FLOSS study by Defense Research and Development Canada

Joseph Potvin wrote:

Please visit the Open Source Software Website of Treasury Board Secretariat's "Chief Information Officer Branch". They have posted a bilingual study "Free and Open Source Software" by Defense Research and Development Canada (a branch of Dept of National Defense) and they have also added a link to the Government of Québec (GoQ) open source website.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

Reply to my feedback from Robert Charpentier of DRDC

THanks for your comments

They will be passed to our Document Review Committee that has endorsed the publication of our document (see validation cycle one). We would like to invite you to support your comments with credible and rigourous scientific references. We do not cite personal opinions in such technical report in order to stay out of "religious thinking". The validation process used to review the content of our report involved more than 65 inviduals from various spheres and technical background (for a period of 14 months). However, we realise that we cannot please everybody with everything we say, all the time and everywhere. By supporting your comments with credible scientific references, you would reatly improve your chances of receiving attention for the drc.

Thanks again for yor interest in GoC publications.

au plaisir

Robert

-----Original Message-----

From: Russell McOrmond

Sent: Friday, April 29, 2005 11:13 AM

To: FOSS -at- drdc-rddc.gc.ca

Cc: doucet.gary -at- tbs-sct.gc.ca; GOSLING members in Ottawa

Subject: Few quick comments on:
http://www.cio-dpi.gc.ca/fap-paf/oss-ll/foss-llo/foss-llo00_e.asp

First a quick introduction: I'm co-coordinator of http://GOSLINGcommunity.org which is a multi-sector community of practice working to "Get Open Source Logic INto Governments". We meet weekly informally each Friday in the Ottawa region, and will soon be moving to a format where one of those Fridays includes a more formal workshop.

I have been a commercial supplier of FLOSS services for more than a decade, before the term "Open Source" was coined. In that full spectrum of development, distribution and software funding models that the government is intended to support, I almost exclusively use peer-production, peer-distribution and pre-payment (vs. post-payment royalties).

The following is not meant to be critical of the work you have done so far, but sent with the hope that the document can be improved. Please take any comments with that in mind.

Right from the Executive Summary I find I am instantly mixed into the "language game" that we waste so much time in in this movement. It is also one of the points of vulnerability that those opposed to a full spectrum (IE: incumbent software-manufacturing vendors such as Microsoft) have been abusing to cause uncertainty in the marketplace.

COTS -- The term "Commercial Off-the-Shelf" is usually used to mean software that is available commercially in a packaged format. This includes a full spectrum of development (software manufacturing to peer-production/FLOSS) and funding models (pre-payment to post-payment).

This is a distribution model (Retail vs peer-distribution/P2P), and should not be contrasted with FLOSS which is a licensing and development model. FLOSS vs. COTS does not make sense considering these two terms are on different axis.

4. Main Definitions

You quickly get into some of the FUD in the sector with the use of the term "practical" to mean non-copyleft and “FSF Idealism” when talking of copyleft. In reality these simply represent a full spectrum of motivations for the creation of software, and it is incorrect (and quite harmful) to consider the motivations of some people to be practical and others to be ideological.

My personal views are opposite to those opinions presented in this document, and consider Copyleft to be a practical anti-theft mechanism and consider support for legacy software manufacturing/post-payment to be old-economy ideological. I am not interested in "giving away" my FLOSS software any more than Microsoft is interested in "giving away" their non-FLOSS software, I just expect a different form of payment. Adherence to my copyleft license supports my business model, which includes legal cost-avoidance involved in post-payment funding models (the so-called "software theft" problem that organizations such as the BSA and CAAST claim they are interested in stopping) which cause some legacy vendors to be motivated to sue their own customers.

While these are my personal views, I would never include them in a government document which should support and encourage a full spectrum of models. In fact, each time you repeated the phrase "(more practical approach)" it grated on me, suggesting that my chosen models from the full spectrum are somehow considered "lesser".

I looked at it for a while, and could not make heads-or-tails of the diagram at the bottom of http://www.cio-dpi.gc.ca/fap-paf/oss-ll/foss-llo/foss-llo03_e.asp

What definitions of the word "proprietary" are you using? There are two (incompatible) ones I have seen:

a) a derogatory term used by many in the FLOSS community to refer to non-FLOSS. In this case the question is not whether there is an owner, but how many owners: proprietary software has a vendor as the owner who retains many exclusive rights, while FLOSS has many owners (each current and future contributor) which could theoretically include all literate beings.

b) A legal term referencing exclusive rights such as copyrights. Software that is in the public domain has "no owners" as there is no copyright holder. Software not in the public domain then has a license agreement. While the public domain is a special case of non-copyleft FLOSS, the vast majority of FLOSS is under copyright and has a specific copyright (and possibly patent and trademark) license agreement.

Neither of these definitions fit your diagram, nor can I make up a definition that makes sense that fits the diagram.

Here is a sample of the diagram I often use which is far simpler and easier to understand: http://www.flora.ca/osw2004/img7.html

5. FOSS Legal Background

"It is expected however that less restrictive license models (such as Mozilla and BSD) will be more popular in the future [4], since hybrid proprietary/FOSS systems are more appropriate to most modern hybrid IT infrastructures."

Ditto above. I wouldn't hold much credibility to Microsoft funded Gartner reports which are pretty ideologically based. My analysis of the marketplace suggests exactly the opposite as the transformative
change that FLOSS represents moves through the marketplace. My
prediction (and like any prediction will be tested by time) is that in a decade that the spectrum choices (software manufacturing, retail, post-payment/royalties) often called "proprietary" will represent a very small niche market where the user-base for a given type of software is too-small to support peer-production or the funding motivations (resource multiplication/etc) behind pre-payment funding schemes.

Again, I would not put either prediction into a document such as this as neither are fully scientific. It would be great if an unbiased group in Industry Canada or another part of government did a study on this, but the point of my comment is to note that currently you simply don't know and writing "it is expected" doesn't help the credibility of the report.

"As opposed to the concept of a copyright, a ‘copyleft’ describes the case where the owner forfeits intellectual property and private licensing. "

This is also not the case as there is no forfeits of intellectual property with copyleft. Most (but not all) of the exclusive rights are forfeited with so-called non-copyleft FLOSS licenses (such as the BSD license), but far fewer exclusive rights are forfeited with copyleft.

"FOSS is intrinsically exposed to the risk of appropriation by commercial vendors."

I have no idea what you are trying to say. With non-copyleft (what you kept calling "practical") you can have non-FLOSS derivatives. This is not considered "appropriation" by those who use those licenses, but included as one of the options they wanted to make possible when they licensed their work that way.

For copyleft licenses, these can be enforced in the courts in the same way that non-FLOSS licenses are enforced. There has been a team working on just this type of work in Germany.

...I will leave things here for now. As you can tell I would have feedback on nearly every page of the document. If you wish to get together and talk more, let me know. I also encourage you to join us in some of our informal meetings where other people can also offer informal feedback.

Thank you.