The Restaurant at the End of the Universe

Matt Asay has an excellent riposte to a singularly wrong-headed post entitled "Open source won't doom traditional enterprise software". As he rightly says, the real question is not the one the above piece thinks to deal with - "Is Enterprise Software Doomed?" - but

"What will be the primary bases for competition once everything is more (or less) open source?"

I believe the answers are also an explanation of why open source does doom traditional enterprise software, because the key differentiators will be things like innovation and serving the customer. Whatever lip-service traditional software companies pay to these ideas, the closed nature of their code, and the fact that customers are locked into their products means that they simply don't deliver either in the way that open source companies will do once they become the norm.

When Elephant Seals Collide

You can't beat a legal battle involving two overlapping pieces of legislation. The sight of lawyers having at each other, secure in the knowledge that the law is on their side, reminds me of nothing so much as two great elephant seals, thwacking each other vigorously, their proboscises all a-jiggle.

We could be in for another of these spectacles, according to this Techdirt article. It seems that the old End-User Licence Agreement (EULA) is being used to trump copyright fair use provisions, and that this might eventually go to the US Supreme Court to sort out (but don't hold your breath for EULAs getting spanked).

Of course, for those of us who use free software, EULAs are but dim memories from some strange, barbaric past, with no question of trumping anything.

Reasons Not to Use Closed Source: No. 471

I've written a couple of times about cases that demonstrate graphically why closed source software is a Bad Thing, but even they pale somewhat beside this story.

The robot that parks cars at the Garden Street Garage in Hoboken, New Jersey, trapped hundreds of its wards last week for several days. But it wasn't the technology car owners had to curse, it was the terms of a software license.

A dispute over the latter meant that the software simply stopped working. And since it was closed source, nothing could be done about it. The results were dramatic:

The Hoboken garage is one of a handful of fully automated parking structures that make more efficient use of space by eliminating ramps and driving lanes, lifting and sliding automobiles into slots and shuffling them as needed. If the robot shuts down, there is no practical way to manually remove parked vehicles.

I bet the garage owners wished they'd chosen open....

Nothing to Fear But Fear Itself

One of the tensions that emerges from time to time in this blog is that between openness and security. In the current climate of the so-called "war on terror", openness is typically characterised as dangerous, irresponsible even, because it gives succour to "them".

Terrorism is not to be trivialised, but it's a question of keeping things in perspective. Magnifying the threat unreasonably and acting disproportionately simply hands victory to those who wish to terrorise. This seems pretty obvious to me, but if you want a rigorously-argued version, you could hardly do better than this one, by John Mueller.

Here's a sample, on the issue of perspective:

[I]t would seem to be reasonable for those in charge of our safety to inform the public about how many airliners would have to crash before flying becomes as dangerous as driving the same distance in an automobile. It turns out that someone has made that calculation: University of Michigan transportation researchers Michael Sivak and Michael Flannagan, in an article last year in American Scientist, wrote that they determined there would have to be one set of September 11 crashes a month for the risks to balance out. More generally, they calculate that an American’s chance of being killed in one nonstop airline flight is about one in 13 million (even taking the September 11 crashes into account). To reach that same level of risk when driving on America’s safest roads — rural interstate highways — one would have to travel a mere 11.2 miles.

(Via Boing Boing.)

Microsoft's Gift to Firefox

Firefox has been incredibly lucky. It has taken Microsoft an extraordinary amount of time to face up to the challenge this free browser represents, during which Firefox has notched up a serious market share that won't be going away any time soon.

However, my great fear was that once Internet Explorer 7 came out, the appeal of Firefox to people who wanted a stable, standards-based browser would diminish considerably. After all, good enough is generally good enough, and surely, I thought, Microsoft will get this one right, and produce what's necessary?

If this report is anything to go by, it seems not.

Incredibly, Microsoft will not be supporting fully the Cascading Style Sheet 2 (CSS 2) standard. As the story explains:

The most critical point in Wilson's post, in my mind, is Microsoft's admission that it will fail the crucial Acid2 browser-compliance test , which the Web Standards Project (WaSP) designed to help browser vendors ensure that their products properly support Web standards. Microsoft apparently disagrees. "Acid2 ... is pointedly not a compliance check," Wilson noted, contradicting the description on the Acid2 Web site. "As a wish list, [Acid2] is really important and useful to my team, but it isn't even intended, in my understanding, as our priority list for IE 7.0." Meanwhile, other browser teams have made significant efforts to comply with Acid2.

If you look at the CSS 2 standard, you'll note that it became a recommendation over eight years ago. And yet Microsoft is still not close to implementing it fully, unlike other browsers. Even if you argue that CSS 2 is only of interest to advanced coders, or at best a standard for the future, it is nonetheless a key test of a browser development team's attitudes and priorities.

This is a tremendous opportunity for Firefox: provided it continues to support standards better than Microsoft - and this now looks likely - it will occupy the high ground with all that this implies in terms of continuing to attract users and designers. Thanks, Microsoft.

Capillary Growth

I see my old chums at OSS Watch have come out with a survey of open source use in higher and further education institutes in the UK, and it makes interesting reading.

The extent to which open source is creeping into higher education almost without anyone noticing is striking. From the summary:

Most institutions (69%) have deployed and will continue to deploy OSS on their servers. Generally, the software on servers is a mix of OSS and proprietary software (PS). The use of OSS is most common for database servers (used by 62% of institutions), web servers (59%) and operating systems (56%).

This is particularly true on the desktop. Although GNU/Linux is not much used there, free software apps are:

Microsoft Office and Internet Explorer are deployed by all institutions on most desktops. Other commonly deployed applications are Microsoft Outlook (82%) and Mozilla/Firefox (68%). The latter's use is now considerably higher than in 2003.

Not mentioned in this summary, is the share for OpenOffice.org (23%) and Thunderbird (22%) both of which are eminently respectable. It's also noteworthy that some 56% of further education establishments surveyed used Moodle.

Turning Back Genomic Time

Bioinformatics allows all kinds of information to be gleaned about the gradual evolution of genomes. For example, it is clear that many genes have arisen from the duplication of an earlier gene, followed by a subsequent divergent specialisation of each duplicate under the pressure of natural selection.

New Scientist describes an interesting experiment to turn back genomic time, and to re-create the original gene that gave rise to two descendants. Moreover, that new "old" gene was shown to work perfectly well, even in today's organisms.

What's impressive about this is not just the way such information can be teased out of the raw genomic data, but that it effectively allows scientists to wind evolution backwards. Note that this is possible because the dynamics of natural selection are reasonably well understood.

Without the idea of natural selection, there would be no explanation for the observed divergent gene pairs, and the experimental fact that their putative ancestor does, indeed function in their stead, as predicted - other than the trivial one of saying that it is so because it was made so. Occam's razor always was the best argument against Intelligent Design.

There's No FUD Like an Old FUD

As readers of these posts may know, I am something of a connoisseur of Microsoft's FUD. So I was interested to come across what looked like a new specimen for my collection:

"One of the beauties of the open-source model is that you get a lot of flexibility and componentization. The big downside is complexity," Ryan Gavin, Microsoft's director of platform strategy, said on the sidelines of the company's worldwide partner conference in Boston last month.

Alas, digging deeper showed this is hardly vintage FUD. Take, for example, the prime witness for the prosecution:

IBS Synergy had started developing products for the Linux platform back in 1998 but gave Linux the boot in early 2004, and now builds its software on the Windows platform. Lim said this was because the company's developers were spending more time hunting for Linux technical support on the Web, and had less time to focus on actual development work.

Right, so these are problems a company had two and half years ago: why is Microsoft raising them now? And is it not just possible that things have moved on somewhat in those 30 months?

So really this is the old "there are too many distributions, you can't get the support" FUD that was so unconvincing that I didn't even bother including it in my FUD timeline above. After all, businesses tend to use, well, Red Hat, SuSE and er, well, that's about it, really. (Via tuxmachines.org.)

Wales's World-Wide Wikia

I wrote about Wikia when it was launched a while back. Now we have WorldWiki, a fairly obvious application of wikis to travel guides - with plenty of advertising potential.

I mention it for two reasons. First, this will be a good test-case of the Wikia idea - if Wales can't get this one up and running, he may have problems with the whole scheme. Secondly, the home page currently has a rather fetching Canaletto-esque view of the Grand Canal, taken from the Rialto if I'm not much mistaken. (Via TechCrunch.)

Blogging the Bloggable

No one has a better bird's eye view of the blogosphere than Dave Sifry, which means that his quarterly report on the same is unmissable. One comment in particular is worth noting.

In the context of the 50 million blog mark being reached on 31 July, he writes:

Will I be posting about the 100 Millionth blog tracked in February of 2007? I can't imagine that things will continue at this blistering pace - it has got to slow down. After all, that would mean that there will be more bloggers around in 7 months than there are bloggers around in total today. I shake my head as I am writing this - the only thing still niggling at my brain is that I'd have been perfectly confident making the same statement 7 months ago when we had tracked our 25 Millionth blog, and I've just proven myself wrong.

For the sake of being wrong, I'll stick my neck out and say that I think he will be reporting 100 million blogs in February next year. The reason is simple - literally.

Blogs are so simple to write, that I think practically everyone who has a Web site will convert unless they have very strong reasons - for example commercial ones - to stick with the free-form Web page. Everyone else - and that's billions of us - just needs a suitable bucket for pouring our thoughts into. And the more basic the bucket, the easier it is to use, and the more that will use it. If this thinking is correct, another 50 million - or even 100 million - blogs is not so hard to achieve.

Resolving the Free Content Licence Madness

Although the most famous example of free content is Wikipedia, it is unusual in that it uses the GNU Free Documentation Licence, rather than one of the better-known Creative Commons licences. And that's a problem, because it makes it hard to mix and match content from different projects.

One man well aware of this - not least because he is the cause of the problem, albeit unwittingly - is Larry Lessig. Heise Online have a good report covering what he said on the topic at the Wikimania conference:

"We need a layer like the TCP/IP layer which facilitates interoperability of content, allows content to move between ´equivalent´ licenses," Mr. Lessig declared, "where what we mean by equivalent is licenses where people mean the same thing. So the GNU Free Documentation License and the Creative Commons Attribution ShareAlike license is saying the same thing: Use my content however you want, to copy, to modify, as long as you give me attribution, as long as the modification is distributed under an equivalent license." The legal differences between the licenses should be bridged, he observed. The various types of licenses could compete with one another, thereby protecting against the weaknesses of any particular license, he stated.

As the two worlds of Wikipedia and CC content continue to grow, addressing this is becoming a matter of some urgency.

Crazy Like a Vixie

As I mentioned elsewhere, I did many interviews when I was writing Rebel Code. Very many. But one key person whom I just could not convince to talk to me was Paul Vixie, Mr BIND.

So when I saw that The Inquirer has instituted a new series of interviews called Internet Gods with none other than Mr V, my heart sank. And then rose.

Why ID Cards Are Stupid, Part 294

Because they can be cracked. So, in exactly how many ways does this scheme have to be found wanting before it is finally taken behind the shed and put out of its misery, Mr Blair?

Who's Afraid of a Terabyte?

So Google Research is offering to send anyone a trillion words - for free. But what does the ready availability of this kind of data deluge mean for the world? Here's what I wrote in an essay called "Digital reality", which appeared in a strange little book entitled Glanglish, back in 1989:

A few hundred years from now the world currency will be the Tera. Short for Terabyte, it represents a million million bytes of digital information. Roughly speaking this corresponds to half a million printed books. The data contained in a Tera will be arbitrary but meaningful: it will be the equivalent of a random selection of a twentieth of the British Library's present holdings. Within such a deluge there will be countless useful facts as well as countless useless ones. The sheer volume will ensure that there are enough of the former in every Tera to provide near parity in value with all other Teras of random structured information.

Surprisingly, a Tera is a small unit. A human being processes about two Teras every hour; multiply that by a world population of tens or hundreds of billions and you have millions of billions of new Teras every year. Add in the billions of computers and their information, as well as the countless billions of Teras from the past, and the quantity becomes unimaginable.

Billions of computers because by this time they will be ubiquitous. The basic models will be as small, cheap, and easy to use as a pen or pencil. Like pencils, they will be thrown away after a couple of uses. But where writing implements can be said only in a metaphorical sense to offer half a million English words or the ability to perform operations in calculus, the pen-sized computers will possess these skills literally, as well as providing myriad other functions.

But nobody will bother using them, any more than people use slide rules or log tables today. The real, state-of-the-art computers will be invisible. They will be the chair you sit in, the wall you lean against, the ground you walk on. The chair will not have a computer as such: it will be one. Or rather, every aspect of it - its shape, its colour, its position - will be the output from one.

Such environmental computers will no longer model reality through simulations: instead, they will offer an infinitely detailed alternative version that merges seamlessly into the old, physical variety. Potentially, every aspect of our world will be formed by computers capable of creating every experience.

Most people will be hooked on this drug of artificial, digital reality. Unlike the already addictive arcade games and television serials of today - which are flint tools in comparison to this future technology - digital reality will not just be a temporary substitute for real life, it will replace it totally, until the latter has no independent meaning. Like all junkies, digital addicts will habituate and constantly demand fresh stimulation in the form of new, manufactured experiences. To provide them, the billions of environmental computers must feed ravenously and unceasingly off the only source of experience's raw material, the Teras of structured data held around the planet. Their competing demands for a limited resource will valorise it; information will become society's most sought-after commodity, its invisible gold, its weightless coinage. Those that control that information, the data lords, will rule the world.

Looks like I was out by several orders of magnitude for both size and time-scale. But then they do say that making predictions is hard, especially about the future.

It's Top of the Wikimania Pops

And talking of Wikimania, if you're short of something to listen to on the beach, there's a bunch of fine MP3s just waiting to be downloaded on this wiki page, which brings together most of the speeches from the recent conference. There's also a few transcripts for those who prefer to mull over the words. (Via Open Access News.)

Of Wikiversity and Diversity

The world of Wikipedia goes from strength to strength. At the Wikimania conference, Jimmy Wales made several interesting announcements, including that of launching Wikiversity, a collection of wikified online courses. This is great news, but I can't help wondering whether it might not help talking to all the other open courseware projects out there. Diversity is all very well, but a little coordination never hurt anyone.

PLoS ONE: Plus and Minus

The good news: the innovative PLoS ON is now live. This should be interesting.

The bad news: its designers have gone bonkers, adopting a Laura Ashley colour scheme and a horrible selection of typefaces. Could we try again, please?

Foul Trademarks

As I wrote recently, I'm not keen on the term "pirates" being bandied about indiscriminately. That applies to things like "bio-piracy" and even the neologism "lingo-piracy":

We’ve heard about biopiracy, the practice of multinational corporations claiming patent rights in the genetic resources of plants and crops in a developing country. Now we are seeing the rise of what might be called lingo-piracy. Brazil is fed up with foreign companies claiming trademarks in common Brazilian words for native fruit, foods and plants. The trademarks give the foreign companies exclusive marketing rights in the words, which in turn inhibits Brazilians from selling their own native foods and fruit in foreign markets.

But I do agree we need a term for the concept so that it can be named and shamed whenever it is encountered. The central issue here is essentially bad trademarks; since we have "fair trade", perhaps we can introduce the concept of "foul trademarks" to cover the situation.

Open Access to Avian Flu Data: One Down

Terima kasih: Indonesia has agreed to provide open access to its avian flu data. Now all we need are for the other couple of dozen affected countries to do the same. (Via Open Access News.)

A Mathematician Writes

One of the first things that children learn in maths is to do a quick check of their answers. Not quite sure if your calculation of 6.9574635 times 4.085647 is correct? Well, 7 times 4 is 28, so your answer really ought to be pretty close to that.

Common sense, right?

Wrong: Amazon says it's a brilliantly-novel idea that no one has ever had before in the history of the universe - and they have a patent to prove it. Words - and numbers - fail me. (Via Techdirt.)

Open Source Citizenship

There's a bit of a public ding-dong being conducted in the pages of some of the IT titles over what constitutes good open source citizenship.

Matt Asay kicked things off:

Aren't Yahoo! and Google missing the point or, rather, conveniently looking past it? Open source isn't about beneficent companies giving code to the impoverished underclass. It's about working on code collaboratively within a community.

To which Yahoo's Jeremy Zawodny replied:

So let's suppose that we decided to release "what we can" into the open source world. Of course, there'd be a lot of legal vetting first. Code licensing is a real mine field, but let's suppose that we cleared that hurdle. It would look as if Yahoo was doing exactly what businesses looking to get into open source are told NOT to do: throwing some half-baked code "over the wall" and slapping a license on it.

But I think that both are being somewhat short-sighted.

Neither Google nor Yahoo is obliged to share their code, since they don't distribute it. They are perfectly entitled to keep it snug within their respective corporate firewalls. In any case, it's unlikely to be widely useful to other projects, so the gift would be large token. But the point is they do both benefit from open source, and it is therefore in their interest to support it as much as possible.

The solution is not to chuck code "over the wall", but rather to help open source in other ways. Google, to its credit, is already doing this, with its Summer of Code projects, its tie-up with Firefox and mostly recently its open source code repository.

As I've written before, Google's track record is not perfect, but it's certainly better than Yahoo's, which might try a little harder at being a good open source citizen in this respect. All it requires is a few high-profile grants to needy free software projects. How about it, Yahoo?

Why It's Called the 'Domesday Book'

The Domesday Book was William the Conqueror's list of swag that he won after the Battle of Hastings. You might, therefore, expect it to be called "Bill's Big Book of Booty" or some such Anglo-Norman equivalent. The actual name chosen is curious, but the explanation is straightforward.

"Domesday" - or "Doomsday" as we would write it - refers to the Last Judgement, the End of Time; and that, apparently, is when the British public is going to gain free access to one of the key documents in its history. The book has just gone online, but it costs an eye-popping £3.50 to see a page. Clearly the National Archives need to modify their tagline "Download your history..." to "Download your history...and pay through your nose for the privilege."

Open Source Oxymoron

For me Web TV is a contradiction in terms. The Web stands for intelligent interactivity, TV for dumb passivity. However, given that TV via the Internet is coming, whether I like it or not, better that it be open rather than closed source. And it looks like that's precisely what will happen. (Via LXer.)

Linus Does Not Scale

One of the darkest moments in the history of free software occurred in September 1998. For perhaps the first - and one hopes the last - time the Linux kernel came perilously close to forking.

The problem was simple: Linus had become a victim of Linux's success. He was unable to cope with the volume of patches that were being sent to him. In the memorable words of Larry McVoy at the time, "Linus does not scale."

That scaling problem was solved by working on a better version control system (what became BitMover, later replaced by the memorably-named Git), as wll as handing off some of Linus's work to others. In the case of the kernel, this could be achieved by mutual agreement, but more generally it is hard to divide up a task among many contributors.

There are now several sites that have sprung up to address this problem. One of them is Amazon's Mechanical Turk, which I wrote about some time back, although I rather missed the key point, which is the use of distributed human intelligence to carry out those kind of tasks that computers presently struggle with. A more recent entrant is Mycroft, discussed in this C|net piece.

Also worth noting is the Crowdsourcing blog, which is a follow-up to the Wired article on the same (and doubtless a feeder to the inevitable book on the subject).

What's interesting about the crowdsourcing idea is that it represents a kind of open source without the openness: that is, participants are essentially computing drones with no way of knowing what the bigger picture is, unlike open source programmers, who can always look at the code. In a sense, then, crowdsourcing is a dilution of the idea at the heart of all the opens, but it's also a broadening in that it enfranchises more or less anybody with basic human processing abilities.

Update: And here's another crowdsourcing blog, called, aptly enough, Crowdsource.

Mashup Journalism

Open source journalism, also called citizen journalism, is nothing new, but I was intrigued to come across something called "SI journalism". This turns out to be re-using data gathered during the journalistic process to create mashups of one kind or another. The proposed name is "Structured Information Journalism", which has all the grace of a dodo in flight.

I'm not quite sure what it should be called - perhaps mashup journalism, which has a suitably tough, streetwise quality about it. Any other suggestions?