Tuesday Nov 03, 2009

☞ Copyright Fascists

Saturday Sep 26, 2009

☞ Topical Music

Wednesday Sep 02, 2009

Sun's Bloggers' License

As a follow-up to my posting on Monday about the new co-ownership license Sun has offered to all the bloggers on blogs.sun.com, I thought it would be good to post a link to the FAQ site and to the license itself (PDF). One interesting extra dimension is that the option to enter into the license is also open to former employees whose blogs are still on display (which is the policy for former employees, unlike some employers I could mention).

I think this is a great and wise step for Sun to have taken. I hope other companies with staff who blog will take the same step.

Monday Aug 31, 2009

Mine, all mine (& theirs too)

Sunset Over The Sierra

One of our design principles for blogs.sun.com over the years has been to allow everything and let good sense and existing rules prevent mishaps - at least until it's clear we need a new rule of some kind. It's been almost entirely effective, and the few cases where it hasn't have been quickly addressed by the Sun blogger community on an internal mailing list that almost every blogger subscribes to. Self-policing definitely beats supervision. Another design principle has been to encourage people to be themselves, and mix up the technical and the personal in their blogging. The resulting blogs have often been compelling and we've grown an unmatched bench of authentic, respected voices.

Of course, those principles leave unanswered questions. One of the questions Sun's present context has raised is, "who owns the blog content?" It's not obvious, since the postings include a mix of personal and Sun content, are posted on a Sun property but often in personal time, and so on. To make it crystal clear, Sun has created a licensing option for every employee that simply shares ownership of everything that's posted equally between Sun and the blogger. That allows Sun to continue to host blogs.sun.com in perpetuity and it allows employees to sort out their own uses for their content. I want to write a book for example, and other want to move their blog to their own domain.

The new license was rolled out today, to accompany the handy new function to export all blog content for use with (for example) WordPress. From now on, every Sun blogger has (if they choose to accept the new license) a clear, documented set of rights to their blogging content. Huge thanks to the team of people that made it happen, especially my favourite lawyer, Tiki Dare, who completely "gets" this stuff and without whose quiet and largely unsung help the open source community would be much the poorer.

Friday May 15, 2009

☞ Excellent satire plus upcoming gigs

Wednesday May 06, 2009

☞ Innovation vs Corporate Effects

Friday May 01, 2009

☞ ODF, Patents and Copyright in NZ & SF

Sunday Feb 22, 2009

Spot On

Best summary of the New Zealand Blackout that I've seen:

Thursday Feb 12, 2009

Old Code and Old Licenses

Brussels Cathedral towers and moon

I was in Brussels at the weekend to attend FOSDEM, one of the handful of "real" Free software developer conferences I attend each year (another is LCA which I went to in Hobart two weeks ago). I was once again honoured to be able to briefly speak to the assembled audience as I did two years ago. This time I announced a small license change to some obscure code, written before the GPL was finalised, to fix a problem for Linux.

Why would that interest anyone? Well, the code in question is the original implementation of Sun RPC, which went on to become RFC 1057 and today is a core part of every UNIX-family operating system. Including Debian GNU/Linux.

The way the code was originally licensed was exceptionally liberal. Written in 1984 or earlier (before the GPL existed), it allowed unfettered use of the Sun RPC implementation in any program for any purpose. The only significant restriction imposed, entirely reasonable to most eyes then, was to say that the module itself could not be sold as-is, only as part of a larger work.

What was liberal is now conservative

Times change. During the 80s, Richard Stallman's Free Software movement established the four freedoms. During the 90s (1994-7), the Debian Free Software Guidelines established a need for the code in their distribution of GNU/Linux to be fully Free software. By the beginning of this decade, Debian maintainers were making a serious effort to audit the millions of lines of code in Debian for true DFSG compliance. And in 2002, they found the old Sun RPC code in core Linux files glibc and portmap.

Reading the history for Debian bug 181493 tells the next part of the story. Inside Sun, the challenge of finding the code in question was Just Too Hard, and the things reached an uneasy impasse.

The issue came back to life last year when the bug was re-asserted as part of the run-up to the Lenny release. I was contacted both by folk at Debian - notably my friend Ean Schuessler - and at Fedora asking if there was anything I could do to accelerate licensing of the old code. Both projects had decided to take a hard line and removing the code from glibc and portmap was going to be a real headache, especially for the stability of glibc.


The task of relicensing old code can be pretty time consuming and involves people who are already much in demand.

  • First, the old code is often very old. The people who wrote it are no longer with the company, it is no longer part of a current product, we sometimes can't even be sure it ever came from Sun. We have to find the original code if we're to make any progress at all. Doing so might mean retrieving crates of paper from long-term storage and crawling through them.
  • Second, once the code is located, a legal expert has to look at the origins of the code and maybe once again crawl through retrieved paperwork to find the contracts behind the code. Their job is to determine if Sun actually has the right to change the license at all.
  • Third, someone has to believe it is their job with respect to the code in question to act on Sun's behalf to evaluate the change, authorise it and bind the company officially.
All this is time-consuming and expensive, and without a current code owner inside Sun it's touch-and-go whether anyone can find either the staff time or the budget to run a license change through to completion.

With help both from Ean and friends at Debian and from the Fedora team at Red Hat, we managed to identify some modern OpenSolaris code that matched the code in Linux. This was a key step. It meant we could trace ownership through the comprehensive records for OpenSolaris and start the process moving. By last week, we finally reached the point where we felt comfortable to relicense the Sun code involved.


On Saturday I was able to tell Europe's Free Software developers that the licenses on the RPC code are no longer a barrier to Free software - we'll change the license to Sun's copyrights in the RPC code to a standard 3-clause BSD license, allowing inheritance of that licensing by both Debian and Fedora. I'm delighted to have been able to fix this problem, which arose not because of failure but because of the success of software freedom over many years and becuase of Sun's early commitment to it.

Monday Jan 26, 2009

Intellectual Privilege

Hobart at Sunrise

Speaking at conferences like linux.conf.au (where I delivered a keynote last Friday) and OSCON is great fun. It's challenging to speak to an audience that's so diverse that it includes both the creator of the Linux kernel and students who just discovered it exists. It's humbling to know that the intelligence and achievement in the audience dwarfs anything I've ever done. And I admit that sometimes it's frustrating that there's a requirement for political correctness!

There are political correctness landmines littering this domain. For example, using the terms "open source" and "free software" is often taken as an indication of either one's cluefulness or of one's affiliation to a particular world-view. Personally, I consider the two expressions complementary - open source communities collaborate on a free software commons - but there's rarely a chance to explain that before I speak.

An especially frustrating one is the expression "intellectual property". The term is used widely in the business and legal communities, and it becomes second nature to speak of patents, copyright, trademarks and trade secrets collectively in this way. The problem with doing so is that the expression is factually wrong, and a legion of open source developers (you know, the ones working on free software) take the use of the phrase "intellectual property" as a genetic marker for "clueless PHB-type" at best and "evil oppressor of geeks" at worst.

Why is it wrong? Well, none of those things is really "property". In particular, copyright and patents are temporary privileges granted to creative people to encourage them to make their work openly available to society. The "social contract" behind them is "we'll grant you a temporary monopoly on your work so you can profit from it; in return you'll turn it over to the commons at the end of a reasonable period so our know-how and culture can grow."

Using the term "intellectual property" is definitely a problem. It encourages a mindset that treats these temporary privileges as an absolute right. This leads to two harmful behaviours:

  • First, people get addicted to them as "property". They build business models that forget the privilege is temporary. They then press for longer and longer terms for the privilege without agreeing in return to any benefit for the commons and society.
  • Second, they forget that one day they'll need to turn the material over to the commons. Software patents in particular contain little, if anything, that will be of value to the commons - no code, no algorithms, really just a list of ways to detect infringement.

Working on the legacy of this sociopathy is the subject for another time, but I believe we need to change the way we talk about the subject. Both Lakoff and Lewis agree; the words we use to describe things change the way we perceive them. The term we use probably needs to allow us to speak casually of "IP", so that we don't find every conversation to be a minefield of political correctness. Various suggestions have been made, but each of them seems to me to be so slanted to the opposite agenda that there's little chance of practitioners using them.

However, the term "intellectual privilege" seems to work. It's got the right initial letters, which is a huge win! But it also correctly describes the actual nature of the temporary rights we're considering. After having written most of this, I then searched to see if anyone else thought the same and found that someone is actually working on a book, endorsed by Lawrence Lessig, that has that as the title!

I doubt I will get the chance to explain all this before my next conference keynote. So if I don't, accept my apologies. When I said "IP" just now, I meant "intellectual privilege", and I think it's the right phrase for the job.

Sunday Mar 16, 2008

Software Freedom: More Than Copyright

New Forest Reflection

I was surprised last week to see a posting from Michael Tiemann, the President of the Open Source Initiative and a VP at Red Hat. Any posting with a subject of line of "Simon Phipps Was Right" is bound to catch my eye, but this one was especially unexpected because in the original discussion I had thought that Michael was largely right! Michael's posting graciously said:

Simon, I'm beginning to think that you were right and I was wrong. You said a standard's process is a crucial aspect of the standard's product, and a process that is not open cannot be trusted to produce a product that can be considered open. I maintained that I had seen and used many wonderful standards that took absolutely zero input from me, and therefore I didn't see my participation as a necessary prerequisite for assuring quality in the future. I believed that no matter what the process, a standard should be judged by the product. Watching the fallout settle from the [ISO ballot resolution meeting] in Geneva, I'm beginning to think that you were right and I was wrong.

I've been thinking about the posting for a week or so now and I've tried to respond thoughtfully. Here is the response I sent to Michael (still awaiting moderation):

Thank-you, Michael - it's not often I see a posting like this. Actually, when we spoke about this at OSCON I found I agreed with many of your arguments, even if that doesn't show in the on-list discussion. The problem is that standards are orthogonal to open source, and attempting to define them in a way that promotes and protects software freedom may be impossible. It's been said that when we create any system we create the game that plays it. The standards system is fully mature and as such is fully gamed, as the DIS29500 debacle you reference is proving.

Maybe a more productive approach going forward is to try to do for the other kinds of so-called intellectual property what the Open Source Definition (OSD) currently does for copyright licensing. Perhaps we need to rename OSD to "Open Source Copyright Definition" and then work on an "Open Source Patent Definition", so that we can avoid the kind of entrapment that software patents can threaten? And as you know I am convinced we need an "Open Source Trademark Definition" to help us as a community of communities to avoid the IceWeasel problem.

If these are interesting, I'd be pleased to spend time exploring them together. Let me know.

New Definitions

The current Open Source Definition doesn't actually define Open Source - rather, it defines a subset of the requirements that protect software freedom, in this case the copyright license. I actually think renaming it ("Open Source Copyright Definition"?) would be good since there's more to Open Source than just the copyright license. I then suggest we explore creating an "Open Source Patent Definition" and an "Open Source Trademark Definition".

What would be in these two new definitions? Both would need to define what promotes software freedom and how it can be protected. Both would need to be pragmatically principled.

  • An Open Source Patent Definition would do for patents what the OS(C)D does for copyrights. I've posted a lot on this subject before, notably in Protecting Developers from Patents and Ten Reasons The World Needs Patent Covenants, so I'd go mining there for my contributions to the discussion. But it may also be that in addition there needs to be a call for patent law reform, maybe as I outlined in Seven Patent Reforms While We Wait For Nirvana.
  • When it comes to an Open Source Trademark Definition, we would need to similarly define the signs that a developer or user needs to know whether software freedom is being promoted in a trademark policy. I've not written about this yet, but I do believe we need to collectively understand the bounds trademark law places on people who have responsibility for trademarks (read: all developers and open source communities as well as all vendors). We then need to construct a path that promotes software freedom without placing impossible demands on trademark owners to behave in ways that are contrary to their responsibilities.

This is not easy stuff. But I do believe that certain recent events between the open and proprietary software worlds mean that it's time for software freedom fighters to get together and work on these things. I'm ready to work on it. What do you say, Michael?


Thoughts and pointers on digital freedoms and technology markets. With a few photos too.


« April 2014