DistroWatch Weekly

MD: I am a Latvian-American from Los Angeles living in Portland, Oregon and have recently wrapped up an eight-year adventure in Latvia. I had no idea that I wanted an open source software environment back in January of 1991 when I first sat down at a UNIX terminal but I did know that I desperately wanted something I could run on my own hardware in my dorm room. The fact that the campus computer labs were windowless was a big motivator in my new quest for software freedom. At the time, the proprietary Coherent operating system was a moderately serviceable Unix clone and I tried both it and a Linux .99 distribution on a 386 that a friend found in the street. It got as far as "Loading Linux..............." and I literally said "I'll get back to you."

Come the second half of the 1990s I was involved with web design, hosting and desktop publishing and confess that I was impressed by Microsoft's NT Server suite on paper. It was ridiculously expensive however and I knew that it would fail to deliver on any promises that mattered to me. My research suggested that I wanted a BSD but Portland was largely a Red Hat town at the time. I tried and gave up on Red Hat 5.1 after finding that the documentation didn't match the software but was quite impressed with Red Hat 5.2. I had finally found the elegant Unix clone that I had sought for nearly a decade.

Alas, the affair didn't last. The subsequent Red Hat 6.0 included GNOME and took the system from a delightful Unix clone to a stunningly bad Windows clone. I'd say that Fedora and Ubuntu are finally passable Windows alternatives but it was a long, bumpy road. My Red Hat experience improved when on-line updates arrived with up2date but sadly those were later revoked. Looking back, everything Red Hat did made sense at the time but we've all learned quite a bit since then.

In 2001 I accepted a job with MandrakeSoft SA on their internal IS team and gave my clients away so I could focus on the work. It was a dream job but alas did not last long thanks to MDK's aggressive growth and high burn rate. My team was terminated and I left with neither a job nor clients. I moved to Latvia where I found myself back on my Unix quest and finally exploring the BSDs.

MD: Linux Fund pioneered the model of raising money for open source using rewards credit cards back in 1999 and has given away over $750,000 under that model. Ironically, I knew the Executive Director David Mandel back then but paid little attention to his work with Linux Fund. After all, I thought that the Unix system I needed already existed, rather than there being a huge need for innovative models for open source project funding and administration.

Come 2007 there was a BSD-related storage project that I needed for a product I was developing and I realized that it really should belong to the commons. It was also way out of my budget. These facts led me to seek out a non-profit that I could partner with to raise money for the work and someone recommended that I talk to Linux Fund. The project itself was shelved but Linux Fund had recently become a full 501(c)(3) non-profit and I worked with Linux Fund to set up BSD Fund for similar projects of cross-BSD interest. One thing lead to another and three years later I have found myself Program Director for both initiatives and loving every minute of it.

Today, both organizations raise money with rewards cards and that money goes largely to community events and overhead. They also raise money for specific projects that we believe fill neglected needs in the community. Some of our recent and current projects are the LiVES video editor, an Ubuntu LoCo community event, some Gnash and Inkscape features, the gEDA/PCB circuit board design tool and some compiler work.

I am pleased to report that I have recently added UK corporate and Canadian personal Linux Fund credit cards to the portfolio and I can safely say this is the last thing I thought I would be the world expert in.

MD: We are striving towards a balance of all revenue sources to help grow during this recession. Linux Fund has always relied on card revenue and is expanding into direct fund raising and event organization. Corporate sponsors have tightened up significantly this last few years and we are relived to see individuals step up and give, especially from outside the US where they might not be able to get a tax write off for their donation. The mission clearly outweighs tax benefits for the majority of our donors and that is very encouraging.

MD: Just as some cards gain you airline points or credit at your favourite store, our cards generate a small contribution every time you use them. The card holder doesn't see any additional fees but as you may know, merchants are charged a number of fees to offset fraud and reward the various financial institutions who are involved in the transaction. Those fees include a percentage for customer rewards and our share happens to benefit open source. Despite the small size, these contributions can really add up. I humbly invite you to purchase your next computer and event travel with one of our cards as we just may be a sponsor of the event you are attending!

MD: We currently offer the Linux Fund and BSD Fund personal cards in the USA, a Linux Fund personal card in Canada and the UK Business Credit Card in the United Kingdom. You can apply for the USA and UK cards on-line and by phone and the questions are quite routine for something of this nature. The Canadian application is by phone and we are negotiating to launch a corporate card in the US and personal card in the UK. Our success with our current programs will directly advance these. We have a few more countries in various stages of negotiation and we would appreciate any and all help making these programs successful. Visit linuxfund.org/cards and bsdfund.org/card for more information about each card and if you get one, do make sure you make your payments on time. Like root access, these are tools you must use with discipline.

MD: Our event funding is simple: We try to fund every volunteer-driven community event out there. This is an ambitious goal but with more cardholders, we could get a modest grant to every major event and quite a few smaller ones. We cannot speak highly enough about community events like SCaLE, LinuxFest NW, Ohio LinuxFest, Florida Linux Show, SE LinuxFest, OggCamp, BSDCan and the like. We're proud to be supporting more and more events outside the US and we feel that such events are critical to the success of software projects. There is no substitution for getting away from the computer and meeting face to face, not to mention over good food and drink.

For our partner projects, I work with a number of peers and advisers to identify projects that appear to fill a pressing need or have never graduated from promising to production status. Examples of pressing needs are the need for competitive engineering tools for Electronic Design Automation and Computer Aided Design. These sound like niches but the resulting systems are used by everyone. Example "1.0" milestone projects include the LiVES video editor, the OGD1 graphics device and the pcc compiler.

With a clear milestone identified, the work needed to complete that milestone is assessed, reviewed and budgeted, and the resulting budget is presented to the board of directors for review and hopefully approval. If approved, we approach the public for money and effectively allow them to vote with their pocketbooks.

We often get requests from projects for funding but would prefer that they actively publish project road maps, development budgets and even hardware and event wish lists. Good budgeting skills can really benefit a project in the long run, especially when it comes time host an event or incorporate by forming a non-profit or joining a conservancy. The Google Summer of Code initiative has actually been a great catalyst for this kind of public budgeting.

MD: I personally would like to see an open source equivalent to Filemaker but I think the entire open source community should first seriously consider some major "no new features" project housekeeping given the number of projects that either lie abandoned, lack critical documentation or sustainable communities. Few projects are truly devoid of merit and there are countless under appreciated gems out there. Human nature and years of itch-scratching are probably to blame but active projects would really benefit from taking stock of their goals, sub-projects and key assets like documentation. As a colleague of mine put it, "a utility without good documentation is of no utility at all." Perhaps an open source idea exchange like www.halfbakery.com is needed for aborted projects to be dropped into before being pruned from the world's repositories while everything else is tightened up. Project resurrection is an exciting endeavour and we should never underestimate the value of old but proven code.

As for a specific gap, again, open hardware from electronics to architecture is a very exciting challenge. As an AutoDesk representative put it, "If God didn't design it, one of our customers did." Having literally all of humanities' technical designs from kitchen utensils to stadiums and jumbo jets tied up in proprietary formats that require proprietary software is arguably a life-threatening oversight. That's without even taking patents into consideration. We've built a technical world that a select few people truly control.

The more I am involved in the community, the more I realize that it is in its infancy. We must accept that software never costs anything to duplicate and that some day, foundations and institutes may very well dominate the software industry. To get to that point will take years of hard work and a general reprogramming of the industry. Seeing the 90% penetration of open source into supercomputing proves that unimaginable change is both possible and logical.

MD: Beside my previous humble requests and humble opinions, I hope that if you don't support Linux Fund, you do support one or more of the dozens of other software foundations. I have compiled a growing list of them at linuxfund.org/foundations and look forward to the time when these foundations share equal mind share and funding with the Red Cross, Doctors Without Borders, the United Way, the Sierra Club and countless other household names.

KM: There are a number of important differences at the very core of BSD and Linux. When you install FreeBSD, you are getting a complete operating system, kernel + userland, which is designed to function together in a very coherent manner. On the Linux side, your "operating system" may vary greatly from distro to distro, or even from install to install, because Linux at its heart is only a kernel and a subset of various tools up to the discretion of the distro packager / installer.

Aside from software itself, there is an important license distinction as well. Both open-source licenses advocate freedom, however freedom means different things to different people. The Linux kernel and many of its components are released under various forms of the GPL license, which requires users / developers to adhere to its terms and conditions in order to keep any development / usage GPL compliant, by giving source code back to the community, restricting DRM (GPL3) and more. The BSD license on the other hand is also open-source, but carries with it no expectation or demands on future usage / development. BSD-Licensed code may be taken and used for any purposes, without having to worry about "staying within compliance".

KM: While at the heart both systems are different, they do share a lot in common with each other. Most of the same applications can and do run on both, from services such as Apache, to desktop and productivity tools such as KDE, OpenOffice, FireFox, Wine and more.

KM: The license is a huge plus for me, plus the way the core operating system is designed feels much more "natural" and intuitive. The stability of their development process is a huge plus, ABI's are very stable and we deal with less "bit-rot" than I've experienced on various flavors of Linux.

KM: Because of Linux changing so rapidly they are often ahead of us in certain areas, like hardware driver support. A lot of open-source desktop applications are developed on Linux, so it can take a bit longer for a release to make it into the FreeBSD ports tree, although this has become much better over the years.

MN: Linux is a re-implementation or copy of the UNIX system that shares a common ancestor with BSD. It was designed to feel familiar to UNIX and BSD users (and in many ways it does), however the underlying architecture to create the system is quite different in some important ways. For example, configuring a firewall, choosing which programs start automatically or installing a driver for your system will be different between BSD and Linux.

MN: The UNIX and FOSS philosophies are the common bond. Many of the command line and graphical tools are the same or function the same in both systems. As an example, developers for both Linux and BSD use OpenSSH to connect to their server, Vim or Emacs to edit their source code and GCC to compile it. Both BSD and Linux servers commonly run the Apache web server, Samba and CUPS for file and printer sharing and MySQL or PostgreSQL for database work. If you have a BSD or Linux desktop then you probably run GNOME or KDE on top of X.org and browse the web with Firefox.

MN: It used to be that Linux was a copy of UNIX, following on the coat-tails of the likes of BSD and Solaris. However, in the last 10 years it has gone ahead in many key ways. The first catalyst for change was better driver support for common PC computer systems which gave it a big boost. Then, as more users and developers adopted Linux, it started to become the platform for innovation and BSD and UNIX trailed behind. Now it is common to see new software updates including new features and improvements released first to Linux and then made compatible with BSD.

MN: There are several variants of BSD, each with different merits. Two excellent examples are OpenBSD and NetBSD.

OpenBSD developers are motivated by a desire to maintain their outstanding security track-record. They review their code and implement features that help ensure that the operating system will withstand even the most motivated attack. Furthermore, it has also historically boasted one of the most robust TCP/IP networking stacks. These two features combine to make it an excellent choice for networking infrastructure.

NetBSD has as a core value the desire to accommodate a great variety of platforms. It runs on a diverse collection of computers so its maintainers strive to create a system that is flexible and portable.