posted by Michael L. Love on Mon 3rd Nov 2003 19:19 UTC

"Bringing GNU/Free Software to all users"
GNU-Darwin is able to bring GNU and other free software to all users

This year the Distribution announced the 1.0 revision. We continue to provide about 270 one_stop style installations per month from the gnu-darwin.org mirror alone (ppc:200,x86:70). This number has been fairly stable (on average) for almost a year. Thus, we are providing many thousands of packages to one_stop users each month and probably around 60,000 (+/-15k) overall per month, as estimated from changes in package deployments since October 2001. Our cd images have regularly appeared in the top 100 files for the whole internet, and our website was also in the top 100 at the commencement of the Free Darwin action last December. This year we enabled shell access for all of our web service users.

A couple of people have noticed that our SourceForge statistics appear to be low. There are a number of factors that affect this. At the center of our work is the package collection, which typically does not show up on the SourceForge activity monitors at all. In another example, I have been doing most of my CVS maintenance using imports instead of commits, but only commits show up on the SourceForge tables. Until recently, our package collection was frozen for stability, development, and strategic purposes, and we were only doing maintenance updates, such as security releases and fixes to key packages. During the freeze, we were focused on publicity and GNU-Darwin Internet Services development among other things. Since that time, we have begun to make our 1.1 package revisions available. The GNU-Darwin statistics at SourceForge are severely skewed, and we are one of the largest, most active, and popular projects there.

Given these remarkable facts and numbers, our support load is incredibly low and manageable, probably due to growing expertise among our users and also the excellent FreeBSD and SourceForge based infrastructure, which supports the GNU-Darwin project. It is difficult to relate these numbers to our total usership, but it appears to be well into the thousands at least, which is verifiable with several independent metrics. Moreover, we are able to scale our numbers to the total OS X usership and whole internet, and everything cross-validates very neatly. It is also safe to assume from all of this that our usership has been growing at a slow steady rate for a long time. This growth was assisted by the Free Darwin and anti-war actions and unhindered by the rudeness of some of our critics.

Our project is aptly named. Look at Darwin itself, and you will see that tons of free software has been added to it since 2000. Nearly all of it was first introduced to the platform by GNU-Darwin, and whatever else they decide to add, it is likely that we have already done it. Pre-2000, the whole system was built with GNU tools, and the process of adding GNU software had already begun. The kernel is even named XNU, which is certainly a knockoff (RMS probably rejected that name). GNU-Darwin is a standalone free OS and software distribution system, so we might never be in the position of RMS, begging for justice in the naming of these things. If you look at our hard media offerings, you will see that we provide about 12 gigabytes in addition to the Darwin installer CD. That is over 20:1, and no one else comes close to matching this, even Apple itself. All of it is either GNU, GPL, or free; so that we are a unique distributor.

From the very beginning, we were about bringing GNU to Darwin. We have come back to the beginning, and that work is finished and bona fide. Take all this together, and it is clear that we have ample room to redefine ourselves in better accordance with the goals of the FSF without breaking our linkage to Darwin. This adds immeasurably to the unique flavor of our Distro, and closes the circle as it were. Moreover, we are now entirely free to compete in the whole technology sector; Apple, Microsoft, RedHat, whatever, and to make our impact on Debian and the rest of the free software community. It is happening already, and we should not be conservative in our aims at this time.

We are the unique OS distributor at the juncture between FSF and Apple, and we already have the software update technology in place to make that count. We can fill the gap entirely, but in order to be credible, there are a couple of things that we should keep in mind. We must minimize our mention of non-free software. When we speak of proprietary software at all, it should be that we intend to replace it with our own offerings. We want to replace M$ and Adobe products with our offerings. We want to replace Windows with GNU-Darwin. We want to replace Mac OS X with GNU-Darwin. We even want to replace RedHat, Suse, and FreeBSD. This consistent approach works fine with new Apple users and advanced *nix users as well. All Mac OS related statements on our website have been moved to the "about page". Moreover, we must now reiterate that our Distribution has always been about bringing GNU to Darwin, so GNU will now have equal billing with Darwin as it were. If we do these things, we will never play second fiddle to anyone in industry or in the free software sector, and GNU-Darwin will spread because of what it is, not because of its relation to other things.

Unlike Debian, GNU-Darwin will have no non-free directories, but rather we will delete non-free items from our servers. There will be a little give and take on what exactly constitutes "free", because we are coming from an entirely different origin than Debian. It is important to note that GNU-Darwin is not a democracy. I have been making most of the crucial decisions without consulting anyone, but each of our users feel their freedom, are at ease making suggestions, etc. Especially for GNU-Darwin, there is difference between freedom and democracy. Overall, there may be differences between GNU-Darwin and Debian, but we can cooperate with them on our shared goals. Metaphorically, GNU-Darwin and Debian can be seen as two barrels of one shotgun.

Thanks to the FSF bounty program, publicity actions by our users, and Apple's recent update, we have had the most visits to our website this month since the time of the Free Darwin action in December and January. Our ad revenue this month was extraordinary, and I'm happy to report that the project has become financially self-sustaining. Last week, we ran Google ads targeting "Linux" users. The exercise was very instructive, and resulted in the new "OpenOffice for ppc" link on the homepage, among other things. (Now, if we could only provide salaries for our developers).

We have stepped up our outreach to free software developers, and improved our cooperation with the community. This effort resulted directly in the soon to be released GNUTLS port, a crucial alternative to OpenSSL and proprietary TLS systems. Our relationship with the Dillo project continues, and we have a little surprise planned (more about that later). If you have a favorite free software project that is not yet available for GNU-Darwin users, be sure and tell the developers that GNU-Darwin wants to help them. Have them email us.

GNU-Darwin appears to be a collegial meritocracy, which also appears to be an excellent way to run a project in the fast paced world of changing technology. Each of our developers was selected (not elected) because of their level of engagement, talents, and contributions. We obviously could not have made the impact that we have, if we were encumbered by the democratic process/bureaucracy of some other distros. Our agility is based on the fact that the people with the greatest experience, know-how, and hands-on ability are making their own decisions and contributions on the fly, seeking each others advice only as required in order to do the job. We are only just getting started. We are not a country. We are building an institution of vast potential and longevity. We will continue our activism in the realms of software freedom and digital liberties, and GNU-Darwin will spread far and wide.

About the author:
Dr. Michael L. Love is currently a postdoctoral associate with the Cornell University high energy synchrotron source at MacCHESS and in the Department of Molecular Biology and Genetics. Dr. Love founded GNU-Darwin in order to advance his work at the facility, although the project is also known for free software advocacy.

Table of contents
  1. "Intro, an activist distribution"
  2. "Bringing GNU/Free Software to all users"
e p (0)    82 Comment(s)

Technology White Papers

See More