OpenServices.org

Well, the name is not that important - I use it just to conceptualize an idea. What the actual implementation should be named, well, that's up for grabs. The idea, though, is phenomenally good. I came up with it, after all. (Where is your sarcasmey when you need one?)

You know all about open source. One of the things you might know about open source that is not about open source is that most of the largest web infrastructure companies use open source software throughout their stack. Be it Yahoo!, Google, Amazon, Facebook, Twitter, and what you have: if they are big, they use open source.

(The only major exception, for obvious reasons, is MSN/Live/Bing.)

The sad thing is that these companies (with the notable exception of Google) rarely give something back to open source. In particular, not even Google releases the source of their main site software. Of course we understand that (sort of). But if there is a market for open source software, why not for open source services?

What is the basic idea of openservices.org? It would be a site or a series of federated sites that not only publish all the software they are written with, but also function in an open source fashion. That means that, while they are run by a group of dedicated and security/stability-conscious admins, the software that runs on them is the result of open collaboration.

You'll say, but we already have something like that! It's called Wikipedia! And I say, well yes, that's exactly what I am thinking of. But more than Wikipedia. Slay the dragon and bypass all the stupid restrictions and costs of the sites you know.

Let's go a step back. In Internet Theory, we classify four types of sites:

1. Anonymous sites - sites that do not make use of the user's identity at all, like www.yahoo.com or www.google.com
2. Registered sites - sites that allow for and possibly require a registration, but that do not store information other than user credentials and preferences from that site, like my.yahoo.com or igoogle.com
3. Private sites - sites that store real world information about a person (instead of just a user), like mail.yahoo.com or gmail.com
4. Restricted sites - private sites that are behind additional protections required by law, by industry standard, or by company policy, like credit card processing sites

There is plenty software available for anonymous sites, and a lot of software for registered sites, but very few projects address the needs of users of private sites. On the other hand, much of the economic action in low-hanging fruits is happening in that very area.

One of the strange anomalies of the Internet is the kinds of sites that can attract large amounts of paying customers. There are porn sites, of course, but there are more free alternatives available now, so that market is drying out.

There are betting sites. They derive their attractiveness mostly from the fact they or their content are banned from several jurisdictions (most notably the United States or part of them). As a result, the demand for such sites can only be met by illegal or semi-legal operations, which as usual create a black market.

There are gaming sites. Here the attractiveness comes from the quality of the product alone, and from the fact that the games become dull after a while. I am talking about sites like World of Warcraft. But the amount of effort that goes into a good online game is way ahead of the pure programming, as places like OpenSim show.

Then there is the last one: personals sites. The application is simple, almost painfully so. From a programmer's perspective, it's a collection of profiles and some way to match them. On many sites, the way to match is simply by searching using a form that might as well have been created in 1994. I would call that the lowest-hanging fruit.

How would OpenServices.org work? There would be four arms:

1. Software development - the creation, publication, and maintenance of all the software on the services
2. Systems administration - the ownership of the servers and the installation of software updates
3. Creative and project management - pushing through the next generation of features
4. Quality - making sure that the new software works reliably

Why would anyone want to work on OpenServices.org? Because it's a great way to introduce yourself to the world. Right now, one of the biggest hindrances for young engineers trying to get into the workforce is lack of experience. I can understand (having been on the hiring side of things) how an employer would feel reluctant to give someone with no experience a chance. At the same time, I can also see that it's really hard to break through the vicious circle of being required to have some form of experience to be able to gain more experience.

Now, imagine a site that works with a volunteer effort. Senior contributors get fancy titles and supervise (which also means, mentor) junior members. Coming out of, say, college, someone might already have several years of system administration experience, or creative design experience.

The only thing required for this to work is enough money for servers and bandwidth. Seems a win-win scenario for everybody. With the possible exception of match.com and eHarmony.