Wikipedia “upgraded” to World Heritage?

On last January we celebrated the 10th Wikipedia birthday. It’s probably one of the most celeb collaborative projects and the best sample of what a community collaborative work can create. And it all happened in just ten years.

Shortly after we attended to a great presentation with Miguel Vidal, and Felipe Ortega (you can review this talk in the MSWL channel, along with a lot of another interesting Libre Software videos)

In the last weeks I have been reading in international press about a shocking initiative started by Wikimedia Germany, as a part of the birthday celebrations: Wikipedia applying to be the first digital site which joins the UNESCO World Heritage listings.

Their main fact: UNESCO protects the World Heritage Sites around the world. But what about the brand new digital world? Shouldn’t UNESCO protect the whole human knowledge exchange (unique in the history of the civilization) and the global open access to it?

Being realistic, it’s hard that the UNESCO consider the Wikimedia request because applicants have to fullfit very strict requirements. For example, only countries can promote a Heritage listing request. Of course, no country owns Wikipedia. It belongs to all the anonymous users that have contributed to make it grow; it doesn’t even belong to Wikimedia Foundation, which only have rights over the trademark. Any country could offer itself as the petition sponsor? Even then, will be Wikipedia able to get UNESCO appreciation? And (most important) people appreciation?

You can read the contents of the petition and watch the Jimmy Wales’ video presentations in order to get the full info.

Will you sign the petition? As Jimmy Wales states, Wikipedia is not technology, it’s culture. I believe in free knowledge and I think it is very important keeping it alive. So, I have already signed it!



I like very much mySQL. I have enjoyed using it intensively during the mswl classes and exercises. This way I have improved a lot my mySQL abilities the last year since Santiago Dueñas, from GSyC/LibreSoft, offered us a good introducton talk and I would be very happy if can choice mySQL when I need to set up a DB engine for some upcoming projects

I like it because:

  • it’s fast, even with huge amounts of data
  • it’s very robust and reliable
  • it’s easy to deploy and use
  • it’s multiplatform
  • it has connection API’s in almost all known languages
  • it’s very secure

And, of course, it’s Open Source (GNU GPL License). But, what is the mySQL status right now after the Sun adcquisition by Oracle? Is there an Open Source future for my favourite DB?

A little bit of history: on April, 2099 Oracle Corporation, the US big enterprise propietary software and DB systems, bought Sun Microsystems; as Sun Microsystems had bought MYSQL AB on January, 2008 they were also purchasing the whole mySQL platform. This way, a propietary solutions only corporation became owner of one of the most important Open Source projects.

There was a lot of buzz those days and lot of people (and, of course, mySQL main creators) alarmed, as it seemed that mySQL could be discontinued or closed. But Oracle, in order to get European Commision approval of the Sun acquisition, accepted to keep mySQL alive and Open Source (with the existing dual commerce and GPL license versions) “…until the fifth anniversary of the closing of the transaction“

How is the situation right now? How is Oracle dealing with mySQL? We got some answers in the last O’Reilly MySQL Conference & Expo on May, 2011 and it seems that mySQL is still strong and alive. Oracle announced in the Conference that they have made big advances in the upcoming mySQL 6.0 line that they started in December, 2010 with mySQL 5.5, introducing the mySQL 5.6 Development Milestones Release. This new version includes nice improvements regarding performance and scalability.

So, it apparently seems that Oracle is doing a great work with mySQL, improving and expanding the platform. The mySQL Community have received with joy this new release and features. Even the formerly CEO of MYSQL AB, Marten Mickos, has stated that mySQL “in better shape than ever under Oracle”.

But, what about the future? Will Oracle giving strong support after 2015, when its agreement with EU Commission is ended?


Several months ago we had a very interesting talk from Roberto Calvo, member of the GSyC/LibreSoft and the development group of LibreGeoSocial, a FLOSS mobile platform which integrates social features and augmented reality. He made us a great introduction to Android, the Open Source and Linux Kernel based software stack for mobile devices. As it’s well known, Google is the flag supporting company (along with the Open Handset Alliance) of this project since they bought Android Inc, the original developer company.

Since Google acquisition and the release of the first Android device (the HTC G1 smartphone, just in US and just on one carrier), Android market share has been rising every month, in a hard battle with Apple’s iPhones. When we attended Roberto’s talk (in Q4 2010), Android market shared has raised from 2,8% (in Q2 2009) to 33%, becoming the most selling mobile platform. I was interested in knowing the current share rate and I got some updated figures at the Google I/O event in May, 2011 and the last Nielsen US in April, 2011 survey:

  • 100 million activated Android devices
  • 400,000 new Android devices activated every day
  • 200,000 free and paid applications available in Android Market
  • 4.5 billion applications installed from Android Market

These figures show Android as the most used mobile device platform, ten points ahead Apple’s iOS. As Roberto told us, Android excedded iOS in 2010 for the first time and it seems that this trend is going on. It has been a huge growth in barely three years.

I think it’s not just a Google or corporate success, it also belongs to the wide developing community and user ecosystem that has been built around the Android project in those three years. But as Roberto early noticed and I have seen in recent readings, developers are facing a big problem that could harm the good figures evolution and it comes from inside the project nature itself: the platform fragmentation

There are hundreds of Android devices, firmwares, OEM distributions and so on (and its combinations). It can be a nightmare for any developer to get sure that his software is going to work as intended. And if developers are not comfortable with a platform or final users are not happy with glitched apps or non supported device features, both could move to another platform which makes their lives easier…

Google (as the flag developers) knows it very well. And in the last Google I/O event in May, 2011 they showed a very ambitious upcoming solution: Ice Cream Sandwich. It was announced as “One OS that runs everywhere.”. The starting point is to melt Android Honeycomb (from tablets) and Android Gingerbread (from smartphones) into a single multipurpose system, running on current generation hardware and it would be released in Q4 2011. If you already think it sounds very promising, keep reading because it could become even better.

When they say “runs everywhere” they mean that it could run in the future in ANY kind of device, including smartphones, tablets, notebooks, embedded systems, home devices, cars, televisions… even custom built devices; check the full keynote coverage for some feature description and demos. And, of course, it would be fully Open Source.

Google didn’t offered many details about Ice Cream Sandwich, as it is a work in progress and it won’t be availiable until the end of the year. But if it is released as they promised it won’t be just a solution for the Android fragmentation (joining current smartphones and tablets OS), it would be one of the most remarked releases from Google’s history and a milestone in the upcoming device (of any kind) automation. We’ll have to wait some months in order to see if it’s really as promising as they noticed. Developers hope so.

Me too!


Earlier this year I attended to a talk about the project ‘Illumos’ (which I knew absolutely nothing), now that it’s been several months since its creation, I had the opportunity to see the video of that talk again.

Solaris was an UNIX-base operating system owned by Sun Microsystems. Solaris was originally released under a propietary license but in 2005 Sun decided to change its business strategy. They created the OpenSolaris project (built upon Solaris 10), released the code under CDDL license (a weak copyleft open source license) and created a community for the development. To manage the project the created a board (like the board in the Apache Software Foundation) and the contributors should transfer the ownership of the code to Sun MicroSystems (similar to the Contributor License Agreement of the ASF). In this way Sun would share the development costs of its operating system but continue releasing proprietary licensed versions (dual licensing business model).

In 2009 Oracle acquired Sun Microsystems, however the purchase was not effective until 2010 because of the monopoly claims against Oracle (as one of the projects of Sun was another DDBB: MySQL). In August 2010, Oracle closed the OpenSolaris repository and announced they would discontinue the development.

The developer community with the sponsorship of some companies whose business model was based on OpenSolaris (like Nexenta) created Illumos. The first idea was not to make a fork but rather a side project to OpenSolaris to allow contributors not depend on the decisions of Oracle. However, after the decisions taken by Oracle regarding OpenSolaris, Illumos had no other choice that to become a fork of OpenSolaris.

What is the status of the project after these months?
Here I leave you an interview with Garrett D’Amore (Illumos’ leader) (you have to jump to minute 3:20 for the part in English) explaining what has been the progress these past months and what are their future plans.


Earlier this year we had a talk about GIT (a distributed version control system), the truth is that I had used a few times for my personal projects but at work I was (and still) using CVS. As this year I had the opportunity to use and learn more about how GIT works, I proposed in my work to migrate our repository and modernize us a bit.

Obviously we will need to access to the whole history and tags from GIT. I’ve been researching a little about the migration process so I made ​​a little ‘manual’ I hope we can implement in the coming weeks. Of course there is a lot of documentation and manuals on how to migrate from CVS to GIT, this is just a summary of the steps we will have follow in our case.

I obtained much of the information from “CVS to git Transition Guide” but this guide included to freeze the CVS archive and we will not need to do so.

We will use git command “git cvsimport” which will allow us to fetch and update the GIT repository from CVS repository. We want to create a central master repository (as we have in CVS) with the master branch.

mkdir project
cd project
git cvsimport -v -k -i -d /path/to/CVSROOT -C project project

Now we will create a new repository without a working copy (project.git) cloning the repository that we have created:

git clone --bare project
rm -rf project

On the other hand, the committers have to create a clone to start working with him:

git clone :project.git

And they will have to set some things:

git config --global "Your Name"
git config --global ""

To make it more comfortable to do pulls is convenient to add what is called ‘remote’:

git remote add origin :project.git
git push origin master

Someday I will tell you how was the experience because we have some reluctance on the team … hopefully we will convince them!

Mono project

A few weeks ago I read a retweet of Miguel de Icaza (@ migueldeicaza) in which a math student posted a chart with data from the Mono project mailing list. He analyzed the whole archive of the standard, developer, bugs and patches mailing lists.

The plot was interesting because it showed a great correlation between the posts to the bug list and the posts to the patch list which according to the author meant a great activity by the mono developers and a great commitment to the project. On the other hand the graph showed a decrease in activity in developer and standard list. The cause of the decrease in the volume of these lists is because (according to the author) to other types of communication like Twitter, but few days later I read other news related to the project that made ​​me doubt about this conclusion.

Attachmate has completely acquired Novell, which was funding some open source projects including the Mono project but it is not clear which will be Attachmate’s strategy regarding Mono and for now it has already announced the dismissal of an undetermined number of developers.

Shortly after Miguel de Icaza announced Xamarin, his new company focused on Mono-based products. They had already thought on spining Mono off from Novell because they believed it was the best for the project, now the decision to carry out this plan has been precipitated because he confirmed that in early May Attachmate laid off all the development teams.

I decided to analyze the repositories of the project to see how these movements have influenced the development of the project and also to compare the results with the graph showing the results from the mailing lists before all these news. (I used CVSAnalY and GNU/R )

There are two interesting things we see in this graph:
1. There is a strong correlation between the number of commits in the repository and the number of posts to the mailing lists (devel and and standard) so I think the decrease on the activity in mailing list in recent months is due to a decrease in the in the volume of development and not because there are other communication channels as suggested by the author.
2. You can observe a sharp decline in number of commits in May over the previous months so we could confirm that (obviously) the latest changes in Novell have adversely affected the development of software in the project.

MeeGo, Qt and Nokia

As told by my colleague Esther on her blog, we recently made a work for the subject “Project Management” about Dawn Foster (Intel’s Community Manager for MeeGo), this has made us to be aware of the latest news on Meego, that have not been few…

Meego is a project created from the merger of two platforms Nokia’s Maemo (which was based on Debian Linux distribution and originally used GTK as its GUI and application framework) and Intel’s Moblin. In January 2008, Nokia acquired Trolltech, the developer of the Qt application framework, and started using Qt library as the default in Maemo.

The first big news regarding Meego was the partnership between Nokia and Microsoft to adopt Windows Phone 7 as Nokia’s primary smartphone platform, which made us wonder about the future of MeeGo and Qt. Shortly after we knew that Nokia was selling off its Qt software licensing and professional business.

What were the reasons for Nokia to make these decisions?

Nokia was losing significant market share against Android and iPhone and needed an operating system for its smartphones to keep up on the market. Symbian was not up to Android or iOS and Meego was taking too long to be ready, this has caused investors get impatient and Nokia has chosen to go for Windows 7. Is it a good choice for Nokia? We do not know by now …

How these choices affect a MeeGo and Qt?

The good news is that Intel has come out in defense of MeeGo and said that it has a future even after Nokia’s decision to use Microsoft’s software in smartphones, MeeGo will be used in tablets and mobile phones this year. In other work we made for “Project Management” we knew that GENIVI (an automotive alliance) has chosen MeeGo for its In-Vehicle Infotainment (IVI) system. So it seems that Meego has enough support to go ahead …

Qt has a dual-license model, providing open LGPL and commercial license alternatives. Digia (the company that has accquired Qt professional business) will now be responsible for issuing all Qt Commercial software licenses and for providing dedicated services and support to licensees. We do not know how this may affect the future of Qt, but as published by Sebastian Nystrom (head of MeeGo in Nokia): “Nokia and Digia have started this collaboration together, and both will be ensuring their work benefits all of the Qt community, not just LGPL or commercial licensees.” Let’s hope so …