At OpenLogic we review open source software prior to adding it to our certified library.  Projects get added to our queue in a number of ways, but more often than not, customers contact us and let us know they are using a project and ask us if we will review, certify and add the project to our library.  The first step in our certification process is to verify that the project is indeed open source.  For example, we automatically disqualify projects that are "free" or "demo" products that aren't provided with source code.  But more importantly, we review the terms under which the package is licensed.  This is probably the best way to determine if the project is really open source or not.  Many times it is easy to tell if it is an open source license, for example if the license is well known (like GPL or the Apache License) or if it is based on a well known license (like BSD) with minor, inconsequential changes.  But often, the licensing is not so cut and dry.

A few weeks ago a customer asked us to review a package.  After reviewing the website I was unable to find any information on licensing.  It is a common practice for community web sites to either include a statement on or link to licensing information.  So, I downloaded the project and unpacked the distribution.  Another common practice is to place the project license in the root directory.  Again, I found nothing.  So I contacted the project owner.  He let me know that he had not given it any thought and that he had not chosen a license yet.  And, yet the project was over 2 years old!!

I gave him some links to resources on licensing open source projects: the Open Source Initiative and the Free Software Foundation.  I also mentioned that Wikipedia has tons of great information on open source licensing.  I also offered some advice on the importance of understanding his goals in respect to his project.  Did he want to be less or more controlling over what ultimately happens with his projects source?

His project is actually a subproject of a very well known open source project.  So he sought the advice of his contact at the parent project and now has plans to officially license his project using a standard open source license very soon.

This story illustrates just one of the many ways that determining whether a project is really open source, and more importantly, how a project is licensed can be very difficult and time consuming.

This is not the first time we've had companies ask us to add "open source" packages to our library that are not a licensed open source project.  I think there are a few lessons in this story.  But ultimately, if your responsibility is to understand how a project is licensed and that your organization is in compliance with the terms and conditions of that license, you can never assume a project is open source until you have the license in hand.  Even if the project is a subproject of a well known open source licensed project.

We've recently made a decision at OpenLogic to focus more on adding novel, valuable, instructive content to OpenLogic Exchange (OLEX).  Not that this hasn't been important in the past, but it's not been a top priority; building out our certified library has.

Here's a list of some of the topics that have been suggested by users in the past.  If you've got input into which of these (or something completely different) you'd like to see, please leave a comment:

• Install, configure, integrate LAMP stacks of various flavors
• Configuring Apache with SSL
• Top ten/twenty steps to take in order to harden MySQL for deployment outside the firewall
• Top ten/twenty steps to take in order to harden Apache for deployment outside the firewall
• Building Apache on AIX
• Open source alternatives to common commercial applications in various domains
• Choosing the right open source tool for the job (various domains)
• A common sense guide to open source licenses
• Creating an open source policy for your enterprise

As you can see, some of these topics have a technical slant, others a business slant.  I'm curious which topics will be the most popular.

So when I talk about what a community manager is, I usually mention leading by influence, advocating, explaining viewpoints, defending the community, representing the community within the company, and in general keeping the community healthy. They are ambassadors that spend lots of time communicating.

Chris Brogan wrote a great post about the Essential Skills of a Community Manager that covers a lot of the same ideas but expresses it much better. I especially liked his analogy:

The best community managers are like a good party host mixed with a fine restaurant host.

 A few other points he added, that I'll have to add to my description when I talk to people are:

• "Community managers are bodyguards." In addition to protecting the community (from the public and the company), he also points out that community managers are bodyguards, they protect the community from trolls.
• "Community managers must build actionable reports."  I think this is really important, especially since:
  • community managers are a relatively new role in the online world so people don't know what they do,
  • without clear goals, it's easy to spend a lot of time just "communicating" and making things run smoothly without a clear goal in mind,
  • having a clear actionable plan shows the value of not only the community manager but also the community they are working with and the company that hired them.

If you are interested in community management, check out Chris' post. He says it well in the Essential Skills of a Community Manager.

In my last post, I question the accuracy of the CIO.com survey. One of the numbers I was skeptical about was:

Nearly half of the survey respondents, 45 percent, are using desktop applications such as OpenOffice.org

That seemed pretty high to me. Until I checked out the Open Source Census site and found OpenOfifice in the top 20 list with a prevelance of … 45%.

Both of these probably poll open source software fans and are in some sense self-selecting but still, 45% of them are using OpenOffice, that's not bad.

Everyone knows about the baggage fees many airlines now charge in an effort to help stem their massive financial losses, but Jeremy Zawodny suggests airlines take it a step further by pricing tickets based on the combined weight of the passenger and his or her baggage. This strikes me as misguided on several levels.

While nobody enjoys getting a seat next to a fellow passenger who struggles to fit between the armrests, a weight-based model would penalize those of us who happen to be tall. My weight is average for my height, but I’m sure I carry a few more pounds than the average person who’s a foot shorter. Why should I have to pay more simply because my genes made me tall? I feel penalized enough every time I have to wedge my long legs into the ever-decreasing amount of leg room allocated for coach-class passenger seats.

Further complicating Jeremy’s suggestion is the thought of adding passenger weigh-ins to the already time-consuming check in process. Just imagine the long line of impatient, disaffected travelers all staring as you glumly step onto the scale, hoping that the light breakfast you took this morning will ensure there’s no increase to the price of your ticket. And wouldn’t you feel awful every time a disapproving ticketing agent notes someone’s weight—and the corresponding price hike—just a little too loudly? I can’t imagine this would boost the already-dismal customer satisfaction ratings airlines earn these days.

I don’t necessarily disagree that baggage weight should be part of the pricing equation, but last time I checked airlines already charge for excess weight. Every checked bag is allowed to weigh up to 50 lbs, and heavier bags result in extra fees. I think that’s incentive enough to pack lightly, especially now that some airlines charge just to check a bag in the first place.

There’s no doubt that the airline industry is in trouble, but more complicated systems of fees and add-ons aren’t going to save their budgets or reputations.

CIO.com did an interesting survey about open source usage in the enterprise. While I question the accuracy of some of the numbers, if you take them as a general guideline, they generate some interesting discussion. For example, as an answer to the question "do we really need support for open source software?"

When it comes to support for open source software, I find that some people agree that it's absolutely critical and others look at you like you are crazy. Those that think you are crazy argue that the open source model itself provides for support. Turns out, if you are a developer working on open source software, the community support is enough:

The quality of support was the biggest obstacle to only 15 percent of developers.

But once your developed application moves into production or is used more widely, community support is often not enough. It's not that the community can't solve your problem, it's just that at 3am when your company's web application is down, you don't want pointers to three other reports that might help solve your problem. You want someone to say "we're on it, we'll have you a work around in an hour." That's what leads to open source software support concerns:

Half the respondents whose companies use open source today (52 percent) cite product support concerns as the greatest barrier to entry. A third (33 percent) of those who don't use open source identified this as a primary problem.

And that's why many companies exist to provide support for open source software products.

OpenLogic takes it one step further and working with partners and the community, provides one number to call support for 400+ open source software products.

I'm very excited - I am now officially joining the GNOME Foundation as Executive Director! Some background for those of you not familiar with GNOME. The GNOME project is:

• a free and open source desktop - an easy to use, intuitive interface for your Linux or Unix computer and
• the GNOME development platform, a developer framework for building applications.

The GNOME Foundation is the nonprofit organization that helps organize and run the GNOME project. The GNOME Foundation members are all contributors to GNOME - hundreds of volunteer developers. They elect a board of directors (from their membership) every 18 months and have a large number of corporate sponsors who participate through an advisory board. My job will be to work with the board of directors, the sponsors and the community to help them further their goals of a creating a free and open source desktop and development platform that is used world-wide by drawing in more developers, sponsors and users.

As some of you probably know, my first experience with open source software was with the GNOME community. Way back in 2000, I was managing the HPUX desktop and we decided that having a free and open source desktop would be advantageous for HPUX users. I ended up working with Ximian (then Helix Code) and attended my first GUADEC in 2001 in Copenhagen. While I was checking into my hotel in Copenhagen, I met some guys in the hotel lobby (Havoc Pennington and some of the Eazel guys) and ended up spending the afternoon exploring Copenhagen with them. The next day at the conference was the day of the famous "You're a girl!" quote. (She was very excited there was another woman at the conference.) I found the GNOME community extremely welcoming and passionate about their project. They got me hooked on open source and I've been spreading the word ever since.

At OpenLogic I've been working on spreading the word of open source in two primary ways.

1. I started the OpenLogic Expert Community, a group of contributors and committers that help OpenLogic provide 24×7 support for over 400 open source software projects. The Expert Community members are developers that are not only passionate about open source software but willing to help enterprises see the benefits of using open source software. OpenLogic's customers say great things about the support model - having the experts on hand is key to making them feel that open source software is an important and reliable part of their infrastructure.
2. I helped launch the Open Source Census - a collaborative, global initiative that aims to count all of the open source software that's out there. Not the number of downloads but the number of installs. The goal is to show people how much open source software is already being used in order to show people they should feel comfortable using it. The Open Source Census is backed by a large number of industry sponsors: IDC, ActiveState, EnterpriseDB, OSU OSL, Open Source Alternative, Navica, Unisys, OSA, OllianceGroup, CollabNet, Microsoft, Carnegie Mellon West, HRO, and OSBF.

I'm happy to say I'll be continuing in advisory role at OpenLogic so I'm not saying good-bye yet to all those folks!

I'm also very excited to be joining the GNOME Foundation to help spread the word about GNOME - a free and open source desktop - by strengthening the Foundation and attracting new industry members and community contributors. 

Replicator quoting research which shows that people who customize their car (from bumper stickers to whatever) are more likely to engage in road rage.  The logic is that the customization is indicative of someone that thinks of their car as personal space and more easily tilts when confronted with bad driving.  I didn't read the study (written by CSU profs and published in the Journal of Applied Social Psychology), but I can only assume that this is a correlation and that no causation was implied in the study.

In the world of web apps, we're all searching for that secret sauce that makes a site "sticky." Features that allow customization are often considered sticky. This logic seems to assume, at minimum, that customization of a thing reinforces strong feelings for the thing - if not outright causation of those feelings.

It would be interesting to know if the more bumper stickers a driver had, the shorter their fuse…  If so, I have another theory to explain the observations.  Perhaps the causal agent of road rage is insanity and all that's really illustrated here is that folks with lots of bumper stickers on their rides are a little nuts.  Perhaps the same can be said of folks with lots of Facebook apps?

Matthew Aslett recently completed a tour of Europe, profiling open source policies and usage in European countries.  Tongue-in-Cheek, he presents a summary of his findings in football tournament format:

http://blogs.the451group.com/opensource/2008/06/30/open-source-champions-of-europe/