Hindi link removed

Thank goodness! An administrator said that link to Devnagri conversion was spam. It's no longer there. I am happy. I was getting angry for no reason. Hehe!

More foolishness on Punjabi Wikipedia

You fools! You pinheads! You bastards! How could you? When I suggested attaching a Gurmukhi-Shahmukhi converter to the Punjabi Wikipedia you acted as if I had inserted a ten-feet long rod covered in spices into your anus! No, you shouted, Western Punjabi and Eastern Punjabi are two separate languages. ARGHH! You same fools are now experimenting with an automatic Gurmukhi-Devnagri converter! The machine is still in beta and it does not function. But that is beyond point! How dare you think you can impose a script on Punjabi? Shahmukhi has been in use for over eight centuries, Gurmukhi has been in use for nearly four centuries. And Devnagri! You stupid Hindu nationalists! First you took away "jugrafiya" (geography), "klas" (class), "student" (student), khla (universe), and hundreds of other words and replaced them with Hindi/Sanskrit words. Now, you want to impose a script no one uses to write Punjabi! That is insane! But why should I care? Yeah!

Russian Wikipedia expands by 14,000 articles in the last month

One month. That's how long it has taken the Russian Wikipedia to close the gap of approximately 6,000 articles with the Spanish Wikipedia. RusWiki has expanded by more than 14,000 articles since Jan. 26 while the SpanWiki managed approximately 7,000. If the Russian tide continues, we will have another million-article Wiki within a couple of months.

Feb. 25 article count

Russian Wikipedia: Сейчас в Википедии 972 032 статьи на русском языке.

Spanish Wikipedia: 971 969 artículos en español.

Italian Wikipedia has million articles

The Italian Wikipedia now has over a million articles. I noticed that two or three days ago. Then I forgot to write about it. So silly.

Here are two images of Wikipedians celebrating this occasion:

THE LOGO

HOMEPAGE

This makes the Italian Wikipedia the fifth wiki to brag about having a million voci.

Race is now between the Spanish and Russian wiki. Both will cross the million mark in the next six months. Although Spanish is leading by about 6,000 for now the rate of creation of articles on the Russian wiki is more.

Spanish Wikipedia:

Russian Wikipedia:

This is going to be interesting. 

Italian Wikipedia to cross one million mark this week

It's been about three weeks since I wrote last time on this blog. I didn't have any thing worthwhile to talk about. I gave up Persian in December, then learnt grade one Kanji and a couple of dozen Kanji from grade two and stopped there, then dabbled into Chinese again only to end up reading an erotic novel in Esperanto.

I am not going to brood about my failures because there is something more interesting happening on Wikipedia.

The Italian Wikipedia now has over 998,500 articles. It's only 1,500 short of the coveted one million mark. The numbers for December 2012 aren't out but this chart shows 357 new articles were being added to the Italian Wikipedia every day during November last year. If that figure is still close to the current rate we should see the Italian version crossing the one million mark within the next four to five days. (1500 / 357 = 4.2)

Following the Italian Wikipedia in article count is the Spanish one. It currently has over 959,000 articles. The latest numbers (Nov. 2012) put the rate of creation of new articles at 260. It needs 41,000 new articles to reach the one million mark. At current rate, it should be there in five to six months.

There is also the Russian Wikipedia which is growing by 423 new articles a day. (Nov 2012) There are currently over 954,000 articles on it. Divide the articles it needs by 423 and it is well on its way to become the fifth (if the rate of creation of new articles doesn't abate) Wikipedia, possibly ahead of the Spanish version, with a million articles within the next four months.

Finally we have the Polish Wikipedia which is churning out new articles at 204/day. (Dec. 2012) It currently has over 944,000 articles. This means we will have to wait until very late this year to see this version go past the 1,000,000 mark.

The Japanese and Portuguese versions are huge too but I don't think they will make it into the Wikipedias with 1,000,000+ articles category this year.

Punjabi Wikipedia - stupid administrators, no unification

Yes, the title tells it all. They are stupid and too entrenched in their comfort zone to even think that unification is possible. One of them says, 'Western Punjabi' is a different language because it has its own ISO code, another claims 'Kashmiri' and 'German' have the "same" grammar and those from Pakistan don't even care to respond! 

I wouldn't have minded had they given some well-thought reasons for not unifying the two Punjabi Wikipedia. It would have been a pleasure to know where I am wrong. But no! They wouldn't have a discussion. Instead, they would aim their guns at me and keep on asking who I was to order them to learn Shahmukhi. 

I tried hard but those fools won't even budge. When they didn't have a good reason (they are too dumb to think of even one good reason to not unify) they turned to saying: 'Wikipedia is no place for activism.', implying I should go away.

Article on Sikhism on the Shahmukhi Wikipedia is titled 'Sikh Mat' (Sikh Religion). The title on Gurmukhi Wikipedia reads, 'Sikhi.' (Sikhism). They are too foolish to realise that doesn't make them two different languages.

When everything failed, I challenged them to come up with a plan, other than unification, to increase the article count by at least 20,000 in the next six months without stretching Wikipedia finances. They didn't respond. Instead they found it easier to attack me, and turn the discussion to other subjects.

I don't know who made them administrators. I know the software isn't perfect, then aren't we here to make it better? Anyway, I am done with it. Let them massacre each other, create ten Punjabi Wikipedias and do as they please.

In case you are interested in the discussion, here is the link.

Unified Punjabi Wikipedia - glimmer of hope

Unification of two Punjabi Wikipedias is still a dream. I have been trying to learn Python for the past few days so that I can create a Shahmukhi<->Gurmukhi conversion program. 

Yesterday, I began to doubt my way of doing things. So I asked on Lernu! what programming language would suit this task the most. There were some helpful answers and it turns out, solution lies in PHP and not Python. I have installed PHP but I haven't yet been able to figure out how to run it. I have to install a mini-server and I don't know how to go about it. I will find out, I know that. 

I phoned Sangam developer Gurpreet Singh Lehal today evening to ask if he will be willing to let Wikipedia use his conversion software. A soft-spoken person, he gladly agreed. This is a great help. It makes things a lot easier.

I have posted two messages announcing this good news to both Wikipedias. On the Indian side, they have expressed their concerns on the efficacy of software and edit wars that will follow on several controversial topics; India and Pakistan are not the friendliest of nations. 

I have yet to receive a message from the Pakistani side. It is presumably because I posted the message there almost two hours later.

Here is an extremely simple image of what Wikipedia will look like after the conversion is complete:

Largest online encyclopaedias

Wikipedia is the largest encyclopaedia ever written. There are more than 22 million (23 July, 2012) articles on it in more than 100 hundred languages. With a little more than four million articles, the English encylopaedia is the largest. Next comes the German Wikipedia with its approximately one and a half million (1.44 million) articles. The French version will soon cross the 1.3 million mark. Currently these are the only Wikipedias (Dutch too, with its more than 100,000 bot generated articles)  to have more than a million articles. The Italian, Spanish and Polish versions are set to join the club before the end of this year.

Compared to this, the Chinese Wikipedia's performance is dismal. Despite having more than a billion speakers, the article count is barely above half-a-million. Even Japanese (with much less speakers) boasts of close to 900,000 articles.

So are the Chinese (just like the Arabs and Indians) disinterested, or are we missing something?

It turns out, they are just as excited about voluntarily contributing online as anyone else in the world. The Chinese Wikipedia may not have an article count comparable to English, German or French, but then it's not the only Chinese collaborative encyclopaedia online.

Baike Baidu (百科百度) and Hudong (互動) are two such giants. There is also Soso (搜搜). Of these, with its 6.4 million (1 July, 2012) articles, Hudong is the largest. Then comes Baike (5 million). It's followed by Soso (900,000).

I got the numbers from this table. This is an article on Baike Baidu on the Chinese Wikipedia.

Here is a screenshot: 

And here are the numbers:

Translation help:

First row from second column onward:

維碁百科：Wikipedia

中文維碁百科：Chinese Wikipedia

百度百科：Baike Baidu

互動百科：Hudong

搜搜百科：Soso

First column, third row from top:

條目數：Article count

So, it turns out the largest online encylopaedia is Hudong. The second largest is Baike Baidu, followed by the English, German and French Wikipedias.

Ugly English on Wikipedia

I read George Orwell's Politics and the English Language a long time ago. I didn't really understand it back then. In fact I couldn't even figure out what he was trying to say in the essay. Then I came across Robert Fisk of The Independent. And it was through his articles syndicated in The Tribune and lectures and speeches on YouTube that I understood what Orwell really meant.

If Orwell found written English of his time "ugly", I am sure he would cry 'horrendous' on seeing the English used on Wikipedia.

Look at this sentence: 

Jew also known as the Jewish people, are a nation and an ethnoreligious group... (Wikipedia article on Jews)

and compare it with this one: 

Jew, Hebrew Yĕhūdhī, or Yehudi,  any person whose religion is Judaism. (Encyclopaedia Britannica on Jews)

Now, I have no idea what does it mean when one says, "XYZ people are a nation!" 

Anyway, this preposterous contortion of language isn't limited to an article or two. It's more or less prevalent, and especially visible in articles on politics.

What worries me more, however, is that this prejudice is carried over to other language Wikipedias through translation. This article on the Russian Revolution on the Chinese Wikipedia doesn't cite a single Chinese reference.
This is rather strange. I don't speak Chinese but my hunch is that they simply translated it from English.

Ask expatriates to contribute to Indian language Wikipedias

An Indian esperantist Giridhar Rao reports (in Esperanto) a suggestion that came out during the visit of a Wikipedia Foundation member to Delhi for a meeting. There the host(?) suggested asking expatriate Indians to contribute to Wikipedia in Indian languages other than English. The person who suggested this thinks they will be suitable to ameliorate local language Wikipedias because of their "nostalgia" and the diligence they show learning their "mother tongue."

Those attending the meeting also discussed the problems they have to face in finding young Indians who can "read and write fluently in Indian languages." Looking for reasons they appear to come to the conclusion that the use of English as a medium of instruction is at the root of the problem. This, a Tamil Wikipedia contributor said, creates a situation in which students are "neither good at expressing themselves in the local language nor in English."

He is right. This is indeed the situation in India. Most youngsters can't even recite the alphabet of their local language in order, their grip on the script is weak, they increasingly always write in Roman characters and invariably all of them are "fluent" in English but have some "hesitation" expressing themselves (= have poor English).

The problem is there but the solution is not to go to the NRIs crying for HELP! There has to be a right diagnosis if we are to weed out the problem. They should understand that:

1. An average Indian doesn't speak even one language fluently is an indication of the dismal state of our antiquated, inefficient and ineffective education system. It is not because of teaching local languages as a subject and using English as the medium of instruction. We can learn from Singapore. And, 
2. Spending scare resources on attracting NRIs at the expense of raising awareness at home isn't going to be a productive strategy.

In the next post I will try to analyse why Indians don't contribute to the Wikipedia, and perhaps propose some "indigenous" solutions! 

Dutch Wikipedia has a million articles now, at least one in ten is computer generated

The Dutch Wikipedia recently crossed the one million mark. It's the fourth edition after English, German and French Wikipedias to pass that number. But despite this it's has been removed from the ten largest Wikipedias that show up on the homepage.

In its place there is now the Chinese Wikipedia and this I believe has happened because the Dutch used bots (automated computer programmes) to create more than 100,000 articles in the spate of only ten days. If I remember correctly that happened between 7 December, 2011 and 17 December, 2011. Unfortunately, I don't remember where I read that neither can I find more information on it.

I find it hard to believe that these computer generated articles add anything of value to an encyclopaedia. Personally speaking, I would like it to cover only 100,000 articles in depth than 1,000,000 superficially. But again, showing off is perhaps a part of the contemporary culture.

 ڈچّ ویکیپیڈیا پر 1,000,000 مضمون، 10% سے زیادہ کمپیوٹروں نے لکھے ہیں

ڈچّ ویکیپیڈیا میں پچھلے دنوں میں 1,000,000 مضمون (آرٹِیکلس) ہو گئے۔ اس سے ڈچّ ویکیپیڈیا انگریزی، جرمن اور فرانسیسی ویکیپیڈیوں کے بعد 1,000,000 کا مقام حاصل کرنے والا چوتھا ویکیپیڈیا بن گیا ہے۔ لیکن اس کے باوجود بھی ڈچّ ویکیپیڈیا اب دس بڑے ویکیپیڈیوں میں شمار نہی ہے۔ اس کو اس لِسٹْ سے بیدخل کر دیا گیا ہے۔

اس کے پیچھے ایک چھوٹی سی وجہ ہے: ڈچّ لوگوں نے مضمونوں کی گنتی بڑھانے کے لئے بوٹس یانی کمپیوٹر پروگراموں کی مدد لی تھی۔ اور ایے سب کرنا اگر ویکیپیڈیا کے قانونوں کے مطابق حرام نہی ہے تو اسے حلال بھی نہی کہا جا سکتا۔ صرف 10 دنوں کے اندر اندر 100,000 سے زیادہ مضمون ڈچّ ویکیپیڈیا پر لکھے گئے اور ایے سب 7 دسمبر، 2011 سے 17 دسمبر، 2011 کے درمیان ہوا۔

مجھے نہی لگتا کہ کمپیوٹر کے ضریئے لکھے گئے مضمونوں میں بہت کچھ پڑھنے کے قابل ہوگا۔ 100,000 مضمونوں کو عمدہ طریقے سے لکھنا کئی گنا بہتر ہے 1,000,000 مضمونوں کو خراب طریقے سے لکھنے کے۔

Is Punjabi going the Hindustani way?

Summario in Interlingua

Panjabi es le lingua del populo del region Panjab que es dividite inter India e Pakistan. Plus que 100 million personas parla le lingua ma le situation currente del linguage non es ben. Il ha duo graphias e parlatores de duo Panjabes non comprende le graphias de le un le altere. Anque il ha un problema de excessive persianisation o sanskritisation. Si le curso non cambia le lingua serea devenir duo separate linguas, como occurreva al hindustani ante 100 annos (hindi e urdu). Ancora religion joca un importante parte. Malgrado le facto que plus que un medio de parlatores es islamic, le lingua es frequentemente ligate a sikhes.

Resumo en Esperanto

La panĝaba estas la lingvo de l'lonĝantoj de l'regiono Panĝabio,  kiu dividiĝas inter Barato kaj Pakistano. Pli ol 100 miliono personoj parolas la lingvon sed la nuntempa situacio de l'lingvo ne estas bona. Ekzistas du skribsistemoj kaj parolantoj de du Panĝaboj ne komprenas la skribsistemojn de l'unu la alian. Ankaŭ estas problemo de uzi neogolismojn el aŭ la persa aŭ la sanskrita. Se la tendenco ne ŝanĝas, la lingvo devenos du apartaj lingvoj, kiel okazis al la hindustana antaŭ 100 jaroj (la hinda kaj la urdua). Ankaŭ religio ludas gravan rolon. Malgraŭ la fakto ke pli ol duono da parolantoj estas islamanoj, oni ofte rilatas la lingvon al siĥoj.

=========================================

A hundred years ago Hindustani was the lingua franca in South Asia. Then surfaced religious tensions and the language was spilt into two. Now depending on which side of the border you are Hindustani is now Hindi or Urdu. A hundred years later, history seems to be repeating itself with Punjabi.

Of the 100 million or more Punjabi speakers, more than 70% live in Pakistan and for most of the remaining India is their home. A Punjabi from Eastern Punjab (India) is more likely to communicate well with a fellow Punjabi from Western Punjab (Pakistan) if he/she has lived most of his life in his/her village and has very little formal education than a "cultured" Punjabi. 

Here a "cultured" Punjabi is generally understood in the cities as someone who doesn't know his language well and talks in Hindi/Urdu or English much of the time. It is these people who have the most difficulty in talking with fellow Punjabis from the across the border in Punjabi.

This is an interesting phenomenon. Because in contrast to European languages, where much of the learned vocabulary is more or less similar even though the colloquial speech may have nothing in common with each other, we fight over the issue of learned vocabulary even though the colloquial speech is virtually indistinguishable.

This is as much true of Punjabi (Western and Eastern) and it is of Hindustani (Hindi and Urdu). Unfortunately I don't speak other Indian languages to comment on what is their situation.

In case of Punjabi the list of issues facing the language is long, but chief among them, from linguistic point of view, are:

1. Use of two Different Scripts

The use of two different scripts to write Punjabi isn't unique to the language and the solution isn't to discard one script in favour of another or adopt a Latin characters based script as I had suggested in an earlier post. We can learn from the Serbians and teach people both scripts on the both sides of the border. If the Japanese can handle three different scripts simultaneously we shouldn't have much of a problem especially when the task of learning Shahmukhi (37 letters) and Gurumukhi (41 letters) is nothing compared to the astronomical amount of effort required to learn Kanji (>2,000 characters).

2. The Issue of Excessive Persianisation or Sanskritisation

When I was a child I used to write a 'student' sṭudenṭ in Punjabi. Now nowadays vidyārthī (Sanskrit/Hindi influence) is more popular on this side of the border and if I'm not wrong the Pakistani Punjabis use tālib-e-ilm (Persian/Urdu influence). I find vidyārthī ugly and tālib-e-ilm is an unnecessary inconvenience. We have borrowed words from English in the past and have successfully indigenised them. No one considers botal (bottle) and haspatāl (hosptial) as foreign. The upshot is when it comes to learned vocabulary we can either get the Western scientific vocabulary adapted or can create new words from existing roots. There is no harm is borrowing from Sanskrit or Persian but the process should be natural.

3. Linking the Language with Religion

Up until the 1930s all Punjabis spoke their tongue. Then a man called Master Tara Singh started promoting Punjabi as the language of Sikhs. This exacerbated the already heightened religious sentiments. As a results Hindus increasingly started identifying themselves with Hindi and Muslims with Urdu. It's just like Bosnians insisting their language is not Serbian. How this tendency can be overcome, I have no idea. But the language will certainly get an impetus if we stopped viewing Punjabi as the language of Sikhs. We must understand they haven't got a copyright!

If the fate of Hindustani befalls Punjabi will depend on factors - politics and religion - that aren't mentioned here. But one thing is for sure: accept the laissez faire and we will have two mutually incomprehensible Punjabies in the coming decades.

Interlingua Wikipedia - Unethically raising the Article Count?

There is something queer happening on the Interlingua Wikipedia. I remember until the beginning of this month, there were about 5,500 (give or take a hundred) articles on the Interlingua version of the Wikipedia. The number today, only three weeks later, stands slightly above 9,100. Roughly speaking, this is an increase of 3,500 articles in less than three weeks. This sounds impressive. Perhaps there is a sudden interest in the language. But wait a minute. 

When you look at the number of active users, who have made at least one edit in the past 30 days, it is only 65 according to the data on the Statisticas page (Esperanto Wikipedia has 468, for comparison) and 50 of them are automated computer programmes or bots. It means either the remaining 15 people wrote all of the 3,500 or so articles (approximately 233 articles per head) or the lovers of the language have turned to playing the same game Volapukists played a couple of years ago. Back then a series of automated computer programmes had created articles on the Volapuk Wikpedia at a breathtaking speed. It now has over 100,000 articles. Of these two possibilities, the second one sounds more plausible.

Almost all the articles created in the past few weeks are one or two line articles and most of them are about cities, towns and villages in Spain and Vietnam. It is unlikely that 15 people suddenly developed an interest for geography and none of them wrote more than one or two sentences.

I believe these bot created articles will add little, if any, value to the Interlingua Wikipedia. The move could indeed backfire as new students may get discouraged on finding these stubs most of the time they look for something on the Interlingua Wikipedia. The editors, or whoever is behind this, should better concentrate on creating a few but quality articles, like these ones. (I wrote them!) ;)

Here is a screen shot of the Statisticas page on 21 September, 2011:

Two Childish Questions about China

Why are there only 340, 230 articles on the Chinese Wikipedia and merely 37, 000 on the Hindi Wikipedia when, in contrast, the English Wikipedia boasts of over 3.5 million articles?

The primary reason for why there are not to be found even half a million articles in a language that is spoken by almost a fifth of world population is simple - the Chinese don't use Wikipedia.  Instead, they turn to two home grown encyclopedias - Hudong and Baike Baidu - for reference. Hudong has over four million(1) articles and Baike Baidu crossed the 2.8 million(2) mark in 2010.

To have a glimpse of these Chinese encyclopedias, compare the articles on Manmohan Singh, the current Indian Prime-minister, on Wikipedia, Hudong and Baike Baidu.

As far as the Hindi Wikipedia is concerned, there are two main reasons for the dearth of articles: 

a. There are more illiterates in India than any other country(3) and 

b. those who are lucky enough to have completed their education contribute mainly to the English Wikipedia(4)

If the Chinese government has banned Google, how  do they look for information on the internet?

Google was never as popular in China as it is in India; the proportion of population in China which used Google was only as significant as the number of people who use Yahoo! or Bing in this country.

Also, as far as I know, there are five countries, including China, where Google is not the most used search engine. They are: China, the Czech Republic, Japan, Russia and South Korea. Except for Japan, the rest of the four have developed their own search engines; Japan uses the Japanese language version of Yahoo! called Yahoo! Japan.

The Chinese love Baidu, the Czechs have developed an affection for Seznam, nostalgic Russians' response to American Google is Yandex and the South Koreans don't let a day pass without using their Naver.

Moreover, these are not small players by any standard. Baidu is the sixth most popular website(5) in the world and Yandex's Alexa Traffic Rank is 24. The tech loving Japanese use Yahoo! Japan so much that Alexa has ranked it the 12th most visited website globally.

References:

(1) = Wikipedia Article on Hudong
(2) = Wikipedia Article on Baike Baidu
(3) = A The Indian Express Report on Illiteracy
(4) = Aapka Apna Wikipedia; Eye, The Sunday Express
(5) = Alexa Traffic Rank - Baidu