UPDATE (3/14/2009 1:07 AM EDT): Check out the blog post from Community Voice Mail addressing my concerns. Oddly, the blogger there claims to have left comments here but I don’t see anything. Just so you all know, I don’t moderate comments except if you put more than one external HTML link in your comment as that’s a sign you may be a spammer. On to the original article…

Launched today, Google Voice is the newest update to Grandcentral,
a service I’ve used since near its inception. It generates a universal phone number that ties together various services such as all your other phone numbers, voicemail, VOIP, SMS and even your Gmail contacts. It’s seamless, it’s convenient, and I love it. The tech press points out that Google Voice is a direct challenge to other established for-profit services such as eBay’s Skype, Vonage and Comcast. They missed out its effect on one nonprofit, Community Voicemail, that offers free voicemail for nonprofit clients.

In the past, I was responsible for handling the technical side of Community Voicemail for New York City. It’s admirable goal was to provide free voicemail accounts for homeless clients throughout the city. At its peak, we had thousands of voicemail accounts being routed out of the office of the Coalition for the Homeless. Over time, it slowly became a burden as the hardware slowly died and then it was down for weeks when replacement hardware was shipped and installed. And then Grandcentral arrived. You could GIVE your clients free voicemail. When it was bought out by Google, Grandcentral unfortunately stopped giving out new accounts. Community Voicemail got a reprieve.

During my time working with them from 2002-2007, there was never an attempt by Community Voicemail to change their client-server delivery method. There was no attempt to build an open API, widgetize it, integrate it with social networks, indeed there wasn’t even a Web client through which you could provision services. Delivery of software for a nationally unified CVM that wouldn’t require direct provisioning of local telephone numbers by a nonprofit was promised but never delivered. You had to have Cisco equipment on-premises just to even start.

There is no doubt that a lot of good was done by CVM before Grandcentral showed up on the scene. Many clients attested to its usefulness. However, Community Voicemail is made redundant in the face of publicly available free voicemail. Indeed, Grandcentral actually offered homeless people in San Francisco free voicemail just like CVM. In 2006, the writing was on the wall and I counseled the Coalition to shut down the New York CVM service and we did. I’m upset that Grandcentral shut down giving out accounts soon afterwards but the launch of Google Voice today ultimately confirms my intuition about voice telephony. Voice is low-bandwidth and the processing of it is hardly more complicated than say email or even IM. It’s so cheap from a data processing point of view that it will be offered for free. Google Voice is just another milestone to a free voice plan for all.

As I said in 2006:

In a deep way, this really gets to the heart of what any non-profit’s true mission is which I believe is to render itself obsolete. If the private sector in the guise of Grandcentral is providing free voicemail, then shouldn’t every non-profit that is currently providing free voicemail in a very serious way ask: “Should we shut down our free voicemail services?” While this may be a sad outcome for many people, we should consider it a victory ultimately for the ability of our sector to step in when no one else did and conversely, to back off when others pick up the slack.

I think Community Voicemail desperately needs a new raison d’etre and indeed there is room in their mission statement to evolve away from voicemail as their only mode of service:

Community Voice Mail (CVM) helps people living in poverty, transition and homelessness rebuild their lives by connecting them to jobs, housing, information and hope. We do this by customizing and distributing communications technology via a national network of community-based services.

They can’t just be a free voicemail provisioner. They need to attack other issues that social services clients face but would be in the same realm of voicemail. Voicemail was ultimately about keeping data in safekeeping for nonprofit clients. Many social services clients don’t just have voicemail as a problem, they also have data safekeeping issues. In other words, it’s really tough for clients to keep all their documentation straight when they’re homeless. I’ve often thought it would be a good idea for homeless clients to also have a one-stop shop where they can could scan in documents such as wedding, birth and naturalization certificates as well as any other government documents so that any nonprofit they’re working with could print them out. Think of it as a electronic folder that makes it easier for clients to keep track of the work they’re doing with nonprofits. With the advent of EC2 and S3, this could easily be a national service that Community Voicemail could start without a large outlay of money.

It’s clear that nonprofits that work with technology will always face the problem of being made obsolete in the face of larger and better-funded ventures. There’s nothing wrong with simply stepping back and reassessing your mission from time to time in the face of that. I really hope that Community Voicemail takes this post to heart and really look into modifying their programs.

Daily Kos unveiled their new server infrastructure just recently:

Hardware is part of the solution, Bingham says. “To handle the traffic better, we moved to a cluster of six quad core Xeons with 8GB RAM for webheads that all boot off a central NFS (Network File System) root, with the capability of adding more webheads as needed,” he said. Daily Kos also added two 16GB eight-core Xeons and a 6×73GB RAID-10 array for database files running a MySQL master/slave setup [...]

Many database-driven sites facing scalability challenges have looked to the distributed caching system memcached, which helps speed dynamic web applications. Bingham says memcached has played an important role in scaling the Daily Kos site. “I greatly expanded memcached usage with 1GB instances memcached running on each webhead, which they all share,” said Bingham. “The backend also places fully rendered pages into memcached, which a hacked up lighttpd running as the front end proxy then serves these pages from memcached directly to anonymous users. This has helped the sites performance immensely, since not only does it spread the work of rendering these pages around to the all of the webheads, but it greatly reduces the amount of work the backend has to do.”

9 million pageviews were served with this setup. memcached turns out to be a piece of technology that Causes uses for its own servers. I’m quite curious as to why they don’t use an EC2-based infrastructure and why they still use lighttpd but who knows?

It turns out that Daily Kos’ widget for serving election results were designed by Free Range Studios. Ok, to be honest, I used the NY Times electoral results more since I was on my Windows Mobile phone. The NY Times election page was superior to the Daily Kos’ Flash widget in that it rendered very quickly and had better information architecture. Normally, I like the work of Free Range Studios but the Daily Kos widget had all the charm of a NORAD map charting incoming Soviet ICBMs a la the movie Wargames.

I love talking about hardware configurations as they can give you a ton of insight as to how you can architect your own solutions for whatever application you create that gets the same level of traffic. Those of you looking to build a Daily Kos-like site can check out Scoop. I don’t recommend using this software unless you’re prepared to do extensive customization. The last stable release of scoop has a 2006 timestamp. That doesn’t bode well for security updates, etc.

_{Original photo by Alan Light}

Oprah Winfrey did a special on philanthropy and giving on September 4th of this year. Highlights were a visit from Bill Clinton hawking his new book appropiately entitled Giving: How Each of Us Can Change the World and Andrew Agassi discussing his Andre Agassi College Preparatory Academy. Also mentioned were Common Cents and Kiva. Below is a chart of the spike in traffic that resulted from the Oprah effect. clintonfoundation.org and kiva.org were well-positioned to take advantage of the Oprah effect. Unfortunately, commoncents.org and the Andrew Agassi College Preparatory Academy don’t register high enough on compete.com’s site analytics right now. Hopefully, stats for those two nonprofits will show up next month.

The Clinton Foundation did well primarily because there were two reasons to drive site traffic, Bill Clinton’s new book and the resulting publicity from it as well as the Oprah show. Kiva did get a boost of 24,000 users in one day before Oprah from being on the Today Show. Notice that for both Kiva and the Clinton Foundation, there were huge gains from the period of 9/4 to 9/7 and traffic was still at a heightened level even a week after after the show. I’m just shaking my head and thinking, “Incredible”. Not only does the Oprah effect cause a huge spike in traffic, it keeps it that way. Kiva was forced to actually impose a maximum donation because all the case listings on its site were completely subscribed. Kiva self-reports 134,000 visits in the for the day of the Oprah show and the day after. Kiva’s weekly average before that period was around 10,000 visitors. Oprah drove near three months of traffic in two days to Kiva!

If anything, this should get IT directors rethinking site scalability. Please see my earlier article on Amazon’s EC2 so that you too can start thinking about how to make your web site more scalable in the future should your organization be lucky enough to achieve an Oprah Effect of its own. Your Oprah effect may be something smaller such as a mention in CNN or the local paper or TV station. Traditional media is still one of the biggest drivers of large surges of traffic to your site. Don’t forget that Kiva was down for four days last year not because of Oprah but because the PBS news show Frontline covered them extensively. It’s quite possible that your Oprah effect may be enough to take your org’s server’s down. It’s actually a great way to gauge the effectiveness of the venue as well because you can certainly track site visits right after the show airs. In a sense, your publicity efforts, your IT efforts and the size of your organization are all aggregated into site statistics. You can’t easily pull those numbers apart but you can certainly tie the effectiveness of a recent public relations push to any significant change in traffic volume and user behavior. Think of site stats in this case as your organization’s public relations heartbeat monitor.

Love them Gym Class Heroes though… Here are new technologies that haven’t been explored nearly enough by nonprofit bloggers in the last few months and part of the reason for that is that it’s not about social networking, a technology that is seemingly easy to blog about since the technical innards are usually hidden behind some friends list. Be aware that some of these technologies are not particularly easy to understand if you don’t have a programming or sysadmin background. If you need help, please comment below. To paraphase from the song, Click Read More And Show Me Some Kind of Sign!

Amazon Flexible Payments Service

Amazon Flexible Payments Service (Amazon FPS) — Amazon FPS is a new payment service built out of Amazon’s e-commerce services. This means that users who are used to Amazon’s payment processing service (who isn’t) can have that same experience on your site. Here is what it can do:

• Create “Payment Instructions” to define conditions and constraints desired for a given transaction, and programmatically obtain payment authorizations or “tokens” that represent these Payment Instructions from customers.
• Execute one-time, multiple, or recurring payments on behalf of customers.
• Aggregate micro-transactions into a single larger transaction using Prepaid and Postpaid capabilities.
• Build payment applications where you are neither the sender nor the recipient of funds. You can build marketplace applications that enable the movement of money between two third parties.

Obviously, this can change the game for even the largest of nonprofits. It’s not recommended yet as a service, but it’s certainly the gamechanger that Google Checkout never became. Why? Simply because nonprofits have one business model on their web sites, the traditional donations model. I have never seen third party transactions or micropayments explored in the nonprofit sector. (If you know of any, please tell me.) It seems fairly clear to me what the business ramifications are, as you can now build a nonprofit business model for earned income on items that are basically virtual. You can sell virtual badges, avatars, icons, e-publications and other virtual paraphernalia that go along with your nonprofit at pennies on the item. If your organization has a lot of research content, you can now create a paywall for it and charge a nominal fee.

You can also create a website that diverts funds from the account of a donor straight to a client. It’s an entirely different business model I know. It changes the very notion of what a nonprofit is if the money is never passed through its accounts and it makes 990-based metrics like Charity Navigator even less relevant. Trust me, old-school orgs will not adopt Amazon FPS but the new nimble ones will.

And even better, there’s already libraries in multiple programming languages for Amazon FPS. And even mo’ better, pricing is reasonable. For transactions over $10, the pricing starts from 2.9% + $0.30 for credit card to 1.9% + $0.30 per transaction if your monthly payment volume is over $100K.

Amazon Elastic Compute Cloud and Amazon Simple Storage Service

Amazon Elastic Compute Cloud (Amazon EC2) and Amazon Simple Storage Service (Amazon S3) are Amazon’s attempt at grid computing. Amazon has thousands of servers that they don’t use all the time. Basically, they let you rent their servers at very low cost but at the same time providing a level of scalability that your nonprofit never had. You want 1,000 servers online tomorrow because your ED is going on Oprah and you don’t want to miss any possible donations? Use EC2. You need to store terabytes of information cheaply and it has to be accessible? Use S3. A lot of new for-profit startups are using this technology but no nonprofits are using it right now. It’s a darn shame because this basically outsources the need for multiple web servers for a nonprofit and at the same time gives nonprofits more reliability and it does it cheaply. It’s that rare time where the old saw about having things fast, easy and cheap but you can only pick two doesn’t apply. Your website’s users will never have to experience a downed site from your org ever again.

Infinite Bits

OK, so let’s say you like the idea of storing your stuff online at Amazon S3. You find out that the pricing is ridiculously low. It’s only $0.15 per GB per month. However, you want your users to use FTP to get to it since it’s kinda hard to use S3 for normal people. Never fear — for $4.95 a month, you can use Infinite Bits which provides an FTP server that your users can upload to. They’ll be providing users their own custom Infinite Bits server so it can be used as a group file sharing mechanism as well. It’s all good here. Now all people have to do is build an online backup system for data and I can see this service showing up at your local datacenter for network admins as well.

inkling markets

I’m afraid I’m not going to have to just introduce a new technology but an entirely new concept here so bear with me. Decision markets are basically the idea that markets represent not just a series of bets but when the bets are all aggregated together is actually an attempt at predicting outcomes. The typical stock market represents an aggregation of all information about a particular set of companies and how well they’re performing. We can use that attribute in a decision market to make important decisions about the outcome of events in the future. I see a problem with adoption of decision markets in that nonprofits usually won’t or can’t publish enough information about an event in order for people to make a good reasonable bet on things but also that people for some reason think of “markets” as not appropriate in working with clients. I’m not talking about placing bets on clients but I could see a world where people could bet on the outcomes of a group of nonprofits in a sector. How can you get a taste of decision markets? Go check out inkling markets.

And how can you get a taste of how not to run a decision market? The way that globalgiving.com experimented briefly with inkling markets is certainly a case in point. The experiment was poorly run and they misunderstood some of the key issues about running a decision market. Global Giving ran a market where 22 projects were placed in a market and the users were asked the question “Which project has the greatest chance of succeeding on GlobalGiving.com?”. It seems innocuous enough but what happened is that the users were given a little Word document about each project and asked to make a decision based only on that one piece of information. There were no metrics to help decide which project could actually succeed, nothing was dynamic so of course, the decision market turned into a popularity contest because the flow of information was too small and too static. It’s as if a stock market were started, you could only read a prospectus, but you never knew how well a company was doing from month to month since no ledgers were kept. Worse, “the greatest chance of succeeding” is a pretty nebulous outcome to predict. What about clear and identifiable goals that people could use?

So what kinds of questions could nonprofits ask in decision markets? They can certainly ask questions relating to future predictions of government policy as much of that information is up on the Web already. They should ask questions about the simple statistics that define their client populations like:

“How many people will there be in the New York City shelter system at the end of October 2007?”
“How many crimes will be committed in Precinct 78?”
“What will be the average listing value in Brooklyn (as measured by Trulia) at the end of 2007?”
“Which charity will have more unique monthly visitors (as measured by Quantcast) in September 2007?”
“Will the Facebook Causes app raise more than $825,000 in December 2007?”

You’ll need to base the decision markets on clear and identifiable statistics. These statistics would be used to help provide your staff with some predictions that they could use to plan how much more capacity they will need in the next year. For more about decision markets, read Alex Kirtland’s great post on decision markets.

So that’s a quick wrap-up of some newer technologies. Next Uncharted Technologies post will cover more unusual technologies that can be used in helping out nonprofits with their missions.