Day 2 of FISL was, if anything, even more eclectic than Day 1. Again, wandering around the booth area and browsing through the program, it was driven home to me that this conference is all about free software, i.e., ANY free software, not necessarily programming related at all. Hence, as mentioned below, I attended a session on using video to communicate about open source software. That's not the kind of topic I've ever found at a conference before, all the more so in the sense that the session focused on using open source software to create videos on open source software. The vision embedded on both ends of the spectrum.
And, in particular, the afternoon sessions I attended can be described in one single word: Awesome. OK, maybe two words: Awesome and Inspiring. But more on that later too. Below, a brief sequential outline of the various things that I observed/attended during today.
Qualipso and OW2
I attended two sessions by Jean-Pierre Laisne today. The main reason for my attendance of his sessions was that I had met Jean-Pierre while standing in line at customs in Sao Paolo. That kind of shared soul destroyingly Kafkaesque experience tends to create a bond that withstands the tests of time. And so, feeling a kindred spirit in him (and only because of having partaken in the same aforementioned custom line angst), I learned about Qualipso, in his first session, and the OW2 Consortium in the second.
Anyone who has used technologies such as JOnAS and, several years ago, CORBA, probably knows about ObjectWeb. OW2 is the result of the 2004 merging of ObjectWeb with a Chinese partner, OrientWare. OW2 focuses on middleware, but then in the broadest possible conception of the term 'middleware'. I.e., JOnAS, that really cool application server, is also considered to be middleware by OW2. In fact, they see everything between the application and the operating system as being middleware, which stretches the term quite far.
His earlier session focused on Qualipso. Here Jean-Pierre focused on the question: "Why might companies be reluctant to adopt free software?" In response to this question, he listed legal issues, business models, interoperability, know how, quality, maturity, industrialization, and support. However, the nub of the problem is that of trust. And, as Jean-Pierre argued, "Trust cannot be claimed. It has to be proved." Rather than being a provider of some kind of ISO quality stamp, Qualipso sees itself as a provider of trust, that is, of providing the tools needed for organizations to produce free software that the consumer can trust. "We want to bring to free software the same level of trust as any technology may have," explained Jean-Pierre.
Qualipso competency centers are being set up to deliver services to people who want it, such as to help them to choose appropriate licenses and to help them to use the right procedures within their organization. These competency centers are envisioned to be set up all over the world. Again, these are not intended to certify open source, but to bring the best tools and methods to the table so as to enable quality in open source. For example, one such center has been set up at the University of Sao Paolo, managed by its Computer Science Department and with research grants from the Brazilian government. In general, the list of members of Qualipso is quite impressive, including big names such as Siemens.
All in all, both presentations were interesting in introducing terms and approaches with which I hadn't been very familiar before.
Video for/by Open Source Software
Next, I attended the aforementioned presentation on creating videos to communicate about open source software. The speakers were Deidre Straughan and Aaron Newcomb, both from Sun Microsystems. Deidre works on http://blogs.sun.com/video, while Aaron does video blogging at http://www.thesourceshow.org. So both come with a lot of practical experience.
They're both very principled in their insistence to only use open source software for the production of their videos. At the start of their presentation, when they were outlining the advantages of video over audio podcasting, Deidre made the interesting point that people respond to faces and that it is easier to learn from someone who you can actually see. On top of that, the unpleasant turn that text-based communication can take (i.e., flamewars) is far less likely to occur if the parties can see each other. One tends to be more sympathetic to someone who you've met, even just once, on top of which one also gets more visual cues about who someone is via video.
Aaron's outline of the open source software he uses (all on Ubuntu) is as follows:
- Editing: Cinelerra
- Graphics: GIMP
- Sound: Audacity
- Video capture: Dvgrab
- Live motion screen capture: Xvidcap
The question that was also addressed was: "But doesn't the creation of video cost a lot of money?" Well, the answer is 'yes', if you're going for a professional TV studio, with a heavy camera on a tripod. On the other hand, you can take a far lighter approach, which is what Deidre and Aaron do. For video shoots you need a camera, microphones, and a tripod, all of which can be cheap. And for production you need a cable, disk space, and software.
Finally, topics such as transcoding (making a video smaller for hosting sites such as YouTube) was discussed. File size is determined by resolution, frames per second, and bit rate. After mentioning various hosting sites that are available, such as YouTube, blip.tv, Vimeo, and Archive.org, the limitations of video were discussed. One of these is the problem of the content of a video not getting indexed by Google. Hence, publishing a transcript together with the video makes the video itself easier to find, since the terms in the text are then indexed by Google.
Clearly, a very practical and throrough introduction to the creation and publishing of videos!
Tribunal de Contas do Tocantis on the NetBeans Platform
Then, literally seconds after having set up my laptop to do some demos at the booth, I met Paulo Canedo (blog here). Paulo works for the Tribunal de Contas do Tocantis, a city council in Brazil, where all the financial management related to the city is done on the NetBeans Platform. The application was started in 2007 and currently has three developers working on it.
In particular, Paulo's team likes the wizard framework of the NetBeans Platform, as well as the update center. Combined, these features enable the application to be responsible for all finances in the city, i.e., purchases and payments. The application processes data via XML and then transforms and stores in a Derby database. Furthermore, the application is used to analyze the data thus obtained and stored.
I had never met Paulo before, nor had I ever heard of this NetBeans Platform application. And I wonder how many more similar unknown NetBeans Platform applications are out there? Paulo will be sending me screenshots of this application, to add to the NetBeans Platform Showcase. (By the way, his blog contains some excellent material on NetBeans, such as this one on JSF/AJAX CRUD application creation via templates in NetBeans IDE.)
Let's do it with LWUIT. Or with JavaFX?
A session on LWUIT (Lightweight UI Toolkit) by Terrence Barr was next on my agenda. LWUIT was presented as an advanced toolkit that lets you create interesting and rich user interfaces on embedded devices. It has the same programming model across all supported platforms and you don't have to wait for it to show up on a device, since you can simply add a library to a device in order to use it there. LWUIT is portable, runs on MIDP, CDC, Java SE, TV platforms, and many more.
If you've programmed in Swing, you'll find LWUIT very familiar. In fact, the architects who developed LWUIT were inspired by Swing. I started wondering how JavaFX might relate to LWUIT, since LWUIT is so closely aligned with Swing, while Swing seems to be continually under fire from JavaFX advocates (e.g., dubious cries of "Swing is legacy" and so on). Terrence unsurprisingly got this question from the audience and argued that, unlike LWUIT, JavaFX is a complete platform (language, runtime, cross-platform media, and audio/video support), while being targeted at high end devices. LWUIT, on the other hand, is limited to component-based user interfaces, while not having tight media integration. It is targeted at today's devices and platforms and is very easy to get started with since it simply means adding a library to a device. However, he pointed to a future where LWUIT could be used on top of JavaFX, since LWUIT runs on Java.
Let's see what happens in that regard, since I can't help feeling that the two seem to be trying to do the same thing, yet in such different ways. Which is most likely to 'win' this battle? LWUIT is really popular already and is in use within a variety of applications: it's been adopted in Spring WTK 2.2 and Java ME SDK 3.0, while there are also a range of LWUIT applications in production (Terrence mentioned Comverse VVM, NUUX, Emoze email, Telmap, and CTIA conference application). Plus, LWUIT is one of the top 5 projects on java.net, in terms of e-mail traffic. It's been doing very well for itself and doesn't require the programmer to learn a new language, unlike JavaFX. Some very specific features are really cool in LWUIT, such as the built-in touch screen support: all LWUIT widgets support touch screen events, such as drag and drop support and gesture scrolling, as popularized by the iPhone.
Still, let's see what happens in the coming period in this space. Basically, I find it hard to believe that LWUIT and JavaFX are not direct competitors.
The Third Wave of Free Software
The next three sessions that I attended, by Simon Phipps, Peter Sunde, and Richard Stallman, respectively, were three of the best sessions I have EVER attended. And they all took place, one after the other, in the same room. So, for the 4 hours that these presentations covered, I didn't get up out of my seat even once. (My seat would have been taken in a split second, that's how busy that room was.)
Simon Phipps (blog here), Sun's open source guru, walked his audience through what he termed 'the three waves of free software'. The first started 25 years ago with IBM being investigated for monopoly abuse, resulting in software being driven out of hardware for the first time. Though the software market was created, the source code remained secret. Two visionaries in particular, Richard Stallman and Bill Joy, led the way to the second wave. Part of Stallman's contribution was the definition of the 4 freedoms ('use', 'study', 'modify', 'distribute'), with the second wave aiming to create a world where 'every geek would be able to use each of the 4 freedoms to solve their problems'. The Apache Software Foundation was the focal point of this second wave, i.e., 'work together to win together', 'synchronized self interest', 'collaborative community', and similar terms, which are reflected in the open source initiative.
The third wave in the story is represented by Ubuntu. Simon Phipps: "Here we can see what happens when a strong and ethical business gets involved in an open source product." In the third wave, which is the one in which we currently find ourselves, business pay staff to work on free software. Gartner has claimed that by 2012, 70% of software will have incorporated open source software. Not long after that, that percentage was adjusted by Gartner to 90%. Phipps' response is that Gartner has it wrong: 90% (or more, probably 100%) of software incorporates open source software already, though by 2012 90% of companies may feel comfortable admitting that that is true for them too!
As a result of this third wave, i.e., the integration of businesses into open source, things need to change. As an example, Phipps pointed to the update of GPL, which now is business friendly for the first time. Then his presentation turned to the questions: "Why does software succeed in the 21st century?" and "How does software enter the enterprise?" Rather than an expensive (in terms of time and money) procurement-driven process, one can now see an adoption-driven market instead. Businesses use open source not out of a shared philosophy with the open source movement, but out of a practical desire to be in control of the procurement process. Typically, nowadays, an iterative adoption process takes place (try something out, see if it works, fix bugs, try it out again, see if it works, then employ some people to maintain the process and the technologies), in which the business is in control, rather than the vendors driving the process. As a result, vendors now need to wait for business to happen, placing businesses in control of the process. This decrease in expense answers the questions of why software is succeeding in the 21st century and how software now enters the enterprise.
"The greatest enemy of freedom," Phipps told us, "is a happy slave." The need to embrace the freedoms that the three waves of free software have brought is important. The fundamental point is that businesses should not be aiming at saving cost, but at gaining freedom. Don't pursue cost savings, but pursue freedoms, which gets you the cost savings you were initially seeking.
The Pirate Bay
Up next was Peter Sunde (Wikipedia entry here), who was just plain wonderful. Read that Wikipedia entry for all the background on him. A cool statistic is that BitTorrent is estimated as being responsible for 80% of all Internet traffic, while The Pirate Bay is used to initiate 55% of all BitTorrent traffic... so that's 40% of all global Internet traffic handled by the three guys in Stockholm who are behind The Pirate Bay. It was quirky, mischievous details such as this that made Peter Sunde's presentation so much fun to attend.
The first part of his presentation (his second of the day) was about the technologies underlying The Pirate Bay. And those are these: lighttpd, MySQL, Sphinx search engine, MemCache, PHP, and OpenTracker. Then, his presentation having been completed in 10 minutes, he repeated the presentation he had done in the morning, which was great since I'd missed it and it was a wonderful experience.
Peter's talk was all about copyright and about how copyright is not good for society at large. Sharing and letting people improve things, on the other hand, is a good thing. Moreover, he argued that copying is a fact, and it is part of human nature, and that he's wanted to copy things since early childhood. (He also pointed out that when someone cut out a photo of him from a newspaper and then placed it on a website... wasn't that also a copyright infringement?)
Out of his mouth, at some point, came the words: "The Internet will not listen to reason." He was talking about the various cool and anarchic activities that people around The Pirate Bay had gotten up to over the years, mostly Internet based. The simple freakishness with which things can snowball thanks to the Internet, despite the absurdity of the activity, prompted the remark about reason (or the lack thereof) and the Internet. For example, by switching the letter "e" in "The Pirate Bay", they set up something called "The Pirat eBay", which eBay frowned upon but could do little to prevent. Adventures relating to Sealand, Post-It notes, court cases, the King Kong defence, the Tor network node for Iran, political hacktavist groups, Epic Win LOL, Manifesta, and the Pirate Party in the European Parliament were related with much relish. It was quirky, it was inspiring, and it was fun.
Fundamentally, underpinning Peter Sunde's story was the message: "Don't fear corporations when they try and squash the fun you're having while exploring boundaries and pushing the envelope." Or maybe not, but that's one thing I got out of it. And I reckon Leonardo di Caprio should play Peter Sunde if a movie about The Pirate Bay ever gets made. Which it should.
Renegotiating Copyright Law
It is hard to figure out what exactly to make of Richard Stallman's (Wikipedia entry here) two hour presentation. What he did was actually incredibly awesome. He picked up on the 4 software freedoms that he had defined way back when, as explained by Simon Phipps earlier in the day, and... attempted to apply them beyond software! Yes, that's the question that he'd been posed numerous times: "Do the 4 software freedoms ALSO apply to books, music, paintings, and other art forms?"
So... his presentation was not about software at all, but about copyright. And, no, he didn't argue that copyright is intrinsically evil. He did, however, argue that in this day and age we should renegotiate the meaning of copyright and look at where things are working and where they are not. And then take steps to address problems arising from copyright law when applied to certain situations where they repress rather than encourage innovation.
For the first hour, I felt like I was back in law school, listening to an exposition of copyright law through the ages. Seriously, that's what it was, starting from an explanation of how people used to copy books by hand, how Italian princes were able to give monopoly on the printing of a book as a reward for services rendered, how the printing press made copying more efficient, and many other similar details that were necessary to build up to the central argument: copyright is now a restriction to every citizen, instead of on publishers. Copyright is now no longer easy to enforce, requiring authorities to enter everyone's computer. Yet, at the same time, the public wants to copy and share information amongst themselves. At this stage of technological development, only draconian laws can actually enforce copyright, to the extent that copyright is no longer beneficial, since the freedoms required to be surrendered are so extreme.
He then turned to the question of what a democratic government would do about copyright law. It would, Stallman argued, reduce copyright power, both in the dimension of length as well as in the dimension of breadth. In terms of length, since the publication cycle is increasingly shortened (within 2 years a book is remaindered, within 3 it is out of print), it would make sense to reconsider the many years that copyright applies. 10 years from the date of publication, is what Stallman suggests. He then referred to a writer whose book had gone out of print, but who was then unable to distribute it privately because of the copyright that the publisher continued to enforce on the book. And that can't be right since it is preventing the writer from making the work public, which is all that the writer wanted to do in the first place.
In terms of breadth, Stallman argued that not all works should be treated in the same way. He pinpointed three categories (1) practical functional works, (2) works that say what someone thinks or thought, (3) art and entertainment. In the first case, which deals with references, recipes, educational material, these should simply be free to be modified, without restriction, and they should be freely redistributed too. All 4 freedoms should apply to this category. In the second case, memoires, autobiographies, and scientific papers, modification would imply misrepresentation, so there is no need for that freedom to be permitted. A minimum freedom would apply in this case: these should be allowed to be shared non-commercially, such as peer to peer on the Internet.
The third category, that of art and entertainment, brought Stallman to an interesting set of arguments. Firstly, he posited that by changing an existing art form one isn't necessarily damaging its artistic integrity or in far fewer cases than might be assumed. He mentioned Shakespeare. How many references (and complete remixing of plays written by others) does Shakespeare have to playwrights before him? Nowadays, Shakespeare might have been sued for copyright infringement and his plays would never have been written or performed. Hence, modification can in fact be a contribution. However, these contributions are not, in Stallman's opinion, urgent and therefore a 10 year wait seems reasonable to him. Therefore, only the freedom to redistribute non-commercial exact copies should be permitted within those 10 years.
Of course, this brings up the question: "Does this not mean that one is taking money from musicians?" Stallman argues that this is not the case, since commercial CDs do not support musicians and that only long established super stars get paid a percentage based on sales. In other words, since musicians get nothing when you buy their CD, they're also losing nothing when you copy it or share it with others. At the same time, he suggests that musicians SHOULD be supported, by one or both of tax money and voluntary payments. The latter is already being done in a number of cases, where you pay a dollar (or some other amount, including a completely voluntary amount) to the website of the artist producing the work.
I really enjoyed the reasoned approach Richard Stallman took to the question of copyright and its increasing impingement on freedoms. Doesn't it make sense to take a step back and renegotiate copyright law in this day and age? Bear in mind, again, that the argument here is not that copyright is evil (or even bad) but that it is too broadly applied and with too little finesse, without any realisation that the world has changed and that the consequences of copyright law, if followed through, entail the promulgation of such draconian laws that individual freedom should perhaps be considered a greater good than the right to prevent others copying things that you have created. (Stallman also spent some time ranting against DRM, against which he is actively campaigning, about which you can find out more at http://www.defectivebydesign.org.) At the very least, these ideas are very thought provoking and the measured approach taken by Richard Stallman in his presentation today was a good thing!
And that's the end of Day 2 of FISL.
It was clearly really great. And I didn't event mention the football match I attended with others, thanks to the great Sun Campus ambassadors (Renato Porto, Daniel Campos, and Dionatan), with Internacional losing to LDU Quito, unfortunately. There was one guy called Filho da Puta, the people in the crowd kept shouting at him, it seemed. He didn't appear to be playing so well, especially near the end of the game.
See you on Day 3 of FISL!