David A. Wheeler's Blog

Sun, 08 Jul 2007

Increasing Government Interest in Free-Libre / Open Source Software (FLOSS)

There’s more on the web now about governments and Free-Libre / Open Source Software (FLOSS), both on my own site and other sites. I think this suggests more and more interest by governments in FLOSS.

I’ve now posted Open Source Software (OSS) in U.S. Government Acquisitions, which is a slightly-updated version of my article which was published in the “DoD Software Tech News” issue devoted to free-libre / open source software (FLOSS). As I noted earlier, this issue has lots of good articles. I was asked to create the article (not the other way around), so I think the very existance of this issue is evidence of increasing interest, not just a personal interest of mine.

I should probably mention why I’m posting a revision. After I submitted my article, the Department of the Navy CIO Robert J. Carey signed a June 5, 2007 memorandum titled “Department of the Navy Open Source Software Guidance” which notes that FLOSS needs to be considered a commercial item when it meets the U.S. government’s standard definition of a commercial item (nearly all extant FLOSS does). That is the same basic point that I raised in my presentation and paper, and I would have gladly mentioned the Navy memo… had it been released at the time. So my modified version now points to the Navy memo, as well as making a few tweaks based on other feedback. This realization that FLOSS programs are (almost always) commercial items is an important point in the U.S. government. Why? Because as noted in Linux.com, that means that extant FLOSS software must be considered when the U.S. government acquires software the same way as other commercial software is considered (i.e., it must be considered before starting a new project to write their own). The Navy memo’s assertion of this makes it worth posting an update.

But think of this updated essay as a only sampler… for the rest of the articles, you’ll still need to go read the “DoD Software Tech News” issue. Getting the issue does require registration, but registration is free and in this case it’s worthwhile.

Anyway, it’s not just one issue about FLOSS in one magazine. Here’s more evidence it’s not just me - on July 5, 2007, the article Open Source Government: good-will needed was posted by Roberto Galoppini’s Commercial Open Source Software blog. He points in turn to various articles, including Matt Asay’s “Open source in government: Leadership needed”, which then leads you back to a very interesting research paper: “Open-Source Collaboration in the Public Sector: The Need for Leadership and Value” by Michael P. Hamel. There’s lots of interesting discussion in those articles about how governments can use FLOSS, and in particular how governments can use FLOSS components and approaches more effectively. I’ll throw in my own two cents here. I would certainly agree that leadership is important in any project, including FLOSS projects; the leadership of Linus Torvalds of the Linux kernel is well-documented, for example. But there are different kinds and levels of leadership. Roberto Galoppini says, “I think that first we need politicians with good-will, willing to put their intellectual potential to work for the overall desires of the general public…”; while that’s good, often what’s important is the ability to lead concept into practice. It’s easy to “lead” by saying words, but I think that often what’s needed is the kind of leadership that rolls up its sleeves and makes actual projects produce useful results. Focusing on specific, measurable products can often get better results. Certainly politicians with goodwill are a good thing, but they can provide value primarily by providing “cover” for those who actually do the hands-on leading of projects, but the latter make or break such efforts.

While I can quibble with some of the stuff Hamel says, some of the his statements seem dead-on with what I see. Hamel notes that “participants in both groups also believe that the creation of value, or products that are appropriate and effective in addressing members’ wants and needs, is critical”… to which I say, Amen. Hamel concludes that “Collaborations with a strong leadership structure, and more importantly a single leader who is persistent, passionate and willing to spend a great deal of time maintaining and improving the organization are much more likely to succeed. Value is also a critical component, and requires that efforts meet the wants and needs of members and clients, whether they be in the form of software, documentation, research or even policy advocacy.” Focusing on a few most useful projects is critical: “a conscious effort to focus energy on a small number of projects in early stages may be an important component in creating value for members of collaborative efforts.” A FLOSS project requires collaboration to be successful, and collaboration requires that the project gain the trust of potential users/developers; “In this research I found that leadership, face-to-face contact, and the legal framework were the primary factors leading to trust. A willingness and ability to evolve, which may be tied to creating products of value to clients and members, might also be an important factor in developing a successful collaboration.” Those statements, at least, seem very sound. These aren’t new ideas, to be sure, but it sure is easy to lose sight of them.

I think there’s a vast opportunity here for governments to use FLOSS. Which is odd in a sense, because in fact, FLOSS is already widely used in governments. MITRE determined back in 2003 that FLOSS was already critically important to the U.S. Department of Defense (DoD). But that widespread use is small compared to its potential, and to how many commercial organizations use FLOSS, and there are a lot of reasons for that. Governments have long aquisition times, often tend to avoid doing anything “risky” (aka “different than the way we did it before”, abetted because there’s no competitor to force them to improve), and many proprietary vendors target governments to try to prevent them from using competing FLOSS products. Which means that even when it makes sense to use FLOSS, it can often take much longer for FLOSS components and approaches to enter into government use. But it has already entered into significant use, and I expect that to accelerate over the years.

There are some other goodies on my web site if you’re interested in this topic (government and FLOSS). My essay is basically a text version of my March 2007 presentation on Free-libre / Open Source Software (FLOSS). It includes snippets from much longer papers that give statistics about FLOSS, explain how to evaluate FLOSS, discuss FLOSS and security, and explain why most FLOSS is commercial software. There’s even a hint of my essay Make Your Open Source Software GPL-Compatible. Or Else, since I note the widespread use of the GPL.

Again, if you have a true PHB, I can’t help you. But many managers are trying to the right thing, and just need reasonable information so that they can do the right thing.

path: /oss | Current Weblog | permanent link to this entry