Our modern Babel?

I wonder what do people think about the potential repercussions of the introduction of IDNs, particularly in terms of fragmentation of the Internet.  In this post I provide some background about the languages on the web, some of my thoughts, and finally questions for which I would love to hear your thoughts.

After many years of debates, International Domain Names (IDNs) have finally become more tangible with the announcement of the Fast Track by ICANN earlier this year.  Right now it is open only to states and territories recognized in the ISO 3166-1 regulation.  A number of countries have already applied for registering their Internet country suffixes in their local languages (IDN ccTLDs).  For example, Egypt announced that they are going to register “.مصر”, which stands for Egypt in Arabic, and Russia started the registration process for “.рф,” which stands for Russian Federation.

Overall, introduction of the IDNs has been met with a lot of enthusiasm.  In the last ICANN meeting in Seol and at the last IGF this was celebrated as the final internationalization of the Internet.  The minister of communication of Egypt was quoted saying that the “Internet now speaks Arabic” and the European Union has also declared that they are going to allow registration of .eu in all 23 official languages of the Union.  People are celebrating the diversity.

At the same time, as expected, not everybody is excited about this development.  It is widely held that the primary opposition to IDNs has been voiced by the trademarks holders.  After sort of figuring out how to protect their trademarks in the current, Roman script dominated, cyberspace (such as the Uniform Domain-Name Dispute-Resolution Policy), they are not particularly psyched about the need to do it again in dozens of other languages and potentially under dozens of other regulatory regimes.

However, not only the trademarks holders are not excited about the new IDNs.  There are also those, who voice concerns about fragmentation of the Internet as a result of adoption of domain names that would be accessible only to speakers of a particular language.  Dwayne Bailey, Research Director of the African Network for Localisation, spoke at the IGF about the danger of monolingual silos or as he put it: “A multilingual world of mono-lingualism.”  Karine Barzliai Nahon wrote a post on this topic, addressing particularly the situation in Israel, but alluding to similar concerns.  I personally had thoughts along the same lines when I first heard about the idea of IDNs and we can find similar arguments even at the very beginning of the debate about IDNs (for example here).

From where I stand as a user of the Internet (and I think most of the people who read those lines share this position), the Internet emerges as this enormous modern (knowledge and information) Tower of Babel.  There is so much information out there and it all is accessible to me at my laptop – all I need to do is to type a query in the search engine or enter a URL.  This is possible primarily because I feel at ease with both the technology and the English language.

Even though English is not the only language online, we can still access most of the content in English.  As some of the stats suggest, in 2008 only 31% of the online content was in English and that percentage was shrinking.  Chinese accounted for 20% and Spanish for 7%.  Between 2000 and 2008, the amount of content in Arabic grew 2064%, in Chinese 755%, and in Portuguese 668%.  However, even if the content itself is in a language that I do not understand, there are automatic translators that are good enough to allow me understanding, and maybe even engaging with, materials in languages other than those that I know.  All I need is to enter a URL of a website into an automatic translator, and here it is at my fingertips.  Isn’t it wonderful?

The “danger” of IDNs thus is fragmentation of content and as a result fragmentation of the Internet itself.  If I am unable to type in a URL of a website I won’t be able to access it, even before I reach the point where I need a translation.  The result could be that different cultural groups will isolate themselves by using the language barrier and we might lose the wealth of information that is out there.  This would be an equivalent of what happened to the Biblical Tower of Babel when all the different languages were introduced – the tower fell.  Our modern (knowledge and information) Tower of Babel might fall as well.

These were some of my initial thoughts and these are the concerns voiced by others as well.  However, the more I think about it the less categorical picture emerges.  Here are some of my more recent thoughts:

• To start with, it is not clear how much attention people pay to the URLs and there is quite a lot of research out there showing that people don’t use URLs for web navigation that much.  I think this is a major point in our thinking about the “threat” and “benefits” of IDNs.  I am not at all convinced that URLs matter.
• Second, I am not sure how much people in fact consume content that is not in languages that they know.  I mean, it may well be that the content online is already segregated and having internationalized URLs will not change much.  I have yet met a native English speaker who was a regular reader of websites in Russian or Chinese (I see a lot of the opposite, but not that).
• Third, I think it is reasonable to assume that just as we have automatic translators that allow browsing entire websites in languages other than those that we know, there will be a technological solution that will make the URLs just as transparent.
• Same goes for keyboards.  If we will insist on typing the URLs, virtual or projection keyboards can allow having an unlimited number of scripts on a single keyboard.  In fact, in this kind of technical solutions, I do believe in letting the markets speak and if there is enough demand for IDNs and enough demand for bypassing the IDNs, the technical solutions will appear.
• Also, as the rhetoric of IDNs suggests, they are aimed not at people who are already online and are comfortable with English, but at those who for various reasons, are not online yet and for whom English is a barrier.  It is easy for us to talk about potential loss of our access to the (dare I say underutilized) wealth of information from a position of relative power.  It is quite different for those who do not have any access at all.
• Finally, it may be natural that we do not understand all the content that is out there.  After all this is how our society became as diverse as it is.  Moreover the effort we need to put into learning and understanding another culture makes the experience even more rewarding.  So, maybe the IDNs are just a natural development?

My bottom line is that while I do share some concerns regarding the IDNs’ potential contribution to the fragmentation of the Internet, I am not at all convinced that this is what will necessarily happen.  Of course, one can think of scenarios where some governments force registration of local domains in a particular language, but even in that case, I am not sure it will work.  Similarly, I am not 100% sure that English is the main barrier to access and effective use of the Web.  I think there are other barriers such as lack of physical infrastructure or lack of technical literacy.  But perhaps more than ever before I think this is a case where we should let the users of the Internet make up their minds whether they want to use internationalized domain names or not.  The history suggests that the currently connected won’t do it, but perhaps the 6 billions of those who are not connected will.

These are some of my thoughts on the subject.  What do you think?  Will IDNs cause further fragmentation of the Internet?  Or will they increase the diversity of the content online and make the Web more accessible?