Monday, April 18, 2011

#Dutch is no longer a top 10 #Wikipedia in reach


The traffic numbers for the top 10 Wikipedias are undergoing a realignment. With German on number 4, Dutch on number 11 room is given to Spanish and Chinese. Given the sheer number of speakers, this re-adjustment makes sense.

All this does not mean that German or Dutch are not doing well. Both are, with twenty percent, growing twice as fast as the English Wikipedia.

With a relatively "big" language moving out of the top 10, the change to the percentage of traffic for the top 10 has gone down by not more then a percent.

I would like to see the top 10 languages getting less then 85% of our traffic. This could be achieved when languages like Arabic, Indonesian or Hindi would get traffic in direct proportion to the population who speak these languages. Then again, Chinese alone has a potential for more spectacular growth. It may be that 85% is just not realistic.
Thanks,
      GerardM

The roar for #Wikipedia in #India

Three Blind Mice
Charles Folkard 1927
When you are #blind, you can not read Wikipedia. You may have a #Braille terminal but when you don't you are out of luck. To be honest, I do not even know if that works for Indic languages.

The best we can do is have Wikipedia articles read to you. However, when people are to record articles, it means a lot of effort that gets you readings that do not get updated.

Ideally we would give you a TTS or Talk To Speech application. Funnily enough this is something we can give you on-line only for the Indic languages thanks to the Silpa TTS application called Dhvani.

It works for Gujurati, Hindi, Odiya among many others and it could be made into a MediaWiki extension.
Thanks,
        GerardM

Alternate view of my #Blogger blog

#Illustrations are not only attention grabbers, they also help get the stories read. What many bloggers do not realise is that when stories are too long, chances are that people will not finish reading them. Illustrations break the text and invite people back to a next paragraph.



Blogger has a novel way of looking at a blog. It shows for instance a mosaic of the illustrations used. The fun part starts when you guess what a story is about. There are more views that are nice to play with:

Thanks,
       GerardM

Sunday, April 17, 2011

#EOL helps all of us better understand life on our planet

At #translatewiki.net we are preparing for the localisation of the Encyclopedia of Life. The message MediaWiki:Twn-project-eol has been created and has been localised in several languages already. This is the motto of the EOL and it will be translated in many more languages.

There is a page about the localisation effort, there is a template that will be added on user pages when the localisation starts in earnest. It is now just a matter of waiting for the EOL developers to get ready with the big changes they are working on.


Aan: judyannebel@gmail.comTheobroma cacao L.

The EOL has every day a featured species and I think it is fun to feature it on my blog post as well when there is some EOL news.
Thanks,
       GerardM

Saturday, April 16, 2011

#Wikimedia #Commons has its first 10 million media files

Commons, the repository of media files, is where you can find illustrations to illustrate almost anything. The media files contained at Commons are used to illustrate all the Wikipedias and all the other projects of the Wikimedia Foundation.

Before Commons, a popular illustration would be stored in all the Wikipedias that used it. This took a lot of storage and, it was really labour intensive. When Commons was created, the technology of sharing media files from a central repository was not available yet but as the benefits were so obvious, it did not take long.

Once Commons was established as a repository, a big effort got under way to move all the freely licensed images to Commons. The files were tagged with license information and they were reduced to a single copy of a picture.


As Commons became established, several projects chose to do away with the option of local storage while others kept this ability to allow for the use of "fair use" material. As the appetite of the Wikimedia projects and MediaWiki projects grew for great illustrations, Commons was discovered as a place where the cultural heritage preserved by our GLAM partners finds a worldwide audience.

Media file number 10,000,000

Having 10 million media files is a nice moment to celebrate Commons. It is a moment to congratulate all the people who uploaded their pictures. It is a moment to celebrate all the people who work so hard to make Commons what it is: a repository of freely usable media files.

The proof of the pudding is in the eating and for Commons the eating is in finding its material in the projects of students and in the illustration of wikis and websites.
Thanks,
      GerardM

Friday, April 15, 2011

A #Toolserver heaven


When God created his first code, he created it in his image. He put in his own repository, called it Open Source, announced it to the world and walked away to do another creation.

Many of the Toolserver tools are abandon ware. They have been thankfully received by the mere mortals who pay daily homage to the Gods by evoking the code in a quest for reaching the heaven of enhanced and divine effectiveness.

As more mere mortals scaled the heights of tool proficiency, their understanding became fractured as more of them started to talk in different tongues. The code lived on and everyone was still expected to continue building the tower that brings knowledge to all of humankind.

The Gods did send their sons to update much of the code; even though the code was in many books, the revelations could be extracted into a single source that would be delivered to the mere mortals. As these new levels of understanding descended on the faithful, the code itself remained a testament to the original structures and coding practices.

The promise of new coding practices descended on the disciples, they theorised and promised an afterlife where all the code will work to its full potential. Code enshrined for all to see and improve upon.
Thanks,
      GerardM

#eol will be the next project for #translatewiki.net

It is a proud moment when you can announce that the Enclyclopedia of Life is the next project for translatewiki.net. We are ready to make the messages available today but, we are going to wait a bit.

As the software is undergoing massive changes at the moment, we were asked to wait a few weeks. This does impress us because the time and effort our localisers will put into it is considered. It is a valuable resource that will be optimally used.

At the question what the next formalities are, we found no need for them. We are willing and happy to start. We agree with EOL and the Library of Alexandria that it will be motivating to our communities when new localisations become often and regularly available.

Anolis carolinensis Voigt, 1832

There will be challenges ahead, one of them will be the static content that is maintained in a CMS. They are the things that describe what EOL is about, how to donate, how to cooperate ... Technically this will be interesting. As there is not the same structure in such text, making sure that we get the updates in and the translations out will be more of a challenge.

The Wikimedia community will find in the Encyclopedia of Life many references to Wikipedia articles. The fusion of many sources is what makes the EOL so interesting. I wonder if the EOL will eventually also support Wikipedia information in other languages. When it does, it will bring even more sources to the Encyclopedia of Life.
Thanks,
        GerardM

Thursday, April 14, 2011

Why #Commons is an option for a #museum

What answers to give to these questions:
These questions are asked to the people who reach out to GLAMs. While there are reasonable standard answers to these questions, a reply needs modification to address specific issues.

[[Wurts brothers]]

The easiest question to answer is Flickr or Commons; you do not have to choose, you can do both. When you choose for Commons, you will find your pictures used in Wikipedia articles and this gets you a worldwide audience that does run each month in the millions for many of our GLAM partners.

We have a track record on crowd sourcing; the most obvious one is the inclusion of images shared by a GLAM in Wikipedia articles. Other activities have been checking and updating the provided annotations and many of these annotations are translated. Our community is also known to digitally restore high resolution images to great effect; many of them became featured pictures. The real opportunities are often achieved in collaboration with local Wikimedians; this is also where Commons differs from Flickr, interested and involved Wikimedians exist in Mexico, France, India and Australia actually, all over the world.

If there is one thing where we let our partners down, it is not in getting them an audience for their collection. It is in the reporting on this audience. There are the stats by Magnus but they are not regularly updated and are a source of frustration with our existing partners.

Our projects are community driven. Outreach to the GLAMs is however an activity where our chapters play an important role. As they are getting better organised, I am sure that they will reach out to more and more organisations in the heritage sector. My hope is that they will ensure that the statistics our GLAM partners rely on will become timely and consistent.
Thanks,
      GerardM

He will not fix your computer but #MediaWiki #mobile

The #Wikimedia Foundation hired in Patrick Reilly as a senior software developer. His task will be to make MediaWiki a platform that is inclusive of support for mobile devices.

The current mobile platform for Wikipedia is doing a great job for generating traffic. However it is dedicated to supporting Wikipedia, it is written in Ruby on Rails and it is in effect a read only rewrite of the MediaWiki code.

The next generation of mobile support will be written in PHP and integrated in MediaWiki itself. As the PHP Evangelism Team Leader Patrick is bound to bring both the expertise and the enthusiasm to make this work.

With mobile support for Wikipedia doing spectacularly well, this transition will not only see Patrick involved. There is an ongoing collaborative study about mobile platforms. This is where the landscape for the new mobile environment is mapped.


It is a big job and with Patrick in a senior position, there is room enough for developers, for localisers and for anyone to get involved. Check out what is already there and make the map as rich as necessary to fit in all the requirements for our next generation mobile support.
Thanks,
       GerardM

#Unicode support is relative

Regularly there are updates to Unicode. Not only are characters added, sometimes attributes of characters are changed. As Unicode is a standard, there is a whole process before changes are finalised. One such step is the "Public review".

A recent public review is about a change to the category of 12 characters. Once this change is accepted, it will change the behaviour for these characters. Organisations that provide Unicode support in their products will have to adopt their products.

The result of all this; displaying Unicode may differ depending on the version of software implementing the standard.
Thanks,
       GerardM

#Lisa is insolvent - #Standards come at a cost

Lisa is an organisation involved in standards for the translation industry. In order to get an industry to accept standards, there is a need for an organisation that not only helps formulate standards but also promotes their use. Lisa organised worldwide conferences where standards and best practices were presented, promoted and defined.

With its insolvency an organisation that played its role in the definition of standards like TMX and TBX. Let's hope that they find a way to continue the good work.
Thanks,
      GerardM

Wednesday, April 13, 2011

A #font for the #Sourashtra script IV

#Firefox on the #Android has a problem with displaying the Pagul font for the Sourashtra script. Brion checked this out and it failed sadly.

"Browser", the default browser on the Android however shows this page nicely. It showed exactly as it does on a desktop computer "so hopefully that means the ligatures / shaping work correctly on both, rather than that both are broken ;) ".

At this moment the WMF is researching the qualities of mobile devices. In a way, the test by Brion shows how complicated it can be. Do you test the default browser on a device or do you test your favourite browser on a device as well.

The good news is that web fonts work on a default Android.
Thanks,
        GerardM

#Dutch expats want Dutch #movies


Almost all Dutch movies and series will be available on ximon.nl. It comes at a cost, but it is the same cost you pay at a rental. For expatiates however this is a God-send as the nearest rental store with Dutch films will likely be in Belgium or the Netherlands.

Getting the rights for all this material has been a big obstacle. For most recent material agreements exist and as always, it is the old material that proves to be a problem. At the moment almost half the pictures can be seen from abroad and this is expected to improve.

There is again some agitation to prevent illegal downloads in the Netherlands. The irony is that the film industry does still struggles to get its act together. What is the point of making something illegal when there is no legal alternative?
Thanks,
      GerardM

Working for the Yankee dollar ..

#Wikipedia doing a #commercial project? Hell, why not! The license allows it and, it is wonderful outreach.

The Indonesian Wikipedia had a writing competition about theology. Publishers are interested in publishing and, the question is: can we do it.

In Commons we explicitly do not accept pictures available under a CC-by-nc license. This ensures that we have the widest possible audience for our material. It follows that the same applies for our articles. When a publisher or a website wants our material and publish it commercially they can. When they do it with us, even better.
Thanks,
       GerardM

#Wikimania 2012


#Washington, DC will be where Wikimania will be held. The expectations have gone up year by year and by moving to the capital of the USA they will be no less. Congratulations to the bidding team; they get what they have asked for..

Wikimania has its own traditions, one of them is people finding fault with the country or whatever. In many ways it is highly amusing as it is so highly useless; once a venue is chosen, that is it.

Every Wikimania has proven its catalytic effects. What I would hope for is what will be effectively a chapter for the US of A.

But first, Haifa !!
Thanks,
       GerardM

Tuesday, April 12, 2011

A #font for the #Sourashtra script III

Great things often happen a step at a time. Each step is an increment in the road travelled and some have the best view. In the story of the "Sourashtra" font several things have happened.
  • the font is now called "Pagul", it means footstep
  • it is licensed under the GPL (General Public License) version 3  with font-exception.
  • it is planned to included this font in the  ttf-indic-fonts package of Debian
  •  it was tested for Firefox, Chrome, IE, Opera and Safari and iPhone
  • some design issues were identified and dealt with
  • IE6 does not provide the best results :)
With the ever increasing number of languages we support, technical issues like fonts and input methods will become increasingly an issue. As the 500 pound gorilla in language support, Wikipedia is increasingly the vector that helps languages find their way to the Internet. The Sourashtra experience shows us that this is manageable and, that we can reach effectively beyond our MediaWiki borders.
Thanks,
        GerardM

Bringing a #Toolserver tool to the next level II

The #SVG translate tool is developing nicely; my pet peeve has been sort of solved. It now uses the Toolserver User Screening Control which is a tool that allows for checking if a user exists on a Wiki.

At best TUSC is a crutch;
  • it seems not to be aware of SUL
  • it introduces a new user with a new password
  • it clutters your talk page with some necessary gibberish
  • the upload is done by a bot anyway
The origin of all this is that the Toolserver is not a Wikimedia project with all its save guards. Add to this that while Wikimedia has SUL, it does not provide authentication services for Wikimedia movement projects like Toolserver. When MediaWiki were to support openID, we would not need additional passwords and have implicit authentication outside of the Toolserver.

This software already exists for MediaWiki for years.


The upload functionality however does work. Now for an invitation to do more localisations; maybe the template can provide information to find more untranslated SVG graphics.
Thanks,
       GerardM

Monday, April 11, 2011

Guest Blog from Mexico

I have recently calculated that between my husband (AlejandroLinaresGarcia)and I, we have uploaded about 9,000 photographs into Wikimedia Commons related to Mexico over the past couple of years. It has come about through a convergence and compatibility of interests. Ive been in Mexico for about seven years and for this American at least, it really is a case of “so close, and yet so far.” Nothing shows the better than the Mexican attitude towards death -- thought of as a living being (Santa Muerte at right) and something to celebrate (Day of the Dead in the cemetery/atrium of the San Andrés Mixquic church below)

Alumbrada7Mixquic

Mexico is right next door, but historically and culturally it is a world away, which gives it a kind of mystery. This is especially true for Mexico City and the highlands in the center, which is off most people’s beaten tourist path. One unknown is the Carnival of Huejotzingo, Puebla pictured below where many kilos of gunpowder are burned over four days....

ZacapoaxtlaFiringRifle

Making use of our weekends, my husband, a Mexican and an artist, and I began exploring towns and other areas in and around Mexico City where we live. On vacations we can go a bit farther, say to the Costa Grande of Guerrero or to Chiapas, which we will do next week. Between this, the love of detail and the blessing of digital photography, we have no trouble snapping hundreds of pictures at a time, like we did in the Sierra Gorda in Querétaro (mission church  in Concá below).

SanMiguelConcaMission1

Instead of just keeping the photos somewhere, or even having a personal account such as Flickr, it seems only natural to upload them into Wikimedia Commons. I have been writing about Mexico in English Wikipedia since 2007, trying to understand what I see and experience. Articles about places NEED photographs, or it’s just a wall of text. While loaded with texts, articles such as Taxco, Traditional fixed markets in Mexico, Franciscan Missions in the Sierra Gorda and Casa de los Azulejos, Mexico City would simply not be appealing without images. Now if I could only find a way not to spend HOURS and HOURS uploading images one at a time….

A #font for the #Sourashtra script II

When you have a font, you can make a web font out of it. Santhosh did just that with the "Sourashtra" font. An example like this speaks louder then words.

Thanks,
      GerardM

What is the point of publishing cultural heritage on the web II

The "Rijksdienst voor het Cultureel Erfgoed" is missing out on the benefits of crowd sourcing. When you have to pay for what is on offer, would you share the result of work done on the images made available on its website?

In a previous blog I mentioned that much of the material shown is out of copyright anyway and, claiming the availability under a restricting license is copy fraud. The illustration was a print by C. van Noordt dated 1756, and it was pointed out to that as it is a print this makes it even more audacious to claim rights.


This digital restored image of the Diever church is a wonderful example of what can be achieved in the crowd space. As an illustration it serves better then the stained original. It is also debatable if the restoration will refer to the cultureel erfgoed website. While any claim to the rights of this print is bogus, a print allows for multiple sources and sourcing it can potentially create problems.

Not sourcing it is however a bad practice.
Thanks.
      GerardM