Enabling languages

@zenodotus But the new lib sector specs don't leverage the I18N improvements in W3C specs.

@zenodotus Not that I am aware of. The library specific standards fall outside of W3C. There were some library sector people involved in linked data specs but honestly, I18N was an uphill battle with those specs. But horizontal reviews make it easier these days.

Library cataloging policy is quietly starting to catch up to a fuller range of languages and scripts that can be represented in the metadata. But don't tell that to @enablelanguages.

@zenodotus Well, if it's broken, it's broken, part or whole. But you know my position is, it's the system. The real question is can it be fixed. And the following question is whether there is anyone in a position to take responsibility for fixing it.

@enablelanguages More as in fundamentally affecting entire systems, not just their parts.

@zenodotus Systemic as in it is categorised as a marginal consideration so less resources? Or systemic as in an artefact of colonialism and néocolonialisme?

@enablelanguages Is it more of a systemic issue?

@zenodotus And vendors are a problem more than a help

@zenodotus Part of the problem is resourcing. Individual institutions aren't necessarily in a position to have an I18N specialist, and there isn't anyone to provide guidance and assistance.

@zenodotus My gut reaction is that existing momentum is built on flawed approaches.

@enablelanguages What do you see as the highest priority shifts to be made?

@zenodotus Not familiar with the Dr

@enablelanguages Are you in touch with Dr. Getaneh Alemu?

@zenodotus I was thinking also in terms of new records as well. The meta framework. Native first, romanisation as an assist. Which is the reversal of a costing practice

@enablelanguages loc.gov/aba/cataloging…

@zenodotus Yes and no, that is an attempt to convert from romanised to native. But considering how much of a sieve tables can be, more robust mechanisms are needed. Data is lacking.

@zenodotus Doesn't address versioning which is a key element.

@enablelanguages loc.gov/marc/mac/2025/…

@zenodotus W3C I18N WG has done a lot of work in this area. What libraries are lacking is people with the knowledge, skills and experience to apply it to this domain

@enablelanguages Can you point to theory that would underlie this?

@zenodotus Unfortunately full bcp-47 is needed. Iso-639-3 by itself doesn't cut it.

@enablelanguages ideas.exlibrisgroup.com/forums/308173-…

@zenodotus There quite a few other things but that's a start. Once interesting area would be getting bidi and vertical text working correctly in UIs, but getting bibframe up to speed with yhos would be the first step.

@zenodotus Internationalisation is poorly understood in library sector and few have the knowledge or skills, and little interest in supporting it.

@zenodotus Bibliographic data cleaning tools that can handle the various I18N issues that n existing records

@zenodotus Sort out the bad encodings mess with Cyrillic romanisations

@zenodotus PyMARC equivalent for Bibframe