Saturday, February 20, 2010

Getting ready for new business

A #wikipedia in the Karachay-Balkar language is imminent. It is waiting in the incubator for the word of the board and, then we will seek its creation. It sports some 900 articles, the complete MediaWiki core has been localised and, what really shows their determination to succeed; they just completed the localisation for the Mobile MediaWiki as well.


What I learned this week:
  • People do not know that http://en.m.wikipedia.org is Wikipedia optimised for mobile phones
  • For 24 Wikipedias the mobile interface (24 messages) is completely localised
  • four languages were completed this week !!
  • The localisation for languages like French and Hebrew is not complete
  • Support for right to left languages like Urdu and Arabic is about to be implemented ...
  • The Indonesian interface is a bit sparse .. how to prettify this
  • People do not know about improved mobile access so they do not use it
  • The strategy project expects a lot of traffic from mobiles in the third world
  • There is no logo for Wikipedia Mobile
  • How do we serve Wiktionary, or other projects on the mobile
  • Wikia has its own mobile implementation
  • ... there is probably much more to know
Thanks,
GerardM

Friday, February 19, 2010

Is the Aramaic #Wikipedia in the Syriac language

I received a letter in Dutch about the Aramaic Wikipedia. It says:

In the language pages of Wikipedia you can find a certain “Aramaic” Wikipedia using arc in the domain. According to ISO arc is a language that was spoken from 700 to 300 BC. It is currently extinct. The current Aramaic language is called Syriac. It has its own script.

I am fluent in the Syriac language but not the Aramaic language. The articles in the arc.wikipedia are all written in the Syriac script, not the Aramaic script.

My question is if it is possible to change the name from arc to syr because the facts indicate that this would be better. Wikipedia is known to provide truthful information. The current situation does not reflect correct information.

I looked into the matter and found that according to SIL the Syriac language is a macro language. It consist of two living languages; Assyrian Neo-Aramaic and Chaldean Neo-Aramaic.

I have no clue what to think of it. The best I can do is ask around, ask you.
Thanks,
GerardM

Mobile #MediaWiki localised in Malayalam and Swahili

I wrote that the Urdu mobile Wikipedia domain is localised and ready for use and, this proved an excellent argument to localise the 24 messages that are part of this skin / interface. Within 24 hours the localisation for the Malayalam and Swahili language has been completed.

There has been a lot of attention in the Strategy project about mobile access. The good news is that three languages have done what they can do now. All it takes is localising 24 messages and have the mobile domain activated.. Getting people to do just that is probably the most effective near term strategy :)


A happy Muddyb Blast Producer asked when the Swahili Wikipedia will have its mobile domain.. I hope soon and, I hope it will raise awareness for the Swahili Wikipedia and generate lots of traffic and bring in many new editors.
Thanks,
GerardM

Closing #wikimedia projects (rant)

Proposals for closure I hate probably the most. They go on for forever, are often requested anonymously and to top it off, there is no clean procedure for finishing of projects because it is not implemented.

When a project is closed, it should be removed. It should not be in the statistics, it should be removed from the Internet. When we actually hope for it to restart at some stage, it should go to the Incubator.


These are the smallest Wikipedias in the traffic statistics. Most of these are closed. They give a bad impression because there is not much you can do once they are "write protected". There is not even a procedure to lift the closure...

The worst part of the continuous bitching is that some anonymous cowards find it in themselves to propose the closure of projects like the Swahili Wikipedia. Such insane proposals are taken seriously and they create a lot of unnecessary stress. In my opinion this is trolling pure and simple and should be dealt with accordingly.

If closure is to be considered in the first place, I would suggest that no activity for over three months would be a requirement and, that new activity resets the clock. It would also be good to define additional limits that prevent a closure request to be considered.. to mind come the finished localisation of the "most used" messages and 1000 articles for a Wikipedia.

I write this rant because people who are working towards activating our projects get really distressed and disheartened. Most of us truly want all our projects to grow and do well and the last thing we want is for the people to move on who have an interest in these projects.
Thanks,
GerardM

Thursday, February 18, 2010

Urdu #wikipedia gives priority to mobile

The Urdu Wikipedia is one of the smaller projects all be it with a big potential; more then 60 million people speak the language. At translatewiki.net there is a lot of activity for this language.

What I find really interesting is that they have given priority to the mobile interface over the most used messages. What this may mean is that for Urdu Wikipedia access is best achieved using mobile telephony.

The yearly growth of traffic is 65% and they currently have some 13K articles. I understand that the Urdu Wikipedia is now available for mobiles. I do wonder how we will learn if this indeed makes a lot of difference for the popularity of Wikipedia in Pakistan.
Thanks,
     GerardM

Wednesday, February 17, 2010

Doing what I am not good at ...

#Translatewiki needs a new main page.. It needs to better reflect the state of the art for MediaWiki and of what translatewiki is about.


I have created this info about the active projects at translatewiki. I had to work a bit on the icons that I used. This info could be appended to the bottom of the main page.

What do you think, what more needs to be done ?
Thanks,
     GerardM

"Psst! Wanna buy an alphabet?"

#Wikipedia is supported in over 250 languages. MediaWiki supports Unicode and, that is one of the reasons of its success. This does not mean that all languages are technically served well. The request for a new Wikipedia, the one for American Sign Language is delayed for such reasons.

I have argued in the past and, I will again that this kind of technical support cannot be expected of amateurs. It takes experience to solve the whole puzzle that makes a language properly supported on computers, on the Internet.


Michael Everson has the kind of expertise that allows him to solve the issues that exist with many of the languages we have Wikipedias for. Malayalam, Burmese, Lingala are some of the languages that would benefit from dedicated support for their fonts. When it turns out that characters are missing in Unicode, he can expedite the inclusion of such characters. There are strategy proposals that could be implemented when Michael can take charge of them (1) .

Michael is a valued member of the language committee, added thousands of characters to Unicode, was responsible for the inclusion of several scripts in Unicode. He is a champion of Blissymbols and he would love to see a Hindi or Malayalam Wikipedia that is bigger then for instance the Dutch one.
Thanks,
GerardM

#Mediawiki seeks to support your language

Icons are part of the user interface and they are not localised in the same way like system messages. A fair bit of design is needed to make them look good and to give them the look that makes sense for the audience of a given language.

I am really happy that Parul Vora, the interaction designer of the MediaWiki Usability Initiative blogged about the need for your support. This way the advanced toolbar can be all Cyrillic, Devanagari or Kannada..


We are really happy with a first contribution for Chinese .. and we are hoping for many more. My hope is that your contributions will all be integrated in MediaWiki itself so that your perfect toolbar will become available on all MediaWiki installations.
Thanks,
      GerardM




Monday, February 15, 2010

Queen Beatrix at centenary celebrations of the Royal Tropics Institute

The queen and princess Máxima will be present on March 11 at the kick-off of the festivities celebrating the first 100 year of the Royal Tropics Institute (KIT) there will be an ethnic fashion show, music and dance.

At this invitation only celebration, a new documentary called "100 years KIT" will have its first showing. In it a journey will be made to Suriname, Mali and Indonesia where the viewer will learn about the projects, and the work of the Royal Tropics Institute. At the same event the annual report 2009 and a new corporate brochure will be presented.

The Royal Tropics Institute is the parent organisation of the Tropenmuseum.

#Usability testing without language restrictions

When the #Wikimedia Foundation accepted the grant that paid for the usability initiative, it came with restrictions. The money was to make the English Wikipedia more usable. It does and it did make significant improvements, the "vector" skin is a vast improvement; I use it where I can. The en.wikinews made it its default skin and I expect that many projects will follow.

While the new developments were not targeted to other projects and languages, they did benefit. Test projects were started for other languages and people were invited to test and to comment. As I understand it, the "vector" skin is better at supporting right to left languages like Arabic.

The multimedia usability project being next, does not have any language restrictions and arguably, if the quality and quantity of Commons is to improve, it is the "other" languages and cultures that are not covered as well as the Western languages like English, German, French and Dutch.

To improve Commons, I would concentrate on one western language; English. One language with a right to left script and one language with a Cyrillic script. The choice for Arabic and Russian would be obvious. Given the expertise at the WMF, I might add Japanese to make it a really nice mix.

Each of these languages pose their issues. Each of these languages represent big vibrant projects that show a healthy growth in traffic. When money is to be spend on usability tests, research in these languages and cultures will have more impact then testing additional western languages. Personally I would consider looking for a partner in a country like Indonesia, they are more likely to have expertise in Asian languages then companies in San Francisco.

In her letter to the board on strategy, Sue argues that our community is to diversify. It seems to me that the multimedia usability project has the potential to have a big impact in reaching out to those places where we want to do better.
Thanks,
     GerardM

Sunday, February 14, 2010

#StatusNet in Arabic

The localisation of StatusNet is well under way in the Arabic language; with 62.46% of the messages localised. The Persian language is only slightly behind with 60.83% and Hebrew is also there but at 17.36%.


As you can see, there is some more work to be done; when the final localisations are in, when the text is properly in the right to left direction, StatusNet will be ready for deployment in these languages.
Thanks,
     GerardM

Saturday, February 13, 2010

A new face for #translatewiki

The main page needs a bit of a redesign; its content does no longer reflect translatewiki.net that well and its default skin is the modern skin. The "modern" skin does not reflect the best of breed in skin technology and "vector" is what defines the best the Wikimedia Foundation has to offer.


There is a new impetus to improve on the existing skinning system by Jack Phoenix and as translatewiki.net is always on the very bleeding edge, it may make sense to make the translatewiki skin one that includes the very best and the very latest in skin design and technology. One reason not to choose "vector" is because it is likely to define the look and feel of WMF projects.

Nikerabbit started this idea for a redesign with a LQT thread. Given that translatewiki.net is the ultimate multi lingual wiki, extra attention has to be given to other scripts including for right to left languages like Arabic, Farsi and Hebrew.

So there are two issues that need to be addressed:
  • update the translatewiki.net main page content
  • update or replace the skin to showcase our best of breed technology
Thanks,
     GerardM

Thursday, February 11, 2010

15 months of progress

At #translatewiki.net there is a page with MediaWiki extensions with localisations in more then 50 languages. It was not updated for some 15 months and now it shows 314 MediaWiki extensions up from 128.

The Babel extension for instance supports 186 languages and, it is an extension that several Wikipedias requested to have installed.

The Wikia extensions only recently found their way to translatewiki.net and, there are already some who have localisations in over 50 languages..

All this is possible thanks to our wonderful community, you can always check out how your language is doing; we can use help in localising, proof reading and helping us maintain and develop our project.
Thanks,
      GerardM

In Friulian the language is called furlan and not friulano

In interwiki links on #Wikipedia, we use the native name of a language to refer to the article in other languages. For the Friulian language Wikipedia, it should say "furlan" but it says "friulano" because it defaults to the Italian translation.

We get these names from the CLDR, or Common Locale Data repository. The CLDR plays a vital role in the support of languages; with a bare minimum of information it becomes possible to make a language version possible in OpenOffice.org. One of the essential bits is the name of the language as it is called in that language.

In MediaWiki, we use the CLDR and consequently people are urged to ensure that the information in their language is correct. This was the message that was given for the Friulian language. It may be cumbersome, but it does ensure an improved support for a language.
Thanks,
     GerardM

#GoogleBuzz is a perfect opportunity for SIP

SIP stands for "Session Initiation Protocol" and it is a key protocol that allows for open internet telephony. Google Talk makes use of this protocol, and consequently it allows you to connect to other SIP providers like VOIPbuster.

Skype is the big, closed and extremely popular internet telephony service that everybody loves to hate. What I would love is when Google made a service available whereby people could connect never mind if they use Skype, Google Talk, VOIPbuster or whatever other service... there are still people who use only POTS.

The benefit would be a much improved telephony service and it would certainly stimulate improvements to the quality of service.. I would love it when the lag on Skype chat would go the way of the dodo.
Thanks,
      GerardM

Wednesday, February 10, 2010

What is the buzz about

Google introduced Buzz in my gmail. It looks interesting; for me it has obvious benefits over Facebook. Facebook is such a time sink, all the crappy games that you are invited to play..

One of the things you can do is include my tweets from twitter, I expect that it is just a matter of time and my identi.ca tweets are included as well. This blogpost is an experiment as well; when posted it should appear as part of the buzz that I generate ...

In Google buzz you are connected to the people that you have contacted with by mail. This makes sense, what I still have to find out how I can include other people to follow.
Thanks,
      GerardM

Tuesday, February 09, 2010

The results of the Kiswhahili #Wikipedia challenge

I received this e-mail from Christine Moon on behalf of the Kiswahili Wikipedia Challenge Team. I think it is of interest to all of us and, I want to congratulate and thank the Challenge Team and the contestants for their important and interesting work.



Since we launched the Kiswahili Wikipedia Challenge in November 2009, there has been a significant growth of the Swahili Wikipedia. In the first month alone, Swahili Wikipedia grew 30% from 2.2MM words to 2.9MM words, which can be compared to single digit organic growth for many African languages like Afrikaans and Zulu.  You can view the progress of the growth published by Wikimedia here:. The statistics are encouraging and show that the challenge was able to spur individuals who would've conventionally been consumers of Wikipedia articles to become creators of new content on Wikipedia. 
As we announce the final prize winners of this Challenge, we're quite thrilled to hear that many participants would like to continue to contribute articles and content to benefit the online community. Some participants said it well, "the winners of this challenge are all of the participants who are walking away with the knowledge gained." As Swahili Wikipedia continues to grow and is referenced by more users, all of you who participated in this Challenge can take pride in the critical role you've played in building up this rich resource.  Thank you to all those who participated and all the judges who made this a success.  Remember, this is just the  beginning! 

Now, please join me in congratulating the final prize winners.  Congratulations for your hard work and passion for Kiswahili!! 

*Final Prize Winners:*

Grand Prize Winners
Kenyatta University - Peter Kanda (Kandyzo)
JKUAT - Samson Maosa (Coolsam)
Strathmore University - Abbas Mahmoud (Abbasjnr)
University of Dar es Salaam - Maria Alphonce (Maria alphonce)

First Prize Winners
IFM - Janeth J Jonathan (Janeth J Jonathan)

JKUAT - Limoke Oscar (Limoke oscar)
Strathmore University - Daniel Kimani (Kimanikihiu)
University of Dar es Salam - Moses Kehengu (Moseske)
University of Nairobi - Peter Kamero (Kamero)
Individual - Neville Tirimba (Perijove)

Second Prize Winners
JKUAT - Eric Kosgey (Erickosgey)
JKUAT - Kenneth Kirop (Kiropkanda)

Kenyatta University - Jacob Kipkoech (Jakibiro)
Strathmore University - Mark Ekisa (Ivermac)
Strathmore University - Erick Njenga (Erick njenga)
UDSM - Mwasapi Kihongosi (Mwasapi kihongosi)

University of Nairobi - Martin Nduati (Nduati)
University of Nairobi - Damiano Wanjohi (Wawanjohi)

Third Prize Winners
Kenyatta University - Chacha Maroa (Chamaroa)
Strathmore University - Alex Wafula (Xelawefs)
Strathmore University - Patrick Kinyua (Rickiny)

One bot does almost 1,0% of all #wikimedia edits

Siebot is the bot operated by Siebrand.. This one bot alone did at the last count 9071569 edits and the counter for the total number of edits is at 952027924 edits. Siebot is used for many functions, it is used heavily for Commons maintenance and for interwiki linking.


The thing with statistics is that you have to be wary of what it is a number represents. One billion edits is enormous and, they are indeed all generated by our community, some do an awesome job and automate much of the drudgery.
Thanks,
     GerardM

Monday, February 08, 2010

#Wikimedia projects counting down to a billion edits.

On the Toolserver there is this counter that shows how many edits are left before we hit a cool billion.
Thanks,
GerardM

#MediaWiki is used for the Haiti relief effort

#Crisiswiki.org is a MediaWiki wiki building a "yellow pages" with everything that is relevant for the Haiti disaster response. This project is organised by Crisis Commons, an organisation that not only aims to provide relief when needed, but also stimulate people all over the world to prepare for an un-hoped for disaster. There is a wiki for that as well.

On their website, you find many geek organisations that have involved themselves in the relief effort. Translatewiki is not that geeky, then again I am biased, but we have already identified two applications that are used for the Haiti relief effort, MediaWiki and OpenStreetMap. Both can do with a massive effort to localise these applications in Haitian Creole

We welcome Open Source applications that are used for disaster management and disaster relief.. When many applications are supported at translatewiki, our community will be strengthened by the people that come specifically to lcoalise this software but equally important, it gives us something real to do, something other then giving money and hoping for the best.
Thanks,
     GerardM

Saturday, February 06, 2010

More on the Haitian relief effort

#OpenStreetMap is used extensively in Haiti, more then the other geo locations. There is a great article in the Guardian that extols the virtues of OpenStreetMap in Haiti. What makes this article so interesting is that it talks about the many relief support applications that are currently in use.

Software like Sahana that helps people find people, managing aid, managing volunteers, tracking camps and liase among the many people and organisations involved. It is the kind of software that should be ready for wherever it is needed.

There is more software that is used, read the article for that.. My point is, that when disaster strikes, you want to be ready. Disaster can and will strike anywhere. When you are prepared, when your software is ready for deployment in the whole world, when it is localised for the people that are there, they can actually use it.

I am happy to learn that we may get people localising OSM at translatewiki.net. I would love it when all the software mentioned is ready to be used when disaster strikes in my backyard,
Thanks,
GerardM

Friday, February 05, 2010

#Wikia localisations are doing well at #translatewiki

The localisation of the Wikia extensions is doing well; a start has been made for many languages. Thirteen languages are now supported for more then 50% of the messages.

The Wikia extensions can now all be localised at translatewiki.net and the software at Wikia has been updated with both improvements to the internationalisation and with the many localisations.

The people at Wikia and translatewiki.net worked hard to get everything working properly. The next step for Wikia will be to implement the LocalisationUpdate extension. They will hopefully do this in their next development cycle.. This will not only make the Wikia extension messages available but also the MediaWiki core messages.

Lately the localisation for languages like Singalese, Urdu, Hausa and Igbo have started or improved considerably and it will be great when people can start their Wikis in these languages at Wikia.
Thanks,
    GerardM

Sunday, January 31, 2010

Commons has 6,000,000 media files

One of the #geograph pictures became the 6 millionth file on Commons.

Geograph is a British organisation that aims to collect geographically representative photographs and information for every square kilometre of Great Britain and Ireland. They do a splendid job and they make their material available under the CC-by-sa licentie.


The picture is called Sailing on Ullswater and was created by James Hearton.

The Geograph contains really nice pictures and, it is a great concept. It would be a great concept for the WMF chapters to adopt. Adopting it on a global scale would be a great way of stimulating local interest for our projects.
Thanks,
GerardM

Thursday, January 28, 2010

Amsterdam fashion week

The #Tropenmuseum currently has an exhibition on traditional clothing, modern designs, video portraits and even a catwalk.






It is fitting that one of the images that the Tropenmuseum asked us to digitally restore was finished this week. Durova blogged about the issues with this work and Susanne recognised it immediately as one of the photos by Tassilo Adam that was featured in an exhibition about the Batak people of Indonesia and the fabrics they use in their clothing..
Thanks,
     GerardM

Wednesday, January 27, 2010

The Gods do not use an iPhone

The #Tropenmuseum new object of the month is this statue of the God Ganesha.

In a break with the past, the object of the month is about objects that have an interesting story to tell. This month, Ganesha is sitting on a money box with a mobile at its ear.


The associated text is in translation:

Ganesha, contemporary appearance of Ganesha as a businessman, sitting on a money box having a conversation on his mobile. The God with the head of an elephant, Ganesha, is the son of Shiva and Parvate, two important deities of the Hindu pantheon. He is the leader of the ganas (earth spirits and kobolds). Ganesha is the God of wisdom and the remover of obstacles of material and psychic nature. As a consequence he is venerated before important business is undertaken. The popular Ganesha can be found in tempels on house shrines, in niches in shops, on dashboards, near a door as a statue, print or as sticker. In many places Ganesha is venerated in the form of a stone painted orange with the contours of en elephant. He is depicted either sitting, standing , crawling or dancing.
Thanks,
GerardM

Sunday, January 24, 2010

#OpenStreetMap is used in Haiti

When a disaster strikes, good information helps understand the situation, it helps the planning and execution of the relief work needed to bring some normalcy back.

On OpenGeoData there is an article that explains how the Fairfax County Urban Search & Rescue Team uses OSM maps on their Garmin units. It is a happy situation that these maps are there for everyone to use.

When you look at the localisation statistics for OSM, you find that Haitian Creole is missing. There is a Haitian Wikipedia so there is support for this language at translatewiki.net. It would be good if we find people willing to work on the OSM localisation as well; this will make the maps useful to the people of Haiti as they are rebuilding their country.
Thanks,
GerardM

Saturday, January 23, 2010

Sue's letter to the board about strategy


On the strategy wiki there is a letter from Sue Gardner and the strategy project team. This letter essentially is a digest of the recent developments of the process that is to produce a strategy for the next five years.

There is much in it and, I am trying to get my head around it. I think I like it, certain parts are brilliant and there are parts where I am not sure. What I really like is that the WMF clearly positions itself as the organisation that makes our projects possible and, much of the local organisations is left to chapters. The problem is what to do when there is no chapter, when a culture, a country is not well covered in our biggest Wikipedias let alone in Wikipedias in local languages.

I am thinking on this, and I am sure to blog again about this in the future.
Thanks,
GerardM

Friday, January 22, 2010

#SIL publishes the ISO-639-3 changes of 2009


Some of the newly published changes are significant for the Wikimedia Foundation; of particular interest is the recognition of the Latgalian language.

With the recognition by SIL, Latgalian is now eligible for a Wikipedia and at translatewiki.net it now merits support as a language. So far Latgalian was considered to be a part of the Latvian language and as a consequence of the new situation, best practices have it that Latvian is now best expressed with the lvs or "standard Latvian" as the current lv / lav codes now refer to a macrolanguage.
Thanks,
      GerardM

#skype works for me again as it should

A functioning skype client is essential for me. Lately the version that came with Ubuntu did not work for me; the sound was broken. I un-installed skype and, eventually learned that skype is now hosted by skype itself. At first this made no difference at all; it was broken, it was awful.

Yesterday, I was notified that a new version of skype was available. I installed it and now it works as it should.. No more rebooting into Windows for me.
Thanks,
    GerardM

#translatewiki.net gets its translation memory

Time is the most valuable resource at translatewiki.net. Everything is done to make our localisers and developers as efficient as possible. Yesterday, we improved the performance of our system thanks to netcup. Today we are really happy to extend our functionality with translation memory functionality.

A translation memory (TM) helps by suggesting a translation based on the work that has been done before. When you localise you get the suggestion what the translation might be and, this will make the process more efficient.

In order to make optimal use of this functionality, it helps when many applications share the same TM. This makes it easier to explain why translatewiki.net is so interested in more open source applications to localise; not only gives it new challenges to our community, it also makes us more efficient.
Thanks,
      GerardM

Wednesday, January 20, 2010

#translatewiki.net is ready for new business

The conversion to the new translatewiki.net server has finished. The DNS entry has been changed.. It can take a day for your ISP to reflect this change..

A big thank you to netcup.de for supporting translatewiki and a big thank you to Siebrand and Niklas for the successful changeover.
Thanks,
     GerardM

#translatewiki.net is being upgraded ....


Monday, January 18, 2010

#Translatewiki.net preparing for an upgrade

With the number of languages, the number of applications, the number of messages per application but best of all the size of our translatewiki.net community growing all the time, it is no wonder that the server capacity ,so graciously donated by netcup, did not fit our requirements any more. The scripts that update our statistics or update the source repositories started to take longer and longer. Our developers were fighting an up hill battle to keep our code sane, efficient and functional and consequently our system suffered.

There are solutions to such problems and, we are really happy to announce that Netcup has been so gracious to provide us with an improved service.

We will start the migration on the 19th of January at 18.00 UTC. The old service will go read only, and once we are done, we hope that you will come and see to what extend we can hammer our system localising some of the best Open Source software around.


If you are more interested to learn if we truly need this new service, you are welcome to peruse the bugs and feature requests and help us ensure that it will take a long time before we need another upgrade. On the other hand, with improved capacity, we may find new challenges for our community any way.
Thanks,
GerardM

Sunday, January 17, 2010

#Tropenmuseum image most used in #wikimedia projects

One of the recent developments around the support to GLAMs is a tool by Magnus Manske called GLAMorous that shows how many images are used from a particular category on Commons.

When you aks for the details for the "Images of the Tropenmuseum", you will find that this image of the Governor-General Mr Andries Cornelis Dirk de Graeff is used 11 times. it is funy to find that the Wikipedias cannot agree among each other what his correct name is; Cornelis or Cornelies..



When you look at the most used images, you find that many of the pictures that have been so lovingly restored are doing really well. Another category that proves popular are part of the royalty of Indonesia, sultans, kings, princes, rayas...

The thing that surprises me most is that the Dutch Wikipedia makes most use of the material of the Tropenmuseum, this may change. I hope that the srn.wikipedia and the id.wikipedia will challenge the Dutch position ..
Thanks,
GerardM

Tuesday, January 05, 2010

Even more complete #Wikipedia statistics


The Wikipedia "Current status" report has become even more complete; it does now report on the localisation of MediaWiki as well.

In order to provide these statistics, some juggling of the numbers is in order; there are for instance languages with more then one localisation. A good example is Portuguese where we have a localisation that caters for the people from Brazil and one for Portugal. The "value" for these localisations is achieved by adding them up and diving by the number of localisations.



The localisation is expressed as a value from 1 to 100 and is comprised of three parts. The "most used" MediaWiki account for 40 points, MediaWiki core is another 30 points, the extensions used by the WMF is 30 points.

When you analyse these numbers, you will find that the "most used" messages are really important. This is only right because of their importance in the usability for a localisation.

What we hope for is that the real wizards will find the relation in numbers for the value of a great localisation for the popularity of our Wikipedias.. When you talk about it, it is obvious. It is however really hard to appreciate its true value.
Thanks,
GerardM

Happy new year



Every year I receive seasons greetings, this one I received from Valerie, I know it is going to be a great year for her. I wish us all happiness, health and wealth for the comming year :)
Thanks,
     GerardM

The #signwriting extension for #mediawiki is getting ready


A lot of recent e-mails did not get the attention they deserved. I received one that makes me happy; it is by Valerie Sutton. It announces that the MediaWiki extension for SignWriting is ready.

SignWriting is a script that enables the writing of sign languages. Given that there are more then 100 sign languages, it is really important that these languages can be written. It is important because for many people, their sign language is their mother tongue and teaching people to read and write in their mother tongue benefits people for the whole of their academic life.

With the availability of a MediaWiki extension, it becomes finally possible to consider a Wikipedia in American Sign Language. Given that SignWriting is a top down language, an extension is required. Given that SignWriting is not supported in Unicode, an extension is required..

Normally when a request for a new language is made, there is a requirement for the localisation of the "most used" MediaWiki messages. Given that MediaWiki itself does not support a top down user interface, it makes sense to give dispensation for this requirement..

The next step will be to create an incubator for ASL. It will be interesting to learn how we will bring the signing world and the wiki world together !
Thanks,
      GerardM

#Translatewiki.net supports #wikia


The new year starts at translatewiki.net with a bang: Wikia, is now supported for its localisation and internationalisation.

This is a big deal! Wikia is after the Wikimedia Foundation, the biggest user of the MediaWiki software. It hosts over 60,000 wikis, has over 3 million content pages and is with 30+ million visitors per month a top 100 world website.

With Wikia making use of translatewiki.net, 68 MediaWiki new extensions are gaining internationalisation and localisation support. For Wikia this is in addition to the 24+ extensions that are already supported at translatewiki.net.

As you may know, the extensions developed by Wikia are as much Open Source as all the other MediaWiki extensions. With the internationalisation and localisation happening by the same community, the same best practices will be used. This means that plural and gender support will become available in the Wikia software, I am sure that this will have its impact on coding practices.

Languages like Kotava may now get active support. Wikia supports more languages, it will be interesting to see how they will be fitted in.

I hope that other companies who provide MediaWiki hosting or products will be interested, it is for them to consider their localisation support. What makes translatewiki.net so valuable is its community. It is not easy to build your own.
Thanks,
     GerardM