Progress and challenges

UA progress in 2019

Our testing in 2019 shows a significant bright spot. Contemporary computer-based browsers, from leading developers, all have the ability to accept, display, and use IDNs. On desktop and laptop computers, with a reasonably up-to-date operating system and a recent browser for the World Wide Web, we find that users can expect IDNs to work and display correctly. While there are still cases where the web server fails to label the content correctly, the browser is not the source of the problem for the end user.

This is a significant gain – especially when viewed in the long-term. Part of the underlying reason for this success is the availability of critical, shared software libraries that have been updated to work properly in the presence of IDNs.  For example, when software designers build browsers into bespoke applications, they get nearly automatic support for IDNs if they use these updated, shared libraries for handling domain names.

It appears, in 2019, that this success is having a trickle-down effect in browsers embedded in mobile, portable devices such as smartphones and tablets. Major operating system manufacturers appear to have learned the hard-won lessons from the desktop and applied those lessons to software libraries for smaller, more portable environments. Even compared to 2018, the quality of IDN support in mobile environments has improved. This is especially true for those developers and end-users who are using Apple’s iOS or the Android ecosystem.

However, as in previous years, progress for the use of IDNs in other settings has been slow. The trend can be simply stated: universal acceptance for IDNs as identifiers in the DNS is increasing, but acceptance for IDNs as identifiers in applications remains woeful.

Validation – the ability to accept that the IDN is a legitimate domain name – remains a critical problem in 2019. Many applications and services on the Internet attempt to validate or inspect domain names as they process those strings. This activity leads those applications to erroneously reject IDNs, especially when the IDNs are in unicode. In 2019 we continue to see widespread use of out-of-date criteria for top-level domains and the use of outdated or non-authoritative lists of top-level domains. This is the most challenging area of Universal Acceptance for IDNs.

Another area of concern in 2019 is linkification. Linkification is an action done by applications. When the applications see text that looks like a URL or other Internet identifier, it automatically creates a hyperlink using the text as a target. Linkification is of interest because a wider variety of applications are taking this action when they see links. However, this activity is not consistent in the presence of IDNs. If Universal Acceptance means that all domain names are treated consistently – so that users receive consistent behaviour from their applications – then we see progress here (especially in social media settings). That consistency does not, in 2019, extend to browsers in smartphones or mobile devices.

While traditional operating systems and mainstream browsers are success stories in 2019, IDN support for non-traditional environments remains elusive. As an example, embedded devices in smart vehicles still suffer from only supporting LDH-based domain names. We look forward to change in this space in the coming years. Since many manufacturers will be embedding systems such as IOS CarPlay or Android Auto, we may look forward to embedded systems that have better IDN support because of their parent operating systems. While not yet a success, we look forward to investigating this development in 2020.