Open Source IS Free Software

It is mostly “open source” on the other side of the Atlantic, and may become “free software” depending to whom you’re talking to. It becomes FOSS or even FLOSS as you fly east, and pass through the old continent. In Turkey you come accross with people calling it “open source” in Turkish (açık kaynak), “free software” in Turkish (özgür yazılım) as in freedom, “open source” (yes, they use English), and even few call it “free software” in Turkish (bedava yazılım) as in free beer. Quite a confusion, huh? You bet…

All in all, it is simply “free software” in Turkish as in freedom. Always and in every context… We have called it “özgür yazılım” from day one, and we continue to do so. We explain why it is “free” as in freedom, and what this freedom means to the user, to the developer and to the society at large.

But in English, there still is this confusion. Being lately exposed to several pieces of work on “free software” (as in freedom) with certain strong business slant, and observing that the authors consistently use the term “open source” to describe it, still meaning “free software” (as in freedom)… and being convinced this will not harm the freedom of the software…

As of today, in my communication in English, I’ll only use the term “open source” to describe the “free software” (as in freedom). After all, “open source” (as in freedom) IS free software!

don’t forget: there’s a new Pardus in town…

Pardus is a unique distribution which has its own solutions which work very well. I think that the distribution deserves far greater renown than it has currently and if it keeps up such innovation and quality, it’ll be an important player in the category of extensive desktop distributions.

gözden geçirmenin tümü burada // the review is here

Draft for GPL v3 is out…

The 3rd discussion draft of the GNU General Public License version 3, which have been worked on for a while, have been published on Free Software Foundation web site. As I expressed before, I don’t think that a license text can solve all the problems related to the freedom of software. Neither I think is necessary to modify the GPL v2, which only recently have begun to be endorsed within the business circles. I think FSF (of which I am an associate member) is acting in a highly ideological manner, which may lead to some very disastrous forking of GPL licenses…

I should change the verbs in the above statements with past perfect. The latest draft is a much more relaxing and soothing (a la GPL v2) text. In stark contrast to the infamous EULA’s protection for the owner and for some cases for the developer of proprietary software, GPL puts the user at the front stage and enforces the freedom of the user. This approach, as I’ve just said, has been approved as a very meaningful distribution and sharing model, not only in the software and IT context, but for the business world as well. The former drafts of GPL v3 have tried to extend this mission to protect and oversee the freedom of society and third parties as well. The clauses against Tivo-ization and DRM have been redefining the borders of the freedom. The present draft has a much softer wording, emphasizing the freedom of the user, again.

The anti-patent approach and the clauses referring to the infamous Microsoft-Novell pact are also new. In fact these were the reason for the draft’s missed deadline. I think these are somewhat unnecessary references to the current headlines, still the text in the present draft is acceptable. These clauses cover the developer and consequently guards and immunizes the user against the possible patent scams.

Among others, Linus Torvalds is pretty pleased with the present draft of GPL v3. Way to go…

Pardus Ideas for Google Summer of Code 2007

This year we have decided to participate to Google Summer of Code and completed our application as of yesterday. Our in-house developers have come out with a number of very nice project ideas, which are good-to-have features in Pardus, either based on Pardus technologies or very closely related to the ease-of-use design goal we have. You may reach our GSoC ideas page here. Any further suggestions will be discussed at and proscpetive students are encouraged to join to the gsoc e-mail list at

il y a un nouveau Pardus en ville

Pardus est une distribution très intéressante. Elle a été bien conçu et vient avec un excellent choix logiciel. Peu, nombreux, les logiciels offerts sont d’une grande qualité et les versions utilisés sont très récentes. Le côté le moins pratique est l’absence ou presque du français, néanmoins, nous espérons que nos astuces pour franciser ce super OS vous aideront car c’est son seul défaut et il serait dommage d’y lever le nez que pour cela. En résumé; facilité, excellente détection matérielle, robuste, véloce, prompt, très stable, codecs/plugins/pilotes propriétaires, bref elle a tout pour plaire!

Voilà plus d’une semaine que nous la testons sur deux ordinateurs et nous ne pourrons nous en départir! Elle devient donc le système principal de notre 64 bits, même si elle est compilé en i686 (32 bits). C’est sans doute la distribution qui nous marquera le plus cette année!

repeat after me: a new … pardus… in … town

Still looking for that perfect desktop distro? Then don't overlook Pardus Linux 2007. It might not be the hot favourite with the mainstream IT publications — and it certainly doesn't have the budget to plaster Linux websites with ads — but this Turkish distro comes with enough innovation and polish to make it one of the best new Linux distributions of 2007.

Sevgili Béranger‘e bağlantı için teşekkürler // Thanks goes to our dear Béranger for the pointer

iDABC related: EUPL, OSOR, things…

Yesterday I attended the IDABC Open Source Software 2007 event in Free Software World Conference in Badajoz.

One of the first presentations of the session was on the EUPL, European Union Public License; which has been approved by the European Commission on January 7th, this year . The final text for the license is here.

Patrice-Emmanuel Schmitz from Unisys made the very-few-slides presentation, and basically told two things:

  • EUPL was necessary to cope with EU languages, EU regulations, and warranty conditions. It’s not put forward as an alternative to GPL, but came out of to relieve some uneasiness from the EU public organizations, and governments.
  • EUPL is “compatible” with GPL, means if you use GPL code in an EUPL program you may license it with GPL, or if you use EUPL code in a GPL program you may license it with GPL; by the way these “may”s are “have to”s from GPL perspective, thus “compatibility” means it may be engulfed in GPL. It is stressed that you have to make some changes to an EUPL code in order to license it under GPL, but this is just a trivial condition.

I have had several concerns regarding EUPL, since the day I’ve first heard about it. I think it is not necessary to invent a new free license, which, eventually might be enforced throughout EU, as an alternative to GPL. I think it is more clever to use a second document for EU public organization contracts, which will be some kind of amendment or addendum to GPL. I addressed two questions to Patrice-Emmanuel, here they are and summary of replies from him and Barbara Held of IDABC:

  • What are the differences between EUPL and GPL? Only minor things related to the three problem areas? Or something in addition? We are not comparing the licenses, the EUPL is a very simple document, it is not accepted to solve all the problems, we are not going to evangelize EUPL as an licensing alternative.
  • What is the position of EUPL regarding the hijacked freedoms that RMS talks about? DRM, patents, Tivo-ization, and such? Where does EUPL stands as compared to GPLv2 and GPLv3? The EUPL came out of certain necessities, so it is not involved with issues which are more open to discussion.

My position regarding EUPL is not changed, I think it is an unnecessary and somehow irrelevant EU digression. For some other views you may have a look at here or here. Just as a reminder, I have negative to very-negative thoughts about the GPLv3, as well. The rest of the session was focused on the infamous OSOR (Open Source Observatory and Repository). The main idea is to collect the free software prepared for public organizations in a central repository, so as other public organizations will be able to reuse them. The project has been tendered and the bid went to Unisys. Now they, and their project partners, are working on OSOR, which will be operational late this year.

OK, I, as always, some concerns regarding this OSOR thing. First, yesterday we’ve heard about several repository projects from all over Europe, some being highly successful and/or ambitious. Among them were three (yes, 3) repositories from three different juntas/xuntas from Spain. Apparently their audience is quite limited, their applicability and effectiveness is highly controversial; still they are insisting to have these local repositories. I don’t think (correct me if I’m wrong) these junta/xunta territories are large enough markets to have some scale economy for reuse of free software only within the territory. They have to get together, I understand their concerns regarding cultural differences and importance of local administrations, still it does not look quite right to me.

The second, and the more important, concern is about the profile of stakeholders. In previous OSOR meetings it was possible to see people from some vendor companies or agencies which have very strong interactions with commercial world. In Badajoz there only are academicians, public organizations, and EU staff. Without the guys sitting on the other side of the table, I don’t think this will work! I’ve asked questions along these lines to the project group in November meeting, and have not got convincing answers. Things have not got better in three months, and the future of OSOR does not seem too bright…

Later edit: The language I’ve used in the last paragraph have been found offensive by the IDABCE/EU officials. I’ve smoothed the terms a bit, though I insist that the OSOR project is not a wise invesment on EU side.

More later edit: The opinions expressed here are solely mine, and does not represent neither Pardus project, UEKAE, TUBITAK nor any other entity. If you are interested in the official Turkish position and policiy regarding OSOR and other issues please visit the related State Planning Organization web site.

stallman in badajoz: RMS as usual

Richard Stallman gave his keynote speech in the Free Software World Conference in Badajoz. The speech was in Spanish, and I was not in the main hall during, so I don’t know about his speech. But I went in for the question & answer part, to see whether there will be an interesting debate going on…

I think RMS has missed the point for all three questions asked, and the outcome was only confusing messages from the “creator” of the free software.

First question was about the internet services we use, especially Google, and the source code that runs beneath. At this front Stallman assumed the conformist position, saying that it is not wrong to have proprietary software based services per se, but it is wise not to use them. It is very clear that the internet services, or RIA and Web 2.0 as the hype goes, will be the next platform where our data lives, and giving away our freedom there will not be much different than using proprietary software on our PCs. Free software movement has to have some proactive initiative in this front as far as freedom is concerned, but Stallman himself admits that this is not the case.

Then there is this announcement/question about some Spanish free software code contest, which RMS somewhat admired; but quickly went on to say that the energy has to be focused on migration, and not on development, since the development is mature enough. Which is completely against what we are all working for, the freedom… Microsoft is saying the same thing in Turkey: “Our platform is robust and advanced enough, don’t try to develop alternatives, develop applications instead, on top of our platform of course…” You cannot tell from where the next “killer-app” will emerge, it is always a wise idea to promote innovation, everywhere and anywhere. Of course, migration is important, but it is not the whole story.

Then there was the so-much-expected question about GPLv3. Stallman mentioned Microsoft-Novell deal, Tivo-ization, patents and DRM as the points where GPLv2 could not been able to secure the freedom of user. Thus a new license was necessary and enters v3. I don’t buy the thesis that a single license should take care of all the relevant freedoms. It’s just a text, after all, and for some, a very important text. If you alienate part, and some very important part, of your audience and allies by singlehandedly changing this text, you are asking for trouble. Which is the case for GPLv3, in my humble opinion. Stallman’s chronic dislike of Linux feeds his stubbornness on GPLv3 case, I guess, but this does not help who are working to make free software a reliable alternative to proprietary world.

… and I went out. It was as expected, Stallman as usual, or a disappointment for the free software world.

PS: The picture is by Sandro Groganz, whom I met today. He has both better equipment and talent for photography 🙂 He blogged about Pardus and put a picture of mine, which -I guess- captures me very elegantly, in his blog.

PPS: Later in the evening, I’ve come accross with this beautiful piece by Tim O’Reilly, which hits the bull’s eye. It’s free as in search freely…

PPPS: … and the Steve Jobs quote about removing DRM from iTunes content, provided that the big four agrees to do so. Putting aside whether this is just rhetoric or he is really gonna do it, this is an example for getting freedoms back, without messing up with certain texts.

off to badajoz!

I’m leaving for Badajoz (here in GoogleEarth) in a couple of hours.

I’ll be doing few things there: First I’ll be on duty for the IDABC Open Source Software Event 2007, where representatives from several EU countries discuss the first results for the Open Source Repository (OSOR) project. Then, I’ll attend the third Free Software World Conference. The main keynote speakers will be Richard Stallman and Miguel de Icaza. I guess they receive lots of heat due to not-so-spectacular publicity and endorsement that GPLv3 drafts are getting and Novell’s recent oh-so-very-close relationship with you-know-who, respectively. I hope to hear and speak a lot about Microsoft’s ISO maneuver for Open Office XML standard as well. The best part, I’ll be able to meet and talk to lots of free software people from all around Europe (or world?). It’s very exciting and fruitful to learn about different perspectives and experiences. Of course, I’ll do some Pardus advertising 😉

Finally, as time permits, I’ll enjoy the nice town of Badajoz, enjoy local wines over tapas (pronounced [TAH-pas] ;-); and of course try to catch up with my to-do list, which becomes longer and longer 🙁

who’s this new pardus in town?

Thanks to the ever-growing success of Pardus 2007 around the globe, we were somewhat pushed into forming an English-speaking planet under I thought it might be a wise idea to toss in a semi-official history of Pardus project for general use. It, inevitably, includes some side info regarding the goals of the project and such. Here it goes…

First, some taxonomy:

  • TÜBİTAK is The Scientific and Technological Research Council of Turkey. Founded in 1963 and headquartered in Ankara, TÜBİTAK has both funding (à la NSF) and research (à la Max Planck Institue) duties. Housing more than one thousand researchers in 15 institues, TÜBİTAK is one of the leading organizations in Turkish R&D.
  • UEKAE, a.k.a. NRIEC, is The National Research Institute of Electronics and Cryptology. Established in early 70s, and functioned under different names in TÜBİTAK Gebze campus (50 kms to İstanbul), NRIEC became what it is now in late 90s. As the name implies, there is not much we can tell about it, we already did 🙂
  • Panthera Pardus Tulliana is a subspecies of leopard, ruled the forests and hills of Anatolia till 70s, but almost or have becmae extinct due to extensive hunting. Recent sightings have been reported, but not confirmed, so it is not possible to tell wheter it is extinct or not.

Pardus project started in Fall 2003 under the auspices of TÜBİTAK UEKAE. This was when I joined the project and UEKAE. The initial drive for Pardus was preparing a feasibility report and project plan for a possible “national operating system” project. The founders of the project were not Linux hackers, so there was no imminent distro in sight by then.

The first hackers joined the project in January 2004, as the project takes a twist, such that we were aiming at a Linux distro, and no more interested in the feasibility and planning thing…

The project has a triad of goals: Developing Linux distribution for widespread use in Turkey and elsewhere, to have some sound financial sustainability even though the mother organization is a not-for-profit, and finally to develop a free software ecosystem in Turkey and elsewhere to innovate in Linux and related technologies. Pardus, the distribution, is the main tool to achive these goals, and has, in turn, three fundamental requirements: Having full Turkish (i.e. mostly UTF-8) support, ease of use and installation, and a human-centric and task-based design approach (instead of a technology-centric and tool-based one).

The first product of the project was Pardus Live CD 1.0, which was available in February 2005, just 5 months after the kick-off of the systematic development. Live CD 1.0 has some of the current Pardus technologies in their prototype form (Comar, automatic Xorg configuration, Zemberek, etc.). Still that was just a demo product, and was missing one of the main components, namely the package manager PiSi. Development of these and refactoring and in some cases recoding of the others took almost a year and the first installable version Pardus 1.0 has been available in late December 2005. This was a much more refined product, but still had some bugs and missing features for big time, and work for the 1.1 version started just afterward. The uphaul was so impressive that it took another year, and we decided to drop the 1.1 version no and go with year label instead, so born Pardus 2007.

Pardus 2007 is the second stable version of the system and since the first month is over after the release 120.000+ copies has already been distributed by three Turkish copies of monthly IT magazines (Chip, PC Magazine, PCNet) besides 120.000+ downloads. Many NGOs including Chamber of Electrical Engineers distributed more than 10.000 copies to their members. We already have deals for February and March issues of some magazines, so there will be a steady channle for distribution. Pardus (and henceforth Linux and free software) covered in national press for approximately 500 times in the last year.

At the moment Pardus has 15 paid developers in UEKAE, besides numerous volunteering developers and hundreds of “dedicated” users who report bugs and test patches, new packages etc. We also have non-Turkish-speaking volunteers from countries such as India, Netherlands, USA, Spain, Poland etc.

We are planning to have major version releases annually for the existing desktop system. In addition, we are working on a server edition, the preview versions of which will be available in the first half of 2007, which will have a release cycle of 18 months. We are planning to support the older versions as customers requested, not being longer than 3 years following the release of the next major version.