Re: OpenSource: build or buy? (fwd)

Kragen Sitaker (
Sat, 19 Dec 1998 23:53:11 -0500 (EST)

Thought y'all might enjoy this.

<>       Kragen Sitaker     <>
"Why are you withholding me?" -- name withheld  "Oh... And dig this:  I
am a fish.  'Nuff said." -- Joe Blaylock (no further explanation)
These are the denizens of the CLUG mailing list.  Their five-year mission:

---------- Forwarded message ---------- Date: Sat, 19 Dec 1998 14:55:31 -0500 From: Frank Hecker <> To: Brian Bartholomew <> Cc: Subject: Re: OpenSource: build or buy?

Brian Bartholomew wrote: > Writing OpenSource programs is neat and all, but perhaps the community > should just band together and *buy* the best-of-breed Unix programs? > Ten million users times any nonzero number of dollars is a very > interesting sum. Perhaps there is a company on this list that could > buy these programs and resell them, in exchange for a profit added to > the resale price. After n million people buy, the programs are GPL'ed.

I can think up many possible reasons as to why this sort of thing wouldn't work, and I'm sure most of the people on this list could do likewise. However, rather than engaging in that sort of effort I think it's a more interesting exercise to try and find comparable initiatives in other areas, and see what it took for them to be successful.

One that comes to mind almost immediately is the Nature Conservancy (TNC), the group that purchases tracts of land for preservation as nature sanctuaries. Their underlying concept is almost identical to what you propose: identify areas (land in their case, code in yours) worth preserving for the benefit of the general public (as opposed to being commercially exploited or otherwise benefiting only a few), and then find ways to buy what you want to preserve. (There are also other ways to do this than just buying something; for example, you can arrange swaps of some sort.)

I think the example of the Nature Conservancy repays close study. Based on a brief review of the TNC web site <> combined with my knowledge of the ways of nonprofit organizations (a side effect of living in the Washington, DC, area), I can pretty confidently state the following:

* What you want to do is in fact possible. TNC is an existence proof for the general concept of converting private property to public property through nonprofit nongovernmental efforts, and I don't believe that there's any essential barrier to applying the same concept to the software field.

Moral: Don't let people discourage you if you really believe in this.

* However, don't underestimate the difficulty involved in getting something like this up and running. TNC has helpfully provided a history of the organization:

Note that five years elapsed from the initial proposal of "direct action" to preserve undeveloped land (1946) to the formal creation of TNC as a nonprofit organization to do this (1951). It took TNC another four years to complete its first land purchase (in 1955), and it wasn't until 14 years from its founding, and almost 20 years from the initial discussions, that TNC secured major monetary support in the form of a foundation grant (1965) and could become a truly effective organization.

Moral: To get something like this done, you need to find a group of people who will dedicate themselves to this for the long term (5 years at a minimum, 10+ years if necessary). And of course, you also need to decide if you want to be one of those people yourself.

* It's extremely difficult to base an effort of this type purely on individual action and donations; you need some serious money to kick the effort into a higher gear, and the places to find that money are corporations, rich people, and nonprofit foundations (which in turn are typically started by rich people, or in some cases by corporations). TNC's first foundation grant (mentioned in the previous paragraph) totaled $550,000 in 1965 (equivalent to a few million dollars today); in 1983 they got a $25,000,000 grant from another foundation. Amounts such as these are what I mean by "serious money".

Also note from the TNC FAQ

that even with a membership base of almost a million people, a third of TNC funding still comes from foundations or corporations.

Moral: Look for rich people or corporations who have extra cash lying around and who have a strong interest in promoting the cause of open source. (But also see the next point.)

* To make an effort like this successful, you need to have people who are professionals at running nonprofit organizations, especially at doing nonprofit fundraising, and _especially_ at raising funds from rich people, corporations, and foundations. If you're pursuing rich people, you need not only to a) find ways to get to them (not all that hard in the software world, you've probably got two degrees of separation at worse) but you also need to b) convince them to part with serious money (say, $100,000 on up) for the cause _and_ c) convince them that you are the person or group to whom they should give that money. Task (b) is hard, especially considering that the high-tech field has a very poor record when it comes to charitable giving of any sort, and task (c) is even harder, especially if you're starting from scratch and you and your group have no positive reputation built up yet.

If you pursue corporations, you'll need to do the same sort of selling job, and if you pursue foundations you'll in addition have to prepare formal grant applications. All this is far from trivial to accomplish. If you then add in all the problems of recruiting and retaining individual donors, you'll definitely need a full-time professional staff of several people at some point.

In this light, note that TNC spends about 15% of its incoming funds on administration of the organization; this is actually quite good for a nonprofit organization. (Some organizations, including some well-known groups with fairly good reputations, have administrative expenses as high as 50 to 75 per cent.) If you're interested in what it takes to run a successful large nonprofit organization in terms of budget, you might look into ordering a copy of TNC's current financial disclosure statement:

(By the way, nonprofit organizations in the US -- so-called 501(c)(3) groups -- are required to file such statements every year.)

Moral: If you get a group that's serious about this, have them try and get advice from someone who knows what they're doing re running and funding nonprofit initiative.

A good place to find such people is at your nearest local chapter of the National Association of Fund Raising Executives (NSFRE) <>, the main professional organization for people doing nonprofit fundraising. (Their web site also has lots of other useful information and pointers to resources.)

Also, if for whatever reason you have bad experiences with nonprofit fundraising efforts in the past (e.g., direct mail you've gotten, telemarketing from nonprofits, people who showed up at your door, whatever), check out the NSFRE's code of ethics and "donor bill of rights" for the way it's supposed to be done:

Final moral: I don't believe that there's any "royal road" to achieving the goal of expanding the pool of libre software, whether by acquisition (as you suggest) or otherwise.

I think that the level of effort in starting a successful project to do what you propose is roughly equivalent to the level of effort needed to form a traditional software startup, produce a plausible business plan and product prototype, and obtain first-round VC funding. That's both bad and good: It's bad in the sense that just discussing this proposal on the fsb list and elsewhere is never going to lead to anything concrete.

But it's good in the sense that people found software startups all the time, with some of them actually succeeding, and if there's at least a moderately large pool of people with equivalent levels of dedication and competence that they want to devote to this type of project, then I think they have a reasonably good chance of getting it done.


Frank Hecker          Pre-sales support, Netscape government sales