Thursday, May 31, 2007

Uploading texts to Wikipedia

A friend of mine had worked on a text for Wikipedia. It was a great text and, I urged him to upload it. This did not happen. I tried to be polite and did not prod him aggressively. At some stage, it was almost a month later, I asked him about it.

His complaint was that it did not work. He had tried it many times only to find that the file was not one of those that could be uploaded. It finally dawned on me that my terminology was the problem; had I said that he had to copy and paste the article everything would have been fine.

Thanks,
GerardM

Sunday, May 27, 2007

A license for SignWriting

The SignWriting script is the product of a process that is already under way for over 30 years. It is a process that has been getting more and more momentum.

The script has been available on line for everyone to use from the start. When people have questions, need support, it has always been provided. People are learning to write their sign language all over the world. The last time I spoke to Valerie Sutton, the creator of the SignWriting script, it had to be short because people from Switzerland wanted to talk to her :)

As the SignWriting movement is ancient in Internet time, it started before licenses and copyright were considered. People needed to be able to use it, so it was made available. In a reply to a previous post on the subject of SignWriting, David Gerard asked about the license and the copyright and asked if there is an organisation behind SignWriting.

Yes, there is an organisation, the "Center for Sutton Movement Writing Inc." which is a USA nonprofit, tax-exempt educational membership organization. The copyright I understand is with Valerie Sutton and, she is considering what license to use. I have been discussing this with Valerie and the license that we are considering is the SIL Open Font License.

The big question she asked me to ask David and my readers is: "If all SignWriting symbols were under the SIL Open Font License (OFL) would you feel free to use the symbols?"

Thanks,
GerardM

Sunday, May 20, 2007

More on scripts, fonts and SignWriting

There were two bits of interesting news this week on scripts and fonts;
  • The Chinese script is much older, it is some 8000 years old. This makes the Chinese script some 3500 years older that previously thought.
  • Redhat made available the "Liberarion fonts", they are fonts that allow for the replacement of the proprietary Microsoft fonts. This is one of the impediments of adopting Linux. It is also one of the more visual aspects where Microsoft shows not to care about interoperability.
The Chinese script is old, it has had a long evolution and it is a living script. It is very much being used and one of the important aspects is that it brings together who speak different languages. The script is very much what unites the Chinese. It is not realistic to change Chinese for the Latin script because it, being based on sounds, will not serve in a same way.

The SignWriting script is young, it represents a revolution in the signing world and it has to be adopted by many signing communities. It is different from other written representations of signing languages because it is actually used for day to day use. As a script, it is different from Chinese because it represent movements like the Latin script represents sounds while a Chinese character represents in essence a concept. For this reason the SignWriting characters will mean different things in different sign languages.

The Liberation fonts make it possible to replace the Microsoft fonts without a need for reformatting the text. When you analyse this, it means that Unicode characters are now available in two interchangeable sets of fonts.

SignWriting does not have Unicode characters and it does not have fonts. The symbols that it uses are complete for most sign languages and some missing characters for the Ethiopian Sign Language are being added at the moment.

One really powerful argument why SignWriting is so important is because it helps deaf people to learn a written language that is foreign to them. English is in essence foreign to a person who grew up with American Sign Language; the written English language is not connected to the every day language of the student. When you learn a second language, you learn the shared concepts quickly. Now in order to know what these shared concepts are, you use a dictionary. Without a dictionary, without a written representation of the primary language, it is extremely hard to learn. It requires a really well trained memory.

It is exactly because of deaf people having to live in a world based on sounds that the bridge that the written word is so important. There is anecdotal evidence that kids who learned to write their sign language are better able to learn the written language of the spoken language that surrounds them.

It is for all these reasons that SignWriting emancipate the deaf. It will emancipate them because as a group they will become better able to communicate in their own world. A world that is both signing and speaking.

Thanks,
GerardM

Thursday, May 17, 2007

Dumb

Over the last month I have become interested in SignWriting. It is a fascinating subject for someone who is passionate about dictionaries. Given that OmegaWiki aims to include all words of all languages, it had bugged me for a long time how to include sign languages as well as written languages.

SignWriting is a recognised script, ISO-15924 Sgnw. It is only recently that it became possible to write a sign language grammatically correct using computer programs. It is written from top to bottom and it does not have its characters included in Unicode. There are some 30.000 characters at the moment and they are being converted from a bitmap into SVG.

I have been watching an instruction video on SignWriting, I have watched a video of some kids singing in sign language. When you see these people sign, I cannot even distinguish the individual signs, it goes way to quick for me. It is quite something, it makes me realise that I am dumb when it comes to signing and illiterate when it comes to writing sign languages. My redeeming quality would be that I am willing to be informed about it.

As the aim of OmegaWiki is to include all languages and as sign languages are as relevant as any other, I hope that the signing communities and particularly the SignWriting community will work with me to achieve this goal. Along this road there is the Unicode challenge and the challenge to get MediaWiki to support SignWriting.

This is likely to happen as SignWriting fulfils its promise and becomes the universally accepted script for sign languages.

Thanks,
GerardM

Tuesday, May 15, 2007

Wiktionary quality issues II

In a previous blog I wrote about the Russian Wiktionary being ostracised by the Polish Wiktionary. There was not enough content and consequently they did not want to have links on the Polish Wiktionary.

Yesterday, I blocked a bot run by an Arab Wiktionarian. I blocked it because it did not comply with the way interwiki links are created, it also did not have a bot flag. A link is only created when the words are exact matches. The problem at the Arab Wiktionary is that they have imported with a bot many words, English words, and they are all upper case.

I have blocked the bot because it is technically wrong. I do run my bot to correct the "damage". The biggest damage however is in the lack of communication between the Wiktionary projects. It is for this reason that I am of the opinion that though courageous efforts, many of them are failures.

Thanks,
GerardM

Saturday, May 12, 2007

Looking at Encyclopedia of Life differently

There has been a lot of comments on the Encyclopedia of Life in the Wiki community lately. All of them miss the most important point. The point is made by the Encyclopedia of Life itself; it will provide "aggregation and will use mash-up technology and Wiki style editing and accumulation of content".

Key in all this is that the Encyclopedia of life is not only Wiki style editing, it also provides for aggregation and mash up technology. It is the one thing that Wikipedia and Citizendium alike are incapable of. With the functionality MediaWiki provides, you are restricted to copying data into the article and having done that, it loses the connection with its origin in a practical way. This reduces these projects to sources of information because of the functionality.

When you look at the examples on the Encyclopedia of Life website, you find that they have information from many sources and present them together. In this way they provide a composite view on the subject that is considered. The data can remain fresh because once data is changed in the sources it consists of, this can be reflected because of the methodology.

One of the exiting things they provide in their mash up is the connection to old literature. This literature is extremely relevant to taxonomy as the oldest name that described a taxon is the one to be used to describe it. The consequence is that many names that were valid once are not valid anymore. It is interesting to learn how they will manage the linking of old valid names to the newer names. This has other applications as well, when an old paper mentions aspects of a species, it may mean that there is no mapping. It is then of relevance to consider the mapping to a subspecies or a different species that is referred to in the old literature.

Mapping the historical names of taxons in time is a hell of a job but an important job because it opens up the understanding of the old literature. For plants, the IPNI resource is of immense value. If anything, I hope that IPNI will be or become part of the mash-up.

Thanks,
GerardM

Tuesday, May 01, 2007

aderente

adarente is an Italian word. This word was defined in OmegaWiki. It has multiple meanings. One of these meanings, not the most common one, became a DefinedMeaning.

When you want to understand all this from what it says, it is extremely relevant to know that indeed "adarente" is the expression that is relevant to all this. When you do not, you will not appreciate that "conforme" is a synonym.

The whole notion of the DefinedMeaning is extremely important to OmegaWiki, it is fundamental. It is sad that, while we do record the word the Expression that goes with the DefinedMeaning, we still not make this visible.

Thanks,
GerardM

Sunday, April 29, 2007

Orientation of text

When you have a text, you write either in a right to left direction, like in Arab or Hebrew, or you write in a left to right direction, like in English or Russian. In OmegaWiki, we are going to support the orientation for right to left once Multilingual MediaWiki is integrated in our software.

I have given SignWriting a lot of thought lately. This is a script that goes top to bottom. What I just realised is, that as as consequence a left to right word mixed into such an environment would also go top to bottom. It makes however as much sense to have a right to left word also go in the same direction.

The problem is that when you have both RTL and LTR in a top down environment, it will seem odd.

Thanks,
GerardM

Friday, April 27, 2007

Google video

For a lark I checked Google Video for Wikipedia articles. It was no surprise to find stuff there. There was all kinds of stuff including this political rant about articles and NPOV on Wikipedia. The guy was insulting and threatening a Wikipedian in a way that I would ban him for a substantial amount of time. This is not freedom of speech is about.

I have no opinion at all about what this guy is on about. What I do know is that he does not help his case. This kind of gangster mentality is really obnoxious. His problem is that by behaving in this way he polarises to the extend that I would not even consider his arguments.

Thursday, April 26, 2007

Skype ..

I have been using Skype for so long now. It is absolutely essential to me. Today I had a first and, the person I spoke to also had a first in the way we were using Skype. To her Skype is an essential tool too.

I spoke with Valerie Sutton, her first was to use Skype and actually listen to it. Valerie is using Skype with video and uses sign language to communicate.

For me it was a first to actually see Valerie while communicating on Skype. It was all really accidental; I was about to disconnect when I found that the video was actually supported as well.

We discussed licenses.. I asked for some pictures to illustrate Wikipedia articles that are about sign languages. She told me that i could have any picture of her that I wanted and have it under a Free license.. We discussed the quality of pictures as well..

Now that the conversation is over, I thought of this, I would like to have a video fragment with every Wikipedia article about a sign language and have someone who is native in that language sign it :)

Thanks,
GerardM

How the Wiktionary interwiki bot is running

In a previous blog about Wiktionary quality issues, in a mail to the Wiktionary mailing list and a post on the English language bear parlour, I informed of the Polish Wiktionaries request / demand not to include the Russian and Vietnamese Wiktionary in the interwiki link.

The general reaction is that the Polish can wish what they want and that I should be willing to accommodate them. It is however not a zero sum game. Whatever I do, it has repercussions and it is in the end for me to make the choice that fits me, as the runner of the bot, best.

I have talked with Andre Engels about this. The result is that I will not update the Polish Wiktionary any more. All other Wiktionaries will still be updated with information with information about all Wiktionaries including the Polish.

I really dislike this situation. The good news is that the Vietnames Wiktionary is aware of this problem and is working on a solution, they just need time. The Russian Wiktionary is in my opinion a wasteland that is not really inviting to work on. Then again it is stubs on a grand scale.. people are working on it though. The Polish .. I am sorry that they feel this way. I do think that this only isolates them and that isolation is the major weakness of the Wiktionary projects.

Thanks,
GerardM

Wednesday, April 25, 2007

The lamest technology mascots ever

Wired has an article with some mascots. The idea was funny enough to get my attention. I got two things out of it. I did not know about the existence of Wikipede or that we might have had a mascot if it had not been for "a slow death by consensus". The other thing is the way our encyclopaedic project is described: " Gangware encyclopaedia Wikipedia".. Funny

Thanks,
     GerardM

Monday, April 23, 2007

Lies, damned lies and statistics

It has often been said that the only "reliable" statistics are the ones that you compile yourself. Strike that, I am not great at this science of statistics, I just use them to blind people with what is made obvious in this way. A week ago we celebrated that we had 15.000 Expressions for German. Today we celebrate that we have over 10.000 Expressions in Italian.

According to Kipcool, the statistics are wrong. His statistics take into account the fact that we occasionally have reason to delete Expressions. According to his figures we have 14.020 German and 9.776 Italian Expressions. This means that we may celebrate the round numbers again at a later date.. :)

For the Destinazione Italia project in OmegaWiki, we will need translations in many languages with an emphasis on European languages. As we already have a lot of content in Italian, it is important to mark what is already there also is known to be part of this collection. In this way do we know how many Expressions are there to process.

It seems to me that numbers, statistics tell a story, they tell as much about what is represented as about the person who uses the data. It is after the collaboration on statistics that you find what the numbers actually mean and what they truly represent. After such a process, they have become the statistics of a project and are owned by its community.

Thanks,
GerardM

Saturday, April 21, 2007

Wiktionary quality issues

On the Wiktionary project I run the interwiki bot. The process is simple; when an article exists in another language spelled exactly the same, I create an "interwiki" link. This allows you to see the information on another language Wiktionary. This process is an automated process, it works on all Wiktionaries and it is an unattended process.

I have received a request from the Polish Wiktionary to stop adding interwiki links for the Russian and for the Vietnamese Wiktionary. The reason given is one of quality. On the Russian Wiktionary many of the articles are created by a bot and they do not provide good information. An example is dispersion, there is nothing really in there. The Vietnamese Wiktionary is more problematic because a bot was used to generate declension and conjugation tables of Russian words and they got it wrong.

The Russian Wiktionary has some 81.000 empty shells and refuse to remove it. The Vietnamese are not willing to remove there incorrect data.

I have been asked to stop including the Russian Wiktionary and the Vietnamese Wiktionary when I run the interwiki process. To be honest, I run the bot as a service and I do not think it is the right thing to do. I think the Vietnamese are wrong not to correct the wrong data that they have. I am less sure about the Russian approach; in essence it is a stub. However, creating a Wiktionary in this way is like stamp collecting; you can look at it but there is not information about it.

Given how the process works, I am not sure that I can exclude either the Russian or the Vietnamese Wiktionary. The way it works is that I run explicitly on all Wiktionaries. When I exclude Russian or Vietnamese, I will probably end up removing all references to these projects. They are the third and fourth Wiktionary is size.

When I do not exclude the Russian and the Vietnamese Wiktionary, the bot may end up being blocked on the Polish Wiktionary. This will also kill off the interwiki process.

From my point of view, using bots to generate content in a Wiktionary only makes sense when there is at least a link to the word in the base language. When the initial creation of stubs is followed by the enrichment of these stubs it is acceptable. For having information that is completely wrong, there is no excuse.

The question is, will there be a discussion about acceptable practices in Wiktionary. The question are:
  • Can the Polish demand what they do?
  • Is having a project that consists mainly of stubs acceptable?
  • Is having incorrect data acceptable?

Thanks,
GerardM

Tuesday, April 17, 2007

The Uyghur‎ user interface

Uyghur‎ is a Turkic language spoken by the Uyghur people in Xinjiang. According to the English Wikipedia, the language is written in the Arabic, Roman and again the Arabic script. Cyrillic is actively used as well.

The officials script in China for this language is Arabic and has been since 1983. The ug.wikipedia has a Latin script user interface. The ug.wiktionary however is right to left. This means that there is at least a need for a user interface that is either in Arab and one in the Latin script. Given that Cyrillic is another actively used script, we need three message files for Uyghur.

When a language is expressed in so many ways, the current MediaWiki software does not allow for supporting Uyghur in a meaningful way. It will be good when Multilingual MediaWiki becomes a reality. The good news is that the prospects for this are really good. The last status report I heard was that some install routines have to be written and then people can start experimenting with this new functionality.

PS Check out what MLMW offers :)

Thanks,
GerardM


Friday, April 13, 2007

SignWriting

SignWriting is one way of expressing signed languages. There are several ways of doing this, signwriting is the one favoured by most people who actually write down the signed languages. There is a request for a Wikipedia for the American Sign Language or ASL expressed in SignWriting.

I am in favour of such a project. There are however a few relevant issues
  • SignWriting is currently not supported in UTF-8 and MediaWiki expects UTF-8.
  • There is software specific to SignWriting, it can be used for a Wikipedia
  • Due to the technical issues, there is no way it can use the Incubator.
Given the objective of the Wikimedia Foundation, supporting this project is very much something that we should support. ASL is a very different language. Given that the organisation behind SignWriting is very much in favour of the creation of a Wikipedia, there is ample scope for the Wikimedia Foundation and the SignWriting organisation to work together and overcome all the relevant issues.

For me it is important that with this Wikipedia project the culture of the deaf will get a boost. When SignWriting becomes even more mainstream, it has the potential to become the script for many of the other sign languages as well.

Practically, I would like to start with this project on WMF servers using the wiki like software that already exists. I would look for programmers and or funding to enable MediaWiki to support SignWriting. I would look for funding to create a full implementation of the SignWriting glyphs in UNICODE.

Thanks,
GerardM

Thursday, April 12, 2007

Braille

When you cannot see, resources like Wikipedia are not available to you. When you cannot see, you need something like braille to make such resources available to you. You still have the issue who is going to convert text to braille.

RoboBraille is an organization that converts to several different formats that can be used by braille readers that work with computers. The software works by receiving an e-mail with the text that needs conversion.

I can imagine that this engine could work for a website as well.. Consider what a difference it would make when Wikipedia would be available in this way.. The converted pages can be cached like any other page. When there is an issue with doing this realtime or near realtime, it would be possible to do this for articles that are featured articles or that are part of a "final version".

I would love to see a solution like this to become part of the service that we provide.. We aim to bring all information to all people.. blind people qualify :)

Thanks,
GerardM

Tuesday, April 10, 2007

Linguists go Wikipedia

In a previous post I mentioned that as part of the funding drive for the Linguist list, the subscribers of the list were asked to vote with their wallet. They did, and an intern will be paid to organise an editorial update of the Wikipedia pages on linguistics. The idea is that the many thousands of linguist that are subscribed to the linguist list will ensure that notable linguists like Eve Clark and Tanya Reinhart will get a mention.

Consider, this is a leading list of some 14,649 linguists that will be urged to help us improve both the quality and the quantity of the coverage of the field of linguistics.

I am really exited about this.

Thanks,
GerardM

Saturday, April 07, 2007

A board of trustees or an executive board

In a post on his blog Tawker suggests that it was "widely reported why" Mr Wool left his job. This is not true. Mr Wool explicitly refrained from explaining why he left his job and in this way prevented a discussion both about the organisation and his role in it.

The organisational consequences in the statements made in this blog are also very much wrong. The Wikimedia Foundation is in the process of setting up an organisation. This effort is hindered by a lack of funding and an unlucky choice in staff. As the staff becomes more professional, the board will be able to distance itself from the day to day affairs.

Thanks,
GerardM

Friday, April 06, 2007

Vietnamese

In OmegaWiki you can have many of the labels used in the data part in your language as well. For this to function, you have to select a language in the "user preferences" and there have to be translations of the words involved. Words like Estonian, Georgian or noun, verb, adjective will be shown in the selected language.

As the user interface is one of the most critical aspects for getting buy-in, it is something I often spend time on. Given that I do not speak languages like tiếng Việt, it is not always easy to find the right translations. For this language I was not able to find Gujarati. More bewildering was that I could not initially find the word Korean, it was there as korean with "tiếng Triều tiên" as its translation.

All in all I have added more than 10 translations in this latest session. I am sure that people who speak this language will have an easier time doing the same job.

Thanks,
GerardM