Open StefanKarpinski opened 5 years ago
There's still a lot of TODOs down at the bottom of the tex file.
We need to get some people to work on those then 😀
Most of the TODO's are done. The main remaining items:
I think this is a great idea and as a new Julia user I'd very much like to see this happen. Having said that, I don't imagine it's an easy thing to do but wanted to let you know that I support this proposal.
The main sticking point right now is the "provision of fonts" question. Proposals are supposed to include contact info for someone willing to provide a font, but we've had no volunteers so far.
Sadly I don't really know what that entails so I'm not of much help I'm affraid.
@cormullion maker of Julia Mono would you be interested in helping with this?
I've already made a font with the required characters - nothing much more I can do... 😄
I've already made a font with the required characters - nothing much more I can do...
That's great! Can you make a PR with your contact information and the information for downloading the draft font?
Feel free to add https://github.com/cormullion/juliamono
and https://github.com/cormullion
at some suitable location in your TEX file. (I'm not much of a TEX-er myself.)
I think they probably want offline contact information too.
It says "the name and contact information", so at least your legal name and email, probably?
Also, do you have a summary of which sub/superscript glyphs were added to https://github.com/cormullion/juliamono?
It's possible that @cormullion may not want to publicize his contact information or identity, in which case perhaps we could use someone else as a proxy to provide offline contact information? I would be happy to do so, for example.
I added these to the Private Use Area:
for i in 0xf0000:8:0xf0000+300
print(string(i, base=16))
for j in i:i+8
print(" "^4, string("x", Char(j)))
end
println()
end
(give or take the odd ϕ that needs changing... 😄) Of course some of these are just copies of the Unicode glyphs as currently defined.
Stefan, you're right - I'm used to keeping OSS work separate. Thank you for your thoughtfulness! If you can't reach me online, I've probably died or have stopped functioning effectively... 😄
Great, I've updated this in the proposal with c06bfa711441c3ed3974779f583002553f5aa90b.
The only remaining to-do is to finalize the list of authors and add the contact information.
@stevengj how would you like to collect endorsements?
This is great work! Excited to see it go forward.
I think we should get the document in a final state and then, once it's in a state where people can know what they're putting their name to, we should start collecting signatures from prominent people in numerical computing.
Just bumping this yet again. A new Unicode version is entering beta, https://www.unicode.org/versions/Unicode14.0.0/ so maybe once the beta lands it would be a good time to give them the proposal.
Got bitten by the lack of proper unicode support for subsuper again :). What needs to be done to get this sent out? Can we finalize the document over the next couple of weeks and use JuliaCon as a vehicle to promote this effort?
Is there anything I — or anyone else who is willing to contribute — can do to push this forward?
Is there anything I — or anyone else who is willing to contribute — can do to push this forward?
(I do not know whether the following idea of mine will help push this forward, but for what it is worth …)
I have been toying with the idea of modifying some (monospace) font with good Unicode coverage (probably Julia Mono or Iosevka). My modification idea:
(1) Add superscripted respectively subscripted versions of as many glyphs as possible (in particular for the Mathematical Alphanumeric Symbols Unicode block).
(2) Use Unicode's variation selectors to render those glyphs.
(3) (Optionally) I believe OpenType has the capacity to provide "substitute" glyphs for sequences of "variation selected" Unicode code points. This could then be used to, for example, display x¹₁
with the ¹
and ₁
horizontally aligned.
Regarding (1): My impression is that this should be quite easy to do, by using something like FontForge's stylistics transformations capabilities. Also, not having (1) is basically the only thing preventing me from using pure text for all my scientific writings.¹
Regarding (2): I believe @cormullion had implemented this as a proof-of-concept?
¹ For scientific writings where JuliaMono suffices I have designed my own markup language. I have not yet made it public but there is a GitHub organization if you want to get updated:
https://github.com/no-markup-markup
(Anyone super-curious I will of course gladly invite to the organization).
(I think the above idea of mine would perhaps help push this forward because a more widespread use of non-standard character encodings could force the Unicode Consortium to make that use standard, similarly to how they were forced to incorporate emojis into Unicode.)
It doesn’t look to me like this proposal is going anywhere. 🤔
If you want any help with the font technology, let me know at github.com/cormullion/juliamono.
This document seems as ready as it's going to get at this point. Time to submit it and see what happens?