Nusantara Scripts Come Online: Indonesia Launches Balinese, Javanese and Pegon Domains Under .id
How a community-led effort brought centuries-old writing systems into the DNS and why Universal Acceptance will decide whether they are truly used.
For most of the world's internet users, typing a web address in their own script is unremarkable. For speakers of Indonesia's hundreds of languages and writing systems, it has until now been largely impossible. That changed at ICANN87 in Bali, where Indonesia's country-code registry announced that domain names can now be registered under .id in Balinese, Javanese and Pegon (Arabic-derived) scripts.
It is a milestone worth pausing on, both for what it achieves and for the question it raises next.
What was launched
During the ICANN87 Annual General Meeting in Bali (17–22 October 2026), the Indonesian registry PANDI (Pengelola Nama Domain Internet Indonesia) used its 19 October welcome reception to showcase the work of local script communities and to announce the registration of domain names in Balinese, Javanese and Pegon scripts under .id. PANDI marked the occasion with a short film, From Cultural Heritage to Digital Identity, framing the launch as the point at which these scripts stop being "merely remembered" and start being used on the open internet.
The Balinese script led the way. PANDI opened Balinese-script registrations under .id in December 2025, and as ICANN notes Balinese-script domains are now not just available but actively registered and used, including by local businesses establishing an online presence in their own writing system. Javanese and Pegon have now followed.
The technical groundwork: Label Generation Rules
The headline is cultural, but the enabling work is deeply technical and it is the part most relevant to how the IDN ecosystem actually functions.
Before a script can be used safely in domain names, the community must define Label Generation Rules (LGRs): the formal rules that determine which code points are allowed, how variant characters relate to one another, and which label forms are valid and secure. Without them, a script is vulnerable to confusable-character attacks and inconsistent handling.
ICANN's account of the Balinese work illustrates the care involved. Local script experts, academic institutions and PANDI trained by ICANN on second-level LGR design analysed the script against both contemporary usage and historical texts preserved on lontar (palm-leaf manuscripts). From the 127 code points in the Unicode Balinese chart, the community short-listed 65 code points, defined 10 variant sets, and set four rules to permit only well-formed, secure labels. One notable decision: allowing only a single instance of below-base character stacking, because the browser address bar cannot reliably present more. The Second-Level Reference LGR for Balinese was published in October 2024 following public comment.
Building on that, PANDI formed panels for the next two scripts. For Javanese, the LGR excludes Javanese digits (several closely resemble letters) and defines variant relationships between the Aksara Murda and Aksara Wyanjana consonant sets, which represent the same sounds in formal versus informal registers. For Pegon, the existing Arabic-script Reference LGR was updated to add two characters and to manage variant relationships so as to avoid confusion for Arabic-script users more broadly. ICANN published the Javanese Second-Level Reference LGR and the updated Arabic LGR supporting Pegon in October 2026, after a public-comment proceeding.
And the work is not finished: technical development is now under way for three further scripts Batak, Buginese and Sundanese.
Why this matters beyond Indonesia
This is a textbook example of the model the multilingual-internet community has been advocating for years: script communities, academics, a national registry and ICANN working together, grounded in Unicode standardisation and formal LGRs, to bring a writing system into the DNS in a way that is both usable and secure. For a country as linguistically dense as Indonesia, it is also a meaningful act of digital preservation giving living scripts a role in everyday digital identity rather than confining them to the archive.
For those of us who track IDN deployment, it is a concrete addition to the delegated-and-usable column at the second level, and a replicable template other registries can study.
The harder half: Universal Acceptance
Here is where the Indonesian community itself has been admirably clear-eyed. Launching the domains is not the finish line, it is actually the starting line.
As Heru Nugroho, chair of PANDI's IDN Working Group, put it, the domains may be available, but Indonesia's digital ecosystem "still faces technical challenges before the scripts can function across applications, websites and email systems." The decisive factor is Universal Acceptance (UA) the principle that all valid domain names and email addresses should work across every digital system, regardless of script. A Balinese or Pegon domain that resolves in the DNS but is rejected by a web form, a sign-up field or an email server is, from the user's point of view, not working at all.
Nugroho tied this directly to the UNESCO–ICANN Universal Acceptance policy brief of 2026, which warns that despite mature IDN and Email Address Internationalisation (EAI) standards, many systems still fail to recognise and process multilingual domains and argues that adopting UA can strengthen local digital economies, widen access to online public services, and help preserve languages and scripts from marginalisation. PANDI's prescription follows the now-familiar UA playbook: governments requiring UA compliance in public-sector IT procurement, developers treating UA as a baseline software standard, and academics strengthening digital-literacy education.
His framing captures the whole challenge in a sentence: success should be measured "not only by whether domain names can display traditional scripts, but also by whether digital applications and services can use them." That is precisely the availability-versus-usability gap at the heart of the UA conversation and it applies to Balinese and Pegon exactly as it applies to every other internationalised script.
The takeaway
Indonesia has done the hard, patient, community-led work of bringing three of its scripts into the domain name system the right way; standardised, secure, and rooted in the communities that use them. That deserves recognition. The next test is whether the wider software ecosystem meets those scripts halfway. A multilingual internet is not achieved when a name can be registered; it is achieved when it can be used - everywhere, by everyone, in their own script.
Credits and sources
This post is based on and draws from the following sources:
PANDI, From Cultural Heritage to Digital Identity: PANDI Brings Nusantara Scripts to the Global Internet (video), published 4 November 2026. https://www.youtube.com/watch?v=e3Sza9egY3U
Pitinan Kooarmornpatana (IDN Programs Director, ICANN), "Celebrating Indonesia's Progress Toward a Multilingual Internet at ICANN87," ICANN Blog, 18 November 2026. https://www.icann.org/en/blogs/details/celebrating-indonesias-progress-toward-a-multilingual-internet-at-icann87-18-11-2026-en
"Indonesia launches domain names in Nusantara scripts," ANTARA News, 28 October 2026 (quoting Heru Nugroho, chair of PANDI's IDN Working Group). https://en.antaranews.com/news/438387/indonesia-launches-domain-names-in-nusantara-scripts
ICANN, "ICANN Publishes Two Additional Second-Level Reference Label Generation Rules" (Javanese; Arabic update for Pegon), 5 October 2026. https://www.icann.org/en/announcements/details/icann-publishes-two-additional-second-level-reference-label-generation-rules-05-10-2026-en
Balinese Script Second-Level Reference LGR and Unicode Balinese code chart, as referenced by ICANN (published October 2024).
UNESCO–ICANN Universal Acceptance policy brief, 2026.
Thanks to PANDI and the Indonesian script communities, and to ICANN's IDN Program, whose published accounts this summary draws on.
