Should Public Funds Mean Public Code?

Lisa points to this article on oreillynet with "two opposing viewpoints on whether all software created by publicly funded research should be licensed as open source, and the chance to weigh in yourself." Open-source software (under whatever license) seems to me like a good way to multiply the investment of tax dollars that public funding relies on, but the counterarguments offered here are interesting.

Comments

FOIA and government source code

[dfelznic]

Hello, I think it would be very interesting to make an FOIA request for the source code to some small insignifigant government application. The source code is produced via public funds and belongs to the people just like any other government document. Anyone ever tried this?

Re:FOIA and government source code

[gorillasoft]

While you could potentially get the source to a "small, insignificant program," it won't necessarily work. There are a variety of exclusions that an agency could use to keep the source code private, and just about any of the ones below could be bended to prevent release.

From the DOJ: The exemptions authorize federal agencies to withhold information covering: (1) classified national defense and foreign relations information; (2) internal agency rules and practices; (3) information that is prohibited from disclosure by another federal law; (4) trade secrets and other confidential business information; (5) inter-agency or intra-agency communications that are protected by legal privileges; (6) information involving matters of personal privacy; (7) certain types of information compiled for law enforcement purposes; (8) information relating to the supervision of financial institutions; and (9) geological information on wells. The three exclusions, which are rarely used, pertain to especially sensitive law enforcement and national security matters.

So, as you can see, the FOIA does *not* mean you have access to everything.

Here is more information:
FOIA Reference Guide

Re:FOIA and government source code

[gnovos]

Sure, but it would look like this:

for (int i=[BLACK];i&lt[BLACK];i++) {

if ([BLACK]) {
callFunction([BLACK]);
} else {
[BLACK];
}
[BLACK](i);
}

To some extent

[RazzleFrog]

I agree with Mr. Dalke in that you can't just apply blanket policies like this. There are always going to be exceptions and fuzzy areas. To absolutely force all public-funded works to provide the source to their projects without considering special cases seems negligent at best.

code is no different

[bokmann]

Code is no different than any other property...

Are Television shows created with public funds available for my use as source material in my own movies?

Are works of art (like the infamous Mapelthorpe photos) considered in the 'public domain'?

I honestly don't know the answer, but I'm sure someone has thought about this in another domain. I wish people would stop thinking that code/cyberspace is really as new and challenging as it seems.

-db

Re:code is no different

[dfenstrate]

I don't know about PBS for copyright information, but in general, any pictures, images, or film taken by State or Federal employees are public domain. There are certain limitations on this- for example, classified information, or the use of military insignia, but the overwhelming majority of the material put out is free for you to use. So yes, you could use it as source material in your own movies, though common decency demands you give credit where it's due.

This picture is a classic example. It was one of the most stunning photos to come out of the montana forest fires- the low resolution of the picture above doesn't do it justice- and any AP photographer would have killed to have the rights to it. But the picture was taken by an on-duty USGS employee, and hence, everyone gets to use it.

So in several cases, yes, publically funded stuff is in the public domain.
I think it all should be, but that would doubtless deprive my University of some much needed cash that the state will never give us.

Re:code is no different

[DGolden]

Bah!

Code, television shows and some works of art are ALL different to most forms of property!

Most forms of property have a physical existence and are not infinitely copyable. They are called "naturally scarce" in the jargon. If I take your car, you don't have it anymore.

Code, television shows, music, films, and the written word, are all simply patterns of information, that ARE infinitely copyable. They are "non-scarce".

Our current social structure sometimes creates "artificial scarcity" out of certain "non-scarce" abstact things like patterns of information - one such artificially scarce social construct is (mis)named "intellectual property".

Unlike real, physical property, you can give me a copy of the underlying information pattern without destroying your own copy, at what is practially near-zero energy/cost. Thus, it is FUNDAMENTALLY DIFFERENT to most things contemporary society badges "property".

One of the things that is happening now is that Joe Soap on the street has non-scarce access to increasing numbers of goods that are currently of merely artificially scarce statys, such as: digital music, films, etc., as well as program code.

The old megacorporations which relied on keeping such patterns of information artificially scarce for their business model are now fighting desperately to keep them that way, while millions of Joe Soaps move to bypass them in copying the non-scarce information patterns.

Remember, intellectual property is only a societal convention - if millions upon millions of people start to ignore it, then bulk society has changed, and the very concept of I.P. is obsolete in its current form - Would you have sided with the scribes when the printing press was invented? Will you side with the factory workers when/if nanotech renders physical items effectively non-scarce, and the factories become obsolete? How about when/if the march of programs effectively eliminates energy/matter scarcity altogether, and things like money and present-day economic systems like communism and capitalism all become obsolete? It could happen, and lots of people are already thinking about it - see Iain Banks' "Culture" novels.

Yes, the abandonment of the profit motive might result in less code/music/films. But personally, having seen the quality of today's code/music/films, I'm one of those people of the view that the best ones will still be made - since the best ones tend not to be made solely for profit.

Author of anti-OSS article has a misconception

[yerricde]

The author of the article opposing release of publicly funded works under an open source license seems to have a misconception as to what common free software licenses say constitutes source code. From the anti-OSS article:

Open source licenses rarely require that local changes be distributed. Open source licenses do not set a limit on the fees charged. Open source licenses set no restriction on when, how, or where the source is distributed (with minor exceptions). As an open source publisher I am free to release my source code only once a year, at a charge of $1 million paid at least two months in advance, and you have to accept it on paper tape while we are both standing under the Eiffel Tower. (I'll cover my own travel arrangements if you take me up on this.) If I am the original copyright holder I'm even allowed to obfuscate the code by removing comments, using nonsense variable names, and other tricks.

This conflicts with the most common definition of source code. The GNU General Public License, one of the most popular free software licenses, specifies the following in section 3: "Accompany it with the complete corresponding machine-readable source code, which must be distributed under the terms of Sections 1 and 2 above on a medium customarily used for software interchange," that is, something other than paper tape. Also, "The source code for a work means the preferred form of the work for making modifications to it," meaning that if reasonable comments aid modification, leave them in.

Government Software Research Black Hole

[Courageous]

I've been working in the government funded software research and development community for a decade and I must have seen as much as $50 million or more of the Public's dollars go into the giant black hole of software research.

One of the major problems of goverment funded research, even when it is contractually bound to be open for government-related inspection and use (which most of it is) is that the various players all jealously guard their turf. This includes other contractors who, even when legitimately approached for copies of the source which they are contractually bound to give you, curiously develop problems getting messages, getting back to you, shipping you source, and providing you access. You'd think it would end there, but no.

The government players themselves jealously guard their turf. Since there is similar and even duplicative work funded across DoD and government, government reps have no desire at all to share. They view the other similar projects as competitive and worry that if one of them gets the upper hand, their own project will be unfunded as redundant or irrelevant. This creates a situation where the government players -- those who are supposed to be working for the Public Trust -- instead drag their feet and use passive resistance in giving up software to even those who are allowed to see it, such as other members of goverment or government contractors working on the government's behalf.

The end result of all of this is that enormous sums of software gets locked up in boxes and never sees the light of day. About the only person who actively looks at the source is the original contractor. For research efforts, its understandable and reasonable that a research project doesn't result in a piece of software that's used by either no one or the very few. However, what's not not reasonable is that the information itself is effectively vaporized.

This is a completely execrable situation and grossly violates the Public Trust. Not only is the system vastly wasteful of the public dollar, it likewise violates the entire basis of public research: the open sharing of information.

For some time now, a sort of jewel in my mind's eye has been glimmering, and it goes like this:

All goverment software development, with the exception of sensitive projects, should be forced into placement under open license into a high profile source repostory such as Source Forge. This, for every government project, would be the primary CVS repository of the project. Project developers would insert code here and be subject to detailed public scrutiny with default anonymous CVS read access.

In my opinion this would blow open the doors of enormous amounts of software development and be of enormous benefit to the general public. Consider how neatly nipped in the bud all the beaureacratic foot-dragging would be. Intermediaries? None. You want the source code? CLICK.

This should be the new standard of non-classified government software development. The money belongs to the People, dammit! So should the research.

C//

Re:YES!

[Artagel]

Having worked in a research lab, I can tell you that who paid for what is often very, very unclear.

Consider a researcher has a research program that has been going for 20 years. He may have been funced at various times by: a) the university (often academic researchers start work on the university's dime), b) government grants, c) private foundation grants, d) his own money. His graduate students may have been supported by any of those various sources or independent scholarship sources. Perhaps a tuition-paying undergraduate contributed some code at some time also.

Just like research equipment, code also accumulates. It may have been traded from other research groups. It is a real mess. Figuring out who owns what can be more or less impossible unless you have a dedicated ground-up code block that is identifiable to a project that has no contributions from anything other than public funding. (Oops, you have a Hertz fellow write 2,000 lines of the 100,000 lines of code... now what?)

Odds are that there are already too many rules regarding the code. Adding another one is just going to bollix up an already intractable mess.

Complete misinterpretation of the copyright clause

[jms]

Andrew Dalke writes:

"To promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries;" -- Article I, Section 8, Clause 8

That seems to me a very clear statement that a public good--the Progress of Science--can be achieved by keeping "Writings" exclusive to the author.

This is a complete misinterpretation of the clause. If you go back to the original copyright laws -- the ones written by the authors of the Constitution, you will find that the Framers required, as a condition of copyright, that the works be published and distributed to the public in order to qualify for copyright. Works that were kept "exclusive" to the author, were ineligible for copyright. This is how patent law works -- if you want a patent, you must disclose to the public how your invention works. You cannot obtain a patent on a device, and simultaneously keep the operation of that device a secret. This is how copyright originally worked, before the 1976 rewrite.

In phrase "exclusive Right" was intended to be understood in the context of required publication. The "exclusive Right" is the right to exclude others from duplicating your invention or writing, not the keeping of writings "exclusive" to the author. The reason why the granting of exclusive rights -- the right to exclude -- was considered a tolerable evil was quite simple -- Such rights were only granted on condition of publication! The public good was that the works were published so that the public could learn from them, and from their examples, create new works! Hardly the case with the "licensed, unpublished, proprietary code" that Dalke is so fond of.

In fact, the keeping of writings exclusively to the author is exactly the problem that copyright was invented to solve!

The first copyright law covered books, charts, and maps. The inclusion of maps was no accident or afterthought. One of the problems in 18th century navigation was a lack of accurate maps. Mapmaking was a difficult, time-consuming, expensive process -- just as software development is today -- and with no way for mapmakers to protect their investments, they resorted to licensing agreements to restrict their users, just as software companies do today. With all of these secret maps, licensed restrictively to ship captains, very little progress was being made in accurate map-making. The problem was that no one could legally compare maps to each other, because all of the maps were locked up under non-disclosure agreements. Copyright was intended to change the situation by granting a monopoly over the reproduction of books, maps, and charts, in exchange for open publication of the works.

Dalke's misinterpretation turns the entire purpose of the Monopoly clause on its head. On the other hand, he can be forgiven for not understanding the purpose of copyright -- most of copyright law has been turned on its head in the last quarter century, beginning with the disasterous rewriting of the copyright code in 1976, and continuing with the disasterous decision to grant copyright protection to object code, and not requiring the publication of source code.

I have a brief analogy. Imagine that you, a young student, aspiring to become a novelist. A good teacher would tell you to read as many novels as you can by your favorite authors, because it's only by reading other people's great works, that you learn how to create your own great works. Imagine if you were told, "If you want to be a novelist, you may not read other people's novels -- that's illegal. You have two choices -- either learn to write from scratch, starting from grammar books, and moving on to short stories, and finally novels -- or alternatively, you can get a job with a book publisher which will permit you to read other people's novels, under strict non-disclosure agreements.

I don't think that such a system would result in very good novels, but that's exactly the situation with computer software. There's plenty of good and bad computer software, and millions of young computer programmers who would like nothing more then to be able to read and learn from that software, but the vast majority of it is locked up, never to be seen by more then a handful of people. Such software does NOT advance the progress of science. Science is advanced by publication, not by secrecy. Dalke's theory seems to be that things are ok, because "real" researchers like himself have access to the source code through their institutions, but for every elite, privileged researcher who has access to the source code, there are thousands of other people who do not, and are unable to contribute anything. They are locked out.

If we really wanted to improve the state of software, and everyone talks about how poor the quality of commercial software is, the first step is to require, as a condition of copyright, the publication of complete source code in conjunction with any object code. The problem is that our copyright law, particularly with respect to software, is so completely dysfunctional, that it no longer serves its purpose -- to build a public domain that others can draw from, learn, and improve upon.

The sole exception to the software copyright fiasco is the software published under the GPL. By mandating public disclosure of source code, programs published under the GPL fulfill the original purpose -- and mandate of copyright. The results validate the original purpose and design of copyright -- to promote scientific progress, by providing an openly published base of work that can be built and expanded on by others.