Why Governments Should Use Open Source Licensing

Here's a wonderful cautionary tale:

Systran created a specially adapted version of its Systran-Unix machine translation software for the [European] Commission, calling it EC-Systran Unix between 1997 and 2002.

On Open Enterprise blog.

What's New about the Novell deal?

The announcement that Attachmate would acquire Novell for $2.2 billion has naturally provoked a flurry of comments and analyses in the free software world. But it's important to pick apart the news to find out what is truly new – and to distinguish between what this changes, and what remains the same.

On The H Open.

Software: What Exactly Can be Copyrighted?

One of the many arguments against allowing patents for software (alongside the principle argument that software is made up of algorithms, which are essentially mathematics, which is pure knowledge and hence is not patentable) is the fact that software is anyway covered by copyright law. This means that others cannot simply copy your code, just as a novelist cannot simply copy large chunks of someone else's writing. But whether copyright law prevents others from copying the underlying ideas of that code by re-implementing them independently is another matter.

On The H Open.

Why Patents are Like Black Holes

When a big enough star dies, it generally implodes, and forms a voracious black hole capable of swallowing anything that comes too close. When a big enough company dies, all that remains is a bunch of patents that can have a similarly negative effect on companies whose business models are too close.

He's Mike Masnick's commentary on the area:

It looks like just about all that's left of former telco equipment giant Nortel is a whole bunch of patents, that are now expected to sell for somewhere in the range of $1.1 billion. The big question, of course, is who ends up with those patents, and what they do with them. Generally speaking, you don't see companies spend $1.1 billion on a bunch of patents, unless they're planning something big. It's entirely possible someone will buy them for defensive purposes, but equally likely that they're used to sue lots of other companies (or, perhaps by the likes of Intellectual Ventures, to scare people into paying up to avoid the possibility of being sued).

And of course, in the field of open source, the really worrying dying star is Novell, as Matt Asay points out:

As reported, as many as 20 organizations have registered bids for Novell, most (or all) of them private equity firms. While an Oracle or a Cisco might acquire Novell for its maintenance streams and product portfolio, it's unclear that private equity firms will have the same motivation. For at least some of these, there will be serious pressure to sell Novell's assets to the highest bidder, regardless of the consequences to Novell's existing customers or to the wider industry.

This wouldn't be so bad if it weren't for the fact that Novell has a treasure trove of patents, with at least 450 patents related to networking, office productivity applications, identity management, and more.

Worth noting is that among those patents are some relating to Unix...

These cases show yet again why patents just don't do what they are supposed to - encourage innovation - but act as very serious threats to other companies that *are* innovating. As more and more of these software stars die, so the number of patent black holes will increase, and with them the unworkability of the patent system. Time to reboot that particular universe...

Follow me @glynmoody on Twitter or identi.ca.

SCO What?

I'm struck by the almost unanimous chorus of indifference that has greeted the news that a court has reversed one part of an ealier ruling regarding who owns the Unix copyright:

For the foregoing reasons, we AFFIRM the district court’s judgment with regards to the royalties due Novell under the 2003 Sun-SCO Agreement, but REVERSE the district court’s entry of summary judgment on (1) the ownership of the UNIX and UnixWare copyrights; (2) SCO’s claim seeking specific performance; (3) the scope of Novell’s rights under Section 4.16 of the APA; (4) the application of the covenant of good faith and fair dealing to Novell’s rights under Section 4.16 of the APA. On these issues, we REMAND for trial.

As well as Groklaw, others that are distinctly unimpressed are Steven J. Vaughan-Nichols, Roy Schestowitz and Eric Bangemann.

And me to that list: SCO still has everything to prove, and very little money to prove it with. And even if it *did* prove anything, all it would gain would be the right to be ground into very fine particles of dust by IBM's legal department....

GNU/Linux Tops TOP500 Supercomputers Again

The fact that GNU/Linux totally dominates the top 500 supercomputing list is hardly news, but the fact that it has managed to *increase* its market share yet further is.

Here are the results for June 2009:

GNU/Linux 443 (88.6%)
Windows 5 (1.0%)
Unix 22 (4.4%)

and here are the figures for six months ago:

GNU/Linux 439 (87.8%)
Windows 5 (1.0%)
Unix 23 (4.6%)

Notice that plucky little Windows, from that small and hopelessly out-gunned company up in Seattle has bravely managed to increase its share by precisely 0%: an impressive result considering the millions of dollars it has spent trying to break into this market.

Snarky? Moi?

Update: More details about the top 20, and GNU/Linux's dominance here.

Follow me @glynmoody on Twitter or identi.ca.

SCO What? It's Patently over for Copyright

Remember SCO? It's a once-important company that developed a death-wish by suing IBM in 2003. As Wikipedia explains...

On Open Enterprise blog.

Finally, MULTICS Goes Open

The source code for the grandfather of Linux, MULTICS, has finally been released:

This is extraordinary news for all nerds, computer scientists and the Open Source community: the source code of the MULTICS operating system (Multiplexed Information and Computing Service), the father of UNIX and all modern OSes, has finally been opened.

Multics was an extremely influential early time-sharing operating system started in 1964 and introduced a large number of new concepts, including dynamic linking and a hierarchical file system. It was extremely powerful, and UNIX can in fact be considered to be a "simplified" successor to MULTICS (the name "Unix" is itself a hack on "Multics"). The last running Multics installation was shut down on October 31, 2000.

(Via OSNews.)

SCO Zombie May Take a Few More Steps

The SCO saga may stagger on awhile:

SCO has gotten a $16 million bid from York Capital for its Unix business. Coupled with the $10 million line of credit York is ready to provide, it's money enough to keep SCO's litigation against Novell and IBM going and to underwrite its budding mobile interests, the company says. SCO's lawyers will be filing papers related to the bid this afternoon with the bankruptcy court in Delaware.

...

The deal SCO has cut with York would reportedly leave SCO with ownership of the litigation and what is being called the "core IP," apparently any IP necessary to the lawsuits. York, a $12 billion Manhattan firm known to buy small software companies with declining revenues and turn them around, would get a 20% interest in any licenses SCO's litigation produces. Say, if circumstances conspired to allow SCO to restart its hated SCOSource Linux licensing scheme. York would also get the right to license SCO's Unix source code and get control of its contracts with licensees. York is expected to invest in the business, which includes a bunch of sterling accounts, and attempt to grow it.

Haven't people got better things to do with their lives?

SCO KO'd, Novell Renewed

Well, we all knew it would happen, and, finally, it has:

Judge Dale Kimball has issued a 102-page ruling [PDF] on the numerous summary judgment motions in SCO v. Novell. Here it is as text. Here is what matters most:

[T]he court concludes that Novell is the owner of the UNIX and UnixWare Copyrights.

That's Aaaaall, Folks! The court also ruled that "SCO is obligated to recognize Novell's waiver of SCO's claims against IBM and Sequent". That's the ball game. There are a couple of loose ends, but the big picture is, SCO lost. Oh, and it owes Novell a lot of money from the Microsoft and Sun licenses.

But there's another interesting aspect to this: SCO lost, and Novell won:

But we must say thank you to Novell and especially to its legal team for the incredible work they have done. I know it's not technically over and there will be more to slog through, but they won what matters most, and it's been a plum pleasin' pleasure watching you work. The entire FOSS community thanks you for your skill and all the hard work and thanks go to Novell for being willing to see this through.

As I've written elsewhere, we really can't let Novell fail, whatever silliness it gets up to with Microsoft: it is simply too important for these kinds of historical reasons.

Update: Here's some nice analysis of the implications.

Virtual GNU/Linux

Virtual Windows systems are familiar enough, but how about this: LINA, a virtual GNU/Linux environment?

With LINA, a single executable written and compiled for Linux can be run with native look and feel on Windows, Mac OS X, and UNIX operating systems.

Released under the GNU GPLv2, LINA sounds pretty interesting. Due out this month. (Via DesktopLinux.com.)

Pipe Dream: Re-wiring the Net

The online world is awash with XML feeds. The great thing about XML is that you can grab it and do stuff with it very easily, because it's basically a structured text file. For example, you can feed one XML stream into another, combine them, and keep on piping them around. A bit like Unix pipes.

Hey, now that's an idea:

Pipes is a hosted service that lets you remix feeds and create new data mashups in a visual programming environment. The name of the service pays tribute to Unix pipes, which let programmers do astonishingly clever things by making it easy to chain simple utilities together on the command line.

What's particularly cool about this new service is the graphical approach, which looks a lot like programming flowcharts. The currently-available pipes are rather limited at the moment - this is still very new - but it's not hard to imagine some very rich stuff coming out of this. Bravo Yahoo. (Via GigaOM.)

Behind the Great (Fire)Wall

Here's something I wish I knew more about:

Zhang Shiliang, who is in charge of the use of open source software in Beijing's Pinggu County government, spoke about the problems of Linux use in his organization. Chinese government is one of the biggest Linux buyers in the country. Since the Pinggu government began to push the use of open source software in 2004, 85% of their 4,680 computers have installed Linux or other open source software. But 53% of them still have to install Microsoft Windows as well, because their superior government uses Windows or other operating systems -- even other incompatible editions of Linux.

And some worrying figures at the end:

According to Lu Shouqun, China's sale of Linux was 175 million yuan ($21 million) in 2005, increasing 81% compared with the previous year. The sale of other open source software that year was 160 million yuan ($19 million). In the operating system market, the share of Linux increased from 4.2% to 9.8% between 2003 and 2005.

But Microsoft also won in that game. "In fact, China's increase of Linux users didn't impair the use of Windows," Lu says. According to his figures, Windows' share of the operating system market increased from 55.1% to 64.8% between 2003 and 2005. Linux mainly took users from Unix, whose share decreased from 30.9% to 19.8%.

OpenStreetMap Takes the Path of Stallman

There's a piece in the Guardian about OpenStreetMap's Isle of Wight effort. I was struck by this wonderful quotation:

The weekend drew around 40 people. By Monday, OpenStreetMap's founder Steve Coast estimated that more than 90% of the island's roads had been recorded. When asked if volunteers used OS [Ordnance Survey] maps, Coast says: "No. It's a taboo." Someone who did pull out an OS map was told to put it away immediately.

Which is precisely analogous to Richard Stallman's attitude when he started GNU, his project to create a benevolent Doppelgänger of the Unix operating system. This is what he told me for Rebel Code:

"I certainly never looked at the source code of Unix. Never. I once accidentally saw a file, and when I realised it was part of Unix source code, I stopped looking at it." The reason was simple: The source code "was a trade secret, and I didn't want to be accused of stealing that trade secret," he says. "I condemn trade secrecy, I think it's an immoral practice, but for the project to succeed, I had to work within the immoral laws that existed."

Publish and Be Damned!

The wilful misunderstanding of Google Books by traditional publishers is truly sad to see. They continue to propagate the idea that Google is somehow going to make the entire text of their titles available, whereas in fact it simply wants to index that text, and make snippets available in its search results.

As a an author I welcome this; nothing makes me happier than see that a search for the phrase "digital code" at Google Books brings up my own title as the top hit. The fact that anyone can dip into the book can only increase sales (assuming the book is worth reading, at least). Yes, it might be possible for a gang of conspirators to obtain scans of the entire book if they had enough members and enough time to waste doing so. But somehow, I think it would be easier to buy the book.

Of course, what is really going on here is a battle for control - as is always the case with open technologies. The old-style publishers are fighting a losing battle against new technologies (and open content) by being as obstructive as possible. Instead, they should be spending their energies working out new business models that let them harness the Internet and search engines to make their books richer and more available to readers.

They are bound to lose: the Internet will continue to add information until it is "good enough" for any given use. This may take time, and the mechanisms for doing so still need some work (just look at Wikipedia), but the amount of useful information is only going in one direction. Traditional publishers will cling on to the few titles that offer something beyond this, but the general public will have learned to turn increasingly to online information that is freely available. More importantly, they will come to expect that free information will be there as a matter of course, and will unlearn the habit of buying expensive stuff printed on dead trees.

It is this dynamic that is driving all of the "opens" - open source, open access, open genomics. The availability of free stuff that slowly but inexorably gets better means that the paid stuff will always be superseded at some point. It happened with the human genome data, when the material made available by the public consortium matched that of Celera's subscription service, which ultimately became irrelevant. It is happening with open source, as GNU/Linux is being swapped in at every level, replacing expensive Unix and Microsoft Windows systems. And it will happen with open content.