Random October 2008 Free Software News

Posted 5 Oct 2008 at 11:23 UTC by lkcl Share This

I'm missing steven's little news posts, so I thought I'd make one up. Things happening recently: Bruce Perens on a landmark appeals decision which strengthens the legitimacy of Free Software Licenses; Venezuela orders 1 million Intel Classmate PCs; SGI relicenses OpenGL; The FSF launches a campaign to highlight the high-priority list, and KDE 4.1.2 codenamed "Codename" is released.

Also coming up: Pyworks will be held at the same time as PHPworks in Atlanta, Nov 12th-11th; FSCon, the Nordic Summit, will be on October 24th - 26th at the IT-University Gothenburg; UKUUG Developer Conference 2008 will be in Manchester, Nov 7th-9th.

That's the summary: if you'd be interested to hear of some random insights into the news above, read on...

Bruce Perens on the landmark appeal strengthening Free Software

Law doesn't get changed or strengthened unless someone challenges it, rather than settling out-of-court. A mistake by a court in the U.S. - not reading fully a ruling by a higher court - resulted some twenty years ago in Software Patents becoming valid. It will take someone else to stand up in court and force a review of that mistaken ruling - which ignored the higher court's ruling that software can only be included in a patent if it is part of a physical (hardware) device.

In the case we're seeing now, Perens describes how Free Software Licenses have not really been challenged, until now. The person trying to get lots of money patented ideas from a Free Software project, then tried to demand patent royalties, then tried to get the license undermined. amongst other things. What he tried to do was to describe Free Software Licenses as "optional" (this is what a company in 1998 tried to do with the Samba project). By saying "I do not agree to this piece of paper", they claim that it can be ignored and thus they can do whatever they like with the source code (forgetting that if the license is ruled as "irrelevant", then standard Copyright law applies, and they're hosed).

Fortunately, the court has taken a dim view of this kind of thing.

Venezuela orders 1 million Intel Classmate PCs

Many people in Free Software may lament the fact that Venezuela is ordering Classmates not OLPCs. The simple fact is that Negroponte has lost the plot. Confidence, Software and means of Communication is far more important than a lovely piece of hardware.

I used to think that it was really bad that Intel is behind the Classmate PC. I held the view that they were a big corporate bully, knocking down the OLPC effort with biased newsreports that weren't vetted or researched properly, even by the BBC. Then I learned that the CEO of Intel had met with Professor Mohammad Yunus, and instantly my view on the matter changed. utterly. If Professor Yunus can inspire Intel to do the same thing that Danone did, I am 100% behind Intel. It's also worth mentioning that I'm 100% behind the SUGAR software, as well, and anything else like it.

What's fantastic about Intel is that they can help fill one important piece of the puzzle, and people - governments - believe and trust them to be able to deliver. With guidance from Professor Yunus, Intel will not be performing a "profix maximisation" exercise on the buyers of the Classmate PCs. This is borne out by the evidence you can see before you: Intel is selling the design of the Classmate PCs to Portugal (for Venezuela to buy 1 million of them, whilst the Portugese government takes a further 500,000).

When the OLPC first came out, and governments announced they were buying orders in the millions, those orders were stalled when they realised that they had no training, and that this "support" issue was simply... non-existent. Of course, the fact that it's often the children who teach the teachers how to use the machines was entirely overlooked - but it's that confidence that needs to be inspired...

The other thing that's great about the Classmate PCs is that the machines can have WIMAX in them. Not the restricted version of WIMAX that's choked off by U.S. companies, so that it only talks to base-stations; the proper version of WIMAX that performs collaborative networking.

Medium-rance Collaborative networking is vital in areas with little to no communications infrastructure, as that wonderful article on an OLPC deployment in a remote village described. The children use it to talk to each other, outside of school hours. However, WIFI is restricted in range - even in areas where there is relatively little metal to get in the way; WIMAX has a range measured in kilometres, opening up the possibility to cover an entire town with only a few machines.

SGI relicenses OpenGL

It's fantastic to hear that SGI is finally releasing OpenGL under a free-software-compatible license. However... don't we have an opengl compatible library, already, because SGI didn't release this years ago?

Haven't we been here before, with Olivetti Labs releasing CORBA back in 1996 but The Open Group's participants only being able to agree to release DCE under the LGPL nearly eight years too late? (yes, the core runtime of DCE - 250,000 lines of code - was always available under a BSD compatible license, but that's not the same as an entire Directory Service and runtime being released).

Haven't we been here before, with Sun releasing Java, ten years too late, under a Free Software License?

Would someone please explain to me why this fantastic decision actually matters? Is SGI planning to make our lives easier, by taking the lead in the development of OpenGL in a cooperative and collaborative manner, or are they planning to do a TrollTech.com or a MySQL.com on us (releasing "free" software, 6 months behind the "proprietary" release, as a "sop").

FSF high-priority software campaign

Following on from Stephen Fry's wonderful assistance in promoting GNU's 25th birthday, the FSF has announced a High Priority list, and a campaign to fund the development of these projects.

The list includes Gnash, an Adobe replacement; a replacement for Google Earth; a replacement for MatLab; a replacement for BIOSes; a replacement for Skype; a replacement for DWG (CAD/CAM library) - with all these replacements, you'd think that there was no room for any real innovation by Free Software people at all - that they have to constantly chase the coat-tails and lap up the dregs of the proprietary software vendors.

But seriously, as the failed acceptance of KDE 4.0 shows, following the lead of another company, who will have taken years to develop and release something, just ... doesn't work. I was utterly disappointed by KDE 4.0 - that it looked so much like Windows XP. Windows XP is old by the time it gets out the door. Why, with the incredibly innovative and exciting infrastructure behind plasma, did the KDE team think that something so old, something that people so desperately want to escape from using, would be a good interface concept to copy???

Whilst I realise that certain features in a product are an inescapable fact, in order for the software to serve a useful purpose, following the lead of proprietary companies just because they've done it isn't necessarily the smartest thing to do - but then, neither is wandering off into surreality and spending man-decades of effort developing something that the average user doesn't want.

An Adobe replacement - most notably the video and microphone capability - is a key strategic requirement. without a shadow of doubt. There is no other cross-platform ubiquitous video and audio communications that comes remotely as close to seamless, quick and easy install and use as Adobe Flash. end of story. Everything else is just a nightmare. Flash brings a new dimension - real-time video and audio communication - to the Internet, in a web browser. any web browser. It's utterly vital that we have this capability, in Free Software form, in many different programming languages. Client and Server. So far, we have red5, which is server-side only. What we need is pyamf, rtmpy and tape to be given as much a priority as Gnash. And swfdec.

A Skype replacement - again, this is about communication. Skype showed us that the issue of easy-install and zero-configuration communications software can be solved. So why the hell is it taking so long for people to produce a free software alternative? Where is the peer-to-peer free software real-time communications network that busts firewalls and provides automatic multi-level proxying? Why, when libnice already exists, is it taking so long for people to integrate this simple library into all of the free software communications applications?

Do free software developers simply not care, or is it that it's actually far more complex an issue than one "spare time" developer can handle? Are we seeing, finally, that the reach of the Free Software "way" is limited - that "itch to scratch" in certain specialist areas is just way beyond even the above-average skilled free software developers' time, resources and ability?

KDE 4.1.2 release

I love KDE. I love the underlying design. I love the way it's not U.S-centric but is more European. Just like that wonderful comparison between Heaven and Hell, KDE has all the designers and Gnome has all the engineers. However, KDE 4 is, in my mind, a significant foot-shooting, Darwin-Awarding-self-head-chainsawing step backwards (if it's in fact possible to take a step backwards with a bullet-hole in your foot).

I like the adoption of the "applets" style thing from MacOSX - the transparent blacking-out of the screen and the raising of applets to the front. I question the usefulness of the wobbly-screen thing but it can definitely be entertaining. I can definitely see the value of transparent screens. I don't like the default "black" colour style. at all.

But... I'm just not loving KDE as much as I feel I should, and I'll stick to installing KDE 3.N for people, and I'll stick to superkaramba for more advanced - and pretty - desktop-widget development, not plasma.

Oh - and the integration with the brain-dead "Network Manager" - the one that takes utter control of all networking and makes it absolutely impossible to do anything other than what it says thou shalt do with your own network devices, doing this by taking away all your flat unix text files in /etc and controlling your devices through worse-than-proprietary binary file formats, is absolutely unforgivable.

(If you've ever tried to configure your KDE-configured laptop as a wireless access point on eth1, to provide fellow conferencees with (non-WEP'd) internet access via eth0, you will know what I'm talking about.)

End of News-topic Rants

Thank you for reading this far. If you find the news links more useful than my own comments, great. If you think I missed something out, that's more worthwhile to hear - tell me, I'll add it. If you believe that something should be done about my comments, please do actually address the issues raised, so that, not least, I don't have to listen to my own voice raving and lamenting the lack of usability of Free Software, for the average person.

The Japanese have a saying, which identifies that point where, if you keep on making the effort that you're making, now, then you will definitely overcome the obstacles that you're presently facing. It says, "don't give up! don't stop! you will get there, but if you slack off even the slightest bit, you won't. So keep going!". Which is really very insightful.

Free software magically reached that point, some time last year, in its aims to bring about change - to undo some of the incalculable damage that proprietary software has wrought, restoring the link between development, innovation and use. It's not over, yet, by a long shot.

Skype replacement, posted 5 Oct 2008 at 21:24 UTC by cdfrey » (Journeyer)

Thanks for the interesting summary. It was more like a news editorial than a piece of journalism, but I liked it, and I think we need more of it here.

Anyway, one of the itches that I would like to see scratched is encrypted VoIP. This is something that I hear very little about, and I wonder if I'm the only one that sees this as an important feature.

From a security standpoint, it seems that going from a POTS to VoIP is a step backward. I'm probably placing too much trust in the telephone companies when I make that comparison, but even so, it seems that we have an opportunity to increase the security of our voice communication dramatically, and we're all missing the boat.

I hope to spend more time learning VoIP in the coming months or year, as I gradually acquire the hardware I need. Hopefully by then I'll have more insight into the technical issues involved. Who knows? I might even scratch.

OpenGL, posted 6 Oct 2008 at 06:42 UTC by trs80 » (Apprentice)

The license SGI used to have for certain parts of OpenGL was mostly open source, but had some revocation provisions. These were minor enough that even Debian ignored them (although gnewsense didn't, in some holier-than-thou nose-cutting).

Isn't NASA's WorldWind a decent open source Google Earth clone? Oh fucknuts, they invented their own license. The high priority list has been around for nearly 3 years now btw - a quick look on archive.org shows the original list was OpenOffice 2.0, GPL Flash, and GNU Java.

Thanks, posted 6 Oct 2008 at 17:31 UTC by joolean » (Journeyer)

Thanks, Luke!

lsbappchk & friends, posted 17 Oct 2008 at 11:59 UTC by chalst » (Master)

Highly appreciated news summary, lkcl. There should maybe be an involuntary rota system for monthly news summaries, where the most recent poster gets to choose the next month's intrepid journalist.

The big FS news of October looks like it is edition 4 of the Linux Standards Base, together with its new release of the LSB ATK suite. The suite tests for possible issues with distributions coded to the standard, and also some other issues that might arise with major distributions, and some gushing tech press coverage suggests that this release has become fairly mature.

I've not found a whole lot of information out there about the suite yet: there is the Linux Foundation's Welcome to the LSB, which introduces the general LSB philosophy of application portability, a wiki page About Linux Application Checker which gives an overview of the architecture of the ATK tool, and there is A short PDF slideshow from IBM's linux division which outlines the LSB certification.

What I looked for and couldn't find:

  • A list of lsbappchk supported source languages;
  • A critical evaluation of the degree to which passing the automated tests actually warrants cross-platform compatibility.

Time will provide these, I guess.

lsbappchk source languages, posted 17 Oct 2008 at 12:11 UTC by chalst » (Master)

Found the first item: at the The Linux Application Checker Compositionwiki page it lists that lsbappchk supports ELF binaries and perl, python and sh scripts. The lsppkgchk tool, the other main engine of the LSB ATK tool, supports only RPMs.

great!, posted 17 Oct 2008 at 21:22 UTC by lkcl » (Master)

thanks for the extra bits, chalst. i do have rather rose-tinted glasses and a strange way of looking at the world so was bound to have missed something :) trs80, thank you for the clarification about opengl - i always wondered why i could get opengl in debian...

Thanks!, posted 30 Oct 2008 at 21:20 UTC by StevenRainwater » (Master)

In theory, mattl of the FSF will continue to post the official FSF news here each month. But the official news leaves out coverage of a lot of the things I used to include such as status of major GNU projects, summaries of FSF branches outside the US, and random other stuff I found interesting. So I enjoyed reading lkcl's summary and I like chalst's idea of a rotating user summarizing the news. I'm afraid this year has turned out to be too busy for me to do anything but bare bones maintenance on Advogato and mod_virgule, much less post news. :(

navel gazing, posted 1 Nov 2008 at 13:05 UTC by lkcl » (Master)

my favourite advogato sport ha ha :) steven, a "news suggestion" box would be something that would be great to have. people could post a title, optional URL, and with a short description. at the end of the month, all the information is collated into one automatically-posted article on the frontpage.

like a "project". in fact, exactly like a project. a collaborative editing "thing", which anyone can edit, and, at the end of the month, the content from it is posted as the "article header". next month, anyone can edit and add to the "header", in preparation for the next automatic publication.

i don't think there's any point in doing "voting" or any kind of thing like that - the lowest-level cert is sufficient.

New Advogato Features

New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.

Keep up with the latest Advogato features by reading the Advogato status blog.

If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!

Share this page