors so far.  I generally don't try to hard, but
I do have to make some effort.  I don't recall any stepping up and volunteering
their interest just based on the sponsor solicitation page.  I've directly
approached a few organizations I knew were heavy users of GDAL and that I
know well.  I also will often mention GDAL and OSGeo sponsorship opportunities
to folks when I answer technical questions about the project and related
software if they look like they might be getting significant value from the
software.  I generally just approach companies under the assumption that
sponsorship is pretty hard to arrange for a government organization.  My
pitch normally goes something like:

"""
If you find you are getting substantial value from GDAL/OGR and you
would like to help sustain the project you might consider becoming
a sponsor:

   http://www.gdal.org/sponsorship.html

Sponsorship helps us fund bug fixing and other efforts to sustain
the project.  Sponsors get some degree of priority in getting
bug fixes, and receive credit on our credits page.  Let me know
if you might be interested, or have any questions about sponsorship.
"""

In a few cases I've offered to waive an hourly fee to fix some bug or
implement some small feature if an organization would like to become
a sponsor.  I've turned what might have been a personal invoice for
$500 into a $3000 silver sponsorship a couple times this way, though
generally these organizations might already have been well disposed
to sponsorship without my sweetening the pot.

I do avoid situations where sponsors are offered specific promises
in return for sponsorship unless it is something I'd be doing out of
my own time.  For instance I discussed possible sponsorship options with
a large company who seemed to want to tie it to efforts to incorporate
or do work on a particular library of theirs and I stressed that
sponsorship would need to be distinct from any efforts in that regard.

Well - thats the collection side.

The other side is making productive use of the money.

In the GDAL project we have only used the money for two things so far.
One has been funding a contract paid maintainer position which is focused
on bug fixing, though it also takes on some other project tasks.  We have
also used it to purchase project t-shirts handed out at the 2007 code
sprint.

We have been very lucky to have Mateusz Loskot available as our paid
maintainer.  He is a skilled self-starter and does not require much
direction, though he has responded well to my priority setting.  We have
a written up RFC describing how the paid maintainer role works:

   http://trac.osgeo.org/gdal/wiki/rfc9_maintainer

My main problem so far in this regard has been Mateusz' limited
availability, and in fact we have accumulated a significant amount of
money in the project fund as he hasn't had enough time available to
consume it all.

With regard to paid maintenance I will note a couple things about the
GDAL project.

1) I have deliberately kept the remuneration fairly modest for this role.
$20/hr or so. This is for a couple reasons.  First, it helps avoid a
sense amoung the unpaid developers that someone is getting rich off the
project.  The maintainer pay isn't enough to make anyone particular
jealous though it may still cause some existing contributors to "leave
it to the paid guy" for some things.  The other reason is I want the
sponsors to feel that their money is being stretched fairly effectively.

Actually, the third reason is that I like to see this maintainer role
as a sort of stepping stone that could lead the maintainer to more
lucrative private contracting related to the project.

2) I have deliberately avoided having the funding go to the existing
core team - especially to the PSC members who actually decide on the
dispersement of the funds.  In particular, I've never taken any GDAL
project money for myself.  I've been concerned about appearance of
self-dealing or our profiteering off the project.  I'm not sure how
reasonable these concerns are - but to me it was important that the
money be seen going to bring additional resources to the project
that wouldn't have been their otherwise and to avoid any appearance
of impropriety in the handling of the money.

--

Mateusz is no longer available for the maintainer role (he has accepted
a related full time job), so now we are having to search for a new
maintainer.  It isn't so easy to find someone with good dev skills,
a pre-apparent interest in GDAL and a willingness to work for modest
financial returns.  Our ability to find a new maintainer may well be
a major test for our sponsorship program.

Overall, the sponsorship program was one of the reasons I was interested
in launching OSGeo and having GDAL join it.  I wanted an orderly, transparent
mechanism where organizations could help fund the project development in
a way beyond funding specific new features.  I think it has gone fairly
well though I hope for a more stable maintainer role in the future.

I will say that I think using some of the funding for team building
activities is a good idea.  We did the t-shirts from this fund, and in
the future I could imagine using the funds to help support a GDAL
code sprint, or perhaps cool prizes for the best bug report, best
patch etc.

--

One other point I'll mention is that 25% of the sponsorship funds
collected go to fund general OSGeo activities.  I have not heard
any complaints about this in the context of the GDAL project (from
sponsors or developers) though I have heard resistance on this from
other projects.  I think it is important to see this as supporting
the underlying infrastructure and community that OSGeo provides to
projects.

But if you see the project sponsorship program as a way to raise
money to fund specific new developments then this cut may seem
painful.  I do think it is important to *not* treat project
sponorship as a mechanism for an organization to fund a specific
development.  Direct contracting with a consultant is still better
for that sort of thing.

I hope this is somewhat helpful.

Best regards,
-- 
---------------------------------------+--------------------------------------
I set the clouds in motion - turn up   | Frank Warmerdam, warmerdam at pobox.com
light and sound - activate the windows | http://pobox.com/~warmerdam
and watch the world go round - Rush    | Geospatial Programmer for Rent