If you don't have time to read through Peter Suber's full explanation of open access, you could always try his haiku version (this isn't new, but I've only just come across it). A sample:I love print, paper.
But I love searching, linking,
using, sharing more.
...
They don't pay authors,
editors or referees.
Then they want the rights.
...
Sure, change copyright
and peer review. But OA
doesn't have to wait.
The latest Netcraft survey shows that Apache has pulled back some of the ground it lost to Microsoft's Web server last month. There have been some pretty massive swings recently, as the oscillations in the graph show: these are largely due to switches in the hosting sector, which can often involve millions of Internet names at a stroke. For example, Go Daddy moved over 1.6 million hostnames from Apache to Microsoft's IIS platform in June.
These new gains for Apache are important, because it suggests that Microsoft's relentless campaign to "convince" hosting companies to switch to its products (and who wouldn't love to be a fly on the wall for those conversations?) may finally have run out of steam. It will be interesting to see what happens next month.
The ever-interesting Steven Vaughan-Nichols, who goes back a long way in the free software world, has a fascinating article about a comparison of two application stacks, one open source, the other from Microsoft. The results were surprising:
The tests showed that such vanilla LAMP (Linux, Apache, MySQL, PHP/Python/PERL) stacks as SLES (SUSE Enterprise Linux Server) 9, Zope, ZODB, and PHP and a pure LAMP based on SLES, produced "C" results. They weren't bad, but they weren't anywhere near as good as an out of the box .NET stack based on Windows Server 2003, IIS (Internet Information Server), SQL Server 2005, ASP (Active Server Pages), and SharePoint Portal Server 2003.
The results mirror those of the Mindcraft tests back in the late 1990s, when GNU/Linux found itself whupped by Microsoft. But the consequence was a range of improvements that soon took free software past Windows. However disappointing the current outcome for the stack tests may be, I'm sure that the same will happen here.
Remember: every bug report makes open source stronger, and the same goes for adverse benchmarks.
Last month I commended Hakon Wium Lie's call for open source typefaces to replace the de facto standard based on Microsoft's fonts. And here's an interesting article about a project called DejaVu that might just do that. The piece has some interesting background information on both DejaVu and its predecessor, Vera.
When I wrote about Microsoft's announcement that it would be sponsoring a project to create an ODF plugin for its Office product, I said the story was big. But I was wrong: it's actually really big, because of a deeply ironic twist to the story, detailed on Groklaw:
It seems that when Microsoft was looking to build its new ODF plugin, it took a short cut. It seems to have grabbed some code from the OpenDocument Fellowship's program that converts ODF to HTML, written by J. David Eisenberg. His code is released under a dual license, the LGPL and the Apache 2.0 license. Microsoft has put it into its ODF plugin, which is licensed under the BSD license.
Is that allowed? It's nice Microsoft endorses the value of the ODF Fellowship code, since they are forever telling us their own code is better. But we're trying to parse out which license Microsoft thinks it is complying with. Not the LGPL, I trust. My question, and I'm no Apache guru, is what about Apache sections 4.1, 4.2, 4.3, and maybe 4.4, plus the required form of notice in the Appendix? It's certainly possible I'm missing something. But it seems it may be Microsoft that neglected to notice some requirements.
Newspapers tend to keep their constituencies firmly in mind when they are writing. So you might expect the Los Angeles Times to be beating the drum for Hollywood fairly unthinkingly. And then up pops this editorial on various US proposals to give entertainment industries even more of a stranglehold over content, which concludes:As they weigh the entertainment industry's pleas, lawmakers shouldn't assume all consumers are bootleggers and every digital device is a hand grenade aimed at Hollywood.
Very interesting: if Hollywood's local newspaper is daring to write this, maybe a few other people in the vicinity are starting to think this way too. (Via TechDirt.)
One of the fascinating things that I learned when I was writing Digital Code of Life is that many diseases - such as obesity, heart disease, diabetes, certain kinds of cancers and neurodegenerative disorders - are not commonly found in the great apes. As I put it then:
In a sense, the human genome has evolved certain advantageous characteristics so quickly that it has not been debugged properly. The major diseases afflicting humans are the outstanding faulty modules in genomic software that Nature was unable to fix in the time since humans evolved as a species.
Another extraordinary fact is that dogs are even more susceptible to these same diseases than humans are, and for the same reason: the domestic breeds have arisen so recently, and from limited populations through inbreeding. But if dogs are like us, only more so, then they also hold out the hope that by investigating the root causes of their afflictions we might be able to understand our own better.
I see that further steps in this direction are now being taken:Melbourne researchers are examining the DNA of dogs in a research project aiming at determining the genetic causes of common pet diseases – and to provide a model for combating diseases such as diabetes and multiple sclerosis in humans.
I've written about what promises to become the patent troll extraordinaire, but here's another, er, specimen. C|net introduces us to the head of Acacia Technologies Group, but don't be fooled by the name. Its patent-based business is anything but fragrant, despite what the interviewee might have you think:
It's the patent system that enabled people like Thomas Edison who actually developed the new technologies, which these companies then want to use to make money without paying for. The invention process is critical to the growth of the US economy and it's the smaller companies that usually come up with the new innovations and disruptive technologies that then the larger companies want to adopt. There's no one forcing them to add these features to their products. Obviously, they're doing it because they can make more money using the new features that were patented by someone else.
For a thoroughgoing refutation of this and other widespread misconceptions about patents and copyright, do read the brilliant Against Intellectual Monopoly by Michele Boldrin and David K. Levine, freely available from here. (Via TechDirt.)
The smiley is a defining icon of the online generation. But as this article points out, it is also likely to go down as yet another enclosure of the virtual commons, a victim of insensate corporate greed, as companies battle it out for the "right" to claim this symbol as their "own".
Wikipedia is deservedly famous, but there is a tendency to conflate Wikipedia with the english version of it. One of Wikipedia's many great achievements - alongside its huge size and the innovative partipation of large numbers of people - is that it is energising communities all around the world to create local versions in languages other than English. There is a list of the main languages at the foot of the main English Wikipedia page.
As Wikipedia explains:Language editions operate independently of one another. Editions are not bound to the content of other language editions, nor are articles on the same subject required to be translations of each other. Automated translation of articles is explicitly disallowed, though multilingual editors of sufficient fluency are encouraged to manually translate articles. The various language editions are held to global policies such as "neutral point of view", though they may diverge on subtler points of policy and practice. Articles and images are shared between Wikipedia editions, the former through "InterWiki" links and pages to request translations, and the latter through the Wikimedia Commons repository. Translated articles represent only a small portion of articles in most editions.
Given this global diversity, and the lack of appreciation of efforts outside the Anglophone world, it's good to see that this three-way interview with leading Wikipedians includes voices from Germany and Japan as well as the obvious English one.
It looks like at least one person at Microsoft gets it:One of the things that I’d like to see us do as a company is release a free, Open-Source, stripped-down version of Windows. There are so many benefits, IMO. We could cut out much of the “integration and innovation” and ship a bare-bones, essentials-only operating system with source that would allow the Open Source community to take a look at our code and really build on it. As SaaS (Software as a Service) and Web 2.0 apps take center stage, there is less and less motivation for customers to plunk down their dollars for a completely proprietary OS, and I see Linux gaining steam in that environment unless we are able to do something significant.
Now, the interesting question is whether this is an officially-sanctioned bit of kite-flying or not. I don't think it is; but I do think we will see an open source Windows one day.... (Via Digg.)
Regular/long-suffering readers will know that I am an implacable foe of the UK Government's scheme to make everyone in the country carry ID cards. As well as being a huge waste of money (all £19 billion of it), they will inevitably make us less secure (just ask Bruce Schneier).
So I was delighted to come across this fantastic scoop by The Sunday Times, which suggests that the scheme is much closer to collapse than even I might have hoped. This is based on some killer emails that were leaked to the newspaper by senior civil servants involved in the doomed project.
A sample of their candid views:This has all the inauspicious signs of a project continuing to be driven by an arbitrary end date rather than reality.
...
I conclude that we are setting ourselves up to fail.
...
Just because ministers say do something does not mean we ignore reality - which is what seems to have happened on ID Card
And don't miss John Lettice's usual lucid analysis of what all this really means.
Use rights over cultural and scientific information are of fundamental political importance to citizens everywhere. These rights will be deeply affected by the kinds of intellectual property rights we allow to develop. This article argues for positive intellectual commons as a means of increasing freedom and diversity in information societies. Selforganized, positive intellectual commons will become more prevalent as citizens conclude that governments, will not deliver the institutions of knowledge that citizens want.
Which is of course what I have been saying for a while in this blog. But this paper by Peter Drahos puts it rather well; moreover, his background makes him rather better-qualified than I am to give some serious academic justifications for the ideas we share. Do read it. (Via Open Access News.)
Over on LWN.net I've an article grandly entitled "The birth of the open source enterprise stack", which has generated a fair amount of comment on the site. At the end, I write:a subsequent feature will explore the surprising richness of the upper layers of the emerging open source enterprise stack, in areas such as systems management, customer relationship management, business intelligence, enterprise content management, enterprise resource planning and communications.
One of the companies I shall be discussing in the context of enterprise content management is Alfresco, so I was intrigued to come across an extensive think-piece by that company's marketing director, Ian Howells.
It, too, has a rather grandiose title: "10 Rules of Open Source Marketing". It draws heavily on Geoffrey Moore's ideas, but contains some interesting insights of its own. The one that I particularly liked was the following:Rule 9: Your Software Infrastructure is Key
Dell transformed the PC industry not by selling cheap PCs but transforming the whole value chain and supply chain for PC production. From an operational perspective Open Source isn't about cheap software but about transforming the whole value chain for software across development, testing, translation, product management, marketing, sales and support.
The number of people downloading your software, asking questions, accessing your Web site, accessing demonstrations, trialing the product, discussing in forums, updating the wiki ... is massive compared to a traditional software start-up company. The extended infrastructure has to be able to support contributions, bug reports, and fixes from other individuals/companies, take feedback from forums and surveys, and be able to support hundreds of thousands people downloading your software. In amongst this, you have to be able to identify those who want to buy support, patches, and updates for a mission-critical environment and those who want to use the open source as part of the community. Open Source companies have to be masters of the whole Open Source software value chain to support the massive growth potential.
I really think this idea is the key to why open source will ultimately prevail: it represents a thorough-going re-invention of the entire process of creating, distributing and supporting code. Responses by traditional software companies are necessarily partial - unless they convert to open source themselves - and so by definition insufficient.
We tend to think of Google as an engine that finds matches for concepts, since that's how we frame our searches. But in fact, it's simply matching patterns - just try typing a few random characters into the Google box and searching for them: you'll be amazed what can turn up.
As this article indicates, those patterns could just as easily be binary sequences that go to make up executable files - which are simply a pattern with a different kind of "meaning". This seems to indicate that Google is not just indexing the manifest content of the Web, but the entire - and much larger - binary universe that is accessible in some way online.
I have been rather remiss in not pointing out that the World eBook Fair started last Tuesday. In celebration of the 35th anniversary of the founding of Project Gutenberg:The World eBook Fair welcomes you to absolutely free access to a variety of eBook unparalleled by any other source. 1/3 million eBooks await you for personal use, all free of charge for the month from July 4 - August 4, 2006, and then 1/2 million eBooks in 2007, 3/4 million in 2008, and ONE million in 2009.
You can either just bung in a search term on the home page given above, or - probably better - go to the full listing of the constituent collections.
I have to say these are pretty impressive. As well as practically every Western classic you could think (already well-covered by Project Gutenberg) there's some interestingly specialist stuff here: for example, Asian classics (don't miss "Response to a Question on the Five Degenerations of the Eon of Strife" - in Tibetan, of course), seriously deep ancient middle eastern texts (Egyptian, Sumerian etc.), tens of thousands of multilingual editions, 8,000 English poems and sheet music.
It's true that these are not all completely open content: many exist in new "editions" which are copyrighted. They also tend to be PDF files, and some scans from books are not very accurate. But it would be churlish to dwell on these deficiencies (none of which applies to the original Project Gutenberg, which is completely open, in the public domain and highly accurate): get downloading and enjoy.
A few months back I wrote about open source's big blunder: its neglect of the education sector. So I was naturally curious when I came across a column that began:I asked for successes at schools using Open Source Software, and I received a wide variety of them.
Alas, the few examples given show a market is that is still, shall we say, learning its ABC. Overall result: could try harder.
I am constantly delighted by the wit and wisdom of TechDirt. The latest example: a nice little meditation on the virtues of "open source" languages like English, where anyone can make up their own words, and that do not have standards bodies à l'Académie française telling people what is and isn't allowed.
It's true that French isn't exactly closed source (I believe you're allowed to write words down across the Channel), but it's a nice conceit.
If you've ever wondered where all these new and not-so-new open source companies are based, but can't be bothered looking them all up online, here's a nice mashup that shows the physical location of many of them. All we need now is a similar map for all the coders.... (Via Matthew Aslett.)
Yesterday I passed on a story about a closed source company unilaterally upping its support prices, and simply locking people out of their files if they refused to pay. Now, here's another good reason not to use proprietary systems.
The UK's shiny new IT system for the National Health Service (NHS) is fast becoming the biggest disaster in the history of computing. The latest area to suffer is that of childhood vaccinations:Child vaccination rates may be falling to risky levels after a new IT system was installed, a health watchdog says.
Ten out of London's 31 primary care trusts have installed new software to manage the vaccine programme as part of a £6.8bn overhaul of NHS computers
...
Richard Bacon, a Tory MP and member of the Commons' Public Accounts Committee, said: "The national vaccination programme has been one of the NHS's greatest successes."
But he added the IT upgrade appeared to be "destroying it at a touch of a button".
And why is this all happening?A spokesman for NHS Connecting for Health said the new system was implemented at short notice because the previous supplier "withdrew support for its ageing system from the market".
Had this "ageing system" been open source, the NHS could simply have called in another third-party contractor and given them the code. Since it was closed source, it was doomed when the supplier abandoned it, leaving the health system up to its neck in the proverbial.
Nor is this a matter of simple inconvenience: children are likely to die, if herd immunity is gradually lost as a result of these IT failings.
According to Wikinews:Yell, the world's biggest yellow pages publisher, today threatened to shut down Yellowikis, the wiki-based yellow pages directory.
Yell accused Yellowikis co-founders Paul Youlten and Rosa Blaus (his 15 year-old daughter) of "misrepresentation", "passing off" and suggested that using the name Yellowikis could "constitute an 'instrument of fraud'."
Yell is demanding that Paul and Rosa close down the website, transfer the domain names to Yell and agree to pay damages to Yell for loss of profits. Yell made $2.4bn in 2005, whereas Yellowikis had a loss of $500. The $500 was used to print T-shirts promoting Yellowikis at the Wikimania conference in Frankfurt.
Since Yell is apparently a UK company, this makes me ashamed to be British.
Let's look at the situation. You have a multi-billion pound company tied to a dead-tree model - just think of the resources it is wasting - bullying an open, volunteer project that is completely online (and innovative, to boot), through legal threats based on totally outrageous accusations.
Well, guess what?
I am now going to put all my Yell directories in for recycling, in an attempt to undo some of the environmental damage they have caused. And if Yell send me any more (as they are bound to do), I will try to refuse them; if I can't, I will promptly recycle those, too.
Henceforth, I will conduct all of my searches through Yellowikis, with the odd bit of Google thrown in where necessary. When I ring up companies I will make it clear that I never use dead-tree directories, and that they really should go online, maybe with something like Yellowikis, which is completely free. (Via TechDirt.)
Biofuels are much in the news lately. Generally, they are presented as a clever way of getting round oil-dependence, with the added bonus of being environmentally sound: after all, what could be greener than plants?
But step back to look at the bigger picture, and you see that biofuels are no solution; worse, they would actually be disastrous to the environmental commons:The United States annually consumes more fossil and nuclear energy than all the energy produced in a year by the country's plant life, including forests and that used for food and fiber, according to figures from the U.S. Department of Energy and David Pimentel, a Cornell University researcher.
...
Corn and soybean production as practiced in the Midwest is ecologically unsustainable. Its effects include massive topsoil erosion, pollution of surface and groundwater with pesticides, and fertilizer runoff that travels down the Mississippi River to deplete oxygen and life from a New Jersey-size portion of the Gulf of Mexico.
After Larry "I'd like to have the complete stack" Ellison, it seems that Sun is joining the Club of Stack Love. Not such a daft idea, actually.
How about this one?
Some doctors who use Dr. Notes' electronic medical records software say they have been denied access to the program and their patients' medical records because they refused to pay increased technical support fees.
(Via LXer and GPL Medicine.)
Well, we've got the open source car, so I suppose it's only fair that there should be open source trains. Sounds like a brilliant solution to the near-total breakdown of rail transport infrastructure in Cambodia. And I'm sure there's a great PhD in there somewhere, tracing the evolution of design as ideas are passed around. (Via Boing Boing.)
Posts
All Comments
To the extent possible under law,
glyn moody
has waived all copyright and related or neighbouring rights to
this work.
This work is published from:
United Kingdom.
