October 26, 2002

Here is the text of the emails I sent to Congressman Tom Davis, the Virginia Republican who chairs the Technology Subcommittee of the House Government Reform Committee, and Congressman Jim Turner, the ranking Democratic member of that committee. This refers to two letters: one dated September 18, 2002, written by Davis and Turner to Richard Clarke, the chair of the President's Critical Infrastructure Protection Board; the other was written by Rep. Adam Smith, a Washington Democrat, to members of the New Democrat Coalition, urging them to sign off in support of the Davis-Turner letter, with the implication that they oppose the use of the GPL to license the results of federal software development and research.

The link to the story and the original letters is here: newsvac.newsforge.com/newsvac/02/10/23/1247236.shtml?tid=4

Dear Congressman:

I am writing in regard to your letter of September 18, 2002 addressed to Richard Clarke, the chairman of the the President's Critical Infrastructure Protection Board, as it pertains to the National Strategy to Secure Cyberspace. I note that you recently had to clarify the meaning of your letter, in light of the attempts of other members of Congress to mischaracterize your message to Mr. Clarke. I am referring to the letter that Representatives Adam Smith, Jim Davis, and Ron Kind sent to members of the New Democrat Coalition, implying that you oppose the use of federal funds to support research and development of computer software that might be released under the General Public License. I commend you for making it plain that your comments referring to "intellectual property licenses that allow for further development and commercialization of that work product" do not in any way imply support for one kind of intellectual property license over another.

As an information technology consultant who uses computer software developed under both proprietary and open source development models, I understand the merits and drawbacks of both approaches. Reasonable people within the information technology community often disagree over whether proprietary or open source software is better in terms of performance, security, and user-friendliness. What I think we can all agree on is that open standards, such as the TCP/IP network protocol, are essential to the advancement of information technology in this country. The research efforts of the federal government to further develop technologies, particularly where it concerns Internet security, are much appreciated by those of us in the private sector who implement technology.

Representative Smith, et. al. seem to be taking the position that if the federal government endorses one kind of intellectual property arrangement over another, it would be better for America's information technology industries. They seem not to understand that new technologies that are developed by the federal government, and which are not under a security classification, are essentially part of the public domain. Thus, there is no intellectual property license involved in the use of these technologies. TCP/IP is a prime example of this model at work: it is the network protocol that runs the Internet and most private networks around the world. As an accepted standard in the public domain, it is owned by nobody, and therefore, intellectual property considerations do not apply. TCP/IP is used as a network protocol in both proprietary and open-source software products, and without such a common standard, the idea of an Internet would be an impossibility.

I would like to express my support for the current policy of the federal government, in releasing the results of non-classified research in information technology for the public consumption. This is a practice that allows for all interested parties to use the technology for the benefit of the broader society. The General Public License, far from contradicting this principle, supports the notion of open standards and the sharing of relevant information. At the same time, the GPL also expressly allows for the commercial sale of software under this license (Part 1), in the form of charging a fee for copying and distribution. The differences between GPL software and software licensed from proprietary software developers is that GPL software may not be patented, and that the source code for the software must be made available to the purchaser (Part 3), in the interest of maintaining open standards and hardware compatibility. In addition, the GPL makes it clear that any non-GPL software that is "aggregated" with GPL software (such as proprietary software bundled with an operating system) does not fall under the terms of the GPL (Part 2).

I would also like to express my opposition to forbidding any licensing arrangement as a pre-condition for using the results of federal technology research. This is what Representative Smith et. al. seem to believe you are calling for in your letter to Mr. Clarke. There is nothing I can think of that would damage America's information technology industries more than to restrict the use of new technologies to certain commercial licensing arrangements. The availability of research results to all interested parties is of paramount importance if the adoption of the best technologies as standards is to be promoted. In addition, such an arrangement promotes competition by lowering barriers to entry for smaller firms who wish to produce derivative software or provide services based on the new technology. Finally, it is in the federal government's interest to assure that the results of this research are open, in keeping with the spirit of the promotion of knowledge in the information technology industry and the economy as a whole.

I thank you for your efforts as a member of the Government Reform Committee's Subcommittee on Technology in promoting the security of the Internet, and I hope you will continue to support federal research efforts whose fruits are available to all.

As you can see, I kept it on the positive side-- Davis and Turner were willing to clarify their position, and they might be amenable to agreeing with those of us that federal research on something as important as Internet security should be open. This is a point I stressed about TCP/IP-- it's a DARPA research project that ended up running the Internet.

I also thought it was important to emphasize how GPL'd software was different from proprietary-license software, and point out that some of the things Rep. Smith was implying about the GPL is not true. I think I stated the case clearly and cogently.

mrkurt

Hello Everyone!

I don't know why I didn't do this sooner-- become a Slashdot user and start journaling, that is. I suppose it took me awhile before I saw the need to actively participate in the Slashdot community. I have been reading the stories and posts for about five months now, and I find the opinions and insights of participants to be interesting, informative, and vital to my understanding the open source community.

My joining Slashdot is my official declaration of my embrace of that community, too. This year has been a progression for me from the world of proprietary software to that of open source. I bought my first Linux distro (Mandrake 8.0) for $25 at Half Price Books in January. I finally got Linux because I didn't want to spend $1000 for Windows 2000 Server-- Uncle Bill doesn't need any more of my money. I first installed it as a dual boot on a machine running Win 98. It took three times to really get the install right-- in other words, where I had a big enough partition for all the software I wanted to install for Linux. I didn't install any of the real server processes until I put in a home network in June. Then, I blew away the Windows partition on that old machine (a Celeron 400 mhz with 64 MB of memory), and reinstalled MDK with all the server processes. Sometimes, the desktop programs run kind of slow, and I am always dealing with "flaky" fonts (guess it's time to copy my TTFs to my Linux machine), but otherwise it has been an enlightening experience. I am glad to be one foot out of the Windows box; hopefully soon, I will be out of Windows altogether. Linux is the server on the network; there are two Windows 2000 machines, one a desktop PC, the other a laptop. I am contemplating buying another PC and installing one of the newer Linux distros; we'll see if my budget can afford it.

I had Apache web server up and running to test both static and CGI pages that I wrote. The CGI pages that took variables from another page didn't send back any dynamically-generated content; I imagine it has to do with settings and a module needing to be enabled. These same pages worked OK on another server's cgi-bin. I also have been able to get Samba running at times; the last time was about two weeks ago, where I had it running for three days. It went down again, after I made the mistake of going to the Microsoft TechNet site to look into the Microsoft Java VM exploit. I went there on behalf of Windows users I support at my church; my "sin" was checking out the patch via Windows Update, to see how large it was. Maybe I'm paranoid, but I got some message that I couldn't access the Windows Update, and all I know is that my Win2k machines couldn't access anything on the Linux machine after that; I can't even ping it from the Windoze boxes. I wonder if it doesn't also have to do with me installing Tiny Firewall about the same time. I later disabled it, but no matter-- I still can't access the Linux machine. =:

As my bio line states, I am a "recovering VB developer"-- I am currently working on my first (and I wish it were my last) VB project. I am doing mostly server-side stuff: DLLs accessing a database and processing data to send to the presentation tier. I have also done some GUIs, mostly forms, but also some front-ends with Excel spreadsheets for reports. I would like to be on the server-side where it comes to programming. Obviously, I saw the need a while ago to move away from Visual Basic to a "real" programming language, if I wanted to develop server-side components. The use of ProgIDs instead of GUIDs in VB is a hindrance to using VB 6 for server-side processes without having to rewrite or unreg/reregister the rest of your software. It's the dumbest thing MS ever did, if they wanted VB to be taken seriously. This problem has been corrected with VB.net, but that's a topic for another Journal entry/rant.

The organization for whom I am currently writing software is a small youth services agency, and I was hired because of alumni ties. The language tool students learn at this univ. (in-depth) is VB; the agency is an all-Windoze enterprise. Ultimately, the agency wants to try to sell this application to other agencies around the state, and it looks like they may already have some buy-in from some folks. There were three other developers who were working on the server-side stuff when we started (almost nine months ago), all of whom have since left. Because I was a consultant who actually knew what he was doing, I reviewed the other developers' code to make sure that it conforms to our needs for using COM/DCOM. Now I'm the only developer standing, aside from another programmer who was hired about two months ago and is working on the front ends. I have now been writing components to generate reports for four months now, and I have pretty much been working on my own to make this stuff work. The components I'm writing are mission-critical to the agency, so we have gone through two iterations already-- this latest version is meant to be real-time, and didn't end up requiring much modification in code. It took me so long (about a month), I suspect, because I have grown wary of the whole MS package.

Well, that is all I will write for now; I don't want you all to get bored. I'd be especially interested in hearing from you if you are in a situation similar to mine-- making that transition to open source. I'd like to hear what your experiences have been like.

mrkurt