And yes, improving people’s lives is an extremely broad goal. There is no magic formula, but there is a heck of information to be collected and analyzed. If we’re curing malaria, how much do different strategies cost and what have been the observed effects on infection rates? If we’re improving education, what’s been tried, and has it worked (this could be answered in a variety of imperfect ways, from looking at attendance to test scores to income 5 years out of school)? What becomes of a person who gets their cleft repaired? These are all things I wish I knew, and I imagine foundations that work in these areas have been working on them for a while. The frustration is not being able to see what they’ve done.

Again, I understand your pragmatic/incremental considerations. You’re focused on what’s possible, as you should be; I’m talking about what’s desired.

I am a bit puzzled as to what you mean by “what sorts of activities improve people’s lives the most” - that could mean a huge variety of types of information, even within the filter of philanthropy sector/foundations.

But you’re right, the “tremendous amount of work [foundations do] deciding whom to give money to” is not up on our website - that’s what we’re building our wiki for.

The page you linked to - our Papers & Reports page - is a reasonably accurate indicator of how much of that information is currently shared (that’s the page where we collate and list everything we can find). The wiki, of course, will be structured quite differently, be more user-friendly and have a broader and more comprehensive scope of information.

As I said in the comment above, this isn’t something that can happen overnight. The ‘culture of privacy’ is just one of the many issues we need to take into account while we’re building the data (relatively) “at hand” into the most effective tool we can.

None of our work on the wiki is open to the public as yet; we’re still building the structure of it at the moment, as well as securing funding so that we can follow the project through. (Philanthropy Australia is a nonprofit organisation as well, though people tend to assume otherwise considering our name and where we sit in the sector!)

Thanks for the discussion.

I didn’t see where else to look for this sort of information, and I didn’t see how to contact you directly, so if you could email me with your thoughts, I’d appreciate it.

For the record, I think the arguments you do list for keeping information secret are uniformly horrendous. But it seems we agree on this.

@Beth: I wouldn’t ever want to “rule out” what technology can do in the future, but the kind of summary I’m talking about seems to be right in the weak spot of current computers’ abilities. Computers do exactly as they are told. If you know in advance which parts of a paper are most important, computers can pull them out for you. But to go through 100-page papers, phone calls, and websites and pick out the important parts, pertaining to the broad/fuzzy goal of helping people - that requires human judgment.

Keep in mind that the early Internet tended to try to have machines do all the work. Early search engines used pure computer logic to pick out the relevant sites - and they were awful. Google revolutionized this not by perfecting the algorithm, but by rewriting it so it depends entirely on the judgment of humans - and most new projects since have been focused on connecting/coordinating *people’s* judgments of what’s important.

That leads to the natural question of whether GiveWell could be done Wikipedia-style, as a social media project rather than a research project. This is how I originally conceived it, but after seeing how much persistence and work (and often money) it takes just to get basic information from a nonprofit - and running a couple experiments - I concluded that this isn’t viable. It can’t work without dedicated researchers. Doesn’t mean the researchers can’t be part-time and plentiful, and it doesn’t mean that the project has to be monolithic - our pages have as many spaces for open discussion as we’ve been able to cram in. But “build the structure and the information will come” can’t work for something like this, where the information largely isn’t even available. I talk more about why I think this in The GiveWell Blog here.

I appreciate the thoughtful and thought-provoking comments.

The privacy issue is a difficult one. The issues involved with private foundations and social media are not dissimilar to corporations and social media, I think; in that corporations are wary of blogging etc (in public, at least) ostensibly because it compromises public image.

Private foundations seem to have similar reluctance to put faith in social media when it compromises their privacy - whether because they believe what they do with their money is their own business, or the projects they fund are controversial, or they don’t want to be flooded with applications for funding… there are many reasons.

But my point is not so much that we should be supporting this attitude, but that the use of social media isn’t just a matter of translating the occasional behind-closed-doors discussion onto an internet medium. We’re keen to make the information we’re gathering and fostering as accessible as possible to the parties who will benefit from it the most.

However, considering the existing attitude – or, I suppose you could say ‘culture of privacy’ – of foundations, we’re aware asking for information from them + making the information completely public = them not disclosing any information to us. So we’re taking baby steps, trying to change the culture gradually – ease them into it rather than shutting down potential positive outcomes by doing too much too fast!

Already in the past year or so we’ve come leaps and bounds by following this course – not that long ago, our website was 10% visible to the public, 90% protected for our Members. Since we’ve reversed those figures, opened up the information on our website and changed how we display ‘protected’ information (and how we choose what is protected information – rather than research documents, now only sensitive info like minutes and contact details), we’ve had immensely positive feedback in the form of our web stats, feedback from Members, and feedback from non-Members, on a local and global scale.

So we’re getting our Members to trust us – but there is still some way to go. Now they’re enthusiastic about what we can do with information, and are very much on the way to trusting us to handle their sensitive information, initially within a relatively closed networking environment (we have about 300 members; trusts, foundations and corporations, of varying scale).

This is a huge step! We’re getting these foundations excited about social media (our wiki, and its collaboration features), and eager to use it. Once they understand more about it through their own experience, we hope that they can better understand who can benefit (and how) from the data that they’re helping us build.

And, personally, that’s one of the things that endears me most to the internet medium, let alone social media – the information and level of access is changeable, evolving. It is not the most difficult thing in the world to gradually modify privacy levels on social media tools! Though those decisions, of course, are not something we can make without the consent of the foundations in question. My hope is that by educating them in social media, the culture will change and making information more available will become something they see the value of themselves – rather than something we force them into.

What’s funny, to me, in nonprofits worrying about this is that people — their donors, their employees — are saying whatever they want. The donors just don’t know about it. And, of course, if they don’t know about it, they don’t have to do anything about it. So that’s what I think the real worry is. What do I do when to respond to what people say.

Do you think that the summary you are talking about can be automated or must it be human generated?

The nonprofit sector is, I think, unique in that there is no good reason for people to hide what they know about what works. If there is a reason I haven’t thought of, please let me know.

Also, regarding whether we will be the “only ones” for long, we may not be, but there is something specific holding back the flood: it is a ton of work. In the end, GiveWell really isn’t about “social media.” We’re using social media because the technology is here and it’s helpful, but if it weren’t here, we’d make a newsletter and Xerox it for everyone we know.

What really defines GiveWell is collecting the information that nonprofits are surprisingly hesitant to share, and centralizing and coherently summarizing it. Information-sharing tools are great, but they don’t magically make this happen. There’s a lot of work to be done, whether the results go on a wiki or a blog or a stone tablet.

http://philanthropy.blogspot.com/2007/01/social-network-philanthropy-redux.html