Hrant Հրանդ Փափազեան Papazian Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard i take it you haven't participated in any Unicode processes for a long time Hrant. Emoji and script work are effectively decoupled. They do their thing, we do our thing, and we meet up every three months to mush it all together

iwsfutcmd Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard Money arises from popular perception. People think Unicode is for emoji. The lion's share of character adoptions are for emoji. Emoji proposals need to demonstrate no precedent... unlike for an invented script that has to establish precedent via... hacking.

Hrant Հրանդ Փափազեան Papazian Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard I fail to see how popular perception and donations coming from emoji funneling money into minority and historic script research is a bad thing.

and like i said, separate silos. they have their process, we have ours. work on emoji doesn't detract from work on scripts...

Hrant Հրանդ Փափազեան Papazian Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard because the set of experts that work on each doesn't have a lot of overlap. it's not like i have to say to myself 'i'd sure like to work on this minority script but Unicode is forcing me to work on emoji', there are emoji experts for that

iwsfutcmd Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard The overall perception that emoji is more important than minority scripts does damage to the... long-term.

Once you admit the dark side of standardization you strike a better balance, and start counterbalancing commerce in favor of more meaningful culture.

iwsfutcmd Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard >>
So even more crucial than fixing something like input methods is correcting the core ideological failing. Or failing that, realize the crucial role of hacking.

Hrant Հրանդ Փափազեան Papazian Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard the key here is Unicode allows room for experimentation *without* violating the integrity of the standard—the PUA. if that's inadequate for your experimentation needs, we should find ways of working within the standard to expand the PUA's capabilities

iwsfutcmd Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard PUA is a... hack.

Hrant Հրանդ Փափազեան Papazian Igor Freiberger Tiro Typeworks Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard perhaps we're using different definitions of 'hack', but the one I'm using is violating a standard to get the results you want, principles be damned.

PUA isn't a hack because it's fully within the standard to use it for custom, not-yet-encoded functionality

iwsfutcmd Hrant Հրանդ Փափազեան Papazian Igor Freiberger Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard With the very large caveats that PUA is all default left-to-right, with no very reliable mechanism to use PUA for right-to-left systems, and PUA characters have no inherent properties for shaping, with no reliable mechanism to assign such properties at the font level.

Tiro Typeworks Hrant Հրանդ Փափազեան Papazian Igor Freiberger Vassil Kateliev @[email protected] Dutch Type Library Adam Twardoch ৵ Nadine Chahine Localfonts André Simard if so, please propose a solution! perhaps a new block of PUA that's default RTL? we already have a precedent for RTL-ing unencoded codepoints, so that wouldn't be a problem

i don't *think* there's any problem with adding more PUA but i'd have to double-check.