Wikimedia is 8 years old. We have built one of the informational wonders of the world, and our mission closes with the essential phrase, "in perpetuity" – yet we have hardly begun to plan for the future.
Yes, we need an endowment fund, and should set one up now, even before we have planned a fundraiser for it (see i.). An endowment is essential to safeguard the future of any long-lived institution. Our annual budget is now many times our basic operating expenses, with over $8M a year in recurring expenses in the current 2009-2010 financial plan (for comparison, $8M is roughly the annual upkeep of the Clinton Presidential Library). Moreover, the plan calls for Foundation spending next year to grow faster than revenue - this is short-term planning. We must prepare for the long term at the same time.
- (i) - Some donors, particularly larger individual donors, may give to an endowment what they would not to a general fund. The endowment should be supported through careful outreach to large donors, and through policies allocating a portion of major unrestricted gifts to the endowment. The greatest investment required to set up an endowment is the definition of what it would support, something that we must do regardless as part of strategic planning. (Some people have questioned whether grantors would give money to an organization with an endowment. Most grants are project based, and not for general planning or support - Universities and other entities with large endowments receive some of the largest grants in the world.)
- (ii) - This is a topic for discussion with the Foundation's finance team. The endowment should be large enough to sustainably support the basic operation of the Projects (see iii. below), able to grow with inflation while supporting any needed central server farms and technical support with its interest, and of a size that we can raise. An initial target of $10M would match current expenses. One of the goals supported by the endowment would be to reduce the future maintenance cost of core services, so the endowment would not need to grow as rapidly as the projects. (While expenses have grown geometrically in years past, there has been no focus on separating core services from research, experiments, and other expenses.)
- (iii) - First we need to work together to definine a core set of services that define our mission. I would include:
- Reliable read/write access to the projects through the Wikimedia domains, and the hardware and bandwidth this requires;
- Reliable access to dumps and statistics
- Reliable backups of private and public data
- These services should be supported as robustly as possible : making it easier for third parties to help support them in-kind with their own time, money, and hardware; having someone keeping the infrastructure involved up and running - either with a dedicated team or with some other effective supporting network; maintaining and improving ways to find and access dumps, from a fileserver and mirrors to actively-seeded torrents of large files; and regularly generating statistics from available sources while protecting private data [which makes this a difficult task to let others share].
- Distributions from the endowment should support all of this, as well as efforts to reduce the future cost of maintaining them.