Wednesday, 17 January 2018

Computer Science Internationalization - Internationalization Workshop

This morning I gave a Computer Science Internationalization to school students. Twelve students attended my workshop which lasted just over two hours. As is my usual practice for such workshops, I made a list of topics to cover and more topics than could be covered in the time available. I then adapt as necessary. I placed emphasis on incorporating world cultures into one's thinking and programming practices. This was a participatory workshop so the students were fully involved.

I started the workshop by saying: Bon Matin, Guten Morgen, 안녕하세요, おはよございます, 早上好. So, right from the start, it is evident that this workshop is international in nature. I explained that one does not need to know multiple languages in order to build software for the world but one does need to understand characteristics of human language scripts. Chinese is much more compact than English which gives a big advantage when using microblog services which have a character limit, twitter now has a 280 character limit. Chinese and Japanese do not use spaces to separate characters and so detecting word boundaries is much harder than English which uses spaces and punctuations characters to separate words.

Topics covered in this workshop included:

  1. McDonalds I explained that common practice with global companies is to have a global gateway which has a set of links to their localised websites. Starting at UK Mcdonalds I asked the students to, without googling, find the McDonalds global gateway. It used to be possible but is no longer. I am not sure why McDonalds removed the link. Next I asked the students to find the global gateway but this time they can google. They soon found it. I explained: the way I found it was by using Opera with VPN set to America, visiting McDonalds USA and on their page was a link to the global gateway corporate.mcdonalds.com/mcd/country/map.html. I then asked the students to explore the different localisations and look for differences in foods and styling to the UK website. Next was the Asia Pacific section of the global gateway. I told the students there is a problem and then asked them what the problem is. The problem is that for China the icon displayed is Youtube but the link is actually for the Chinese video service Youku i.youku.com/u/UMzg1MDY4MTg0. I explained: China blocks access to many sites and services, youtube and twitter are blocked in China.
  2. Writing names correctly: Next was how to write my name correctly ie André and not Andre. First I asked the students to write André. They discovered that by holding down the e key down a popup appears showing e combined with various diacritics. Next, using the keyboard viewer, I showed the general procedure: hold down the alt key to see all the diacritics supported by that keyboard mapping, release the alt key to see all the letter/diacritic combinations supported by the keyboard mapping. We used the ABC - Extended keyboard mapping.
  3. Writing Chinese: Next was the basics of writing Chinese using the pinyin Simplified Chinese Input Method. I tasked the students with writing: xiao shan (小山), dian you (电邮) and zai xian (在线). Some of you will have already worked out which way this is going😀 Put these Chinese characters together and we have my Chinese email address 小山@电邮.在线. I explained: "We are on the threshold of a huge growth in registration and usage of internationalised email addresses. Rajasthan recently launched a Hindi email service whereby residents can have a free hindi email address. There is a group dedicated to promoting usage of internationalised domain names and internationalised email addresses by the name of UASG (Universal Acceptance Steering Group)." I then tasked the students to read and browse uasg.tech.
  4. Character counting: Next was character counting. I had 2 text strings: ① one (two)! [three] four five. ② pear(plum)grape!apple. I asked: from a computer's point of view, how many characters are there in these text strings. Reader: I will let you count for yourselves😀 I explained: "the brackets are full width forms and so what might be a space character followed by an opening bracket is in actual fact is just a single character. Such characters are used in Chinese and Japanese as every character, including punctuation, has the same width."
  5. Numbered webpage lists: The next task was a numbered list. I had a pre-prepared html template which the students used. We started with a standard decimal ordered list. The I showed the students the inline CSS — <ol style="list-style-type:thai;">. I then tasked the students: "Without googling, guess at language names and try them with the list-style-type." Next task was: "You can now google. Find the valid list-style-type language names." Several quickly found developer.mozilla.org/en-US/docs/Web/CSS/list-style-type which is a page I frequently use. I also showed codepoints.net/search?gc=Nd to show how many different numbering systems there are which will all, hopefully, be supported at some time in the future.
  6. Rotating Unicode Characters: This programming challenge was, using CSS, change the Unicode character 👀   from right looking to left looking. I allowed googling right from the start. I did not use the word "rotate" or rotating when giving this challenge. Yesterday I wrote a blog article for this challenge. I showed this blog article briefly towards the end of the workshop   👀 schappo.blogspot.co.uk/2018/01/computer-science-internationalization_16.html. In my blog article I demonstrate a cultural application.
  7. Language Adaptive Web App: In my introductory web programming module, I write demonstrator code for lectures which I then make the source available to the students to use or not use, as they wish. The web app I used in this workshop was one that is language adaptive, English and Chinese. I firstly showed the school students how to change the "preferred language for webpages" which is set by users in preferences. I used Firefox for this demo. I firstly showed my Web App with my Firefox set to preferred language = English. So the text in my buttons was in English. I then changed the preferred language to Chinese. Refresh the page and my button text is now in Chinese. I emphasised that this does not happen automatically and it does require programming to make it happen and again with most everything else I did in this workshop it is simple to do. I showed my source code. The crucial statement is: if(/zh|zh-CN|zh-TW|zh-SG|zh-HK|cmn/i.test(navigator.languages[0])){.... One thing I forgot to do was show my CJK version of my web app in which I have changed every identifier to Chinese, Japanese or Korean😀
  8. Unicode: I explained: "There are 136000+ characters in Unicode. In ASCII there are just 128 characters. Each Unicode character has a unique codepoint and when represented is usually prefixed with U+. The Unicode consortium guarantee that once a character is officially included into Unicode, it's codepoint will never change. The Unicode character set is continually being developed and a new version is released every summer. Anyone can submit a proposal for character(s) to be included into the Unicode character set. There is a Unicode Consortium group which deals specifically with Emoji proposals as Emoji are hugely popular." I then tasked the students to browse the Unicode character set using the "Emoji & Symbols" viewer.
  9. Regular Expressions (regex): We visited speakerdeck.com/andre_schappo/unicode-regular-expressions. I quickly went through some of the ASCII based slides. I stated: "If you search the net you will finds thousands upon thousands of ASCII based regex examples and explanations but what you will not see is...". I then proceeded to slide 11. I explained the cultural references and humour in slides 11,12 and 13. This is a good example of how world cultures can be incorporated into one's programming practices. 〖 I notice in slide 11 that i have inadvertently missed out the + character. The first regex should be: '^[一二三四五六七八九]+$'. I will fix sometime but not now.〗
  10. A programming challenge: I finished with a programming challenge ➜ jsfiddle.net/coas/wda45gLp. I briefly explained the code but did not have time to show my solutions to the challenge. In 2 or 3 weeks time I will put my solutions in my jsfiddle.

During the workshop I said: "One problem is that there is no culture of programming internationalisation in school, college and university Computer Science departments. Therefore, staff and students do not even think to ask themselves questions such as: "I wonder if it is possible to number html list with numbers from other languages?". If they did ask themselves such a question then a quick google and they would soon discover that it possible and it is so easy to do.

I did not have much opportunity to chat with the students but one student told me: "In his school, they are only taught ASCII programming and ASCII text processing. He learned about Unicode by his own efforts and outside of the classroom." This is typical of schools, colleges and Universities. Unicode is not being taught but it most definitely should be taught.

I concluded with: "I hope you will teach your classmates and teachers about Computer Science internationalisation. If your school would like me to visit and give a workshop or presentation, I am more than happy to do so. All I ask is that my visit expenses are paid by your school.

Readers of this article: I extend the same offer to you. If you would like me to give a workshop or presentation on Computer Science internationalisation at your company, organisation, school, university, I am happy to do so. All I ask is that my expenses are paid. Computer Science internationalisation is my specialism and my passion. I can talk for hours, days, weeks, months, years about Computer Science internationalisation😀 I am not hard to find on the internet but one way of contacting me is to tweet me at twitter.com/andreschappo

Attendees of this workshop: I hope you enjoyed this workshop. I enjoyed giving it. If there is anything I have forgotten that you would like to see included in this article, please email or tweet me. During the workshop there was not sufficient time to show you a youtube video in which Chris Broad explains one of the ways in which Japanese people celebrate Christmas. See ➜ youtube.com/watch?v=SFw-TZzqX8M. Can you spot the word play in the name of his youtube channel?

〖 Note: I wrote this article quickly whilst the workshop is fresh in my mind. In a week or so I will revisit this article and tidy it up. 〗

Tuesday, 16 January 2018

Computer Science Internationalization - Rotating Unicode Characters

I sometimes encounter Unicode characters that have direction and sometimes that direction is not the direction I want. Letʼs take the character 👀 . The eyes are looking to the right. Recently I wanted the eyes to look to the left. Using CSS we can change the direction so that we have left looking eyes  👀 . Here is the inline CSS I am using in this blog article to produce left looking eyes  👀 .

<span style="display:inline-block;transform:rotate(180deg);vertical-align:15%;">
👀
</span>

福 is considered an auspicious character in the Chinese culture. If you search the net google.co.uk/search?q=福&tbm=isch you will find many images of 福 and many of these images are of an upside down 福。The CSS I use to display is exactly the same as above.

Letʼs display a that has a bit more impact.

Here is the CSS I use to display the big red .

<span style="display:block;transform:rotate(180deg);text-align:center;
text-shadow: -4px -4px 4px Grey;text-weight:800;font-size:100pt;
font-family:'Hannotate SC',cursive;color:Red;">
福
</span>

Why ? It is all to do with homophones, words that sound the same(ish). Here is an explanation from en.wikipedia.org/wiki/Fu_(character)

When displayed as a Chinese ideograph, Fú (福) is often displayed upside-down on diagonal red squares. The reasoning is based on a wordplay: in nearly all varieties of Chinese: the words for "upside-down" (倒, Pinyin: dào) and "to arrive" (到, Pinyin: dào) are homophonous. Therefore, the phrase an "upside-down Fú (福)" sounds nearly identical to the phrase "Good luck arrives". Pasting the character upside-down on a door or doorpost thus translates into a wish for prosperity to descend upon a dwelling.

〖 At some future time I will experiment a bit more with vertical-align values. 〗

Sunday, 14 January 2018

Computer Science Internationalization - Email Address Internationalization (EAI)

An internationalised email address is of the form Unicode@IDN where Unicode is a name consisting of Unicode characters (excluding the ASCII subset of Unicode) and an IDN (Internationalised Domain Name) consists of Unicode characters (again excluding the ASCII subset of Unicode). There are hybrid email addresses, such as, ASCII@IDN. My focus for this blog article is fully internationalised email addresses eg a fully Hindi email address.

Currently, internationalised email addresses are somewhat of a rarity and few people know they even exist or that they can register such an email. I believe, at time of writing, that we are on the threshold of a significant growth in registration and usage of internationalised email addresses.

I consider it is time to raise awareness of internationalised email addresses and that is the reason for this blog article. I will publish internationalised email addresses. I will only publish those emails which are in the public domain and which are fully internationalised. I will publish non public domain, personal email addresses but only if I am given explicit permission by the holder to publish in this blog article. This will be an ongoing article and I will endeavour to update on a regular basis.

I am aiming for some ten emails per language. When I reach my quota I will endeavour to change some of the emails so that if you revisit you will see a different set of emails.

If you do come across any fully internationalised email addresses in the public domain, please do let me know. You can tweet me at twitter.com/andreschappo.

  1. Chinese 中文
    1. 小山@电邮.在线 — my Chinese email address. 小山 is my adopted Chinese name. There is a story behind this name ➜ schappo.blogspot.co.uk/2012/07/my-adopted-chinese-name.html
    2. 阿賈伊@电邮.在线 — Ajay Data अजय डाटा, CEO of Data Xgen Technologies. 阿賈伊 is a transliteration of अजय and written in hanyu pinyin is a jia yi.
  2. Hindi हिन्दी
    1. वसुंधरा@राजस्थान.भारत — Vasundhara Raje वसुंधरा राजे - At time of writing, the 13th Chief Minister of Rajasthan, India. Here is a video, from December 2017, of the official launch of the Hindi email service and of this Hindi email address ➜ youtube.com/watch?v=XmUW4n9UmFk
    2. अजय@डाटा.भारत — Ajay Data अजय डाटा, CEO of Data Xgen Technologies.
  3. Russian Русский

Friday, 12 January 2018

Copy and Paste — A Tip

My standard practice for mouse based copy & paste is:

  1. move the mouse to the text I want to copy
  2. mouse down and drag to select the target text
  3. release the mouse
  4. press the key combination ⌘C to copy the text
  5. paste to destination

I have been using this method for many many years and so have billions of other people. My very first use of a mouse was on an early Apple Mac computer so I have been using a computer mouse since around 1985.

There is though a problem with this method which I usually encounter when selecting text on a webpage. Sometimes this copy & paste method can result in the unintentional activation of a link or too much text or too little text being selected. It can be quite irritating, especially when one is in a hurry.

Several months ago I accidentally discovered a solution to this problem which works every time, or at least it has for me. The solution requires a small change to the copy & paste method. The new method is:

  1. move the mouse to the text I want to copy
  2. mouse down and drag to select the target text
  3. do NOT release the mouse ie keep your finger pressed on the mouse
  4. press the key combination ⌘C to copy the text
  5. paste to destination

On occasions, people have said that I am a slow learner. This certainly proves them right as it has taken me some 32 years to discover this copy & paste method 😀

Environment: OSX High Sierra 10.13.2

Wednesday, 10 January 2018

Computer Science Internationalization - A Programming Challenge

I recently issued a simple (or is it?) programming challenge to 140+ university students. I will be giving this same challenge to school students at a participatory internationalisation workshop. The challenge is on jsfiddle. You too can try this challenge. You do not need to join jsfiddle and when you connect to jsfiddle you will be given your own copy of my code. You can thus modify the code as you wish and no changes will be made to my master version. You can also run your code on jsfiddle.

Here is the challenge. Either click on this link ➜ jsfiddle.net/coas/wda45gLp or click "Edit in JSFiddle" in the top right of the following window. Once connected to jsfiddle you can execute your code by clicking Run in the top left corner.

There are several solutions to this challenge. I will put at least two solutions in this blog article but not until after my internationalisation workshop. If you would like to share your solution(s) with me, tweet me at twitter.com/andreschappo.

Sunday, 7 January 2018

Computer Science Internationalization - Presentation of Links

All of us encounter links in documents, webpages and email. They are readily identified as they are usually coloured blue if not visited and red if visited. In the case of IDNs (Internationalised Domain Names) there are several ways of presenting links to users and I have encountered all of these variants. Letʼs look at Korea University's IDN 고려대학교.한국. This can be presented as www.고려대학교.한국, http://고려대학교.한국, http://www.고려대학교.한국, www.xn--299a9hr4mn4fgs6b.xn--3e0b707e, http://xn--299a9hr4mn4fgs6b.xn--3e0b707e, http://www.xn--299a9hr4mn4fgs6b.xn--3e0b707e. If you click on these links you will see that they all work. I do not like any of these ways of presenting links. xn--299a9hr4mn4fgs6b.xn--3e0b707e is the punycode form of the domain name and should never be presented to users. It is used for behind the scenes communication between internet devices. (I still prefer the name punnycode 😁 )

I have used different forms over the years and I do consider there is a best way of presenting links and I have done this on many occasions. But I have mostly done it by way of experimentation and I have not done it consistently. As of yesterday, I have decided to have a consistent working practice for presenting both IDN and ASCII links. My personal rules for presenting links are:-

  1. I will not use the www or http(s) prefix and will most definitely not use the punycode form. The link for Korea University now becomes ➜ 고려대학교.한국. We now have a simple and elegant presentation of the link. Note also that it is a single human language script which in this case is Korean Hangeul. Therefore when one is typing this link there is no need to switch between English and Korean on oneʼs device. This, to me, is the most important part of presenting links as a single human language script.
  2. Presentation of email addresses is well established and presented email address links are not prefixed with the "mailto://" scheme name. We can do the same thing with internationalised email addresses. The link for my Chinese email address is 小山@电邮.在线?Subject=你好小山😜. We can easily distinguish between website links and email links because an email link has the @ symbol. If your email client works correctly, then, when you click my Chinese email link the To: field should be filled in with 小山@电邮.在线 and the Subject: field with 你好小山😜
  3. I will always show the real address in the link eg ko.wikipedia.org/wiki/고려대학교. I will never use anything of the form " … please click here for further information. ". I consider this to be extremely bad security. I abandoned this practice many years ago. How many of you hover over a link to determine the real address before you click the link? I do sometimes, but mostly I do not. With my links there are no surprises, what you see is what you will get. The one thing I have no control over is redirection. A website can, at anytime, redirect to a different web address. Such redirection does sometimes happen though not very often and usually it is for legitimate reasons such as redirection to a new version of a website.
  4. When the url is extremely long which I cannot reasonably fit into say a presentation slide, I will use ellipsis to indicate this is not the complete address eg once.upon/a/time/there/was/a/beautiful/…
  5. So far we have only considered the two most common schemes, http(s) and mailto. There are many other schemes, such as smb, sftp and imap. There is a list of the registered schemes at iana.org/assignments/uri-schemes/uri-schemes.xhtml. In cases like this I will include the scheme prefix in my link so that it can be easily distinguished from the aforementioned type of links eg sftp://some.fileserver.somewhere/freestuff/user-manual.txt"

Some systems and apps will break my working practice as they will not allow me to present my links as I consider they should be presented. I will endeavour to find work arounds for such systems. First on my list is twitter.

Techie Tip: When I was setting up my Chinese email address in my email signature, my email client insisted on decoding the domain name 电邮.在线 to the punycode form xn--wny099c.xn--3ds443g. When something like this happens I add an extra level of encoding so that the system decodes to the level I want not the level the system wants. This sometimes works and sometimes not. In this case it worked. I added percent encoding. I percent encoded to %E5%B0%8F%E5%B1%B1@%E7%94%B5%E9%82%AE.%E5%9C%A8%E7%BA%BF and then my email client decoded to 小山@电邮.在线 which is precisely what I wanted. A really useful web app for doing such conversions is Richard Ishidaʼs Unicode code converter r12a.github.io/apps/conversion

Friday, 5 January 2018

Computer Science Internationalization - Composed v Decomposed Text

My given name André, has the diacritic, acute accent over the letter e. In Unicode there are two ways of constructing and storing é. Either as a single composed character U+00E9 LATIN SMALL LETTER E WITH ACUTE or decomposed as the two characters e U+0065 LATIN SMALL LETTER E and ´ U+0301 COMBINING ACUTE ACCENT. Processes can and do convert between composed and decomposed forms. How can we know which of these forms is being stored?

Letʼs take the case of pasting into a browser web form. I will use André as my test data and I will make use of Richard Ishidaʼs Uniview r12a.github.io/uniview. To use Uniview, paste text into the largish white textbox and then click the white down arrow . You will then see information about all the characters in the textbox. Pasting André into Uniview gives the following results:-

  1. Chrome — é is decomposed
  2. Firefox — é is decomposed
  3. Safari — é is composed

The text André in this blog article is in decomposed form, except where I indicate otherwise. In my rather limited tests, Chrome and Firefox do not convert the text, so if the text starts as decomposed it arrives in Uniview as decomposed. Safari, on the other hand converts decomposed text to composed.

Letʼs now go back a step. The copy operation involves copying text to the clipboard and the paste operation takes text from the clipboard. Can we determine which form is in the clipboard? Yes we can and here is one way of doing it.

We are now going to use the terminal app. Typing the command  pbpaste|hexdump -C  will show the contents of the clipboard at byte level. Copying André and running the command  pbpaste|hexdump -C  will give  41 6e 64 72 65 cc 81 . This is André displayed in Unicode UTF-8 encoding, where  41 = A; 6e = n; 64 = d; 72 = r; 65 = e; cc 81 = combining acute accent. If we copy the composed form of André (⬅︎ composed ⬅︎ André) we get  41 6e 64 72 c3 a9  where  c3 a9 = composed é.

Conclusion: Itʼs complicated! Not being aware of these different forms of text can lead to, very difficult to find, bugs in code. Given different combinations of apps, versions and processes the results may well be different. One certainty though, is that one needs to have a good understanding of Unicode.

〖 In a future blog article I will give a programming example where it is essential that one knows whether text is composed or decomposed 〗

Environment: OSX High Sierra 10.13.2, Chrome 63.0.3239.84, Firefox 57.0.4, Safari 11.0.2