Internationalizing computer science in Africa

Last year I posted on whether Unicode and internationalization (i18n) is included in any computer science curriculum in Africa. A recent comment to that post by Andre Schappo asking whether there are any organizations in Africa promoting internationalization of university curricula more generally offers another angle to approach this issue.

Part of Unicode charts for Ethiopic/Ge'ez

Andre's question follows a post on his blog about two organizations that promote internationalization of teaching curricula, one in the UK and the other in Australia. Depending on how one defines promotion of internationalization in higher education, one might add many other initiatives and consortia which seek in one way or another to develop and support international or global studies. The degree to which such efforts overlap with or might impact the content of computer science courses is an interesting question. In my limited experience, international/global studies mainly addresses disciplines in other areas (social sciences, humanities, certain applied disciplines). It certainly is worth asking how a program of internationalization at a university would apply to computer science and see how the discussion goes.

However, in the case of Africa - and also Asia - internationalization of the computer science curriculum would seem to follow as much from attention to localization as to international and global perspectives.

In any event, this issue of how Unicode and i18n figure in computer science instruction - worldwide as well as in Africa - is one that is important for technical and language planning reasons as well as for the same reasons that motivate attention to internationalization in the higher education generally.

Thoughts on linking L10n and ICT4D in Africa

It is striking that in Africa - multilingual as it is - discussions of uses of information and communication technology (ICT) for development (ICT4D) generally pay little attention to how one might optimally adapt the technology and the content to national linguistic realities. There are explanations for this, including for example:

• An outward-looking focus on the internet as way to tap international markets and knowledge (contrasted with a relatively limited local market for internet content, although that is changing somewhat with increased access to smart mobile devices)
• The roles of English / French / Portuguese as official languages, and higher status of these Europhone languages relative to African languages
• Attitudes about language (the notion that African languages are not adapted to science or technology, the impression that there are too many African languages to deal with)
• Issues relating to the written form of languages (in some cases lacking or not standardized, or standard orthographies may not be widely taught; computer systems from abroad may not have fonts for the alphabets used to write the languages)

However, explicit attention to potential uses of the first languages and local lingua francas in ICTs could address such issues and lead to various benefits: some in areas on which traditional ICT and ICT4D policies focus (such as enhancing user skills in ICTs; increasing relevant web content); and some in other areas important to development that fall outside the usual ICT discussions (language development, links with indigenous knowledge, new kinds of creativity).

Discussions of localization (L10n) and multilingual computing in the context of ICTs for national development could be built around several themes, such as:

1. Dialogue between ICT policy and language policy processes. There is apparently little overlap between ICT policy and language policy in most countries of Africa, hence little institutional support for discussion of "localization policy" or, as it is called in some Asian countries, "local language computing policy." This structural divide can go all the way down to the level of computer technicians and linguists. Bringing together discussions of language policy and planning, and education policy as regards languages on the one hand, and ICT policy as concerns expansion of use of the technologies and their contribution to development on the other, could yield some new insights and foster collaborations among efforts to promote ICT4D and develop languages.
2. Approaching localization as a way of adding value to the impact and potential of ICT for national development. Localization is not about replacing one language with another, of course, but rather adding diverse language capacities and content to computer systems and the way we use them. ICT is in many ways inherently additive or positive sum - adding one more option in terms of language interface or content need not take away from another, but rather adds to the benefits potential users can get out of ICT, and what they can use the technology to accomplish. In a multilingual context, promoting ways of enhancing language capacities of computer systems (localizing software, developing tools to make that possible) and increasing diverse language content (through localizing content and creating local content) are relatively inexpensive ways of giving ICT more dimensions, (soft) access points, and meaning - hence value.
3. Linking with broadcast technologies. Broadcast media - the older forms of "ICTs" in effect - have been "localized" from the beginning, so the interface with technology that is not localized will be limited. Localization of web content about important topics in health and livelihoods, for example, could be used directly by community radio stations, without the need for locally (and potentially inaccurately) translating that information.
4. Leveraging the benefits of internationalization of ICT and of language technology for Africa. A lot of work has been going into making computer systems and software more able to handle diverse languages and scripts: Unicode's accommodation of all writing systems; tools to develop language resources (from keyboards to corpora); open-source opportunities for localization; research on advanced applications (such as computer assisted translation), etc. Multilingual Africa stands to benefit a lot from these advances (and contribute in turn to them), but it has not, on official levels, been systematically seeking either to take advantage of the opportunities or to place itself in a position to participate more fully in the future (e.g., training of computer technicians in aspects of language technologies). There are some initiatives on NGO, academic, or project levels, but official support has in most countries appears to be minimal. This is an area that ICT policy can address.

---

(Adapted from a message to the Afrik-IT list entitled "Re: Ghana and other open access network model and SAT3," 23 August 2007. This blog entry was originally posted on 30 September 2015 and inadvertently deleted.)

Unicode and the architecture of ICT

Latin omega, used in Kulango, and now
included in Unicode's latest version.

Unicode released its 8th version earlier this month, so maybe it's a good time to take stock of what the effort to encode all the world's scripts in a single system - which includes the Universal Coded Character Set (ISO/IEC 10646) - has meant for African languages.

The attention with a coding system for characters is naturally on the characters and scripts they are part of, and among the changes in Unicode 8 are some Latin-based characters used to write the Kulango language of Ivory Coast and the endangered Ik language of Uganda. These have been added to Unicode's "Latin Extended-D" block. (The Unicode standard also includes technical specifications for handling text.)

In past years there have been similar small additions to Unicode, as well as major additions of whole African scripts like Osmanya (2003), Tifinagh (2005), N'Ko (2006), Vai (2008), Bamum (2009), Mende Kikakui (2014), and Bassa Vah (2014).

The Ethiopic/Ge'ez script - used for several languages of Ethiopia and Eritrea, such as Amharic and Tigrinya - was first added in 1999, before the Unicode standard was widely adopted. At the time, different scripts used different coding systems, hindering use of languages with complex or non-Latin scripts, and posing particular problems for anyone wanting to combine scripts in a document or on a webpage.

Unicode as an enabling architecture

With that I wanted to quote from an unsigned blog posting entitled "What are we missing out on" (9 Dec. 2014) that was part of an American University course on international communication, as I think it offers an important perspective on why encoding scripts in Unicode is important:

"Tonight’s presentation on Ethiopic language and its inclusion in Unicode presented an important element about the global digital divide because it asks the question: how can, even with access to information communication technologies and internet access, someone utilize technology if it is not available in their native language? In short, they can’t. This is an important element to consider in regard to technology and the reality of, to borrow from Laura DeNardis’s description, its architecture. As the group described in their literature related to their case study, the architecture of something has the power to include and exclude. The analogy we have used in class in class is that of bridges that are built low enough to prevent busses driving under them.

"In the case of Ethiopia, technology was created in a way that excludes the nation’s 90 languages because American companies created the technology in English with a western cultural perspective. Further, their commercial interests drive their actions, and there is no financial incentive to include languages in which there is no commercial demand. Therefore, not only are these groups of people excluded from the benefits of technology, we are also denied the benefit of their knowledge. As the group noted, we “feel” like we are so much more connected, but cannot assume that the majority of the information is in English and unless everyone is able to put information in the digital realm, we are missing out. After this presentation, I cannot help but to believe that we are indeed missing out. Tonight we discussed Ethiopic, but what other languages are we missing out on?"

The "architecture" of information and communication technology (ICT) in this case starts with the internationalization piece ("i18n"), which is addressed in part by Unicode. It would also include availability or not of localized ("L10n") software, which is especially important for major languages, and content. Localization is also conditioned by factors such as education, policy (of national governments as well as of development organizations), and, as alluded to above, economics. As such, once one moves beyond the enabling architecture, the interplay of factors looks more like an "ecology."

Unicode, by adding scripts and amending characters, makes the written form of African languages theoretically accessible on modern computing devices and across the internet. That's huge progress beyond where things were a decade and a half ago, when various 8-bit encodings dominated (at which time I noted some issues of lack of support for African language scripts were stuck at practically the same place they had been ten years before!). But it's still only part of the process of fully addressing the linguistic dimensions of the digital divide in Africa. Until that happens, we all will be "missing out" in different ways.

N'Ko on the web: Review of experience with ebola FAQ

Here's a quick recap of the demonstration of (experiment with) posting the N'Ko text of a WHO FAQ on ebola. Shortly after I posted it, I wrote "There would seem to be no reason not to use the internet for dissemination of webpages and mobile content about ebola in N'Ko, ..." Then Charles Riley of Yale University and Athinkra LLC pointed out some problems. And I found others.

I stand by that estimation, but after some tedious work on formatting and font coding in HTML, do so with greater emphasis on the caveats: "... although there would need to be attention to testing of commonly used systems and of ways to feed or facilitate loading of fonts that include N'Ko."

A summary of findings and lessons learned follows, but first a quick recap of what I did. My first attempt was to copy-past N'Ko text from a PDF of the FAQ on ebola, which as previously mentioned, did not work (RTL text was switched to LTR; character order was mixed, and combining diacritics sometimes not combined). Working from the Word document original, in contrast, was deceptively easy since I was working in Firefox (FF) ver. 32 - copy-paste into the "Compose" screen on Blogger, fix some bullets and do some minor formatting and voilà!

Problems were identified in Microsoft Internet Explorer 8 and Chrome 37 (mostly white boxes but some strings of text), and FF (some alignment of combining diacritics used to mark tones). Varying font commands inherited from the Word source document caused the irregular display of characters, and were fixed (an alternative to try later would be an unformatted text paste). 

However, while this work facilitated a correct display of diacritics on FF 32.02 and MSIE 8 on a computer running Windows XP, the diacritics were all off the mark in FF 32.03 and MSIE 11 on a Windows 7 system (the latter did not have the DejaVu fonts installed, which might be the problem there).

Lessons learned so far:

• align="left" - Full justification of N'Ko text may space the combining diacritics as if they were characters, leading to misalignment.
• Font choice/availability - N'Ko text display seems especially sensitive to font commands. Do not have an answer to why formatted text from a Word document pasted into Blogger looked fine in FF but not in MSIE 8 or Chrome. Installation of a font with N'Ko support may help.
• Browsers may not be the main issue - Despite initially encountering display issues in MSIE 8 and Chrome more than FF, those could be corrected in the source code. Also, the fact that the most current FF (32.03) and MSIE (11) misplace all diacritics on a computer without a particular font (DejaVu) points back to the font issue. 
• But browsers are not not an issue - The N'Ko script normally shows liaisons between characters within words (somewhat like Arabic), but while these show on FF, they did not on MSIE
• No bold or italics - Bold N'Ko is apparently not supported by the DejaVu font. Italics are, but while those did display on FF 32, they would not on MSIE 8.

Next steps

This has been a learning experience, but it should be pointed out that with sites like Kanjamadi, N'Ko on the internet is a reality and a potential to be pursued, even as there are issues. (Kanjamadi displays impressively on the same FF32 on XP that did great with the WHO ebola FAQ, but cannot load in MSIE 8 and has the same diacritic problems on MSIE 11 on the Windows 7 machine mentioned above).

Before posting more in N'Ko, it would be helpful to have more feedback on display issues for this script in function of available fonts, browsers, and operating systems.

Next steps should have as their goal a simple how-to for organizations wanting to display text in N'Ko on the web - whether for ebola education or any other useful purpose. Similar localization guides could be developed for other West African languages as well.

So, what about the last year?

Busy. Some work highlights:

• Vastly expanded the Pan African Localisation wiki (and that's still in progress). The idea with this is to develop a comprehensive information resource for people of a range of backgrounds approaching localization in African languages. First of all, localizers - people working on or planning a project for some aspect of localization in one or more African langauges. Also policymakers (ICT, language, and increasingly localization), development program planners, local activists, etc. It involves bringing a lot of information together (in many cases just basic with outlinks) and interlinking it. Localization to a certain degree involves bringing together of "previously unrelated skills, or matrices, of thought" (per the well known formulation of Arthur Koestler in another context some years ago) so this wiki is combining information on African languages and sociolinguistics, along with language technology tools, ICT4D information, mention of policies etc. In fact, all these "skils or matrices of thought" can also be understood as interacting parts of a larger "localization ecology," which is a concept I'm working on.


• Presented at and participated in "Regional Consultation on Local Language Computing Policy in Developing Asia" and "PAN Localization Project Phase II Meeting"meetings of the PAN Localisation Project, held in Thimphu, Bhutan in January. The PAN L10n project is funded by IDRC and has been particularly successful in facilitating localization work in several countries: Afghanistan, Bangladesh, Bhutan, Cambodia, Laos, Mongolia, Nepal, Pakistan, and Sri Lanka. The discussion of "local language computing policy" involved representatives from several other countries in the region.


• Attended a one-day seminar on "Recent Experiences on Measuring Languages in Cyberspace" at UNESCO in Paris. This was on 22 February, the second day of a two day observance of "International Mother Language Day."


• Helped organize and participated in the "Pan African Research on L10N Workshop & Localization Blitz" held in Marrakech, Morocco in February 2007. This is another piece in IDRC's strategy to develop a network for localization in Africa. The workshop was organized by the Tactical Technology Collective - same organization that put together Africa Source I (Okahandja, Namibia, 2004) & II (Kalangala, Uganda, 2006) and the Localisation Dev workshop (Warsaw, Poland, 2004). Many of the participants also had been at the PanAfrican Localisation Workshop in Casablanca, Morocco (2005); three others and I came from the seminar in Paris mentioned above.


• Attended the 38th ACAL 2007 & 11th ALTA 2007 Conference held at the University of Florida in Gainesville, Florida, US, in March 2007. This was an opportunity to connect with various people involved in research on and teaching of African languages. It has been my thought for a while that linguists and language instructors can be involved more with localization efforts.


• Submitted extensive comments on the Africa section of the "International Assessment of Agricultural Science and Technology for Development" in May. More about this in a later post.


• Taught a distance course on "Language and Development Communication for the Payson School of Tulane University. This was an interesting challenge and opportunity to learn as well as teach. I hope I have the opportunity to give this course again.


• Continued to interact online with various people working on aspects of localization and internationalization. This is not really news though - have been doing this in one form or another for several years.