OSL

Congrats go to Corey and the rest of the OSL crew for winning the InfoWorld 100 Award for Education.

With the rackathon humming along, the new data center on-line and lots of hosted projects, the OSL continues to do amazingly awesome things. I’m really glad to see them continuing to thrive and I’m sure they are happy I’m not in there mucking with people’s systems anymore * … hahaha … :-)

Way to go OSL! Keep up the great work!

* - I never actually “mucked” with anyone’s systems … more like fiddled. And I’m an excellent “mucker” BTW.

I wanted to let people know that as of June 30th, 2006 I’ll no longer be working at the OSL. I’m moving on to head up a company in Portland called JanRain, Inc. You can read more about it from OSU’s press release.

I want to make it clear this is an amicable split and one that was simply a function of a great opportunity that I had to take. The OSL is in great shape and the University is still completely backing our on-going efforts here. The OSL will continue on and they will probably find somebody to replace me that makes me look bad … Ha!

This isn’t the end, just a change. I’m excited about the new endeavor but it was really hard to leave such an amazing team of individuals. Most of all, I have to thank Curt Pederson and more importantly Shay Dakan (sorry Curt, Shay’s got you beat on this one) … In all honesty, Shay has done more for the open source community than that community knows. She was the one who took the chance on my crazy idea and trusted Jason and I to run it. The OSL, all of its employees and the hosted partners all owe a deep debt of gratitude to Shay. Thanks Shay.

I’ve setup a new blog to start talking about the new gig and I really hope to keep in touch with all of the great people I’ve worked with here at the OSL and in the open source community. Hope to see all of you in Portland this Summer for OSCON!!!

For the past two years I’ve been hearing about grid computing and how utility computing is going to revolutionize the industry. I never really saw the application for it. When I saw that Sun’s grid play had zero customers by October of last year, I was feeling pretty smug about it. Does grid computing even matter?

Flash back over the past couple of months and watching projects like Mozilla, Meebo, Flickr, Zooomr and others try to keep up with demand under increasing load and the “slashdot” and “digg” effects. The barrier to entry for some new projects can be painful if a) the project is resource intensive and b) they experience exponential growth in a short time period. How can companies ramp up their infrastructure in a timely fashion while still doing work at-the-speed-of-business (not to use a horrible cliche but it seems apt here)?

The barrier to entry is still pretty low for new Web 2.0 companies. There was so much sunk cost left over from the bubble that things like bandwidth, office space and others are much cheaper to acquire. Coupling that with maturing open source tools and you have a formula for getting your service out to users much more quickly. The one thing that hasn’t really changed though is the time to purchase, configure and deploy the “servers” for driving these services. I’ll give a few examples.

Sometime in September 2005, Meebo launched to great acclaim. If you’ve never seen Meebo, think of it as an AJAX-ified web-based instant messenger client for AIM, Google Talk, Yahoo! Instant Messenger, etc. It actually is a web front-end to Gaim running on Linux from what I can gather. Over the course of the last few months Meebo has been working hard to ramp up their infrastructure to meet the growing demand. It appears they have upwards of 35+ application servers (or instances) serving up their instant messaging goodness. It has taken months to get this infrastructure up and running not to mention the sheer cost of the gear. Their setup is fairly specialized with the Gaim usage but most of it appears to be LAMP stack based.

In October 2005, another Web 2.0 application launched; Zooomr. Zooomr has been best described (not by me) as Flickr on steroids. Just last week they were dugg which rendered the service almost completely unusable. Looking at their infrastructure, it is definitely based on a LAMP stack (without the A and P == Python). The bummer here is there is this great new web service that simply could not scale when the demand hit. The lost opportunity cost could be pretty significant. How many people will remember to go back other than the really core users?

Looking at the above two examples, I started to think there has to be a better way. What if you could just leverage some sort of grid of LAMP machines to make this happen and then pay per cycle while you bring up your new infrastructure? Folks like Meebo and Zooomr would be able to launch very quickly, meet the Slashdot and Digg effects with ease while continually increasing usage.

What would it take to implement this? Well, you could have this as a value-added service of a co-location facility. They have a cluster of machines ready-to-go plus some staging infrastructure for people to test what they want to deploy. That could get pretty cost intensive for one co-location facility pretty quickly. However, what if you did this across multiple companies and co-location facilities? Now you’d have your grid component (there would have to be some standardization here) and the ability to fail across the infrastructure very easily. Hook in some caching and redundant database services and now you’ve got a pretty compelling tool.

Taking the above examples, they could have deployed across this LAMP grid and been able to scale that much further, that much faster right when they released. This would remove a serious constraint for these companies (and many others) if it could be done economically.

I’ve pitched this idea to a few folks and they think its pretty neat. Of course you can make anything look good on a white board. The big question is, can this be done and more importantly would it be economical and effective for these small companies to use and for the co-location facilities to deploy?

We’ve been having a rousing discussion on the POSSE list over the last couple of days spurred on by David Myers, the founder of openforest.org. He posted a great set of questions that I took the time to follow-up on. I’m posting the questions and my responses to them here as a chance to vet some of the ideas to a wider audience.

1. In his Open for Business column, Jon Perr writes “Much remains to be done, of course, to bring about Oregon’s success as an open source mecca.” What are some of the top items on that to-do list?

We have so many components in place right now; OSDL, OSL, OTBC, PSU, OSU, etc … More acronyms than you can shake a stick at. However, what we’re lacking are the new companies to take advantage of these assets.

During RecentChangesCamp, there was a group of students visiting from France that came to see about this “open technology cluster” here in Oregon. They had done quite a bit of research and were traveling around Oregon visiting all of the people/groups in the cluster. Their conclusion? “Where’s the beef?” There isn’t a cluster here because there are almost no startups. I actually had to concur with them; they were dead on.

I almost feel like we don’t have anybody thinking big enough (myself included). I’m not saying we need an uber huge company here or that we need to sacrifice what makes this Oregon. Just the opposite. I believe that our way of life here really melds with this always-connected, merit-based economy that is emerging on the Internet. If you could encapsulate the Oregon way of life in a web service, I think you’d have a gold mine. It would just suck if somebody did it in the bay area first …

So what’s on the todo list? Great question … But to me its finding where the rubber meets the road in Portland; connect the movers/shakers, stir it all together and get them to think big. We’ve got all of the pieces; now we just need to put it all together.

(I think I just answered your question with another question — sorry ’bout that)

2. Does the democratic-yet-meritocratic nature of the open source community make it an effective organizing principle for extending the development model beyond software? Like developing a grassroots-powered media network, for instance. Or perhaps writing legislation? Couldn’t the US Constitution be thought of as an early open source development project?

I saw a great presentation from Mark Shuttleworth (of Ubuntu-fame) who talked about this specific issue. He said that the reason we had open source instead of open law or open medicine first was because of the tools. As soon as email, diff and patch revealed themselves on the Internet, we had the most rudimentary tools of collaboration. As these tools have matured over time, they are starting to have applicability to folks outside of the technology space. Its no longer just geeks writing code to help them write more code. Its becoming technology for the sake of X where X is democracy, music sharing and anything else you can imagine. Take 10 lawyers, put them in a room with a 200 page word document and say “collaborate”! No wonder lawyers are so miserable; the tools for collaboration for them stink. Imagine using a wiki instead of a word doc? Holy crap … Now we’re talkin’.

In Thomas Friedman’s book “The World is Flat” (if you haven’t read this you really should — IMHO its one of the #1 drivers for open source adoption at the C-level right now) there is a quote (and I can’t remember it exactly) about how the industrial revolution was all about the innovations of single individuals. Take Henry Ford as an example. The knowledge revolution will be about the innovations of communities; people coming together to do great things. I think open source is just one great example of this. However, I believe the best is yet to come.

3. If Oregon attempts to pitch itself as an “open source” clone of the technology hubs in Silicon Valley or Redmond would we risk endangering the sort of collaborative, cooperative and sustainably-minded ethos seemingly indigenous to our corner of the country? Isn’t this the kind of sensibility that plays to the strengths of open source development?

I don’t think we want to pitch ourselves as a clone of anything. We’re too small of a State to just try and copy something or even to compare ourselves to Silicon Valley or Redmond. I think we have to play on our strengths you list above; its what got us here.

The on-line identity space has always been fascinating to me. We all have on-line identities but today they are siloed in places like Yahoo!, Amazon and eBay. My identity (and subsequent reputation) is locked in these sites (sometimes referred to Identity v1.0). Every site you go to wants information from you to “confirm” who you are or at least attach your contributions/efforts to some single identity. How many forums, websites and on-line merchants do you have login information for? I believe the number you’re looking for is a gazillion. That’s a lot of zeros.

IMHO, the keys to successful Internet identity are being light weight and decentralized. We’ve seen efforts in the past like Passport and others that have failed because they were either a) too ambitious, b) didn’t have the users’ best interest in mind or c) were centralized. De-centralization speaks to the nature of the Internet. Being light weight helps get deployments out quickly giving insights into how people use the technology. This helps shape the next generations of the tools as well as things that get layered on top of it. Dictating the entire stack doesn’t work in today’s marketplace.

Looking across the identity landscape, I see two contenders that match the keys to success from above. One is OpenID and the other is the Higgins Project. OpenID was proposed by the folks at Danga (aka LiveJournal) and the Higgins Project was announced today but has been in the works for quite some time.

From what I have read about the two technologies, they look quite similar. A thin layer of identity, coupled with a way to describe the services tied to that identity. No assumptions are made about what you’ll layer on top of the identity. Again, I think this speaks well of the Internet. Imagine going back to 1985 and trying to define not only what the Internet is, but how you’d use it and the framework around making it successful, scalable, etc. You couldn’t do it. I think the same is true with identity. Define a thin layer that acts as a platform that people can use effectively. How they use it and the services/tools you (and more importantly they) layer on top of that are the opportunities for commercialization.

One of my favorite complaints of OpenID was that it lacked a trust mechanism. Of course it does. Its just an identity component. Trust can be layered on later if the identity piece is a platform. The problem with identity in the past has been everyone has tried to engineer the Itanium. What we really need though is the Opteron. The fact is, trying to design the entire stack in one go is a major pain. OpenID addresses this by being a thin layer of identity on top of which people can place anything they want. From the looks of it, Higgins will do the same thing.

Okay, great. Higgins and OpenID are possible options. The next thing I hear in all of my conversations with people is, “Amazon|Yahoo!|eBay will never adopt this. It will never take off.” Really? LiveJournal already has 9 million OpenID users. That’s a pretty good start. So how do we get the “big” sites to adopt something like this and break the identity 1.0 barrier? I’m wondering if there is a way that we can get the tail to wag the dog here.

Of course this is all pie-in-the-sky stuff. You’d need an army of developers and years to implement libraries for something like OpenID for all of these different applications. Hmm, actually that’s not true. Most of it is already done. Holy crap. Libraries for PHP, Perl, Ruby, Python and .NET as well as patches for tools like Mailman and MediaWiki?! Painting the full picture, we see that we have millions of enabled users (LiveJournal) and plenty of tools to make it happen.

How many open source tools do you know of that require a login? The list is endless; phpBB, Wordpress, Drupal, Moodle, every-single-other-CMS, every-single-other-blog-tool, etc. What if you could get OpenID or Higgins enabled by default in them? I know I’d love to be able to serve up my own identity with this Wordpress instance. Not only that, each of these tools wants to have things like social networking (I know this user and that one but I hate that guy, etc). In each instance, they are tied to one tool or a specific site. Again, siloed identity 1.0. Its almost like these sites are making their software do acrobatics above and beyond their “main thing”. Imagine if these sites were OpenID or Higgins-enabled? Not only could you login to them seamlessly, you could tie all sorts of other services in.

What else could you layer on this? What about reputation or social networking? I love Orkut and LinkedIn but again, they are coupled to only those sites. If I had my OpenID and could build a network of other OpenID users that would be pretty powerful. Enable it through open web services (a la Google Maps) and now you can do more effective messaging and others. What if you could take the reputation or trust or social networking and apply it to your Digg login? Some might say that would hurt Digg’s traffic; I disagree. I actually think it would augment it. If users can easily traverse many different web sites and engage in conversations across them, that frees up those sites to actually focus on making tangible content (or aggregating it as the case may be). Add in a Firefox extension or Internet Explorer plugin to ease the process and maybe a little Grease Monkey action and things get very, very interesting. The possibilities are endless and I guarantee the best ones won’t be thought of until this thing is going at a thousand miles an hour.

So where do we go from here? What if we start a grassroots campaign to get OpenID enabled in the top-tier open source applications? Enable OpenID or Higgins across these projects and I guarantee you’ll see ideas behind identity 2.0 realized very quickly.

There is a trend that is occurring in the enterprise open source space that could become very lucrative in the coming years. For those of you attending OSBC this week, listen up.

The open source development model continues its march into just about every space that software exists in. Today, I want to talk about one that I think could be very lucrative for the companies involved; enterprise applications.

Recent years (or is just months?!) have seen Oracle turn into even bigger than the 800 lb. gorilla. Acquisitions have made Oracle and enterprise powerhouse but I believe they will be coming at a cost. Over the next 2 to 3 years, Oracle will be busy digesting these companies into the fold and integrating product lines across the board. This will be tough for customers as the pace of innovation for Oracle will slow while this happens. If you’re a new player to this market, you’ve got a chance to steal some market share.

The fact is, if you’re not in the game today, 3 years isn’t enough to jump in and gain any traction. The enterprise is finicky and even more so when it comes to critical applications. So who is going to make out like bandits? I think its the existing players in the market; Salesforce.com, Compiere, SugarCRM and Project.net.

Everybody has been talking about these companies for quite some time. What is most interesting to me is that each one represents business opportunities for the other. That is if they can all come together to work together.

Imagine that you are the CIO or more likely the Director of some portion of your IT organization. You have been using a CRM solution or maybe an ERP system for the past 10 or 15 years. Its time to re-up your license and maybe take the plunge from Oracle on VMS to Oracle on Unix or Linux. Guess what? Lots of people are re-evaluating the status quo and looking at new solutions like SugarCRM or Compiere. The best part about this is that they usually deploy one of these solutions in a small corner of the office, usually to replace just one component or application that is end-of-life.

What is very interesting about going with an open source (and more importantly open standards-based) product like SugarCRM, Compiere or Project.net, is that these products are built to work with other tools. Grandpa’s CRM solution wasn’t designed to play nice with anybody. Now, you get something like SugarCRM in the door, you can now look at deploying other products like Compiere or Project.net. Guess what? They will all work together with very little heavy lifting (that is in comparison to making that COBOL monster directly leverage one of those apps).

Each one of these players in the enterprise space has a unique opportunity to work together to drive business to one another. Here’s to hoping that they all realize that and execute on it.

(I’m telling the whole story here simply to get everyone up to speed — I often forget that I see more of this story than others so I want to share completely)

Update: FLOSS == Free/Libre/Open Source Software

During OSCON 2005, I had a chance to meet with some folks involved with the Drupal project. At the time, the group was trying to figure out a way to fund some improvements in their infrastructure as well as how to take in money from the Google Summer of Code program. The problem was that Drupal was not a non-profit and could not act as its own fiscal agent. The OSL helped out where we could by taking the donations they received for equipment and putting them towards that but the Google funds were much more complicated.

Just a few days before that, I had attended the FLOSS Foundation summit. This was organized by Alison Randal (of Perl and O’Reilly fame) and was a chance for the project leads for several FLOSS projects to gather and talk about some of the mechanics of managing and growing their communities. There were people there from the very mature (Eclipse) to the just-getting-started (Dojo Foundation). What struck me as interesting is that each project had their own path but each one struggled to get there.

Here at the OSL we have a pretty unique view into lots of different projects. Our focus is on helping projects with hosting and development. More importantly, we help projects focus on their “main thing”. This rarely ever involves hosting or scaling of infrastructure or getting bits to end users. These are boring, but essential, components to open source projects. Talking with Drupal and the attendees at the FLOSS summit it occurred to me that it would fantastic if there was something that could managing the “hosting” of the administrative side of the house for these projects.

Around about this time (or shortly thereafter), Chris Messina and I had started a dialog on the above. We had both seen what was happening and we came up with the idea of a “Web 2.0 Foundation”. We quickly found this had some limitations (the biggest IMHO was WTF do we do when Web 3.0 comes out?!). The conversation then led to what if we had a FLOSS Foundation that could do some simple things for projects to help them get to where they are going quicker? Again, every project has its own path, and this isn’t about incubating the projects but more providing some services and leveraging some social networks to help these communities get-to-where-they-are-going.

On January 10th, 2006, Chris, myself, Jason, Raven, Phil and Tara all gathered at the OSAF to talk about how to make the above happen. In our discussions, we couldn’t quite figure out a place to start and after several hours in front of a white board we were all pretty wiped out. (Fortunately MacWorld was going on so everybody was in high spirits with the uber cool new Macs … ha!) Leading up to this discussion, both Chris and I had talked with Raven at length about the need for some type of organization like this.

I was a little set back by the talks at the OSAF and was even starting to question the need for yet-another-organization. That is until I got a phone call from one of the projects we host who asked if we could act as a fiscal agent for them. You see, they had not received their non-profit status from the IRS (this can take up to six months). Here at the OSL and as a part of Oregon State University we can’t really do something like that so I had to tell them we could not help. However, it sparked my interest once again in this concept.

We’re now up to last week. Pat Mochel and I are on the technical board of advisors for LinuxFund and we had planned to get together at the OSDL to brainstorm on some ideas for LinuxFund (and so I could get him up to speed on the above). Bear in mind that Pat and I have no direct bearing on governance at LinuxFund other than we know the Executive Director and a few of the board members. Other than that, we just give our thoughts. In talking with Pat, he and I hashed over some general ideas of what this FLOSS organization would look like. It should be pretty lean and mean; farm out what you can and keep a minimal staff. Leverage the community and the eco-system to hire things like FLOSS-friendly accountants and legal help. I love pro-bono but nothing gets things done like cold, hard cash.

An hour of conversation and some time in front of a white board yielded this “org chart”:

Mind you this is a first run and one that actually has a bunch of my doodles on it (that reminds me … I need to call David Pool). It’s also full of fits and starts about who should actually be the umbrella (if you look closely there is an “OSL?” in there which we quickly determined would not make sense due to our organizational structure at the University here).

Alright, so what is this thing anyways? Start with something simple like a “fiscal agent” and do that for the open source community. This would be in the same vein as what the Tides Foundation does for traditional non-profits (traditional being anything that doesn’t involve software I guess). You don’t have to do it permanently, just help these specific communities get to where they are going. If they want to stay under the umbrella, great, we can do that too. If not? Fantastic! Go forth and do good things. Spin out as a non-profit, etc. The goal of this umbrella org would be to enable these communities and keep them focused on their main thing.

As part of the fiscal agent, you manage accounts for these projects. Let’s say the project does something through fundable.org and then wants to spend the money on servers to be hosted somewhere. Great. Let’s make sure we have some accounts payable people that can do the work of paying the tab and let the project do pick out what they need from some website. We do this at the OSL to a lessor extent already; we help projects choose the infra and they go off and buy it. Yes, I’m being mildly self-serving here because the OSL would like to be a close partner to the as-yet-defined organization.

Also, it should be noted that I randomly picked out projects for the left hand side of that thing. I have no illusions that the Apache Software Foundation would want to use something like this; its simply a frame of reference. I do know that Drupal and the Participatory Culture Foundation could both use a fiscal agent.

I think the problem I was having before this was trying to design the “uber” org all in one go. Looking back at the OSL, we started by simply addressing some needs for open source projects that were unmet. From there, we found our main thing. If we had sat around a table for months we wouldn’t have designed the OSL as it is today. When it comes to open source and their communities, you can’t “design” you have to feel your way through. Execute, learn and repeat.

With all of that said, where do we go from here? Well, I’ll be at RecentChangesCamp.org this weekend and would love to discuss this with other folks. As the OSL, we’d like to help with this in any way that we can. It’s obvious to me that there is a need; now we’re in search of the people to make it happen.

Jason McKerr and I will be down in San Francisco on Monday through Wednesday night spreading open source goodness and hopefully getting some traction on the creation of a FOSS Foundation. More on that below.

Chris Messina has been kind enough to organize a meetup of folks in the Bay Area to have a beer or three and talk all good things around open source, etc. So if you’re downtown Monday evening, come find us at the Thirsty Bear.

On Tuesday afternoon, I’ll meet with several other folks will be meeting at the OSAF to talk about the possibility of creating a FOSS umbrella foundation. Originally we had talked of creating a Web 2.0 foundation to help the fledgling projects in this space but then realized we’d feel like fools when either a) the bubble burst or b) Web 3.0 hits.

At the OSL we help people with hosting their projects. From Mozilla to Debian to Freenode and many, many others. Hosting is just one more thing that rapidly growing projects should not have to deal with. It takes away from their “main thing” of developing a great piece of software or platform, etc. In our workings with these projects, we see them in the same place in terms of managing their finances, yearly accounting, defining their governance and even in the formation of their own non-profit foundations.

The goal of this as-of-yet-to-be-determined FOSS Foundation is to help with the other things that projects need in order to get to where they are going. Every project will have a different journey. Some will become their own non-profits, some will want to stay under the umbrella. Others might want to land under another umbrella group like the Apache Software Foundation. What ever it is, let’s keep people focused on their “main thing” and continue to accelerate the pace of innovation within open source.

Please contact Chris Messina or me if you’re interested in attending the meeting on Tuesday. Hope to see you there.

Free Geek is one of the most progressive community organizations I have ever seen. I had heard about it years ago through a friend and they said “you have to see this place”. At the time I wasn’t interested; a bunch of geeks recycling computers? That didn’t sound like my idea of fun.

For those that don’t know, Free Geek is a non-profit community organization that is based in SE Portland, OR. They have a 14,000 sq ft facility that is used for recycling used computer equipment as well as spaces for training, classes and a small conference room. The donated computer equipment is refurbished and then given to volunteers at Free Geek

About a year ago I had a meeting scheduled for me at Free Geek so I was finally going to get a chance to see the facility. Again, I was a little skeptical (for whatever reason I don’t know). I showed up about 10 minutes before Free Geek opened and I was amazed … what were all of these people doing out in front of the store front? Oh, they must want in on some cheap equipment. That makes sense.

At 11am the doors opened and I was surprised to see that nobody went into the thrift shop; they were all there to volunteer. I went into the thrift shop and looked around a bit. Lots of decent computer equipment at really reasonable prices. The proceeds from the shop go to help run the non-profit. I stepped out of the shop and headed towards the main facility and now, there was a line of cars with people dropping computer equipment off. The volunteers were taking the equipment from the cars and heading it into the facility. Now my curiosity was thoroughly piqued.

I quickly found the conference room and had my meeting (about helping organize a Penguin Day for October 2005) and then asked for the full tour. Within minutes I could see this was an amazing organization. Machines come in from cars pulling up. Volunteers in the “triage” area determine where to route the incoming equipment. From there machines get moved through out the facility and either completely rebuilt or off to the recycling area where components are salvaged. Guess what operating system they run on the newly refurbished machines? You guessed it, Linux.

The hook for the volunteers is that with 24 hours of volunteer time, they get a refurbished computer. Wow. Genius.

If you ever get a chance to visit the lovely city of Portland, OR you have to put a visit to Free Geek at the top of your list of sites to see.

What if we had a “best viewed in Firefox” day? The idea here is to have people sign up to switch their sites for one day only to a page that says “best viewed in Firefox”. On that page they would have links to download the latest version of Firefox and a link to the main site. Ideally you’d have some magic in there to just let the people who already have Firefox to by-pass onto the main site. Imagine if Google did this for just one day?

The goal of this campaign would be to raise awareness of Firefox to users of all of the other browsers. We’ve hit a critical mass and doing something like this on some bigger sites might actually help drive further adoption.

Happy New Year Everybody!